|
1
|
Park S, Chung S, Kim Y, Yang SA, Kwon S, Cho JM, Lee MJ, Cho E, Ryu J, Kim S, Lee J, Yoon HJ, Choi E, Kim K, Lee H. A deep-learning algorithm using real-time collected intraoperative vital sign signals for predicting acute kidney injury after major non-cardiac surgeries: A modelling study. PLoS Med 2025; 22:e1004566. [PMID: 40299885 PMCID: PMC12040160 DOI: 10.1371/journal.pmed.1004566] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 08/18/2024] [Accepted: 02/24/2025] [Indexed: 05/01/2025] Open
Abstract
BACKGROUND Postoperative acute kidney injury (PO-AKI) prediction models for non-cardiac major surgeries typically rely solely on preoperative clinical characteristics. METHODS AND FINDINGS In this study, we developed and externally validated a deep-learning-based model that integrates preoperative data with minute-scale intraoperative vital signs to predict PO-AKI. Using data from three hospitals, we constructed a convolutional neural network-based EfficientNet framework to analyze intraoperative data and created an ensemble model incorporating 103 baseline variables of demographics, medication use, comorbidities, and surgery-related characteristics. Model performance was compared with the conventional SPARK model from a previous study. Among 110,696 patients, 51,345 were included in the development cohort, and 59,351 in the external validation cohorts. The median age of the cohorts was 60, 61, and 66 years, respectively, with males comprising 54.9%, 50.8%, and 42.7% of each cohort. The intraoperative vital sign-based model demonstrated comparable predictive power (AUROC (Area Under the Receiver Operating Characteristic Curve): discovery cohort 0.707, validation cohort 0.637 and 0.607) to preoperative-only models (AUROC: discovery cohort 0.724, validation cohort 0.697 and 0.745). Adding 11 key clinical variables (e.g., age, sex, estimated glomerular filtration rate (eGFR), albuminuria, hyponatremia, hypoalbuminemia, anemia, diabetes, renin-angiotensin-aldosterone inhibitors, emergency surgery, and the estimated surgery time) improved the model's performance (AUROC: discovery cohort 0.765, validation cohort 0.716 and 0.761). The ensembled deep-learning model integrating both preoperative and intraoperative data achieved the highest predictive accuracy (AUROC: discovery cohort 0.795, validation cohort 0.762 and 0.786), outperforming the conventional SPARK model. The retrospective design in a single-nation cohort with non-inclusion of some potential AKI-associated variables is the main limitation of this study. CONCLUSIONS This deep-learning-based PO-AKI risk prediction model provides a comprehensive approach to evaluating PO-AKI risk prediction by combining preoperative clinical data with real-time intraoperative vital sign information, offering enhanced predictive performance for better clinical decision-making.
Collapse
Affiliation(s)
- Sehoon Park
- Department of Internal Medicine, Seoul National University Hospital, Seoul, Korea
| | - Soomin Chung
- Interdisciplinary Program in Bioengineering, Seoul National University, Seoul, Korea
| | - Yisak Kim
- Interdisciplinary Program in Bioengineering, Seoul National University, Seoul, Korea
- Department of Radiology, Seoul National University Hospital, Seoul, Korea
| | - Sun-Ah Yang
- Department of Transdisciplinary Medicine, Seoul National University Hospital, Seoul, Korea
| | - Soie Kwon
- Department of Internal Medicine, Chung Ang University Hospital, Seoul, Korea
| | - Jeong Min Cho
- Department of Internal Medicine, Seoul National University Hospital, Seoul, Korea
- Department of Internal Medicine, Chung-Ang University Gwangmyeong Hospital, Gwangmyeong, Korea
| | - Min Jae Lee
- KAIST: Korea Advanced Institute of Science and Technology, Daejeon, Korea
| | - Eunbyeol Cho
- KAIST: Korea Advanced Institute of Science and Technology, Daejeon, Korea
| | - Jiwon Ryu
- Department of Internal Medicine, Seoul National University Bundang Hospital, Gyeonggi-do, Korea
| | - Sejoong Kim
- Department of Internal Medicine, Seoul National University Bundang Hospital, Gyeonggi-do, Korea
- Department of Internal Medicine, Seoul National University College of Medicine, Seoul, Korea
| | - Jeonghwan Lee
- Department of Internal Medicine, Seoul National University College of Medicine, Seoul, Korea
- Department of Internal Medicine, Seoul National University Boramae Medical Center, Seoul, Korea
| | - Hyung Jin Yoon
- Department of Biomedical Engineering, Seoul National University College of Medicine, Seoul, Korea
| | - Edward Choi
- KAIST: Korea Advanced Institute of Science and Technology, Daejeon, Korea
| | - Kwangsoo Kim
- Department of Transdisciplinary Medicine, Seoul National University Hospital, Seoul, Korea
- Department of Medicine, Seoul National University, Seoul, Korea
| | - Hajeong Lee
- Department of Internal Medicine, Seoul National University Hospital, Seoul, Korea
- Department of Internal Medicine, Seoul National University College of Medicine, Seoul, Korea
| |
Collapse
|
|
2
|
Balch JA, Ruppert MM, Guan Z, Buchanan TR, Abbott KL, Shickel B, Bihorac A, Liang M, Upchurch GR, Tignanelli CJ, Loftus TJ. Risk-Specific Training Cohorts to Address Class Imbalance in Surgical Risk Prediction. JAMA Surg 2024; 159:1424-1431. [PMID: 39382865 PMCID: PMC11465118 DOI: 10.1001/jamasurg.2024.4299] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/01/2024] [Accepted: 08/05/2024] [Indexed: 10/10/2024]
Abstract
Importance Machine learning tools are increasingly deployed for risk prediction and clinical decision support in surgery. Class imbalance adversely impacts predictive performance, especially for low-incidence complications. Objective To evaluate risk-prediction model performance when trained on risk-specific cohorts. Design, Setting, and Participants This cross-sectional study performed from February 2024 to July 2024 deployed a deep learning model, which generated risk scores for common postoperative complications. A total of 109 445 inpatient operations performed at 2 University of Florida Health hospitals from June 1, 2014, to May 5, 2021 were examined. Exposures The model was trained de novo on separate cohorts for high-risk, medium-risk, and low-risk Common Procedure Terminology codes defined empirically by incidence of 5 postoperative complications: (1) in-hospital mortality; (2) prolonged intensive care unit (ICU) stay (≥48 hours); (3) prolonged mechanical ventilation (≥48 hours); (4) sepsis; and (5) acute kidney injury (AKI). Low-risk and high-risk cutoffs for complications were defined by the lower-third and upper-third prevalence in the dataset, except for mortality, cutoffs for which were set at 1% or less and greater than 3%, respectively. Main Outcomes and Measures Model performance metrics were assessed for each risk-specific cohort alongside the baseline model. Metrics included area under the receiver operating characteristic curve (AUROC), area under the precision-recall curve (AUPRC), F1 scores, and accuracy for each model. Results A total of 109 445 inpatient operations were examined among patients treated at 2 University of Florida Health hospitals in Gainesville (77 921 procedures [71.2%]) and Jacksonville (31 524 procedures [28.8%]). Median (IQR) patient age was 58 (43-68) years, and median (IQR) Charlson Comorbidity Index score was 2 (0-4). Among 109 445 operations, 55 646 patients were male (50.8%), and 66 495 patients (60.8%) underwent a nonemergent, inpatient operation. Training on the high-risk cohort had variable impact on AUROC, but significantly improved AUPRC (as assessed by nonoverlapping 95% confidence intervals) for predicting mortality (0.53; 95% CI, 0.43-0.64), AKI (0.61; 95% CI, 0.58-0.65), and prolonged ICU stay (0.91; 95% CI, 0.89-0.92). It also significantly improved F1 score for mortality (0.42; 95% CI, 0.36-0.49), prolonged mechanical ventilation (0.55; 95% CI, 0.52-0.58), sepsis (0.46; 95% CI, 0.43-0.49), and AKI (0.57; 95% CI, 0.54-0.59). After controlling for baseline model performance on high-risk cohorts, AUPRC increased significantly for in-hospital mortality only (0.53; 95% CI, 0.42-0.65 vs 0.29; 95% CI, 0.21-0.40). Conclusion and Relevance In this cross-sectional study, by training separate models using a priori knowledge for procedure-specific risk classes, improved performance in standard evaluation metrics was observed, especially for low-prevalence complications like in-hospital mortality. Used cautiously, this approach may represent an optimal training strategy for surgical risk-prediction models.
Collapse
Affiliation(s)
- Jeremy A. Balch
- Department of Surgery, University of Florida, Gainesville
- Department of Health Outcomes and Biomedical Informatics, University of Florida, Gainesville
- Intelligent Clinical Care Center, College of Medicine, University of Florida, Gainesville
| | | | - Ziyuan Guan
- Intelligent Clinical Care Center, College of Medicine, University of Florida, Gainesville
| | | | | | - Benjamin Shickel
- Intelligent Clinical Care Center, College of Medicine, University of Florida, Gainesville
| | - Azra Bihorac
- Intelligent Clinical Care Center, College of Medicine, University of Florida, Gainesville
| | - Muxuan Liang
- College of Medicine, University of Florida, Gainesville
| | | | | | - Tyler J. Loftus
- Department of Surgery, University of Florida, Gainesville
- Intelligent Clinical Care Center, College of Medicine, University of Florida, Gainesville
| |
Collapse
|
|
3
|
De Santo LS, Rubino AS, Montella AP, Golini Petrarcone C, Palmieri L, Galbiati D, Pisano A, De Feo M. Incidence and risk factors of acute kidney injury in redo cardiac surgery: a single center analysis. Sci Rep 2024; 14:27267. [PMID: 39516304 PMCID: PMC11549308 DOI: 10.1038/s41598-024-78990-3] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Download PDF] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/01/2024] [Accepted: 11/05/2024] [Indexed: 11/16/2024] Open
Abstract
The incidence, risk factors and prognostic implications of acute kidney injury (AKI) in patients undergoing redo cardiac surgery are still poorly defined. We prospectively collected data on 394 consecutive redo patients between January 2011 and October 2020. Patients were divided into groups according to the occurrence of different degrees of postoperative AKI (No AKI vs. Any AKI; No AKI-AKI 1 vs. AKI 2-3). The relationship between AKI and other major complications was also investigated. Postoperatively, AKI 1 occurred in 124 (31.5%), AKI 2 in 36 (9.1%) and AKI 3 in 64 (16.2%). Higher KDIGO classes were associated with increased in-hospital mortality: 5.3% among patients with no postoperative AKI and 8.9%, 13.9% and 64.1% in patients with AKI 1, 2 and 3, respectively (p < 0.001). Age, baseline hemoglobin, comorbidity, EuroSCORE II, operative time and transfusion during CPB proved to be significantly associated to the occurrence of AKI. Our study confirms the burden and prognostic role of AKI in a large, all comers, single center database of redo cardiac procedures.
Collapse
Affiliation(s)
- Luca Salvatore De Santo
- Department of Translational Medical Sciences, Monaldi Hospital, University of Campania "Luigi Vanvitelli", Via Leonardo Bianchi, 80131, Naples, Italy
| | | | - Antonio Pio Montella
- Department of Translational Medical Sciences, Monaldi Hospital, University of Campania "Luigi Vanvitelli", Via Leonardo Bianchi, 80131, Naples, Italy
| | - Caterina Golini Petrarcone
- Department of Translational Medical Sciences, Monaldi Hospital, University of Campania "Luigi Vanvitelli", Via Leonardo Bianchi, 80131, Naples, Italy
| | - Lucrezia Palmieri
- Department of Translational Medical Sciences, Monaldi Hospital, University of Campania "Luigi Vanvitelli", Via Leonardo Bianchi, 80131, Naples, Italy
| | - Denise Galbiati
- Cardiovascular Department, Cardiac Surgery Unit of the IRCCS Humanitas Research Hospital, Rozzano, Italy
| | - Antonio Pisano
- Cardiac Anesthesia and Intensive Care Unit, Monaldi Hospital, Via Leonardo Bianchi, 80131, Naples, Italy
| | - Marisa De Feo
- Department of Translational Medical Sciences, Monaldi Hospital, University of Campania "Luigi Vanvitelli", Via Leonardo Bianchi, 80131, Naples, Italy
| |
Collapse
|
|
4
|
Quickfall D, La AM, Koyner JL. 10 tips on how to use dynamic risk assessment and alerts for AKI. Clin Kidney J 2024; 17:sfae325. [PMID: 39588357 PMCID: PMC11586629 DOI: 10.1093/ckj/sfae325] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/24/2024] [Indexed: 11/27/2024] Open
Abstract
Acute kidney injury (AKI) is a common syndrome in hospitalized patients and is associated with increased morbidity and mortality. The focus of AKI care requires a shift away from strictly supportive management of established injury to the early identification and timely prevention of worsening renal injury. Identifying patients at risk for developing or progression of severe AKI is crucial for improving patient outcomes, reducing the length of hospitalization and minimizing resource utilization. Implementation of dynamic risk scores and incorporation of novel biomarkers show promise for early detection and minimizing progression of AKI. Like any risk assessment tools, these require further external validation in a variety of clinical settings prior to widespread implementation. Additionally, alerts that may minimize exposure to a variety of nephrotoxic medications or prompt early nephrology consultation are shown to reduce the incidence and progression of AKI severity and enhance renal recovery. While dynamic risk scores and alerts are valuable, implementation requires thoughtfulness and should be used in conjunction with the overall clinical picture in certain situations, particularly when considering the initiation of fluid and diuretic administration or renal replacement therapy. Despite the contemporary challenges encountered with alert fatigue, implementing an alert-based bundle to improve AKI care is associated with improved outcomes, even when implementation is incomplete. Lastly, all alert-based interventions should be validated at an institutional level and assessed for their ability to improve institutionally relevant and clinically meaningful outcomes, reduce resource utilization and provide cost-effective interventions.
Collapse
Affiliation(s)
- Danica Quickfall
- Committee on Clinical Pharmacology and Pharmacogenomics, Biological Science Division, University of Chicago, Chicago, IL, USA
| | - Ashley M La
- Section of Nephrology, Department of Medicine, University of Chicago, Chicago, IL, USA
| | - Jay L Koyner
- Committee on Clinical Pharmacology and Pharmacogenomics, Biological Science Division, University of Chicago, Chicago, IL, USA
- Section of Nephrology, Department of Medicine, University of Chicago, Chicago, IL, USA
| |
Collapse
|
|
5
|
Cui C, Qiu L, Li L, Chen FL, Liu X, Sun H, Liu XC, Bao L, Li LQ. A time series algorithm to predict surgery in neonatal necrotizing enterocolitis. BMC Med Inform Decis Mak 2024; 24:304. [PMID: 39425161 PMCID: PMC11487704 DOI: 10.1186/s12911-024-02695-w] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/10/2024] [Accepted: 09/25/2024] [Indexed: 10/21/2024] Open
Abstract
BACKGROUND Determining the optimal timing of surgical intervention for Neonatal necrotizing enterocolitis (NEC) poses significant challenges. This study develops a predictive model using the long short-term memory network (LSTM) with a focal loss (FL) to identify infants at risk of developing Bell IIB + NEC early and issue timely surgical warnings. METHODS Data from 791 neonates diagnosed with NEC are gathered from the Neonatal Intensive Care Unit (NICU), encompassing 35 selected features. Infants are categorized into those requiring surgical intervention (n = 257) and those managed medically (n = 534) based on the Mod-Bell criteria. A fivefold cross-validation approach is employed for training and testing. The LSTM algorithm is utilized to capture and utilize temporal relationships in the dataset, with FL employed as a loss function to address class imbalance. Model performance metrics include precision, recall, F1 score, and average precision (AP). RESULTS The model tested on a real dataset demonstrated high performance. Predicting surgical risk 1 day in advance achieved precision (0.913 ± 0.034), recall (0.841 ± 0.053), F1 score (0.874 ± 0.029), and AP (0.917 ± 0.025). The 2-days-in-advance predictions yielded (0.905 ± 0.036), recall (0.815 ± 0.057), F1 score (0.857 ± 0.035), and AP (0.905 ± 0.029). CONCLUSION The LSTM model with FL exhibits high precision and recall in forecasting the need for surgical intervention 1 or 2 days ahead. This predictive capability holds promise for enhancing infants' outcomes by facilitating timely clinical decisions.
Collapse
Affiliation(s)
- Cheng Cui
- Neonatal Diagnosis and Treatment Center of Children's Hospital of Chongqing Medical University, National Clinical Research Center for Child Health and Disorders, Ministry of Education Key Laboratory of Child Development and Disorders, Chongqing Key Laboratory of Child Rare Diseases in Infection and Immunity, Chongqing, 400014, China
| | - Ling Qiu
- The First People's Hospital Of Longquanyi District, Chengdu, 610100, China
| | - Ling Li
- Guang'an District Maternal and Child Health Care and Family Planning Service Center, Chengdu, 638000, China
| | - Fei-Long Chen
- College of Computer Science and Technology, Chongqing University of Posts and Telecommunications, Chongqing, 400065, China
| | - Xiao Liu
- College of Safety Engineering, China University of Mining and Technology, Beijing, 221116, China
| | - Huan Sun
- Neonatal Diagnosis and Treatment Center of Children's Hospital of Chongqing Medical University, National Clinical Research Center for Child Health and Disorders, Ministry of Education Key Laboratory of Child Development and Disorders, Chongqing Key Laboratory of Child Rare Diseases in Infection and Immunity, Chongqing, 400014, China
| | - Xiao-Chen Liu
- Neonatal Diagnosis and Treatment Center of Children's Hospital of Chongqing Medical University, National Clinical Research Center for Child Health and Disorders, Ministry of Education Key Laboratory of Child Development and Disorders, Chongqing Key Laboratory of Child Rare Diseases in Infection and Immunity, Chongqing, 400014, China
| | - Lei Bao
- Neonatal Diagnosis and Treatment Center of Children's Hospital of Chongqing Medical University, National Clinical Research Center for Child Health and Disorders, Ministry of Education Key Laboratory of Child Development and Disorders, Chongqing Key Laboratory of Child Rare Diseases in Infection and Immunity, Chongqing, 400014, China.
| | - Lu-Quan Li
- Neonatal Diagnosis and Treatment Center of Children's Hospital of Chongqing Medical University, National Clinical Research Center for Child Health and Disorders, Ministry of Education Key Laboratory of Child Development and Disorders, Chongqing Key Laboratory of Child Rare Diseases in Infection and Immunity, Chongqing, 400014, China.
| |
Collapse
|
|
6
|
Li Q, Shen J, Lv H, Chen Y, Zhou C, Shi J. Features selection in a predictive model for cardiac surgery-associated acute kidney injury. Perfusion 2024:2676591241289364. [PMID: 39382228 DOI: 10.1177/02676591241289364] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/10/2024]
Abstract
BACKGROUND Cardiac surgery-associated acute kidney injury (CSA-AKI) is related to increased morbidity and mortality. However, limited studies have explored the influence of different feature selection (FS) methods on the predictive performance of CSA-AKI. Therefore, we aimed to compare the impact of different FS methods for CSA-AKI. METHODS CSA-AKI is defined according to the kidney disease: Improving Global Outcomes (KDIGO) criteria. Both traditional logistic regression and machine learning methods were used to select the potential risk factors for CSA-AKI. The area under the receiver operating characteristic curve (AUC) was used to evaluate the performance of the models. In addition, the importance matrix plot by random forest was used to rank the features' importance. RESULTS A total of 1977 patients undergoing cardiac surgery at Fuwai hospital from December 2018 to April 2021 were enrolled. The incidence of CSA-AKI during the first postoperative week was 27.8%. We concluded that different enrolled numbers of features impact the final selected feature number. The more you input, the more likely its output with all FS methods. In terms of performance, all selected features by various FS methods demonstrated excellent AUCs. Meanwhile, the embedded method demonstrated the highest accuracy compared with the LR method, while the filter method showed the lowest accuracy. Furthermore, NT-proBNP was found to be strongly associated with AKI. Our results confirmed some features that previous studies have reported and found some novel clinical parameters. CONCLUSIONS In our study, FS was as suitable as LR for predicting CSA-AKI. For FS, the embedded method demonstrated better efficacy than the other methods. Furthermore, NT-proBNP was confirmed to be strongly associated with AKI.
Collapse
Affiliation(s)
- Qian Li
- Department of Anesthesiology, State Key Laboratory of Cardiovascular Disease, Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China
| | - Jingjia Shen
- Department of Anesthesiology, State Key Laboratory of Cardiovascular Disease, Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China
| | - Hong Lv
- Department of Anesthesiology, State Key Laboratory of Cardiovascular Disease, Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China
| | - Yuye Chen
- Department of Anesthesiology, State Key Laboratory of Cardiovascular Disease, Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China
| | - Chenghui Zhou
- Department of Anesthesiology, State Key Laboratory of Cardiovascular Disease, Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China
| | - Jia Shi
- Department of Anesthesiology, State Key Laboratory of Cardiovascular Disease, Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China
| |
Collapse
|
|
7
|
Potosnak W, Dufendach KA, Nagpal C, Kaczorowski DJ, Yoon P, Bonatti J, Miller JK, Dubrawski AW. Intraoperative Features Improve Model Risk Predictions After Coronary Artery Bypass Grafting. ANNALS OF THORACIC SURGERY SHORT REPORTS 2024; 2:336-340. [PMID: 39790416 PMCID: PMC11708469 DOI: 10.1016/j.atssr.2024.02.005] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Accepted: 02/26/2024] [Indexed: 01/12/2025]
Abstract
Background Intraoperative physiologic parameters could offer predictive utility in evaluating risk of adverse postoperative events yet are not included in current standard risk models. This study examined whether the inclusion of continuous intraoperative data improved machine learning model predictions for multiple outcomes after coronary artery bypass grafting, including 30-day mortality, renal failure, reoperation, prolonged ventilation, and combined morbidity and mortality (MM). Methods The Society of Thoracic Surgeons (STS) database features and risk scores were combined with retrospectively gathered continuous intraoperative data from patients. Risk models were developed for each outcome by training a logistic regression classifier on intraoperative data using 5-fold cross-validation. STS risk scores were included as offset terms in the models. Results Compared with the STS Risk Calculator, models developed using a combination of the intraoperative features and the STS preoperative risk score had improved mean area under the receiver operating characteristic curve for prolonged ventilation (0.750 [95% CI, 0.690-0.809] vs 0.800 [95% CI, 0.750-0.851]) and MM (0.695 [95% CI, 0.644-0.746] vs 0.724 [95% CI, 0.673-0.775]). Additionally, models developed using intraoperative features had improved calibration, measured with Brier score, for prolonged ventilation (0.060 [95% CI, 0.050-0.070] vs 0.055 [95% CI, 0.045-0.065]) and MM (0.092 [95% CI, 0.081-0.103] vs 0.087 [95% CI, 0.075-0.098]). Conclusions The inclusion of time series intraoperative data in risk models may improve early postoperative care by identifying patients who require closer monitoring postoperatively.
Collapse
Affiliation(s)
- Willa Potosnak
- Auton Lab, School of Computer Science, Carnegie Mellon University, Pittsburgh, Pennsylvania
| | - Keith A. Dufendach
- Auton Lab, School of Computer Science, Carnegie Mellon University, Pittsburgh, Pennsylvania
- Division of Cardiac Surgery, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania
| | - Chirag Nagpal
- Auton Lab, School of Computer Science, Carnegie Mellon University, Pittsburgh, Pennsylvania
| | - David J. Kaczorowski
- Division of Cardiac Surgery, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania
| | - Pyongsoo Yoon
- Division of Cardiac Surgery, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania
| | - Johannes Bonatti
- Division of Cardiac Surgery, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania
| | - James K. Miller
- Auton Lab, School of Computer Science, Carnegie Mellon University, Pittsburgh, Pennsylvania
| | - Artur W. Dubrawski
- Auton Lab, School of Computer Science, Carnegie Mellon University, Pittsburgh, Pennsylvania
| |
Collapse
|
|
8
|
Oh M, Jung YM, Kim W, Lee H, Kim TK, Ko S, Lim J, Lee SM. Prediction for Perioperative Stroke Using Intraoperative Parameters. J Am Heart Assoc 2024; 13:e032216. [PMID: 39119968 PMCID: PMC11963952 DOI: 10.1161/jaha.123.032216] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 08/19/2023] [Accepted: 06/20/2024] [Indexed: 08/10/2024]
Abstract
BACKGROUND Perioperative stroke is a severe complication following surgery. To identify patients at risk for perioperative stroke, several prediction models based on the preoperative factors were suggested. Prediction models often focus on preoperative patient characteristics to assess stroke risk. However, most existing models primarily base their predictions on the patient's baseline characteristics before surgery. We aimed to develop a machine-learning model incorporating both pre- and intraoperative variables to predict perioperative stroke. METHODS AND RESULTS This study included patients who underwent noncardiac surgery at 2 hospitals with the data of 15 752 patients from Seoul National University Hospital used for development and temporal internal validation, and the data of 449 patients from Boramae Medical Center used for external validation. Perioperative stroke was defined as a newly developed ischemic lesion on diffusion-weighted imaging within 30 days of surgery. We developed a prediction model composed of pre- and intraoperative factors (integrated model) and compared it with a model consisting of preoperative features alone (preoperative model). Perioperative stroke developed in 109 (0.69%) patients in the Seoul National University Hospital group and 11 patients (2.45%) in the Boramae Medical Center group. The integrated model demonstrated superior predictive performance with area under the curve values of 0.824 (95% CI, 0.762-0.880) versus 0.584 (95% CI, 0.499-0.667; P<0.001) in the internal validation; and 0.716 (95% CI, 0.560-0.859) versus 0.505 (95% CI, 0.343-0.654; P=0.018) in the external validation, compared to the preoperative model. CONCLUSIONS We suggest that incorporating intraoperative factors into perioperative stroke prediction models can improve their accuracy.
Collapse
Affiliation(s)
- Mi‐Young Oh
- Department of NeurologyBucheon Sejong HospitalBucheon‐siGyeonggi‐doSouth Korea
| | - Young Mi Jung
- Department of Obstetrics and GynecologySeoul National University College of MedicineSeoulSouth Korea
- Department of Obstetrics and Gynecology, Guro HospitalKorea University College of MedicineSeoulSouth Korea
| | | | - Hyung‐Chul Lee
- Department of Anesthesiology and Pain MedicineSeoul National University College of MedicineSeoulSouth Korea
- Department of Anesthesiology and Pain MedicineSeoul National University HospitalSeoulSouth Korea
| | - Tae Kyong Kim
- Department of Anesthesiology and Pain MedicineSeoul National University College of MedicineSeoulSouth Korea
- Department of Anesthesiology and Pain MedicineMetropolitan Government Seoul National University Boramae Medical CenterSeoulSouth Korea
| | - Sang‐Bae Ko
- Department of NeurologySeoul National University HospitalSeoulSouth Korea
| | | | - Seung Mi Lee
- Department of Obstetrics and GynecologySeoul National University College of MedicineSeoulSouth Korea
- Department of Obstetrics and GynecologySeoul National University HospitalSeoulSouth Korea
- Innovative Medical Technology Research InstituteSeoul National University HospitalSeoulSouth Korea
- Institute of Reproductive Medicine and Population & Medical Big Data Research Center, Medical Research CenterSeoul National UniversitySeoulSouth Korea
| |
Collapse
|
|
9
|
Lima DL, Kasakewitch J, Nguyen DQ, Nogueira R, Cavazzola LT, Heniford BT, Malcher F. Machine learning, deep learning and hernia surgery. Are we pushing the limits of abdominal core health? A qualitative systematic review. Hernia 2024; 28:1405-1412. [PMID: 38761300 DOI: 10.1007/s10029-024-03069-x] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/27/2024] [Accepted: 04/29/2024] [Indexed: 05/20/2024]
Abstract
INTRODUCTION This systematic review aims to evaluate the use of machine learning and artificial intelligence in hernia surgery. METHODS The PRISMA guidelines were followed throughout this systematic review. The ROBINS-I and Rob 2 tools were used to perform qualitative assessment of all studies included in this review. Recommendations were then summarized for the following pre-defined key items: protocol, research question, search strategy, study eligibility, data extraction, study design, risk of bias, publication bias, and statistical analysis. RESULTS A total of 13 articles were ultimately included for this review, describing the use of machine learning and deep learning for hernia surgery. All studies were published from 2020 to 2023. Articles varied regarding the population studied, type of machine learning or Deep Learning Model (DLM) used, and hernia type. Of the thirteen included studies, all included either inguinal, ventral, or incisional hernias. Four studies evaluated recognition of surgical steps during inguinal hernia repair videos. Two studies predicted outcomes using image-based DMLs. Seven studies developed and validated deep learning algorithms to predict outcomes and identify factors associated with postoperative complications. CONCLUSION The use of ML for abdominal wall reconstruction has been shown to be a promising tool for predicting outcomes and identifying factors that could lead to postoperative complications.
Collapse
Affiliation(s)
- D L Lima
- Department of Surgery, Montefiore Medical Center, New York, NY, USA.
| | - J Kasakewitch
- Department of Surgery, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, USA
| | - D Q Nguyen
- Albert Einstein, College of Medicine, New York, USA
| | - R Nogueira
- Department of Surgery, Montefiore Medical Center, New York, NY, USA
| | - L T Cavazzola
- Federal University of Rio Grande Do Sul, Porto Alegre, Brazil
| | - B T Heniford
- Division of Gastrointestinal and Minimally Invasive Surgery, Department of Surgery, Carolinas Medical Center, Charlotte, NC, USA
| | - F Malcher
- Division of General Surgery, NYU Langone, New York, USA
| |
Collapse
|
|
10
|
Namavarian A, Gabinet-Equihua A, Deng Y, Khalid S, Ziai H, Deutsch K, Huang J, Gilbert RW, Goldstein DP, Yao CMKL, Irish JC, Enepekides DJ, Higgins KM, Rudzicz F, Eskander A, Xu W, de Almeida JR. Length of Stay Prediction Models for Oral Cancer Surgery: Machine Learning, Statistical and ACS-NSQIP. Laryngoscope 2024; 134:3664-3672. [PMID: 38651539 DOI: 10.1002/lary.31443] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/27/2024] [Revised: 03/17/2024] [Accepted: 03/27/2024] [Indexed: 04/25/2024]
Abstract
OBJECTIVE Accurate prediction of hospital length of stay (LOS) following surgical management of oral cavity cancer (OCC) may be associated with improved patient counseling, hospital resource utilization and cost. The objective of this study was to compare the performance of statistical models, a machine learning (ML) model, and The American College of Surgeons National Surgical Quality Improvement Program's (ACS-NSQIP) calculator in predicting LOS following surgery for OCC. MATERIALS AND METHODS A retrospective multicenter database study was performed at two major academic head and neck cancer centers. Patients with OCC who underwent major free flap reconstructive surgery between January 2008 and June 2019 surgery were selected. Data were pooled and split into training and validation datasets. Statistical and ML models were developed, and performance was evaluated by comparing predicted and actual LOS using correlation coefficient values and percent accuracy. RESULTS Totally 837 patients were selected with mean patient age being 62.5 ± 11.7 [SD] years and 67% being male. The ML model demonstrated the best accuracy (validation correlation 0.48, 4-day accuracy 70%), compared with the statistical models: multivariate analysis (0.45, 67%) and least absolute shrinkage and selection operator (0.42, 70%). All were superior to the ACS-NSQIP calculator's performance (0.23, 59%). CONCLUSION We developed statistical and ML models that predicted LOS following major free flap reconstructive surgery for OCC. Our models demonstrated superior predictive performance to the ACS-NSQIP calculator. The ML model identified several novel predictors of LOS. These models must be validated in other institutions before being used in clinical practice. LEVEL OF EVIDENCE 3 Laryngoscope, 134:3664-3672, 2024.
Collapse
Affiliation(s)
- Amirpouyan Namavarian
- Department of Otolaryngology-Head & Neck Surgery, University of Toronto, Toronto, Ontario, Canada
| | | | - Yangqing Deng
- Department of Biostatistics, Princess Margaret Cancer Center-University Health Network, Toronto, Ontario, Canada
| | - Shuja Khalid
- Department of Computer Science, University of Toronto, Toronto, Ontario, Canada
| | - Hedyeh Ziai
- Department of Otolaryngology-Head & Neck Surgery, University of Toronto, Toronto, Ontario, Canada
| | - Konrado Deutsch
- Department of Otolaryngology-Head & Neck Surgery, University of Toronto, Toronto, Ontario, Canada
| | - Jingyue Huang
- Department of Biostatistics, Princess Margaret Cancer Center-University Health Network, Toronto, Ontario, Canada
| | - Ralph W Gilbert
- Department of Otolaryngology-Head & Neck Surgery, University of Toronto, Toronto, Ontario, Canada
- Department of Otolaryngology-Head & Neck Surgery, Princess Margaret Cancer Center-University Health Network, Toronto, Ontario, Canada
| | - David P Goldstein
- Department of Otolaryngology-Head & Neck Surgery, University of Toronto, Toronto, Ontario, Canada
- Department of Otolaryngology-Head & Neck Surgery, Princess Margaret Cancer Center-University Health Network, Toronto, Ontario, Canada
| | - Christopher M K L Yao
- Department of Otolaryngology-Head & Neck Surgery, University of Toronto, Toronto, Ontario, Canada
- Department of Otolaryngology-Head & Neck Surgery, Princess Margaret Cancer Center-University Health Network, Toronto, Ontario, Canada
| | - Jonathan C Irish
- Department of Otolaryngology-Head & Neck Surgery, University of Toronto, Toronto, Ontario, Canada
- Department of Otolaryngology-Head & Neck Surgery, Princess Margaret Cancer Center-University Health Network, Toronto, Ontario, Canada
| | - Danny J Enepekides
- Department of Otolaryngology-Head & Neck Surgery, University of Toronto, Toronto, Ontario, Canada
- Department of Otolaryngology-Head & Neck Surgery, Sunnybrook Health Sciences Center, Toronto, Ontario, Canada
| | - Kevin M Higgins
- Department of Otolaryngology-Head & Neck Surgery, University of Toronto, Toronto, Ontario, Canada
- Department of Otolaryngology-Head & Neck Surgery, Sunnybrook Health Sciences Center, Toronto, Ontario, Canada
| | - Frank Rudzicz
- Department of Computer Science, University of Toronto, Toronto, Ontario, Canada
- International Centre for Surgical Safety, Li Ka Shing Knowledge Institute, St. Michael's Hospital, Toronto, Ontario, Canada
- Vector Institute for Artificial Intelligence, Toronto, Ontario, Canada
| | - Antoine Eskander
- Department of Otolaryngology-Head & Neck Surgery, University of Toronto, Toronto, Ontario, Canada
- Department of Otolaryngology-Head & Neck Surgery, Sunnybrook Health Sciences Center, Toronto, Ontario, Canada
| | - Wei Xu
- Department of Biostatistics, Princess Margaret Cancer Center-University Health Network, Toronto, Ontario, Canada
- Department of Otolaryngology-Head & Neck Surgery, Princess Margaret Cancer Center-University Health Network, Toronto, Ontario, Canada
- Department of Biostatistics, Dalla Lana School of Public Health, University of Toronto, Toronto, Ontario, Canada
| | - John R de Almeida
- Department of Otolaryngology-Head & Neck Surgery, University of Toronto, Toronto, Ontario, Canada
- Department of Otolaryngology-Head & Neck Surgery, Princess Margaret Cancer Center-University Health Network, Toronto, Ontario, Canada
- Department of Otolaryngology-Head & Neck Surgery, Sinai Health System, Toronto, Ontario, Canada
- Institute of Health Policy, Management, and Evaluation, University of Toronto, Toronto, Ontario, Canada
| |
Collapse
|
|
11
|
Song Y, Zhai W, Ma S, Wu Y, Ren M, Van den Eynde J, Nardi P, Pang PYK, Ali JM, Han J, Guo Z. Machine learning-based prediction of off-pump coronary artery bypass grafting-associated acute kidney injury. J Thorac Dis 2024; 16:4535-4542. [PMID: 39144311 PMCID: PMC11320255 DOI: 10.21037/jtd-24-711] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/01/2024] [Accepted: 06/27/2024] [Indexed: 08/16/2024]
Abstract
Background The cardiac surgery-associated acute kidney injury (CSA-AKI) occurs in up to 1 out of 3 patients. Off-pump coronary artery bypass grafting (OPCABG) is one of the major cardiac surgeries leading to CSA-AKI. Early identification and timely intervention are of clinical significance for CSA-AKI. In this study, we aimed to establish a prediction model of off-pump coronary artery bypass grafting-associated acute kidney injury (OPCABG-AKI) after surgery based on machine learning methods. Methods The preoperative and intraoperative data of 1,041 patients who underwent OPCABG in Chest Hospital, Tianjin University from June 1, 2021 to April 30, 2023 were retrospectively collected. The definition of OPCABG-AKI was based on the 2012 Kidney Disease Improving Global Outcomes (KDIGO) criteria. The baseline data and intraoperative time series data were included in the dataset, which were preprocessed separately. A total of eight machine learning models were constructed based on the baseline data: logistic regression (LR), gradient-boosting decision tree (GBDT), eXtreme gradient boosting (XGBoost), adaptive boosting (AdaBoost), random forest (RF), support vector machine (SVM), k-nearest neighbor (KNN), and decision tree (DT). The intraoperative time series data were extracted using a long short-term memory (LSTM) deep learning model. The baseline data and intraoperative features were then integrated through transfer learning and fused into each of the eight machine learning models for training. Based on the calculation of accuracy and area under the curve (AUC) of the prediction model, the best model was selected to establish the final OPCABG-AKI risk prediction model. The importance of features was calculated and ranked by DT model, to identify the main risk factors. Results Among 701 patients included in the study, 73 patients (10.4%) developed OPCABG-AKI. The GBDT model was shown to have the best predictions, both based on baseline data only (AUC =0.739, accuracy: 0.943) as well as based on baseline and intraoperative datasets (AUC =0.861, accuracy: 0.936). The ranking of importance of features of the GBDT model showed that use of insulin aspart was the most important predictor of OPCABG-AKI, followed by use of acarbose, spironolactone, alfentanil, dezocine, levosimendan, clindamycin, history of myocardial infarction, and gender. Conclusions A GBDT-based model showed excellent performance for the prediction of OPCABG-AKI. The fusion of preoperative and intraoperative data can improve the accuracy of predicting OPCABG-AKI.
Collapse
Affiliation(s)
- Yuezi Song
- Department of Anesthesiology, Chest Hospital, Tianjin University, Tianjin, China
| | - Wenqian Zhai
- Department of Anesthesiology, Chest Hospital, Tianjin University, Tianjin, China
- Tianjin Key Laboratory of Cardiovascular Emergencies and Critical Care, Tianjin, China
| | - Songnan Ma
- Academy of Medical Engineering and Translational Medicine, Tianjin University, Tianjin, China
| | - Yubo Wu
- Academy of Medical Engineering and Translational Medicine, Tianjin University, Tianjin, China
| | - Min Ren
- Tianjin Institute of Cardiovascular Disease, Tianjin, China
| | | | - Paolo Nardi
- Department of Cardiac Surgery, Tor Vergata University Hospital of Rome, Rome, Italy
| | - Philip Y. K. Pang
- Department of Cardiothoracic Surgery, National Heart Centre Singapore, Singapore, Singapore
| | - Jason M. Ali
- Department of Cardiothoracic Surgery, Royal Papworth Hospital, Cambridge, UK
| | - Jiange Han
- Department of Anesthesiology, Chest Hospital, Tianjin University, Tianjin, China
- Tianjin Key Laboratory of Cardiovascular Emergencies and Critical Care, Tianjin, China
| | - Zhigang Guo
- Tianjin Key Laboratory of Cardiovascular Emergencies and Critical Care, Tianjin, China
- Department of Cardiovascular Surgery, Chest Hospital, Tianjin University, Tianjin, China
| |
Collapse
|
|
12
|
Shin S, Choi TY, Han DH, Choi B, Cho E, Seog Y, Koo BN. An explainable machine learning model to predict early and late acute kidney injury after major hepatectomy. HPB (Oxford) 2024; 26:949-959. [PMID: 38705794 DOI: 10.1016/j.hpb.2024.04.005] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 05/19/2023] [Revised: 12/13/2023] [Accepted: 04/19/2024] [Indexed: 05/07/2024]
Abstract
BACKGROUND Risk assessment models for acute kidney injury (AKI) after major hepatectomy that differentiate between early and late AKI are lacking. This retrospective study aimed to create a model predicting AKI through machine learning and identify features that contribute to the development of early and late AKI. METHODS Patients that underwent major hepatectomy were categorized into the No-AKI, Early-AKI (within 48 h) or Late-AKI group (between 48 h and 7 days). Modeling was done with 20 perioperative features and the performance of prediction models were measured by the area under the receiver operating characteristic curve (AUROCC). Shapley Additive Explanation (SHAP) values were utilized to explain the outcome of the prediction model. RESULTS Of the 1383 patients included in this study, 1229, 110 and 44 patients were categorized into the No-AKI, Early-AKI and Late-AKI group, respectively. The CatBoost classifier exhibited the greatest AUROCC of 0.758 (95% CI: 0.671-0.847) and was found to differentiate well between Early and Late-AKI. We identified different perioperative features for predicting each outcome and found 1-year mortality to be greater for Early-AKI. CONCLUSIONS Our results suggest that risk factors are different for Early and Late-AKI after major hepatectomy, and 1-year mortality is greater for Early-AKI.
Collapse
Affiliation(s)
- Seokyung Shin
- Department of Anesthesiology and Pain Medicine, Severance Hospital, Anesthesia and Pain Research Institute, Yonsei University College of Medicine, 50-1 Yonsei-ro, Seodamun-gu, Seoul 03722, South Korea
| | - Tae Y Choi
- Department of Anesthesiology and Pain Medicine, Severance Hospital, Anesthesia and Pain Research Institute, Yonsei University College of Medicine, 50-1 Yonsei-ro, Seodamun-gu, Seoul 03722, South Korea
| | - Dai H Han
- Department of Surgery, Division of Hepato-biliary and Pancreatic Surgery, Yonsei University College of Medicine, 50-1 Yonsei-ro, Seodamun-gu, Seoul 03722, South Korea
| | - Boin Choi
- Severance Hospital, 50-1 Yonsei-ro, Seodamun-gu, Seoul 03722, South Korea
| | - Eunsung Cho
- Severance Hospital, 50-1 Yonsei-ro, Seodamun-gu, Seoul 03722, South Korea
| | - Yeong Seog
- Yonsei University College of Medicine, 50-1 Yonsei-ro, Seodamun-gu, Seoul 03722, South Korea
| | - Bon-Nyeo Koo
- Department of Anesthesiology and Pain Medicine, Severance Hospital, Anesthesia and Pain Research Institute, Yonsei University College of Medicine, 50-1 Yonsei-ro, Seodamun-gu, Seoul 03722, South Korea.
| |
Collapse
|
|
13
|
Arina P, Kaczorek MR, Hofmaenner DA, Pisciotta W, Refinetti P, Singer M, Mazomenos EB, Whittle J. Prediction of Complications and Prognostication in Perioperative Medicine: A Systematic Review and PROBAST Assessment of Machine Learning Tools. Anesthesiology 2024; 140:85-101. [PMID: 37944114 PMCID: PMC11146190 DOI: 10.1097/aln.0000000000004764] [Citation(s) in RCA: 15] [Impact Index Per Article: 15.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/22/2023] [Indexed: 11/12/2023]
Abstract
BACKGROUND The utilization of artificial intelligence and machine learning as diagnostic and predictive tools in perioperative medicine holds great promise. Indeed, many studies have been performed in recent years to explore the potential. The purpose of this systematic review is to assess the current state of machine learning in perioperative medicine, its utility in prediction of complications and prognostication, and limitations related to bias and validation. METHODS A multidisciplinary team of clinicians and engineers conducted a systematic review using the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) protocol. Multiple databases were searched, including Scopus, Cumulative Index to Nursing and Allied Health Literature (CINAHL), the Cochrane Library, PubMed, Medline, Embase, and Web of Science. The systematic review focused on study design, type of machine learning model used, validation techniques applied, and reported model performance on prediction of complications and prognostication. This review further classified outcomes and machine learning applications using an ad hoc classification system. The Prediction model Risk Of Bias Assessment Tool (PROBAST) was used to assess risk of bias and applicability of the studies. RESULTS A total of 103 studies were identified. The models reported in the literature were primarily based on single-center validations (75%), with only 13% being externally validated across multiple centers. Most of the mortality models demonstrated a limited ability to discriminate and classify effectively. The PROBAST assessment indicated a high risk of systematic errors in predicted outcomes and artificial intelligence or machine learning applications. CONCLUSIONS The findings indicate that the development of this field is still in its early stages. This systematic review indicates that application of machine learning in perioperative medicine is still at an early stage. While many studies suggest potential utility, several key challenges must be first overcome before their introduction into clinical practice. EDITOR’S PERSPECTIVE
Collapse
Affiliation(s)
- Pietro Arina
- Bloomsbury Institute of Intensive Care Medicine and Human Physiology and Performance Laboratory, Centre for Perioperative Medicine, Department of Targeted Intervention, University College London, London, United Kingdom
| | - Maciej R. Kaczorek
- Wellcome/EPSRC Centre of Interventional and Surgical Sciences and Department of Medical Physics and Biomedical Engineering, University College London, London, United Kingdom
| | - Daniel A. Hofmaenner
- Bloomsbury Institute of Intensive Care Medicine, University College London, London, United Kingdom; and Institute of Intensive Care Medicine, University Hospital Zurich, Zurich, Switzerland
| | - Walter Pisciotta
- Bloomsbury Institute of Intensive Care Medicine, University College London, London, United Kingdom
| | - Patricia Refinetti
- Human Physiology and Performance Laboratory, Centre for Perioperative Medicine, Department of Targeted Intervention, University College London, London, United Kingdom
| | - Mervyn Singer
- Bloomsbury Institute of Intensive Care Medicine, University College London, London, United Kingdom
| | - Evangelos B. Mazomenos
- Wellcome/EPSRC Centre of Interventional and Surgical Sciences and Department of Medical Physics and Biomedical Engineering, University College London, London, United Kingdom
| | - John Whittle
- Human Physiology and Performance Laboratory, Centre for Perioperative Medicine, Department of Targeted Intervention, University College London, London, United Kingdom
| |
Collapse
|
|
14
|
Choi E, Leonard KW, Jassal JS, Levin AM, Ramachandra V, Jones LR. Artificial Intelligence in Facial Plastic Surgery: A Review of Current Applications, Future Applications, and Ethical Considerations. Facial Plast Surg 2023; 39:454-459. [PMID: 37353051 DOI: 10.1055/s-0043-1770160] [Citation(s) in RCA: 6] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 06/25/2023] Open
Abstract
From virtual chat assistants to self-driving cars, artificial intelligence (AI) is often heralded as the technology that has and will continue to transform this generation. Among widely adopted applications in other industries, its potential use in medicine is being increasingly explored, where the vast amounts of data present in electronic health records and need for continuous improvements in patient care and workflow efficiency present many opportunities for AI implementation. Indeed, AI has already demonstrated capabilities for assisting in tasks such as documentation, image classification, and surgical outcome prediction. More specifically, this technology can be harnessed in facial plastic surgery, where the unique characteristics of the field lends itself well to specific applications. AI is not without its limitations, however, and the further adoption of AI in medicine and facial plastic surgery must necessarily be accompanied by discussion on the ethical implications and proper usage of AI in healthcare. In this article, we review current and potential uses of AI in facial plastic surgery, as well as its ethical ramifications.
Collapse
Affiliation(s)
- Elizabeth Choi
- Wayne State University School of Medicine, Detroit, Michigan
| | - Kyle W Leonard
- Department of Otolaryngology, Henry Ford Hospital, Detroit, Michigan
| | - Japnam S Jassal
- Department of Otolaryngology, Henry Ford Hospital, Detroit, Michigan
| | - Albert M Levin
- Department of Public Health Science, Henry Ford Health, Detroit, Michigan
- Center for Bioinformatics, Henry Ford Health, Detroit, Michigan
| | - Vikas Ramachandra
- Department of Public Health Science, Henry Ford Health, Detroit, Michigan
- Center for Bioinformatics, Henry Ford Health, Detroit, Michigan
| | - Lamont R Jones
- Department of Otolaryngology, Henry Ford Hospital, Detroit, Michigan
| |
Collapse
|
|
15
|
Nolde JM, Schlaich MP, Sessler DI, Mian A, Corcoran TB, Chow CK, Chan MTV, Borges FK, McGillion MH, Myles PS, Mills NL, Devereaux PJ, Hillis GS. Machine learning to predict myocardial injury and death after non-cardiac surgery. Anaesthesia 2023; 78:853-860. [PMID: 37070957 DOI: 10.1111/anae.16024] [Citation(s) in RCA: 4] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Accepted: 03/17/2023] [Indexed: 04/19/2023]
Abstract
Myocardial injury due to ischaemia within 30 days of non-cardiac surgery is prognostically relevant. We aimed to determine the discrimination, calibration, accuracy, sensitivity and specificity of single-layer and multiple-layer neural networks for myocardial injury and death within 30 postoperative days. We analysed data from 24,589 participants in the Vascular Events in Non-cardiac Surgery Patients Cohort Evaluation study. Validation was performed on a randomly selected subset of the study population. Discrimination for myocardial injury by single-layer vs. multiple-layer models generated areas (95%CI) under the receiver operating characteristic curve of: 0.70 (0.69-0.72) vs. 0.71 (0.70-0.73) with variables available before surgical referral, p < 0.001; 0.73 (0.72-0.75) vs. 0.75 (0.74-0.76) with additional variables available on admission, but before surgery, p < 0.001; and 0.76 (0.75-0.77) vs. 0.77 (0.76-0.78) with the addition of subsequent variables, p < 0.001. Discrimination for death by single-layer vs. multiple-layer models generated areas (95%CI) under the receiver operating characteristic curve of: 0.71 (0.66-0.76) vs. 0.74 (0.71-0.77) with variables available before surgical referral, p = 0.04; 0.78 (0.73-0.82) vs. 0.83 (0.79-0.86) with additional variables available on admission but before surgery, p = 0.01; and 0.87 (0.83-0.89) vs. 0.87 (0.85-0.90) with the addition of subsequent variables, p = 0.52. The accuracy of the multiple-layer model for myocardial injury and death with all variables was 70% and 89%, respectively.
Collapse
Affiliation(s)
- J M Nolde
- Dobney Hypertension Centre, Royal Perth Hospital Research Foundation, Perth, Australia
| | - M P Schlaich
- Dobney Hypertension Centre, Royal Perth Hospital Research Foundation, Perth, Australia
| | - D I Sessler
- Department of Outcomes Research, Cleveland Clinic, Cleveland, OH, USA
| | - A Mian
- School of Computer Science and Software Engineering, University of Western Australia, Perth, Australia
| | - T B Corcoran
- Department of Anaesthesia and Pain Medicine, Royal Perth Hospital and Medical School, University of Western Australia and Department of Anaesthesiology and Peri-operative Medicine, Alfred Hospital and Monash University, Melbourne, Australia
| | - C K Chow
- Westmead Applied Research Centre, Faculty of Medicine and Health, University of Sydney, and Department of Cardiology, Westmead Hospital, Sydney, Australia
| | - M T V Chan
- Department of Anaesthesia and Intensive Care, The Chinese University of Hong Kong, Hong Kong Special Administrative Region, China
| | - F K Borges
- McMaster University, Faculty of Health Sciences and Population Health Research Institute, Hamilton, ON, Canada
| | - M H McGillion
- McMaster University, Faculty of Health Sciences and Population Health Research Institute, Hamilton, ON, Canada
| | - P S Myles
- Department of Anaesthesiology and Peri-operative Medicine, Alfred Hospital and Monash University, Melbourne, Australia
| | - N L Mills
- British Heart Foundation Centre for Cardiovascular Science, University of Edinburgh and Usher Institute, Edinburgh, UK
| | - P J Devereaux
- McMaster University, Faculty of Health Sciences and Population Health Research Institute, Hamilton, ON, Canada
| | - G S Hillis
- Medical School, University of Western Australia and Department of Cardiology, Royal Perth Hospital, Perth, Australia
| |
Collapse
|
|
16
|
Reich DA, Adiyeke E, Ozrazgat-Baslanti T, Rabley AK, Bozorgmehri S, Bihorac A, Bird VG. Clinical Considerations for Patients Experiencing Acute Kidney Injury Following Percutaneous Nephrolithotomy. Biomedicines 2023; 11:1712. [PMID: 37371807 PMCID: PMC10296554 DOI: 10.3390/biomedicines11061712] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/16/2023] [Revised: 06/06/2023] [Accepted: 06/08/2023] [Indexed: 06/29/2023] Open
Abstract
Acute kidney injury (AKI) is a common postoperative outcome in urology patients undergoing surgery for nephrolithiasis. The objective of this study was to determine the prevalence of postoperative AKI and its degrees of severity, identify risk factors, and understand the resultant outcomes of AKI in patients with nephrolithiasis undergoing percutaneous nephrolithotomy (PCNL). A cohort of patients admitted between 2012 and 2019 to a single tertiary-care institution who had undergone PCNL was retrospectively analyzed. Among 417 (n = 326 patients) encounters, 24.9% (n = 104) had AKI. Approximately one-quarter of AKI patients (n = 18) progressed to Stage 2 or higher AKI. Hypertension, peripheral vascular disease, chronic kidney disease, and chronic anemia were significant risk factors of post-PCNL AKI. Corticosteroids and antifungals were associated with increased odds of AKI. Cardiovascular, neurologic complications, sepsis, and prolonged intensive care unit (ICU) stay percentages were higher in AKI patients. Hospital and ICU length of stay was greater in the AKI group. Provided the limited literature regarding postoperative AKI following PCNL, and the detriment that AKI can have on clinical outcomes, it is important to continue studying this topic to better understand how to optimize patient care to address patient- and procedure-specific risk factors.
Collapse
Affiliation(s)
- Daniel A. Reich
- University of Florida College of Medicine, Gainesville, FL 32610, USA; (D.A.R.); (E.A.); (T.O.-B.); (S.B.); (A.B.)
| | - Esra Adiyeke
- University of Florida College of Medicine, Gainesville, FL 32610, USA; (D.A.R.); (E.A.); (T.O.-B.); (S.B.); (A.B.)
- Division of Nephrology, Hypertension, and Renal Transplantation, University of Florida College of Medicine, Gainesville, FL 32610, USA
- Intelligent Critical Care Center (IC3), University of Florida, Gainesville, FL 32610, USA
| | - Tezcan Ozrazgat-Baslanti
- University of Florida College of Medicine, Gainesville, FL 32610, USA; (D.A.R.); (E.A.); (T.O.-B.); (S.B.); (A.B.)
- Division of Nephrology, Hypertension, and Renal Transplantation, University of Florida College of Medicine, Gainesville, FL 32610, USA
- Intelligent Critical Care Center (IC3), University of Florida, Gainesville, FL 32610, USA
| | - Andrew K. Rabley
- Department of Urology, University of Florida College of Medicine, Gainesville, FL 32610, USA;
| | - Shahab Bozorgmehri
- University of Florida College of Medicine, Gainesville, FL 32610, USA; (D.A.R.); (E.A.); (T.O.-B.); (S.B.); (A.B.)
| | - Azra Bihorac
- University of Florida College of Medicine, Gainesville, FL 32610, USA; (D.A.R.); (E.A.); (T.O.-B.); (S.B.); (A.B.)
- Division of Nephrology, Hypertension, and Renal Transplantation, University of Florida College of Medicine, Gainesville, FL 32610, USA
- Intelligent Critical Care Center (IC3), University of Florida, Gainesville, FL 32610, USA
| | - Vincent G. Bird
- University of Florida College of Medicine, Gainesville, FL 32610, USA; (D.A.R.); (E.A.); (T.O.-B.); (S.B.); (A.B.)
- Department of Urology, University of Florida College of Medicine, Gainesville, FL 32610, USA;
| |
Collapse
|
|
17
|
Schwager E, Ghosh E, Eshelman L, Pasupathy KS, Barreto EF, Kashani K. Accurate and interpretable prediction of ICU-acquired AKI. J Crit Care 2023; 75:154278. [PMID: 36774817 PMCID: PMC10121926 DOI: 10.1016/j.jcrc.2023.154278] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/12/2022] [Revised: 01/17/2023] [Accepted: 02/03/2023] [Indexed: 02/12/2023]
Abstract
PURPOSE We developed and validated two parsimonious algorithms to predict the time of diagnosis of any stage of acute kidney injury (any-AKI) or moderate-to-severe AKI in clinically actionable prediction windows. MATERIALS AND METHODS In this retrospective single-center cohort of adult ICU admissions, we trained two gradient-boosting models: 1) any-AKI model, predicting the risk of any-AKI at least 6 h before diagnosis (50,342 admissions), and 2) moderate-to-severe AKI model, predicting the risk of moderate-to-severe AKI at least 12 h before diagnosis (39,087 admissions). Performance was assessed before disease diagnosis and validated prospectively. RESULTS The models achieved an area under the receiver operating characteristic curve (AUROC) of 0.756 at six hours (any-AKI) and 0.721 at 12 h (moderate-to-severe AKI) prior. Prospectively, both models had high positive predictive values (0.796 and 0.546 for any-AKI and moderate-to-severe AKI models, respectively) and triggered more in patients who developed AKI vs. those who did not (median of 1.82 [IQR 0-4.71] vs. 0 [IQR 0-0.73] and 2.35 [IQR 0.14-4.96] vs. 0 [IQR 0-0.8] triggers per 8 h for any-AKI and moderate-to-severe AKI models, respectively). CONCLUSIONS The two AKI prediction models have good discriminative performance using common features, which can aid in accurately and informatively monitoring AKI risk in ICU patients.
Collapse
Affiliation(s)
| | - Erina Ghosh
- Philips Research North America, Cambridge, MA, USA
| | | | - Kalyan S Pasupathy
- Department of Biomedical & Health Information Sciences, University of Illinois, Chicago, IL, USA; Center for Clinical & Translational Science, University of Illinois, Chicago, IL, USA
| | | | - Kianoush Kashani
- Division of Nephrology and Hypertension, Department of Medicine, Mayo Clinic, Rochester, MN, USA; Division of Pulmonary and Critical Care Medicine, Department of Medicine, Mayo Clinic, Rochester, MN, USA.
| |
Collapse
|
|
18
|
Feng Y, Wang AY, Jun M, Pu L, Weisbord SD, Bellomo R, Hong D, Gallagher M. Characterization of Risk Prediction Models for Acute Kidney Injury: A Systematic Review and Meta-analysis. JAMA Netw Open 2023; 6:e2313359. [PMID: 37184837 PMCID: PMC12011341 DOI: 10.1001/jamanetworkopen.2023.13359] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 01/24/2023] [Accepted: 03/30/2023] [Indexed: 05/16/2023] Open
Abstract
Importance Despite the expansion of published prediction models for acute kidney injury (AKI), there is little evidence of uptake of these models beyond their local derivation nor data on their association with patient outcomes. Objective To systematically review published AKI prediction models across all clinical subsettings. Data Sources MEDLINE via PubMed (January 1946 to April 2021) and Embase (January 1947 to April 2021) were searched using medical subject headings and text words related to AKI and prediction models. Study Selection All studies that developed a prediction model for AKI, defined as a statistical model with at least 2 predictive variables to estimate future occurrence of AKI, were eligible for inclusion. There was no limitation on study populations or methodological designs. Data Extraction and Synthesis Two authors independently searched the literature, screened the studies, and extracted and analyzed the data following the Preferred Reporting Items for Systematic Review and Meta-analyses guideline. The data were pooled using a random-effects model, with subgroups defined by 4 clinical settings. Between-study heterogeneity was explored using multiple methods, and funnel plot analysis was used to identify publication bias. Main Outcomes and Measures C statistic was used to measure the discrimination of prediction models. Results Of the 6955 studies initially identified through literature searching, 150 studies, with 14.4 million participants, met the inclusion criteria. The study characteristics differed widely in design, population, AKI definition, and model performance assessments. The overall pooled C statistic was 0.80 (95% CI, 0.79-0.81), with pooled C statistics in different clinical subsettings ranging from 0.78 (95% CI, 0.75-0.80) to 0.82 (95% CI, 0.78-0.86). Between-study heterogeneity was high overall and in the different clinical settings (eg, contrast medium-associated AKI: I2 = 99.9%; P < .001), and multiple methods did not identify any clear sources. A high proportion of models had a high risk of bias (126 [84.4%]) according to the Prediction Model Risk Of Bias Assessment Tool. Conclusions and Relevance In this study, the discrimination of the published AKI prediction models was good, reflected by high C statistics; however, the wide variation in the clinical settings, populations, and predictive variables likely drives the highly heterogenous findings that limit clinical utility. Standardized procedures for development and validation of prediction models are urgently needed.
Collapse
Affiliation(s)
- Yunlin Feng
- Department of Nephrology, Sichuan Provincial People’s Hospital, University of Electronic Science and Technology of China, Chengdu, China
- The George Institute for Global Health, University of New South Wales, Sydney, Australia
| | - Amanda Y. Wang
- The George Institute for Global Health, University of New South Wales, Sydney, Australia
- Concord Clinical School, University of Sydney, Sydney, Australia
- The Faculty of Medicine and Health Sciences, Macquarie University, Sydney, Australia
| | - Min Jun
- The George Institute for Global Health, University of New South Wales, Sydney, Australia
| | - Lei Pu
- Department of Nephrology, Sichuan Provincial People’s Hospital, University of Electronic Science and Technology of China, Chengdu, China
| | - Steven D. Weisbord
- Renal Section, Medicine Service, Veterans Affairs Pittsburgh Healthcare System, Pittsburgh, Pennsylvania
- Renal-Electrolyte Division, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania
| | - Rinaldo Bellomo
- Department of Critical Care, University of Melbourne, Melbourne, Australia
| | - Daqing Hong
- Department of Nephrology, Sichuan Provincial People’s Hospital, University of Electronic Science and Technology of China, Chengdu, China
| | - Martin Gallagher
- The George Institute for Global Health, University of New South Wales, Sydney, Australia
- South Western Sydney Clinical School, University of New South Wales, Sydney, Australia
| |
Collapse
|
|
19
|
Wu M, Jiang X, Du K, Xu Y, Zhang W. Ensemble machine learning algorithm for predicting acute kidney injury in patients admitted to the neurointensive care unit following brain surgery. Sci Rep 2023; 13:6705. [PMID: 37185782 PMCID: PMC10130041 DOI: 10.1038/s41598-023-33930-5] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/17/2022] [Accepted: 04/20/2023] [Indexed: 05/17/2023] Open
Abstract
Acute kidney injury (AKI) is a common postoperative complication among patients in the neurological intensive care unit (NICU), often resulting in poor prognosis and high mortality. In this retrospective cohort study, we established a model for predicting AKI following brain surgery based on an ensemble machine learning algorithm using data from 582 postoperative patients admitted to the NICU at the Dongyang People's Hospital from March 1, 2017, to January 31, 2020. Demographic, clinical, and intraoperative data were collected. Four machine learning algorithms (C5.0, support vector machine, Bayes, and XGBoost) were used to develop the ensemble algorithm. The AKI incidence in critically ill patients after brain surgery was 20.8%. Intraoperative blood pressure; postoperative oxygenation index; oxygen saturation; and creatinine, albumin, urea, and calcium levels were associated with the postoperative AKI occurrence. The area under the curve value for the ensembled model was 0.85. The accuracy, precision, specificity, recall, and balanced accuracy values were 0.81, 0.86, 0.44, 0.91, and 0.68, respectively, indicating good predictive ability. Ultimately, the models using perioperative variables exhibited good discriminatory ability for early prediction of postoperative AKI risk in patients admitted to the NICU. Thus, the ensemble machine learning algorithm may be a valuable tool for forecasting AKI.
Collapse
Affiliation(s)
- Muying Wu
- Intensive Care Unit, Affiliated Dongyang Hospital of Wenzhou Medical University, No. 60 Wuning West Road, Jinhua, Dongyang, Zhejiang, People's Republic of China
| | - Xuandong Jiang
- Intensive Care Unit, Affiliated Dongyang Hospital of Wenzhou Medical University, No. 60 Wuning West Road, Jinhua, Dongyang, Zhejiang, People's Republic of China.
| | - Kailei Du
- Intensive Care Unit, Affiliated Dongyang Hospital of Wenzhou Medical University, No. 60 Wuning West Road, Jinhua, Dongyang, Zhejiang, People's Republic of China
| | - Yingting Xu
- Intensive Care Unit, Affiliated Dongyang Hospital of Wenzhou Medical University, No. 60 Wuning West Road, Jinhua, Dongyang, Zhejiang, People's Republic of China
| | - Weimin Zhang
- Intensive Care Unit, Affiliated Dongyang Hospital of Wenzhou Medical University, No. 60 Wuning West Road, Jinhua, Dongyang, Zhejiang, People's Republic of China
| |
Collapse
|
|
20
|
Begum M F, Narayan S. A Pattern mixture model with long short-term memory network for oliguric acute kidney injury prediction. JOURNAL OF KING SAUD UNIVERSITY - COMPUTER AND INFORMATION SCIENCES 2023. [DOI: 10.1016/j.jksuci.2023.03.007] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 03/19/2023]
|
|
21
|
Balch JA, Ruppert MM, Shickel B, Ozrazgat-Baslanti T, Tighe PJ, Efron PA, Upchurch GR, Rashidi P, Bihorac A, Loftus TJ. Building an automated, machine learning-enabled platform for predicting post-operative complications. Physiol Meas 2023; 44:024001. [PMID: 36657179 PMCID: PMC9910093 DOI: 10.1088/1361-6579/acb4db] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/04/2022] [Revised: 12/29/2022] [Accepted: 01/19/2023] [Indexed: 01/21/2023]
Abstract
Objective. In 2019, the University of Florida College of Medicine launched theMySurgeryRiskalgorithm to predict eight major post-operative complications using automatically extracted data from the electronic health record.Approach. This project was developed in parallel with our Intelligent Critical Care Center and represents a culmination of efforts to build an efficient and accurate model for data processing and predictive analytics.Main Results and Significance. This paper discusses how our model was constructed and improved upon. We highlight the consolidation of the database, processing of fixed and time-series physiologic measurements, development and training of predictive models, and expansion of those models into different aspects of patient assessment and treatment. We end by discussing future directions of the model.
Collapse
Affiliation(s)
- Jeremy A Balch
- Intelligent Critical Care Center, University of Florida, Gainesville, FL, United States of America
- Department of Surgery, University of Florida, Gainesville, Florida, United States of America
| | - Matthew M Ruppert
- Intelligent Critical Care Center, University of Florida, Gainesville, FL, United States of America
- Department of Medicine, University of Florida, Gainesville, Florida, United States of America
| | - Benjamin Shickel
- Intelligent Critical Care Center, University of Florida, Gainesville, FL, United States of America
- Department of Medicine, University of Florida, Gainesville, Florida, United States of America
| | - Tezcan Ozrazgat-Baslanti
- Intelligent Critical Care Center, University of Florida, Gainesville, FL, United States of America
- Department of Medicine, University of Florida, Gainesville, Florida, United States of America
| | - Patrick J Tighe
- Department of Anesthesiology, University of Florida, Gainesville, Florida, United States of America
| | - Philip A Efron
- Department of Surgery, University of Florida, Gainesville, Florida, United States of America
| | - Gilbert R Upchurch
- Department of Surgery, University of Florida, Gainesville, Florida, United States of America
| | - Parisa Rashidi
- Intelligent Critical Care Center, University of Florida, Gainesville, FL, United States of America
- Department of Biomedical Engineering, University of Florida, Gainesville, Florida, United States of America
| | - Azra Bihorac
- Intelligent Critical Care Center, University of Florida, Gainesville, FL, United States of America
- Department of Medicine, University of Florida, Gainesville, Florida, United States of America
| | - Tyler J Loftus
- Intelligent Critical Care Center, University of Florida, Gainesville, FL, United States of America
- Department of Surgery, University of Florida, Gainesville, Florida, United States of America
| |
Collapse
|
|
22
|
Eysenbach G, Kang YX, Duan SB, Yan P, Song GB, Zhang NY, Yang SK, Li JX, Zhang H. Machine Learning-Based Prediction of Acute Kidney Injury Following Pediatric Cardiac Surgery: Model Development and Validation Study. J Med Internet Res 2023; 25:e41142. [PMID: 36603200 PMCID: PMC9893730 DOI: 10.2196/41142] [Citation(s) in RCA: 10] [Impact Index Per Article: 5.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/17/2022] [Revised: 11/11/2022] [Accepted: 11/15/2022] [Indexed: 01/07/2023] Open
Abstract
BACKGROUND Cardiac surgery-associated acute kidney injury (CSA-AKI) is a major complication following pediatric cardiac surgery, which is associated with increased morbidity and mortality. The early prediction of CSA-AKI before and immediately after surgery could significantly improve the implementation of preventive and therapeutic strategies during the perioperative periods. However, there is limited clinical information on how to identify pediatric patients at high risk of CSA-AKI. OBJECTIVE The study aims to develop and validate machine learning models to predict the development of CSA-AKI in the pediatric population. METHODS This retrospective cohort study enrolled patients aged 1 month to 18 years who underwent cardiac surgery with cardiopulmonary bypass at 3 medical centers of Central South University in China. CSA-AKI was defined according to the 2012 Kidney Disease: Improving Global Outcomes criteria. Feature selection was applied separately to 2 data sets: the preoperative data set and the combined preoperative and intraoperative data set. Multiple machine learning algorithms were tested, including K-nearest neighbor, naive Bayes, support vector machines, random forest, extreme gradient boosting (XGBoost), and neural networks. The best performing model was identified in cross-validation by using the area under the receiver operating characteristic curve (AUROC). Model interpretations were generated using the Shapley additive explanations (SHAP) method. RESULTS A total of 3278 patients from one of the centers were used for model derivation, while 585 patients from another 2 centers served as the external validation cohort. CSA-AKI occurred in 564 (17.2%) patients in the derivation cohort and 51 (8.7%) patients in the external validation cohort. Among the considered machine learning models, the XGBoost models achieved the best predictive performance in cross-validation. The AUROC of the XGBoost model using only the preoperative variables was 0.890 (95% CI 0.876-0.906) in the derivation cohort and 0.857 (95% CI 0.800-0.903) in the external validation cohort. When the intraoperative variables were included, the AUROC increased to 0.912 (95% CI 0.899-0.924) and 0.889 (95% CI 0.844-0.920) in the 2 cohorts, respectively. The SHAP method revealed that baseline serum creatinine level, perfusion time, body length, operation time, and intraoperative blood loss were the top 5 predictors of CSA-AKI. CONCLUSIONS The interpretable XGBoost models provide practical tools for the early prediction of CSA-AKI, which are valuable for risk stratification and perioperative management of pediatric patients undergoing cardiac surgery.
Collapse
Affiliation(s)
| | - Yi-Xin Kang
- Department of Nephrology, The Second Xiangya Hospital of Central South University, Changsha, China
| | - Shao-Bin Duan
- Department of Nephrology, The Second Xiangya Hospital of Central South University, Changsha, China
| | - Ping Yan
- Department of Nephrology, The Second Xiangya Hospital of Central South University, Changsha, China
| | - Guo-Bao Song
- Department of Cardiovascular Surgery, The Second Xiangya Hospital of Central South University, Changsha, China
| | - Ning-Ya Zhang
- Information Center, The Second Xiangya Hospital of Central South University, Changsha, China
| | - Shi-Kun Yang
- Department of Nephrology, The Third Xiangya Hospital of Central South University, Changsha, China
| | - Jing-Xin Li
- Department of Cardiovascular Surgery, Xiangya Hospital of Central South University, Changsha, China
| | - Hui Zhang
- Department of Pediatrics, Xiangya Hospital of Central South University, Changsha, China
| |
Collapse
|
|
23
|
Takkavatakarn K, Hofer IS. Artificial Intelligence and Machine Learning in Perioperative Acute Kidney Injury. ADVANCES IN KIDNEY DISEASE AND HEALTH 2023; 30:53-60. [PMID: 36723283 DOI: 10.1053/j.akdh.2022.10.001] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Received: 07/21/2022] [Revised: 09/30/2022] [Accepted: 10/28/2022] [Indexed: 12/24/2022]
Abstract
Acute kidney injury (AKI) is a common complication after a surgery, especially in cardiac and aortic procedures, and has a significant impact on morbidity and mortality. Early identification of high-risk patients and providing effective prevention and therapeutic approach are the main strategies for reducing the possibility of perioperative AKI. Consequently, several risk-prediction models and risk assessment scores have been developed for the prediction of perioperative AKI. However, a majority of these risk scores are only derived from preoperative data while the intraoperative time-series monitoring data such as heart rate and blood pressure were not included. Moreover, the complexity of the pathophysiology of AKI, as well as its nonlinear and heterogeneous nature, imposes limitations on the use of linear statistical techniques. The development of clinical medicine's digitization, the widespread availability of electronic medical records, and the increase in the use of continuous monitoring have generated vast quantities of data. Machine learning has recently shown promise as a method for automatically integrating large amounts of data in predicting the risk of perioperative outcomes. In this article, we discussed the development, limitations of existing work, and the potential future direction of models using machine learning techniques to predict AKI after a surgery.
Collapse
Affiliation(s)
- Kullaya Takkavatakarn
- Division of Nephrology, Department of Medicine, Icahn School of Medicine at Mount Sinai, New York, NY; Division of Nephrology, Department of Medicine, Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand
| | - Ira S Hofer
- Department of Anesthesiology, Pain and Perioperative Medicine, Icahn School of Medicine at Mount, Sinai, NY.
| |
Collapse
|
|
24
|
Aslani N, Galehdar N, Garavand A. A systematic review of data mining applications in kidney transplantation. INFORMATICS IN MEDICINE UNLOCKED 2023. [DOI: 10.1016/j.imu.2023.101165] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/13/2023] Open
|
|
25
|
Zhang H, Wang AY, Wu S, Ngo J, Feng Y, He X, Zhang Y, Wu X, Hong D. Artificial intelligence for the prediction of acute kidney injury during the perioperative period: systematic review and Meta-analysis of diagnostic test accuracy. BMC Nephrol 2022; 23:405. [PMID: 36536317 PMCID: PMC9761969 DOI: 10.1186/s12882-022-03025-w] [Citation(s) in RCA: 14] [Impact Index Per Article: 4.7] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/11/2022] [Accepted: 11/30/2022] [Indexed: 12/23/2022] Open
Abstract
BACKGROUND Acute kidney injury (AKI) is independently associated with morbidity and mortality in a wide range of surgical settings. Nowadays, with the increasing use of electronic health records (EHR), advances in patient information retrieval, and cost reduction in clinical informatics, artificial intelligence is increasingly being used to improve early recognition and management for perioperative AKI. However, there is no quantitative synthesis of the performance of these methods. We conducted this systematic review and meta-analysis to estimate the sensitivity and specificity of artificial intelligence for the prediction of acute kidney injury during the perioperative period. METHODS Pubmed, Embase, and Cochrane Library were searched to 2nd October 2021. Studies presenting diagnostic performance of artificial intelligence in the early detection of perioperative acute kidney injury were included. True positives, false positives, true negatives and false negatives were pooled to collate specificity and sensitivity with 95% CIs and results were portrayed in forest plots. The risk of bias of eligible studies was assessed using the PROBAST tool. RESULTS Nineteen studies involving 304,076 patients were included. Quantitative random-effects meta-analysis using the Rutter and Gatsonis hierarchical summary receiver operating characteristics (HSROC) model revealed pooled sensitivity, specificity, and diagnostic odds ratio of 0.77 (95% CI: 0.73 to 0.81),0.75 (95% CI: 0.71 to 0.80), and 10.7 (95% CI 8.5 to 13.5), respectively. Threshold effect was found to be the only source of heterogeneity, and there was no evidence of publication bias. CONCLUSIONS Our review demonstrates the promising performance of artificial intelligence for early prediction of perioperative AKI. The limitations of lacking external validation performance and being conducted only at a single center should be overcome. TRIAL REGISTRATION This study was not registered with PROSPERO.
Collapse
Affiliation(s)
- Hanfei Zhang
- grid.54549.390000 0004 0369 4060School of Medicine, University of Electronic Science and Technology of China, Chengdu, China ,grid.54549.390000 0004 0369 4060Department of Nephrology, Sichuan Provincial People’s Hospital, University of Electronic Science and Technology of China, Chengdu, China
| | - Amanda Y. Wang
- grid.1004.50000 0001 2158 5405The faculty of medicine and health sciences, Macquarie University, Sydney, NSW Australia
| | - Shukun Wu
- grid.54549.390000 0004 0369 4060School of Medicine, University of Electronic Science and Technology of China, Chengdu, China ,grid.54549.390000 0004 0369 4060Department of Nephrology, Sichuan Provincial People’s Hospital, University of Electronic Science and Technology of China, Chengdu, China
| | - Johnathan Ngo
- grid.1013.30000 0004 1936 834XConcord Clinical School, University of Sydney, Sydney, Australia
| | - Yunlin Feng
- grid.54549.390000 0004 0369 4060School of Medicine, University of Electronic Science and Technology of China, Chengdu, China ,grid.54549.390000 0004 0369 4060Department of Nephrology, Sichuan Provincial People’s Hospital, University of Electronic Science and Technology of China, Chengdu, China
| | - Xin He
- grid.54549.390000 0004 0369 4060Department of Nephrology, Sichuan Provincial People’s Hospital, University of Electronic Science and Technology of China, Chengdu, China ,grid.488387.8Department of Nephrology, Affiliated Hospital of Southwest Medical University, Luzhou, China
| | - Yingfeng Zhang
- grid.54549.390000 0004 0369 4060School of Medicine, University of Electronic Science and Technology of China, Chengdu, China
| | - Xingwei Wu
- grid.54549.390000 0004 0369 4060School of Medicine, University of Electronic Science and Technology of China, Chengdu, China ,grid.54549.390000 0004 0369 4060Department of Nephrology, Sichuan Provincial People’s Hospital, University of Electronic Science and Technology of China, Chengdu, China ,grid.54549.390000 0004 0369 4060Department of Pharmacy, Sichuan Provincial Peoples Hospital, School of Medicine, University of Electronic Science and Technology of China, Chengdu, China
| | - Daqing Hong
- grid.54549.390000 0004 0369 4060School of Medicine, University of Electronic Science and Technology of China, Chengdu, China ,grid.54549.390000 0004 0369 4060Department of Nephrology, Sichuan Provincial People’s Hospital, University of Electronic Science and Technology of China, Chengdu, China ,grid.54549.390000 0004 0369 4060Renal Department and Nephrology Institute, Sichuan Provincial People’s Hospital, School of Medicine, University of Electronic Science and Technology of China, Chengdu, China
| |
Collapse
|
|
26
|
Vagliano I, Chesnaye NC, Leopold JH, Jager KJ, Abu-Hanna A, Schut MC. Machine learning models for predicting acute kidney injury: a systematic review and critical appraisal. Clin Kidney J 2022; 15:2266-2280. [PMID: 36381375 PMCID: PMC9664575 DOI: 10.1093/ckj/sfac181] [Citation(s) in RCA: 13] [Impact Index Per Article: 4.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/08/2022] [Indexed: 09/08/2023] Open
Abstract
BACKGROUND The number of studies applying machine learning (ML) to predict acute kidney injury (AKI) has grown steadily over the past decade. We assess and critically appraise the state of the art in ML models for AKI prediction, considering performance, methodological soundness, and applicability. METHODS We searched PubMed and ArXiv, extracted data, and critically appraised studies based on the Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis (TRIPOD), Checklist for Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies (CHARMS), and Prediction Model Risk of Bias Assessment Tool (PROBAST) guidelines. RESULTS Forty-six studies from 3166 titles were included. Thirty-eight studies developed a model, five developed and externally validated one, and three studies externally validated one. Flexible ML methods were used more often than deep learning, although the latter was common with temporal variables and text as predictors. Predictive performance showed an area under receiver operating curves ranging from 0.49 to 0.99. Our critical appraisal identified a high risk of bias in 39 studies. Some studies lacked internal validation, whereas external validation and interpretability of results were rarely considered. Fifteen studies focused on AKI prediction in the intensive care setting, and the US-derived Medical Information Mart for Intensive Care (MIMIC) data set was commonly used. Reproducibility was limited as data and code were usually unavailable. CONCLUSIONS Flexible ML methods are popular for the prediction of AKI, although more complex models based on deep learning are emerging. Our critical appraisal identified a high risk of bias in most models: Studies should use calibration measures and external validation more often, improve model interpretability, and share data and code to improve reproducibility.
Collapse
Affiliation(s)
- Iacopo Vagliano
- Deptartment of Medical Informatics, Amsterdam UMC, University of Amsterdam, Amsterdam, Amsterdam Public Health Research Institute, Amsterdam, The Netherlands
| | - Nicholas C Chesnaye
- ERA Registry, Department of Medical Informatics, Amsterdam UMC, University of Amsterdam, Amsterdam Public Health Research Institute, Amsterdam, The Netherlands
| | - Jan Hendrik Leopold
- Deptartment of Medical Informatics, Amsterdam UMC, University of Amsterdam, Amsterdam, Amsterdam Public Health Research Institute, Amsterdam, The Netherlands
| | - Kitty J Jager
- ERA Registry, Department of Medical Informatics, Amsterdam UMC, University of Amsterdam, Amsterdam Public Health Research Institute, Amsterdam, The Netherlands
| | - Ameen Abu-Hanna
- Deptartment of Medical Informatics, Amsterdam UMC, University of Amsterdam, Amsterdam, Amsterdam Public Health Research Institute, Amsterdam, The Netherlands
| | - Martijn C Schut
- Deptartment of Medical Informatics, Amsterdam UMC, University of Amsterdam, Amsterdam, Amsterdam Public Health Research Institute, Amsterdam, The Netherlands
| |
Collapse
|
|
27
|
Abstract
OBJECTIVE To describe the frequency and patterns of postoperative complications and FTR after inpatient pediatric surgical procedures and to evaluate the association between number of complications and FTR. SUMMARY AND BACKGROUND FTR, or a postoperative death after a complication, is currently a nationally endorsed quality measure for adults. Although it is a contributing factor to variation in mortality, relatively little is known about FTR after pediatric surgery. METHODS Cohort study of 200,554 patients within the National Surgical Quality Improvement Program-Pediatric database (2012-2016) who underwent a high (≥ 1%) or low (< 1%) mortality risk inpatient surgical procedures. Patients were stratified based on number of postoperative complications (0, 1, 2, or ≥3) and further categorized as having undergone either a low- or high-risk procedure. The association between the number of postoperative complications and FTR was evaluated with multivariable logistic regression. RESULTS Among patients who underwent a low- (89.4%) or high-risk (10.6%) procedures, 14.0% and 12.5% had at least 1 postoperative complication, respectively. FTR rates after low- and high-risk procedures demonstrated step-wise increases as the number of complications accrued (eg, low-risk- 9.2% in patients with ≥3 complications; high-risk-36.9% in patients with ≥ 3 complications). Relative to patients who had no complications, there was a dose-response relationship between mortality and the number of complications after low-risk [1 complication - odds ratio (OR) 3.34 (95% CI 2.62-4.27); 2 - OR 10.15 (95% CI 7.40-13.92); ≥3-27.48 (95% CI 19.06-39.62)] and high-risk operations [1 - OR 3.29 (2.61-4.16); 2-7.24 (5.14-10.19); ≥3-20.73 (12.62-34.04)]. CONCLUSIONS There is a dose-response relationship between the number of postoperative complications after inpatient surgery and FTR, ever after common, "minor" surgical procedures. These findings suggest FTR may be a potential quality measure for pediatric surgical care.
Collapse
|
|
28
|
Mehl SC, Portuondo JI, Pettit RW, Fallon SC, Wesson DE, Shah SR, Vogel AM, Lopez ME, Massarweh NN. Association of prematurity with complications and failure to rescue in neonatal surgery. J Pediatr Surg 2022; 57:268-276. [PMID: 34857374 PMCID: PMC9125744 DOI: 10.1016/j.jpedsurg.2021.10.050] [Citation(s) in RCA: 6] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 09/22/2021] [Revised: 10/15/2021] [Accepted: 10/25/2021] [Indexed: 10/19/2022]
Abstract
BACKGROUND The majority of failure to rescue (FTR), or death after a postoperative complication, in pediatric surgery occurs among infants and neonates. The purpose of this study is to evaluate the association between gestational age (GA) and FTR in infants and neonates. METHODS National cohort study of 46,452 patients < 1 year old within the National Surgical Quality Improvement Program-Pediatric database who underwent inpatient surgery. Patients were categorized as preterm neonates, term neonates, or infants. Neonates were stratified based on GA. Surgical procedures were classified as low- (< 1% mortality) or high-risk (≥ 1%). Multivariable logistic regression and cubic splines were used to evaluate the association between GA and FTR. RESULTS Preterm neonates had the highest FTR (28%) rates. Among neonates, FTR increased with decreasing GA (≥ 37 weeks, 12%; 33-36 weeks, 15%; 29-32 weeks, 30%; 25-28 weeks 41%; ≤ 24 weeks, 57%). For both low- and high-risk procedures, FTR significantly (trend test, p < 0.01) increased with decreasing GA. When stratifying preterm neonates by GA, all GAs ≤ 28 weeks were associated with significantly higher odds of FTR for low- (OR 2.47, 95% CI [1.38-4.41]) and high-risk (OR 2.27, 95% CI [1.33-3.87]) procedures. A lone inflection point for FTR was identified at 31-32 weeks with cubic spline analysis. CONCLUSIONS The dose-dependent relationship between decreasing GA and FTR as well as the FTR inflection point noted at GA 31-32 weeks can be used by stakeholders in designing quality improvement initiatives and directing perioperative care. LEVEL OF EVIDENCE Level IV, Retrospective cohort study.
Collapse
Affiliation(s)
- Steven C. Mehl
- Michael E. DeBakey Department of Surgery, Baylor College of Medicine, 1 Baylor Plaza, MS390, Houston, TX 77030, United States,Department of Surgery, Division of Pediatric Surgery, Texas Children’s Hospital, Houston, TX, United States,Corresponding author at: Michael E. DeBakey Department of Surgery, Baylor College of Medicine, 1 Baylor Plaza, MS390, Houston, TX 77030, United States. (S.C. Mehl)
| | - Jorge I. Portuondo
- Michael E. DeBakey Department of Surgery, Baylor College of Medicine, 1 Baylor Plaza, MS390, Houston, TX 77030, United States
| | - Rowland W. Pettit
- Michael E. DeBakey Department of Surgery, Baylor College of Medicine, 1 Baylor Plaza, MS390, Houston, TX 77030, United States
| | - Sara C. Fallon
- Michael E. DeBakey Department of Surgery, Baylor College of Medicine, 1 Baylor Plaza, MS390, Houston, TX 77030, United States,Department of Surgery, Division of Pediatric Surgery, Texas Children’s Hospital, Houston, TX, United States
| | - David E. Wesson
- Michael E. DeBakey Department of Surgery, Baylor College of Medicine, 1 Baylor Plaza, MS390, Houston, TX 77030, United States,Department of Surgery, Division of Pediatric Surgery, Texas Children’s Hospital, Houston, TX, United States
| | - Sohail R. Shah
- Michael E. DeBakey Department of Surgery, Baylor College of Medicine, 1 Baylor Plaza, MS390, Houston, TX 77030, United States,Department of Surgery, Division of Pediatric Surgery, Texas Children’s Hospital, Houston, TX, United States
| | - Adam M. Vogel
- Michael E. DeBakey Department of Surgery, Baylor College of Medicine, 1 Baylor Plaza, MS390, Houston, TX 77030, United States,Department of Surgery, Division of Pediatric Surgery, Texas Children’s Hospital, Houston, TX, United States
| | - Monica E. Lopez
- Michael E. DeBakey Department of Surgery, Baylor College of Medicine, 1 Baylor Plaza, MS390, Houston, TX 77030, United States,Department of Surgery, Division of Pediatric Surgery, Texas Children’s Hospital, Houston, TX, United States
| | - Nader N. Massarweh
- Atlanta VA Health Care System, Decatur, GA, United States,Department of Surgery, Division of Surgical Oncology, Emory University School of Medicine, Atlanta, GA, United States,Department of Surgery, Morehouse School of Medicine, Atlanta, GA, United States
| |
Collapse
|
|
29
|
Bajaj T, Koyner JL. Artificial Intelligence in Acute Kidney Injury Prediction. Adv Chronic Kidney Dis 2022; 29:450-460. [PMID: 36253028 PMCID: PMC10259199 DOI: 10.1053/j.ackd.2022.07.009] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/10/2022] [Revised: 07/20/2022] [Accepted: 07/27/2022] [Indexed: 01/25/2023]
Abstract
The use of artificial intelligence (AI) in nephrology and its associated clinical research is growing. Recent years have seen increased interest in utilizing AI to predict the development of hospital-based acute kidney injury (AKI). Several AI techniques have been employed to improve the ability to detect AKI across a variety of hospitalized settings. This review discusses the evolutions of AKI risk prediction discussing the static risk assessment models of yesteryear as well as the more recent trend toward AI and advanced learning techniques. We discuss the relative improvement in AKI detection as well as the relative dearth of data around the clinical implementation and patient outcomes using these models. The use of AI for AKI detection and clinical care is in its infancy, and this review describes how we arrived at our current position and hints at the promise of the future.
Collapse
Affiliation(s)
- Tushar Bajaj
- Section of Nephrology, Department of Medicine University of Chicago, Chicago, IL, USA
| | - Jay L Koyner
- Section of Nephrology, Department of Medicine University of Chicago, Chicago, IL, USA.
| |
Collapse
|
|
30
|
Loftus TJ, Shickel B, Ozrazgat-Baslanti T, Ren Y, Glicksberg BS, Cao J, Singh K, Chan L, Nadkarni GN, Bihorac A. Artificial intelligence-enabled decision support in nephrology. Nat Rev Nephrol 2022; 18:452-465. [PMID: 35459850 PMCID: PMC9379375 DOI: 10.1038/s41581-022-00562-3] [Citation(s) in RCA: 32] [Impact Index Per Article: 10.7] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Accepted: 03/16/2022] [Indexed: 12/12/2022]
Abstract
Kidney pathophysiology is often complex, nonlinear and heterogeneous, which limits the utility of hypothetical-deductive reasoning and linear, statistical approaches to diagnosis and treatment. Emerging evidence suggests that artificial intelligence (AI)-enabled decision support systems - which use algorithms based on learned examples - may have an important role in nephrology. Contemporary AI applications can accurately predict the onset of acute kidney injury before notable biochemical changes occur; can identify modifiable risk factors for chronic kidney disease onset and progression; can match or exceed human accuracy in recognizing renal tumours on imaging studies; and may augment prognostication and decision-making following renal transplantation. Future AI applications have the potential to make real-time, continuous recommendations for discrete actions and yield the greatest probability of achieving optimal kidney health outcomes. Realizing the clinical integration of AI applications will require cooperative, multidisciplinary commitment to ensure algorithm fairness, overcome barriers to clinical implementation, and build an AI-competent workforce. AI-enabled decision support should preserve the pre-eminence of wisdom and augment rather than replace human decision-making. By anchoring intuition with objective predictions and classifications, this approach should favour clinician intuition when it is honed by experience.
Collapse
Affiliation(s)
- Tyler J Loftus
- Department of Surgery, University of Florida Health, Gainesville, FL, USA
| | - Benjamin Shickel
- Department of Medicine, University of Florida Health, Gainesville, FL, USA
| | | | - Yuanfang Ren
- Department of Medicine, University of Florida Health, Gainesville, FL, USA
| | - Benjamin S Glicksberg
- Department of Genetics and Genomic Sciences, Icahn School of Medicine at Mount Sinai, New York, NY, USA
- Hasso Plattner Institute for Digital Health at Mount Sinai, Icahn School of Medicine at Mount Sinai, New York, NY, USA
| | - Jie Cao
- Department of Computational Medicine and Bioinformatics, University of Michigan Medical School, Ann Arbor, MI, USA
| | - Karandeep Singh
- Department of Learning Health Sciences and Internal Medicine, University of Michigan Medical School, Ann Arbor, MI, USA
| | - Lili Chan
- The Mount Sinai Clinical Intelligence Center, Icahn School of Medicine at Mount Sinai, New York, NY, USA
- Division of Nephrology, Icahn School of Medicine at Mount Sinai, New York, NY, USA
- Charles Bronfman Institute of Personalized Medicine, Icahn School of Medicine at Mount Sinai, New York, NY, USA
| | - Girish N Nadkarni
- The Mount Sinai Clinical Intelligence Center, Icahn School of Medicine at Mount Sinai, New York, NY, USA
- The Division of Data-Driven and Digital Medicine (D3M), Icahn School of Medicine at Mount Sinai, New York, NY, USA
| | - Azra Bihorac
- Department of Medicine, University of Florida Health, Gainesville, FL, USA.
| |
Collapse
|
|
31
|
Kalisnik JM, Bauer A, Vogt FA, Stickl FJ, Zibert J, Fittkau M, Bertsch T, Kounev S, Fischlein T. Artificial intelligence-based early detection of acute kidney injury after cardiac surgery. Eur J Cardiothorac Surg 2022; 62:6581706. [PMID: 35521994 DOI: 10.1093/ejcts/ezac289] [Citation(s) in RCA: 7] [Impact Index Per Article: 2.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 11/15/2021] [Revised: 04/14/2022] [Accepted: 05/03/2022] [Indexed: 11/14/2022] Open
Abstract
OBJECTIVES This study aims to improve early detection of cardiac surgery-associated acute kidney injury using artificial intelligence-based algorithms. METHODS Data from consecutive patients undergoing cardiac surgery between 2008 and 2018 in our institution served as the source for artificial intelligence-based modeling. Cardiac surgery-associated acute kidney injury was defined according to the Kidney Disease Improving Global Outcomes criteria. Different machine learning algorithms were trained and validated to detect cardiac surgery-associated acute kidney injury within 12 hours after surgery. Demographic characteristics, comorbidities, preoperative cardiac status, intra- and postoperative variables including creatinine and hemoglobin values were retrieved for analysis. RESULTS From 7507 patients analyzed, 1699 patients (22.6%) developed cardiac surgery-associated acute kidney injury. The ultimate detection model, 'Detect-A(K)I', recognizes cardiac surgery-associated acute kidney injury within 12 hours with an area under the curve of 88.0%, sensitivity of 78.0%, specificity of 78.9%, and accuracy of 82.1%. The optimal parameter set includes serial changes of creatinine and hemoglobin, operative emergency, bleeding-associated variables, cardiac ischaemic time and cardiac function-associated variables, age, diuretics and active infection, chronic obstructive lung and peripheral vascular disease. CONCLUSIONS The 'Detect-A(K)I' model successfully detects cardiac surgery-associated acute kidney injury within 12 hours after surgery with the best discriminatory characteristics reported so far.
Collapse
Affiliation(s)
- Jurij Matija Kalisnik
- Department of Cardiac Surgery, Klinikum Nuremberg, Paracelsus Medical University, Nuremberg, Germany.,Medical School, University of Ljubljana, Slovenia
| | - André Bauer
- Department of Computer Science, Julius Maximillian University of Wuerzburg, Germany
| | - Ferdinand Aurel Vogt
- Department of Cardiac Surgery, Klinikum Nuremberg, Paracelsus Medical University, Nuremberg, Germany.,Artemed Clinic Munich-South, Munich, Germany
| | | | - Janez Zibert
- Faculty of Health Sciences, University of Ljubljana, Ljubljana, Slovenia
| | - Matthias Fittkau
- Department of Cardiac Surgery, Klinikum Nuremberg, Paracelsus Medical University, Nuremberg, Germany
| | - Thomas Bertsch
- Institute of Clinical Chemistry, Laboratory Medicine and Transfusion Medicine, Paracelsus Medical University, Nuremberg, Germany
| | - Samuel Kounev
- Department of Computer Science, Julius Maximillian University of Wuerzburg, Germany
| | - Theodor Fischlein
- Department of Cardiac Surgery, Klinikum Nuremberg, Paracelsus Medical University, Nuremberg, Germany.,Paracelsus Medical University, Nuremberg, Germany
| |
Collapse
|
|
32
|
Ren Y, Loftus TJ, Datta S, Ruppert MM, Guan Z, Miao S, Shickel B, Feng Z, Giordano C, Upchurch GR, Rashidi P, Ozrazgat-Baslanti T, Bihorac A. Performance of a Machine Learning Algorithm Using Electronic Health Record Data to Predict Postoperative Complications and Report on a Mobile Platform. JAMA Netw Open 2022; 5:e2211973. [PMID: 35576007 PMCID: PMC9112066 DOI: 10.1001/jamanetworkopen.2022.11973] [Citation(s) in RCA: 47] [Impact Index Per Article: 15.7] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 01/22/2023] Open
Abstract
IMPORTANCE Predicting postoperative complications has the potential to inform shared decisions regarding the appropriateness of surgical procedures, targeted risk-reduction strategies, and postoperative resource use. Realizing these advantages requires that accurate real-time predictions be integrated with clinical and digital workflows; artificial intelligence predictive analytic platforms using automated electronic health record (EHR) data inputs offer an intriguing possibility for achieving this, but there is a lack of high-level evidence from prospective studies supporting their use. OBJECTIVE To examine whether the MySurgeryRisk artificial intelligence system has stable predictive performance between development and prospective validation phases and whether it is feasible to provide automated outputs directly to surgeons' mobile devices. DESIGN, SETTING, AND PARTICIPANTS In this prognostic study, the platform used automated EHR data inputs and machine learning algorithms to predict postoperative complications and provide predictions to surgeons, previously through a web portal and currently through a mobile device application. All patients 18 years or older who were admitted for any type of inpatient surgical procedure (74 417 total procedures involving 58 236 patients) between June 1, 2014, and September 20, 2020, were included. Models were developed using retrospective data from 52 117 inpatient surgical procedures performed between June 1, 2014, and November 27, 2018. Validation was performed using data from 22 300 inpatient surgical procedures collected prospectively from November 28, 2018, to September 20, 2020. MAIN OUTCOMES AND MEASURES Algorithms for generalized additive models and random forest models were developed and validated using real-time EHR data. Model predictive performance was evaluated primarily using area under the receiver operating characteristic curve (AUROC) values. RESULTS Among 58 236 total adult patients who received 74 417 major inpatient surgical procedures, the mean (SD) age was 57 (17) years; 29 226 patients (50.2%) were male. Results reported in this article focus primarily on the validation cohort. The validation cohort included 22 300 inpatient surgical procedures involving 19 132 patients (mean [SD] age, 58 [17] years; 9672 [50.6%] male). A total of 2765 patients (14.5%) were Black or African American, 14 777 (77.2%) were White, 1235 (6.5%) were of other races (including American Indian or Alaska Native, Asian, Native Hawaiian or Pacific Islander, and multiracial), and 355 (1.9%) were of unknown race because of missing data; 979 patients (5.1%) were Hispanic, 17 663 (92.3%) were non-Hispanic, and 490 (2.6%) were of unknown ethnicity because of missing data. A greater number of input features was associated with stable or improved model performance. For example, the random forest model trained with 135 input features had the highest AUROC values for predicting acute kidney injury (0.82; 95% CI, 0.82-0.83); cardiovascular complications (0.81; 95% CI, 0.81-0.82); neurological complications, including delirium (0.87; 95% CI, 0.87-0.88); prolonged intensive care unit stay (0.89; 95% CI, 0.88-0.89); prolonged mechanical ventilation (0.91; 95% CI, 0.90-0.91); sepsis (0.86; 95% CI, 0.85-0.87); venous thromboembolism (0.82; 95% CI, 0.81-0.83); wound complications (0.78; 95% CI, 0.78-0.79); 30-day mortality (0.84; 95% CI, 0.82-0.86); and 90-day mortality (0.84; 95% CI, 0.82-0.85), with accuracy similar to surgeons' predictions. Compared with the original web portal, the mobile device application allowed efficient fingerprint login access and loaded data approximately 10 times faster. The application output displayed patient information, risk of postoperative complications, top 3 risk factors for each complication, and patterns of complications for individual surgeons compared with their colleagues. CONCLUSIONS AND RELEVANCE In this study, automated real-time predictions of postoperative complications with mobile device outputs had good performance in clinical settings with prospective validation, matching surgeons' predictive accuracy.
Collapse
Affiliation(s)
- Yuanfang Ren
- Intelligent Critical Care Center, University of Florida, Gainesville
- Division of Nephrology, Hypertension, and Renal Transplantation, Department of Medicine, University of Florida, Gainesville
| | - Tyler J. Loftus
- Intelligent Critical Care Center, University of Florida, Gainesville
- Department of Surgery, University of Florida, Gainesville
| | - Shounak Datta
- Intelligent Critical Care Center, University of Florida, Gainesville
- Division of Nephrology, Hypertension, and Renal Transplantation, Department of Medicine, University of Florida, Gainesville
| | - Matthew M. Ruppert
- Intelligent Critical Care Center, University of Florida, Gainesville
- Division of Nephrology, Hypertension, and Renal Transplantation, Department of Medicine, University of Florida, Gainesville
| | - Ziyuan Guan
- Intelligent Critical Care Center, University of Florida, Gainesville
- Division of Nephrology, Hypertension, and Renal Transplantation, Department of Medicine, University of Florida, Gainesville
| | - Shunshun Miao
- Intelligent Critical Care Center, University of Florida, Gainesville
- Division of Nephrology, Hypertension, and Renal Transplantation, Department of Medicine, University of Florida, Gainesville
| | - Benjamin Shickel
- Intelligent Critical Care Center, University of Florida, Gainesville
- Division of Nephrology, Hypertension, and Renal Transplantation, Department of Medicine, University of Florida, Gainesville
| | - Zheng Feng
- Intelligent Critical Care Center, University of Florida, Gainesville
- Department of Electrical and Computer Engineering, University of Florida, Gainesville
| | - Chris Giordano
- Intelligent Critical Care Center, University of Florida, Gainesville
- Department of Anesthesiology, University of Florida, Gainesville
| | - Gilbert R. Upchurch
- Intelligent Critical Care Center, University of Florida, Gainesville
- Department of Surgery, University of Florida, Gainesville
| | - Parisa Rashidi
- Intelligent Critical Care Center, University of Florida, Gainesville
- Department of Biomedical Engineering, University of Florida, Gainesville
| | - Tezcan Ozrazgat-Baslanti
- Intelligent Critical Care Center, University of Florida, Gainesville
- Division of Nephrology, Hypertension, and Renal Transplantation, Department of Medicine, University of Florida, Gainesville
| | - Azra Bihorac
- Intelligent Critical Care Center, University of Florida, Gainesville
- Division of Nephrology, Hypertension, and Renal Transplantation, Department of Medicine, University of Florida, Gainesville
| |
Collapse
|
|
33
|
Busnatu Ș, Niculescu AG, Bolocan A, Petrescu GED, Păduraru DN, Năstasă I, Lupușoru M, Geantă M, Andronic O, Grumezescu AM, Martins H. Clinical Applications of Artificial Intelligence-An Updated Overview. J Clin Med 2022; 11:jcm11082265. [PMID: 35456357 PMCID: PMC9031863 DOI: 10.3390/jcm11082265] [Citation(s) in RCA: 56] [Impact Index Per Article: 18.7] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/10/2022] [Revised: 04/09/2022] [Accepted: 04/14/2022] [Indexed: 12/16/2022] Open
Abstract
Artificial intelligence has the potential to revolutionize modern society in all its aspects. Encouraged by the variety and vast amount of data that can be gathered from patients (e.g., medical images, text, and electronic health records), researchers have recently increased their interest in developing AI solutions for clinical care. Moreover, a diverse repertoire of methods can be chosen towards creating performant models for use in medical applications, ranging from disease prediction, diagnosis, and prognosis to opting for the most appropriate treatment for an individual patient. In this respect, the present paper aims to review the advancements reported at the convergence of AI and clinical care. Thus, this work presents AI clinical applications in a comprehensive manner, discussing the recent literature studies classified according to medical specialties. In addition, the challenges and limitations hindering AI integration in the clinical setting are further pointed out.
Collapse
Affiliation(s)
- Ștefan Busnatu
- “Carol Davila” University of Medicine and Pharmacy, 050474 Bucharest, Romania; (Ș.B.); (A.B.); (G.E.D.P.); (D.N.P.); (I.N.); (M.L.); (O.A.)
| | - Adelina-Gabriela Niculescu
- Department of Science and Engineering of Oxide Materials and Nanomaterials, Faculty of Applied Chemistry and Materials Science, Politehnica University of Bucharest, 011061 Bucharest, Romania;
| | - Alexandra Bolocan
- “Carol Davila” University of Medicine and Pharmacy, 050474 Bucharest, Romania; (Ș.B.); (A.B.); (G.E.D.P.); (D.N.P.); (I.N.); (M.L.); (O.A.)
| | - George E. D. Petrescu
- “Carol Davila” University of Medicine and Pharmacy, 050474 Bucharest, Romania; (Ș.B.); (A.B.); (G.E.D.P.); (D.N.P.); (I.N.); (M.L.); (O.A.)
| | - Dan Nicolae Păduraru
- “Carol Davila” University of Medicine and Pharmacy, 050474 Bucharest, Romania; (Ș.B.); (A.B.); (G.E.D.P.); (D.N.P.); (I.N.); (M.L.); (O.A.)
| | - Iulian Năstasă
- “Carol Davila” University of Medicine and Pharmacy, 050474 Bucharest, Romania; (Ș.B.); (A.B.); (G.E.D.P.); (D.N.P.); (I.N.); (M.L.); (O.A.)
| | - Mircea Lupușoru
- “Carol Davila” University of Medicine and Pharmacy, 050474 Bucharest, Romania; (Ș.B.); (A.B.); (G.E.D.P.); (D.N.P.); (I.N.); (M.L.); (O.A.)
| | - Marius Geantă
- Centre for Innovation in Medicine, “Carol Davila” University of Medicine and Pharmacy, 050474 Bucharest, Romania;
| | - Octavian Andronic
- “Carol Davila” University of Medicine and Pharmacy, 050474 Bucharest, Romania; (Ș.B.); (A.B.); (G.E.D.P.); (D.N.P.); (I.N.); (M.L.); (O.A.)
| | - Alexandru Mihai Grumezescu
- Department of Science and Engineering of Oxide Materials and Nanomaterials, Faculty of Applied Chemistry and Materials Science, Politehnica University of Bucharest, 011061 Bucharest, Romania;
- Research Institute of the University of Bucharest—ICUB, University of Bucharest, 050657 Bucharest, Romania
- Academy of Romanian Scientists, Ilfov No. 3, 50044 Bucharest, Romania
- Correspondence:
| | - Henrique Martins
- Faculty of Health Sciences, Universidade da Beira Interior, 6200-506 Covilha, Portugal;
| |
Collapse
|
|
34
|
Yoon HK, Yang HL, Jung CW, Lee HC. Artificial intelligence in perioperative medicine - a narrative review. Korean J Anesthesiol 2022; 75:202-215. [PMID: 35345305 PMCID: PMC9171545 DOI: 10.4097/kja.22157] [Citation(s) in RCA: 9] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/08/2022] [Accepted: 03/15/2022] [Indexed: 11/23/2022] Open
Abstract
Recent advancements in artificial intelligence (AI) techniques have enabled the development of accurate prediction models using clinical big data. AI models for perioperative risk stratification, intraoperative event prediction, biosignal analyses, and intensive care medicine have been developed in the field of perioperative medicine. Some of these models have been validated using external datasets and randomized controlled trials. Once these models are implemented in electronic health record systems or software medical devices, they could help anesthesiologists improve clinical outcomes by accurately predicting complications and suggesting optimal treatment strategies in real-time. This review provides an overview of the AI techniques used in perioperative medicine and a summary of the studies that have been published using these techniques. Understanding these techniques will aid in their appropriate application in clinical practice.
Collapse
Affiliation(s)
- Hyun-Kyu Yoon
- Department of Anesthesiology and Pain Medicine, Seoul National University Hospital, Seoul, Republic of Korea
| | - Hyun-Lim Yang
- Department of Anesthesiology and Pain Medicine, Seoul National University Hospital, Seoul, Republic of Korea.,Biomedical Research Institute, Seoul National University Hospital, Seoul, Republic of Korea
| | - Chul-Woo Jung
- Department of Anesthesiology and Pain Medicine, Seoul National University Hospital, Seoul, Republic of Korea.,Department of Anesthesiology and Pain Medicine, Seoul National University College of Medicine, Seoul, Republic of Korea
| | - Hyung-Chul Lee
- Department of Anesthesiology and Pain Medicine, Seoul National University Hospital, Seoul, Republic of Korea.,Department of Anesthesiology and Pain Medicine, Seoul National University College of Medicine, Seoul, Republic of Korea
| |
Collapse
|
|
35
|
Yang L, Gabriel N, Hernandez I, Vouri SM, Kimmel SE, Bian J, Guo J. Identifying Patients at Risk of Acute Kidney Injury Among Medicare Beneficiaries With Type 2 Diabetes Initiating SGLT2 Inhibitors: A Machine Learning Approach. Front Pharmacol 2022; 13:834743. [PMID: 35359843 PMCID: PMC8961669 DOI: 10.3389/fphar.2022.834743] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/13/2021] [Accepted: 01/20/2022] [Indexed: 11/13/2022] Open
Abstract
Introduction: To predict acute kidney injury (AKI) risk in patients with type 2 diabetes (T2D) prescribed sodium-glucose cotransporter two inhibitors (SGLT2i). Methods: Using a 5% random sample of Medicare claims data, we identified 17,694 patients who filled ≥1 prescriptions for canagliflozin, dapagliflozin and empagliflozin in 2013–2016. The cohort was split randomly and equally into training and testing sets. We measured 65 predictor candidates using claims data from the year prior to SGLT2i initiation. We then applied three machine learning models, including random forests (RF), elastic net and least absolute shrinkage and selection operator (LASSO) for risk prediction. Results: The incidence rate of AKI was 1.1% over a median 1.5 year follow up. Among three machine learning methods, RF produced the best prediction (C-statistic = 0.72), followed by LASSO and elastic net (both C-statistics = 0.69). Among individuals classified in the top 10% of the RF risk score (i.e., high risk group), the actual incidence rate of AKI was as high as 3.7%. In the logistic regression model including 14 important risk factors selected by LASSO, use of loop diuretics [adjusted odds ratio (95% confidence interval): 3.72 (2.44–5.76)] had the strongest association with AKI incidence. Disscusion: Our machine learning model efficiently identified patients at risk of AKI among Medicare beneficiaries with T2D undergoing SGLT2i treatment.
Collapse
Affiliation(s)
- Lanting Yang
- Department of Pharmacy and Therapeutics, University of Pittsburgh, Pittsburgh, PA, United States
| | - Nico Gabriel
- Division of Clinical Pharmacy, Skaggs School of Pharmacy and Pharmaceutical Sciences, University of California, San Diego, San Diego, CA, United States
| | - Inmaculada Hernandez
- Division of Clinical Pharmacy, Skaggs School of Pharmacy and Pharmaceutical Sciences, University of California, San Diego, San Diego, CA, United States
| | - Scott M. Vouri
- Department of Pharmaceutical Outcomes and Policy, University of Florida, Gainesville, FL, United States
| | - Stephen E. Kimmel
- Department of Epidemiology, University of Florida, Gainesville, FL, United States
| | - Jiang Bian
- Department of Health Outcomes and Biomedical Informatics, University of Florida, Gainesville, FL, United States
| | - Jingchuan Guo
- Department of Pharmaceutical Outcomes and Policy, University of Florida, Gainesville, FL, United States
- *Correspondence: Jingchuan Guo,
| |
Collapse
|
|
36
|
Xu J, Hu Y, Liu H, Mi W, Li G, Guo J, Feng Y. A Novel Multivariable Time Series Prediction Model for Acute Kidney Injury in General Hospitalization. Int J Med Inform 2022; 161:104729. [DOI: 10.1016/j.ijmedinf.2022.104729] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/02/2021] [Revised: 01/28/2022] [Accepted: 02/21/2022] [Indexed: 10/19/2022]
|
|
37
|
Loftus TJ, Tighe PJ, Ozrazgat-Baslanti T, Davis JP, Ruppert MM, Ren Y, Shickel B, Kamaleswaran R, Hogan WR, Moorman JR, Upchurch GR, Rashidi P, Bihorac A. Ideal algorithms in healthcare: Explainable, dynamic, precise, autonomous, fair, and reproducible. PLOS DIGITAL HEALTH 2022; 1:e0000006. [PMID: 36532301 PMCID: PMC9754299 DOI: 10.1371/journal.pdig.0000006] [Citation(s) in RCA: 21] [Impact Index Per Article: 7.0] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 12/22/2022]
Abstract
Established guidelines describe minimum requirements for reporting algorithms in healthcare; it is equally important to objectify the characteristics of ideal algorithms that confer maximum potential benefits to patients, clinicians, and investigators. We propose a framework for ideal algorithms, including 6 desiderata: explainable (convey the relative importance of features in determining outputs), dynamic (capture temporal changes in physiologic signals and clinical events), precise (use high-resolution, multimodal data and aptly complex architecture), autonomous (learn with minimal supervision and execute without human input), fair (evaluate and mitigate implicit bias and social inequity), and reproducible (validated externally and prospectively and shared with academic communities). We present an ideal algorithms checklist and apply it to highly cited algorithms. Strategies and tools such as the predictive, descriptive, relevant (PDR) framework, the Standard Protocol Items: Recommendations for Interventional Trials-Artificial Intelligence (SPIRIT-AI) extension, sparse regression methods, and minimizing concept drift can help healthcare algorithms achieve these objectives, toward ideal algorithms in healthcare.
Collapse
Affiliation(s)
- Tyler J. Loftus
- Department of Surgery, University of Florida Health, Gainesville, Florida, United States of America
- Precision and Intelligent Systems in Medicine (PrismaP), University of Florida, Gainesville, Florida, United States of America
| | - Patrick J. Tighe
- Departments of Anesthesiology, Orthopedics, and Information Systems/Operations Management, University of Florida Health, Gainesville, Florida, United States of America
| | - Tezcan Ozrazgat-Baslanti
- Precision and Intelligent Systems in Medicine (PrismaP), University of Florida, Gainesville, Florida, United States of America
- Department of Medicine, University of Florida Health, Gainesville, Florida, United States of America
| | - John P. Davis
- Department of Surgery, University of Virginia, Charlottesville, Virginia, United States of America
| | - Matthew M. Ruppert
- Precision and Intelligent Systems in Medicine (PrismaP), University of Florida, Gainesville, Florida, United States of America
- Department of Medicine, University of Florida Health, Gainesville, Florida, United States of America
| | - Yuanfang Ren
- Precision and Intelligent Systems in Medicine (PrismaP), University of Florida, Gainesville, Florida, United States of America
- Department of Medicine, University of Florida Health, Gainesville, Florida, United States of America
| | - Benjamin Shickel
- Precision and Intelligent Systems in Medicine (PrismaP), University of Florida, Gainesville, Florida, United States of America
- Department of Medicine, University of Florida Health, Gainesville, Florida, United States of America
| | - Rishikesan Kamaleswaran
- Department of Biomedical Informatics, Emory University School of Medicine, Atlanta, Georgia, United States of America
| | - William R. Hogan
- Department of Health Outcomes & Biomedical Informatics, College of Medicine, University of Florida, Gainesville, Florida, United States of America
| | - J. Randall Moorman
- Department of Medicine, University of Virginia, Charlottesville, Virginia, United States of America
| | - Gilbert R. Upchurch
- Department of Surgery, University of Florida Health, Gainesville, Florida, United States of America
| | - Parisa Rashidi
- Precision and Intelligent Systems in Medicine (PrismaP), University of Florida, Gainesville, Florida, United States of America
- Departments of Biomedical Engineering, Computer and Information Science and Engineering, and Electrical and Computer Engineering, University of Florida, Gainesville, Florida, United States of America
| | - Azra Bihorac
- Precision and Intelligent Systems in Medicine (PrismaP), University of Florida, Gainesville, Florida, United States of America
- Department of Medicine, University of Florida Health, Gainesville, Florida, United States of America
| |
Collapse
|
|
38
|
Legouis D, Criton G, Assouline B, Le Terrier C, Sgardello S, Pugin J, Marchi E, Sangla F. Unsupervised clustering reveals phenotypes of AKI in ICU COVID-19 patients. Front Med (Lausanne) 2022; 9:980160. [PMID: 36275817 PMCID: PMC9579431 DOI: 10.3389/fmed.2022.980160] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/28/2022] [Accepted: 09/20/2022] [Indexed: 11/13/2022] Open
Abstract
Background Acute Kidney Injury (AKI) is a very frequent condition, occurring in about one in three patients admitted to an intensive care unit (ICU). AKI is a syndrome defined as a sudden decrease in glomerular filtration rate. However, this unified definition does not reflect the various mechanisms involved in AKI pathophysiology, each with its own characteristics and sensitivity to therapy. In this study, we aimed at developing an innovative machine learning based method able to subphenotype AKI according to its pattern of risk factors. Methods We adopted a three-step pipeline of analyses. First, we looked for factors associated with AKI using a generalized additive model. Second, we calculated the importance of each identified AKI related factor in the estimated AKI risk to find the main risk factor for AKI, at the single patient level. Lastly, we clusterized AKI patients according to their profile of risk factors and compared the clinical characteristics and outcome of every cluster. We applied this method to a cohort of severe COVID-19 patients hospitalized in the ICU of the Geneva University Hospitals. Results Among the 248 patients analyzed, we found 7 factors associated with AKI development. Using the individual expression of these factors, we identified three groups of AKI patients, based on the use of Lopinavir/Ritonavir, baseline eGFR, use of dexamethasone and AKI severity. The three clusters expressed distinct characteristics in terms of AKI severity and recovery, metabolic patterns and hospital mortality. Conclusion We propose here a new method to phenotype AKI patients according to their most important individual risk factors for AKI development. When applied to an ICU cohort of COVID-19 patients, we were able to differentiate three groups of patients. Each expressed specific AKI characteristics and outcomes, which probably reflect a distinct pathophysiology.
Collapse
Affiliation(s)
- David Legouis
- Division of Intensive Care, Department of Acute Medicine, University Hospital of Geneva, Geneva, Switzerland
- Laboratory of Nephrology, Department of Medicine and Cell Physiology, University Hospital of Geneva, Geneva, Switzerland
- *Correspondence: David Legouis
| | - Gilles Criton
- Geneva School of Economics and Management, University of Geneva, Geneva, Switzerland
| | - Benjamin Assouline
- Division of Intensive Care, Department of Acute Medicine, University Hospital of Geneva, Geneva, Switzerland
| | - Christophe Le Terrier
- Division of Intensive Care, Department of Acute Medicine, University Hospital of Geneva, Geneva, Switzerland
| | - Sebastian Sgardello
- Department of Surgery, Center Hospitalier du Valais Romand, Sion, Switzerland
| | - Jérôme Pugin
- Division of Intensive Care, Department of Acute Medicine, University Hospital of Geneva, Geneva, Switzerland
| | - Elisa Marchi
- Division of Intensive Care, Department of Acute Medicine, University Hospital of Geneva, Geneva, Switzerland
| | - Frédéric Sangla
- Division of Intensive Care, Department of Acute Medicine, University Hospital of Geneva, Geneva, Switzerland
| |
Collapse
|
|
39
|
Uchino E, Sato N, Okuno Y. Artificial Intelligence in Predicting Kidney Function and Acute Kidney Injury. Artif Intell Med 2022. [DOI: 10.1007/978-3-030-64573-1_270] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/29/2022]
|
|
40
|
Does Artificial Intelligence Make Clinical Decision Better? A Review of Artificial Intelligence and Machine Learning in Acute Kidney Injury Prediction. Healthcare (Basel) 2021; 9:healthcare9121662. [PMID: 34946388 PMCID: PMC8701097 DOI: 10.3390/healthcare9121662] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/19/2021] [Revised: 11/19/2021] [Accepted: 11/26/2021] [Indexed: 02/06/2023] Open
Abstract
Acute kidney injury (AKI) is a common complication of hospitalization that greatly and negatively affects the short-term and long-term outcomes of patients. Current guidelines use serum creatinine level and urine output rate for defining AKI and as the staging criteria of AKI. However, because they are not sensitive or specific markers of AKI, clinicians find it difficult to predict the occurrence of AKI and prescribe timely treatment. Advances in computing technology have led to the recent use of machine learning and artificial intelligence in AKI prediction, recent research reported that by using electronic health records (EHR) the AKI prediction via machine-learning models can reach AUROC over 0.80, in some studies even reach 0.93. Our review begins with the background and history of the definition of AKI, and the evolution of AKI risk factors and prediction models is also appraised. Then, we summarize the current evidence regarding the application of e-alert systems and machine-learning models in AKI prediction.
Collapse
|
|
41
|
Soranno DE, Bihorac A, Goldstein SL, Kashani KB, Menon S, Nadkarni GN, Neyra JA, Pannu NI, Singh K, Cerda J, Koyner JL. Artificial Intelligence for AKI!Now: Let's Not Await Plato's Utopian Republic. KIDNEY360 2021; 3:376-381. [PMID: 35373136 PMCID: PMC8967630 DOI: 10.34067/kid.0003472021] [Citation(s) in RCA: 11] [Impact Index Per Article: 2.8] [Reference Citation Analysis] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Received: 06/18/2021] [Accepted: 11/17/2021] [Indexed: 01/10/2023]
Affiliation(s)
- Danielle E. Soranno
- Departments of Pediatrics, Bioengineering and Medicine, University of Colorado, Aurora, Colorado
| | - Azra Bihorac
- Department of Medicine and Precision and Intelligent Systems in Medicine (PrismaP), University of Florida, Gainesville, Florida
| | - Stuart L. Goldstein
- University of Cincinnati College of Medicine and Cincinnati Children’s Hospital, Cincinnati, Ohio
| | - Kianoush B. Kashani
- Division of Nephrology and Hypertension, Division of Pulmonary and Critical Care Medicine, Department of Medicine, Mayo Clinic, Rochester, Minnesota
| | - Shina Menon
- University of Washington and Seattle Children’s Hospital, Seattle, Washington
| | - Girish N. Nadkarni
- Division of Data Driven and Digital Medicine (D3M) and Division of Nephrology, Icahn School of Medicine at Mount Sinai, New York, New York
| | - Javier A. Neyra
- Division of Nephrology, Bone and Mineral Metabolism, University of Kentucky Medical Center, Lexington, Kentucky
| | | | - Karandeep Singh
- Department of Internal Medicine and School of Information, University of Michigan, Ann Arbor, Michigan
| | - Jorge Cerda
- Department of Medicine, Albany Medical Center, Albany, New York
| | - Jay L. Koyner
- Section of Nephrology, Department of Medicine, University of Chicago, Chicago, Illinois
| |
Collapse
|
|
42
|
Filiberto AC, Loftus TJ, Elder CT, Hensley S, Frantz A, Efron P, Ozrazgat-Baslanti T, Bihorac A, Upchurch GR, Cooper MA. Intraoperative hypotension and complications after vascular surgery: A scoping review. Surgery 2021; 170:311-317. [PMID: 33972092 PMCID: PMC8318382 DOI: 10.1016/j.surg.2021.03.054] [Citation(s) in RCA: 9] [Impact Index Per Article: 2.3] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/16/2021] [Revised: 03/24/2021] [Accepted: 03/25/2021] [Indexed: 11/30/2022]
Abstract
BACKGROUND Intraoperative hypotension during major surgery is associated with adverse health outcomes. This phenomenon represents a potentially important therapeutic target for vascular surgery patients, who may be uniquely vulnerable to intraoperative hypotension. This review summarizes current evidence regarding the impact of intraoperative hypotension on postoperative complications in patients undergoing vascular surgery, focusing on potentially modifiable procedure- and patient-specific risk factors. METHODS A scoping review of the literature from Embase, MEDLINE, and PubMed databases was conducted from inception to December 2019 to identify articles related to the effects of intraoperative hypotension on patients undergoing vascular surgery. RESULTS Ninety-two studies met screening criteria; 9 studies met quality and inclusion criteria. Among the 9 studies that defined intraoperative hypotension objectively, there were 9 different definitions. Accordingly, the reported incidence of intraoperative hypotension ranged from 8% to 88% (when defined as a fall in systolic blood pressure of >30 mm Hg or mean arterial pressure <65). The results demonstrated that intraoperative hypotension is an independent risk factor for longer hospital length of stay, myocardial injury, acute kidney injury, postoperative mechanical ventilation, and early mortality. Vascular surgery patients with comorbid conditions that confer increased vulnerability to hypoperfusion and ischemia appear to be susceptible to the adverse effects of intraoperative hypotension. CONCLUSION There is no validated, consensus definition of intraoperative hypotension or other hemodynamic parameters associated with increased risk for adverse outcomes. Despite these limitations, the weight of evidence suggests that intraoperative hypotension is common and associated with major postoperative complications in vascular surgery patients.
Collapse
Affiliation(s)
| | - Tyler J Loftus
- Department of Surgery, University of Florida, Gainesville, FL; Precision and Intelligent Systems in Medicine (PrismaP), University of Florida, Gainesville, FL
| | - Craig T Elder
- Department of Surgery, University of Florida, Gainesville, FL
| | - Sara Hensley
- Department of Surgery, University of Florida, Gainesville, FL
| | - Amanda Frantz
- Department of Anesthesia, University of Florida, Gainesville, FL
| | - Phillip Efron
- Department of Surgery, University of Florida, Gainesville, FL
| | - Tezcan Ozrazgat-Baslanti
- Precision and Intelligent Systems in Medicine (PrismaP), University of Florida, Gainesville, FL; Division of Nephrology, Hypertension and Renal Transplantation, Department of Medicine, University of Florida, Gainesville, FL
| | - Azra Bihorac
- Precision and Intelligent Systems in Medicine (PrismaP), University of Florida, Gainesville, FL; Division of Nephrology, Hypertension and Renal Transplantation, Department of Medicine, University of Florida, Gainesville, FL
| | | | - Michol A Cooper
- Department of Surgery, University of Florida, Gainesville, FL.
| |
Collapse
|
|
43
|
Filiberto AC, Ozrazgat-Baslanti T, Loftus TJ, Peng YC, Datta S, Efron P, Upchurch GR, Bihorac A, Cooper MA. Optimizing predictive strategies for acute kidney injury after major vascular surgery. Surgery 2021; 170:298-303. [PMID: 33648766 PMCID: PMC8276529 DOI: 10.1016/j.surg.2021.01.030] [Citation(s) in RCA: 9] [Impact Index Per Article: 2.3] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/14/2020] [Revised: 01/18/2021] [Accepted: 01/23/2021] [Indexed: 12/29/2022]
Abstract
BACKGROUND Postoperative acute kidney injury is common after major vascular surgery and is associated with increased morbidity, mortality, and cost. High-performance risk stratification using a machine learning model can inform strategies that mitigate harm and optimize resource use. It is hypothesized that incorporating intraoperative data would improve machine learning model accuracy, discrimination, and precision in predicting acute kidney injury among patients undergoing major vascular surgery. METHODS A single-center retrospective cohort of 1,531 adult patients who underwent nonemergency major vascular surgery, including open aortic, endovascular aortic, and lower extremity bypass procedures, was evaluated. The validated, automated MySurgeryRisk analytics platform used electronic health record data to forecast patient-level probabilistic risk scores for postoperative acute kidney injury using random forest models with preoperative data alone and perioperative data (preoperative plus intraoperative). The MySurgeryRisk predictions were compared with each other as well as with the American Society of Anesthesiologists physical status classification. RESULTS Machine learning models using perioperative data had greater accuracy, discrimination, and precision than models using either preoperative data alone or the American Society of Anesthesiologists physical status classification (accuracy: 0.70 vs 0.64 vs 0.62, area under the receiver operating characteristics curve: 0.77 vs 0.68 vs 0.61, area under the precision-recall curve: 0.70 vs 0.58 vs 0.48). CONCLUSION In predicting acute kidney injury after major vascular surgery, machine learning approaches that incorporate dynamic intraoperative data had greater accuracy, discrimination, and precision than models using either preoperative data alone or the American Society of Anesthesiologists physical status classification. Machine learning methods have the potential for real-time identification of high-risk patients who may benefit from personalized risk-reduction strategies.
Collapse
Affiliation(s)
| | - Tezcan Ozrazgat-Baslanti
- Department of Medicine, University of Florida, Gainesville, FL; Precision and Intelligent Systems in Medicine (PrismaP), University of Florida, Gainesville FL
| | - Tyler J Loftus
- Department of Surgery, University of Florida, Gainesville, FL; Precision and Intelligent Systems in Medicine (PrismaP), University of Florida, Gainesville FL
| | - Ying-Chih Peng
- Precision and Intelligent Systems in Medicine (PrismaP), University of Florida, Gainesville FL
| | - Shounak Datta
- Department of Medicine, University of Florida, Gainesville, FL; Precision and Intelligent Systems in Medicine (PrismaP), University of Florida, Gainesville FL
| | - Philip Efron
- Department of Surgery, University of Florida, Gainesville, FL; Department of Anesthesia, University of Florida, Gainesville, FL
| | | | - Azra Bihorac
- Department of Medicine, University of Florida, Gainesville, FL; Precision and Intelligent Systems in Medicine (PrismaP), University of Florida, Gainesville FL
| | - Michol A Cooper
- Department of Surgery, University of Florida, Gainesville, FL.
| |
Collapse
|
|
44
|
Solanki SL, Pandrowala S, Nayak A, Bhandare M, Ambulkar RP, Shrikhande SV. Artificial intelligence in perioperative management of major gastrointestinal surgeries. World J Gastroenterol 2021; 27:2758-2770. [PMID: 34135552 PMCID: PMC8173379 DOI: 10.3748/wjg.v27.i21.2758] [Citation(s) in RCA: 3] [Impact Index Per Article: 0.8] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Download PDF] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 01/16/2021] [Revised: 04/06/2021] [Accepted: 04/28/2021] [Indexed: 02/06/2023] Open
Abstract
Artificial intelligence (AI) demonstrated by machines is based on reinforcement learning and revolves around the usage of algorithms. The purpose of this review was to summarize concepts, the scope, applications, and limitations in major gastrointestinal surgery. This is a narrative review of the available literature on the key capabilities of AI to help anesthesiologists, surgeons, and other physicians to understand and critically evaluate ongoing and new AI applications in perioperative management. AI uses available databases called "big data" to formulate an algorithm. Analysis of other data based on these algorithms can help in early diagnosis, accurate risk assessment, intraoperative management, automated drug delivery, predicting anesthesia and surgical complications and postoperative outcomes and can thus lead to effective perioperative management as well as to reduce the cost of treatment. Perioperative physicians, anesthesiologists, and surgeons are well-positioned to help integrate AI into modern surgical practice. We all need to partner and collaborate with data scientists to collect and analyze data across all phases of perioperative care to provide clinical scenarios and context. Careful implementation and use of AI along with real-time human interpretation will revolutionize perioperative care, and is the way forward in future perioperative management of major surgery.
Collapse
Affiliation(s)
- Sohan Lal Solanki
- Department of Anesthesiology, Critical Care and Pain, Tata Memorial Hospital, Homi Bhabha National Institute, Mumbai 400012, Maharashtra, India
| | - Saneya Pandrowala
- Gastro-Intestinal Services, Department of Surgical Oncology, Tata Memorial Hospital, Homi Bhabha National Institute, Mumbai 400012, Maharashtra, India
| | - Abhirup Nayak
- Department of Anesthesiology, Critical Care and Pain, Tata Memorial Hospital, Homi Bhabha National Institute, Mumbai 400012, Maharashtra, India
| | - Manish Bhandare
- Gastro-Intestinal Services, Department of Surgical Oncology, Tata Memorial Hospital, Homi Bhabha National Institute, Mumbai 400012, Maharashtra, India
| | - Reshma P Ambulkar
- Department of Anesthesiology, Critical Care and Pain, Tata Memorial Hospital, Homi Bhabha National Institute, Mumbai 400012, Maharashtra, India
| | - Shailesh V Shrikhande
- Gastro-Intestinal Services, Department of Surgical Oncology, Tata Memorial Hospital, Homi Bhabha National Institute, Mumbai 400012, Maharashtra, India
| |
Collapse
|
|
45
|
He ZL, Zhou JB, Liu ZK, Dong SY, Zhang YT, Shen T, Zheng SS, Xu X. Application of machine learning models for predicting acute kidney injury following donation after cardiac death liver transplantation. Hepatobiliary Pancreat Dis Int 2021; 20:222-231. [PMID: 33726966 DOI: 10.1016/j.hbpd.2021.02.001] [Citation(s) in RCA: 15] [Impact Index Per Article: 3.8] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 07/27/2020] [Accepted: 02/02/2021] [Indexed: 02/05/2023]
Abstract
BACKGROUND Acute kidney injury (AKI) is a common complication after liver transplantation (LT) and is an indicator of poor prognosis. The establishment of a more accurate preoperative prediction model of AKI could help to improve the prognosis of LT. Machine learning algorithms provide a potentially effective approach. METHODS A total of 493 patients with donation after cardiac death LT (DCDLT) were enrolled. AKI was defined according to the clinical practice guidelines of kidney disease: improving global outcomes (KDIGO). The clinical data of patients with AKI (AKI group) and without AKI (non-AKI group) were compared. With logistic regression analysis as a conventional model, four predictive machine learning models were developed using the following algorithms: random forest, support vector machine, classical decision tree, and conditional inference tree. The predictive power of these models was then evaluated using the area under the receiver operating characteristic curve (AUC). RESULTS The incidence of AKI was 35.7% (176/493) during the follow-up period. Compared with the non-AKI group, the AKI group showed a remarkably lower survival rate (P < 0.001). The random forest model demonstrated the highest prediction accuracy of 0.79 with AUC of 0.850 [95% confidence interval (CI): 0.794-0.905], which was significantly higher than the AUCs of the other machine learning algorithms and logistic regression models (P < 0.001). CONCLUSIONS The random forest model based on machine learning algorithms for predicting AKI occurring after DCDLT demonstrated stronger predictive power than other models in our study. This suggests that machine learning methods may provide feasible tools for forecasting AKI after DCDLT.
Collapse
Affiliation(s)
- Zeng-Lei He
- Division of Hepatobiliary and Pancreatic Surgery, the First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310003, China; Key Laboratory of Combined Multi-organ Transplantation, Ministry of Public Health, the First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310003, China
| | - Jun-Bin Zhou
- Division of Hepatobiliary and Pancreatic Surgery, the First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310003, China; Key Laboratory of Combined Multi-organ Transplantation, Ministry of Public Health, the First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310003, China
| | - Zhi-Kun Liu
- Division of Hepatobiliary and Pancreatic Surgery, the First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310003, China; Key Laboratory of Combined Multi-organ Transplantation, Ministry of Public Health, the First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310003, China
| | - Si-Yi Dong
- Division of Hepatobiliary and Pancreatic Surgery, the First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310003, China; Key Laboratory of Combined Multi-organ Transplantation, Ministry of Public Health, the First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310003, China
| | - Yun-Tao Zhang
- Division of Hepatobiliary and Pancreatic Surgery, the First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310003, China; Key Laboratory of Combined Multi-organ Transplantation, Ministry of Public Health, the First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310003, China
| | - Tian Shen
- Division of Hepatobiliary and Pancreatic Surgery, the First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310003, China; Key Laboratory of Combined Multi-organ Transplantation, Ministry of Public Health, the First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310003, China
| | - Shu-Sen Zheng
- Division of Hepatobiliary and Pancreatic Surgery, the First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310003, China; Key Laboratory of Combined Multi-organ Transplantation, Ministry of Public Health, the First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310003, China
| | - Xiao Xu
- Division of Hepatobiliary and Pancreatic Surgery, the First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310003, China; Key Laboratory of Combined Multi-organ Transplantation, Ministry of Public Health, the First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310003, China.
| |
Collapse
|
|
46
|
Song X, Liu X, Liu F, Wang C. Comparison of machine learning and logistic regression models in predicting acute kidney injury: A systematic review and meta-analysis. Int J Med Inform 2021; 151:104484. [PMID: 33991886 DOI: 10.1016/j.ijmedinf.2021.104484] [Citation(s) in RCA: 105] [Impact Index Per Article: 26.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/20/2021] [Revised: 04/10/2021] [Accepted: 05/06/2021] [Indexed: 12/16/2022]
Abstract
INTRODUCTION We aimed to assess whether machine learning models are superior at predicting acute kidney injury (AKI) compared to logistic regression (LR), a conventional prediction model. METHODS Eligible studies were identified using PubMed and Embase. A total of 24 studies consisting of 84 prediction models met inclusion criteria. Independent samples t-test was performed to detect mean differences in area under the curve (AUC) between ML and LR models. One-way ANOVA and post-hoc t-tests were performed to assess mean differences in AUC between ML methods. RESULTS AUC data were similar between ML (0.736 ± 0.116) and LR (0.748 ± 0.057) models (p = 0.538). However, specific ML models, such as gradient boosting (0.838 ± 0.077), exhibited superior performance at predicting AKI as compared to other ML models in the literature (p < 0.05). Creatinine and urine output, standard variables assessed for AKI staging, were classified as significant predictors across multiple ML models, although the majority of significant predictors were unique and study specific. CONCLUSIONS These data suggest that ML models perform equally to that of LR, however ML models exhibit variable performance with some ML models displaying exceptional performance. The variability in ML prediction of AKI can be attributed, in part, to the specific ML model utilized, variable selection and processing, study and subject characteristics, and the steps associated with model training, validation, testing, and calibration.
Collapse
Affiliation(s)
- Xuan Song
- ICU, DongE Hospital Affiliated to Shandong First Medical University, Shandong, China
| | - Xinyan Liu
- ICU, DongE Hospital Affiliated to Shandong First Medical University, Shandong, China
| | - Fei Liu
- Urology Department, Tai'an Traditional Chinese Medicine Hospital Affiliated to Shandong University of Traditional Chinese Medicine, Shandong, China
| | - Chunting Wang
- ICU, Shandong Provincial Hospital Affiliated to Shandong First Medical University, Shandong, China.
| |
Collapse
|
|
47
|
Kim M, Li G, Mohan S, Turnbull ZA, Kiran RP, Li G. Intraoperative Data Enhance the Detection of High-Risk Acute Kidney Injury Patients When Added to a Baseline Prediction Model. Anesth Analg 2021; 132:430-441. [PMID: 32769380 DOI: 10.1213/ane.0000000000005057] [Citation(s) in RCA: 12] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/05/2022]
Abstract
BACKGROUND Aspects of intraoperative management (eg, hypotension) are associated with acute kidney injury (AKI) in noncardiac surgery patients. However, it is unclear if and how the addition of intraoperative data affects a baseline risk prediction model for postoperative AKI. METHODS With institutional review board (IRB) approval, an institutional cohort (2005-2015) of inpatient intra-abdominal surgery patients without preoperative AKI was identified. Data from the American College of Surgeons National Surgical Quality Improvement Program (preoperative and procedure data), Anesthesia Information Management System (intraoperative data), and electronic health record (postoperative laboratory data) were linked. The sample was split into derivation/validation (70%/30%) cohorts. AKI was defined as an increase in serum creatinine ≥0.3 mg/dL within 48 hours or >50% within 7 days of surgery. Forward logistic regression fit a baseline model incorporating preoperative variables and surgical procedure. Forward logistic regression fit a second model incorporating the previously selected baseline variables, as well as additional intraoperative variables. Intraoperative variables reflected the following aspects of intraoperative management: anesthetics, beta-blockers, blood pressure, diuretics, fluids, operative time, opioids, and vasopressors. The baseline and intraoperative models were evaluated based on statistical significance and discriminative ability (c-statistic). The risk threshold equalizing sensitivity and specificity in the intraoperative model was identified. RESULTS Of 2691 patients in the derivation cohort, 234 (8.7%) developed AKI. The baseline model had c-statistic 0.77 (95% confidence interval [CI], 0.74-0.80). The additional variables added to the intraoperative model were significantly associated with AKI (P < .0001) and the intraoperative model had c-statistic 0.81 (95% CI, 0.78-0.83). Sensitivity and specificity were equalized at a risk threshold of 9.0% in the intraoperative model. At this threshold, the baseline model had sensitivity and specificity of 71% (95% CI, 65-76) and 69% (95% CI, 67-70), respectively, and the intraoperative model had sensitivity and specificity of 74% (95% CI, 69-80) and 74% (95% CI, 73-76), respectively. The high-risk group had an AKI risk of 18% (95% CI, 15-20) in the baseline model and 22% (95% CI, 19-25) in the intraoperative model. CONCLUSIONS Intraoperative data, when added to a baseline risk prediction model for postoperative AKI in intra-abdominal surgery patients, improves the performance of the model.
Collapse
Affiliation(s)
- Minjae Kim
- From the Department of Anesthesiology, Columbia University Medical Center, New York, New York.,Department of Epidemiology
| | - Gen Li
- Department of Biostatistics, Mailman School of Public Health, Columbia University, New York, New York
| | - Sumit Mohan
- Department of Epidemiology.,Division of Nephrology, Department of Medicine, Columbia University Medical Center, New York, New York
| | - Zachary A Turnbull
- Department of Anesthesiology, Weill Cornell Medicine, New York, New York
| | - Ravi P Kiran
- Department of Epidemiology.,Division of Colorectal Surgery, Department of Surgery, Columbia University Medical Center, New York, New York
| | - Guohua Li
- From the Department of Anesthesiology, Columbia University Medical Center, New York, New York.,Department of Epidemiology
| |
Collapse
|
|
48
|
Artificial intelligence to guide management of acute kidney injury in the ICU: a narrative review. Curr Opin Crit Care 2021; 26:563-573. [PMID: 33027147 DOI: 10.1097/mcc.0000000000000775] [Citation(s) in RCA: 12] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/31/2022]
Abstract
PURPOSE OF REVIEW Acute kidney injury (AKI) frequently complicates hospital admission, especially in the ICU or after major surgery, and is associated with high morbidity and mortality. The risk of developing AKI depends on the presence of preexisting comorbidities and the cause of the current disease. Besides, many other parameters affect the kidney function, such as the state of other vital organs, the host response, and the initiated treatment. Advancements in the field of informatics have led to the opportunity to store and utilize the patient-related data to train and validate models to detect specific patterns and, as such, predict disease states or outcomes. RECENT FINDINGS Machine-learning techniques have also been applied to predict AKI, as well as the patients' outcomes related to their AKI, such as mortality or the need for kidney replacement therapy. Several models have recently been developed, but only a few of them have been validated in external cohorts. SUMMARY In this article, we provide an overview of the machine-learning prediction models for AKI and its outcomes in critically ill patients and individuals undergoing major surgery. We also discuss the pitfalls and the opportunities related to the implementation of these models in clinical practices.
Collapse
|
|
49
|
Artificial Intelligence in Predicting Kidney Function and Acute Kidney Injury. Artif Intell Med 2021. [DOI: 10.1007/978-3-030-58080-3_270-1] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/26/2022]
|
|
50
|
Yao L, Zhang H, Zhang M, Chen X, Zhang J, Huang J, Zhang L. Application of artificial intelligence in renal disease. CLINICAL EHEALTH 2021. [DOI: 10.1016/j.ceh.2021.11.003] [Citation(s) in RCA: 3] [Impact Index Per Article: 0.8] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/22/2022] Open
|