|
1
|
Kurtcehajic A, Zerem E, Bokun T, Alibegovic E, Kunosic S, Hujdurovic A, Tursunovic A, Ljuca K. Could near focus endoscopy, narrow-band imaging, and acetic acid improve the visualization of microscopic features of stomach mucosa? World J Gastrointest Endosc 2024; 16:157-167. [PMID: 38577642 PMCID: PMC10989255 DOI: 10.4253/wjge.v16.i3.157] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 10/18/2023] [Revised: 01/07/2024] [Accepted: 02/18/2024] [Indexed: 03/14/2024] Open
Abstract
BACKGROUND Conventional magnifying endoscopy with narrow-band imaging (NBI) observation of the gastric body mucosa shows dominant patterns in relation to the regular arrangement of collecting venules, subepithelial capillary network, and gastric pits. AIM To evaluate the effectiveness of a new one-dual (near) focus, NBI mode in the assessment of the microscopic features of gastric body mucosa compared to conventional magnification. METHODS During 2021 and 2022, 68 patients underwent proximal gastrointestinal endoscopy using magnification endoscopic modalities subsequently applying acetic acid (AA). The GIF-190HQ series NBI system with dual focus capability was used for the investigation of gastric mucosa. At the time of the endoscopy, the gastric body mucosa of all enrolled patients was photographed using the white light endoscopy (WLE), near focus (NF), NF-NBI, AA-NF, and AA-NF-NBI modes. RESULTS The WLE, NF and NF-NBI endoscopic modes for all patients (204 images) were classified in the same order into three groups. Two images from each patient for the AA-NF and AA-NF-NBI endoscopic modes were classified in the same order. According to all three observers who completed the work independently, NF magnification was significantly superior to WLE (P < 0.01), and the NF-NBI mode was significantly superior to NF magnification (P < 0.01). After applying AA, the three observers confirmed that AA-NF-NBI was significantly superior to AA-NF (P < 0.01). Interobserver kappa values for WLE were 0.609, 0.704, and 0.598, respectively and were 0.600, 0.721, and 0.637, respectively, for NF magnification. For the NF-NBI mode, the values were 0.378, 0.471, and 0.553, respectively. For AA-NF, they were 0.453, 0.603, and 0.480, respectively, and for AA-NF-NBI, they were 0.643, 0.506, and 0.354, respectively. CONCLUSION When investigating gastric mucosa in microscopic detail, NF-NBI was the most powerful endoscopic mode for assessing regular arrangement of collecting venules, subepithelial capillary network, and gastric pits among the five endoscopic modalities investigated in this study. AA-NF-NBI was the most powerful endoscopic mode for analyzing crypt opening and intervening part.
Collapse
Affiliation(s)
- Admir Kurtcehajic
- Department of Gastroenterology and Hepatology, Blue Medical Group, Tuzla 75000, Tuzla Kanton, Bosnia and Herzegovina
| | - Enver Zerem
- Department of Medical Sciences, The Academy of Sciences and Arts of Bosnia and Herzegovina, Sarajevo 71000, Bosnia and Herzegovina
| | - Tomislav Bokun
- Department of Gastroenterology and Hepatology, University Clinical Hospital Dubrava, Zagreb 10000, Croatia
| | - Ervin Alibegovic
- Department of Gastroenterology and Hepatology, University Clinical Center Tuzla, Tuzla 75000, Tuzla Kanton, Bosnia and Herzegovina
| | - Suad Kunosic
- Department of Physics, Faculty of Natural Sciences and Mathematics, University of Tuzla, Tuzla 75000, Tuzla Kanton, Bosnia and Herzegovina
| | - Ahmed Hujdurovic
- Department of Internal Medicine, Blue Medical Group, Tuzla 75000, Tuzla Kanton, Bosnia and Herzegovina
| | - Amir Tursunovic
- Department of Surgery, University Clinical Center Tuzla, Tuzla 75000, Bosnia and Herzegovina
| | - Kenana Ljuca
- School of Medicine, University of Tuzla, Tuzla 75000, Bosnia and Herzegovina
| |
Collapse
|
|
2
|
Zhang JQ, Mi JJ, Wang R. Application of convolutional neural network-based endoscopic imaging in esophageal cancer or high-grade dysplasia: A systematic review and meta-analysis. World J Gastrointest Oncol 2023; 15:1998-2016. [DOI: 10.4251/wjgo.v15.i11.1998] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 08/01/2023] [Revised: 09/05/2023] [Accepted: 10/11/2023] [Indexed: 11/15/2023] Open
Abstract
BACKGROUND Esophageal cancer is the seventh-most common cancer type worldwide, accounting for 5% of death from malignancy. Development of novel diagnostic techniques has facilitated screening, early detection, and improved prognosis. Convolutional neural network (CNN)-based image analysis promises great potential for diagnosing and determining the prognosis of esophageal cancer, enabling even early detection of dysplasia.
AIM To conduct a meta-analysis of the diagnostic accuracy of CNN models for the diagnosis of esophageal cancer and high-grade dysplasia (HGD).
METHODS PubMed, EMBASE, Web of Science and Cochrane Library databases were searched for articles published up to November 30, 2022. We evaluated the diagnostic accuracy of using the CNN model with still image-based analysis and with video-based analysis for esophageal cancer or HGD, as well as for the invasion depth of esophageal cancer. The pooled sensitivity, pooled specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR), diagnostic odds ratio (DOR) and area under the curve (AUC) were estimated, together with the 95% confidence intervals (CI). A bivariate method and hierarchical summary receiver operating characteristic method were used to calculate the diagnostic test accuracy of the CNN model. Meta-regression and subgroup analyses were used to identify sources of heterogeneity.
RESULTS A total of 28 studies were included in this systematic review and meta-analysis. Using still image-based analysis for the diagnosis of esophageal cancer or HGD provided a pooled sensitivity of 0.95 (95%CI: 0.92-0.97), pooled specificity of 0.92 (0.89-0.94), PLR of 11.5 (8.3-16.0), NLR of 0.06 (0.04-0.09), DOR of 205 (115-365), and AUC of 0.98 (0.96-0.99). When video-based analysis was used, a pooled sensitivity of 0.85 (0.77-0.91), pooled specificity of 0.73 (0.59-0.83), PLR of 3.1 (1.9-5.0), NLR of 0.20 (0.12-0.34), DOR of 15 (6-38) and AUC of 0.87 (0.84-0.90) were found. Prediction of invasion depth resulted in a pooled sensitivity of 0.90 (0.87-0.92), pooled specificity of 0.83 (95%CI: 0.76-0.88), PLR of 7.8 (1.9-32.0), NLR of 0.10 (0.41-0.25), DOR of 118 (11-1305), and AUC of 0.95 (0.92-0.96).
CONCLUSION CNN-based image analysis in diagnosing esophageal cancer and HGD is an excellent diagnostic method with high sensitivity and specificity that merits further investigation in large, multicenter clinical trials.
Collapse
Affiliation(s)
- Jun-Qi Zhang
- The Fifth Clinical Medical College, Shanxi Medical University, Taiyuan 030001, Shanxi Province, China
| | - Jun-Jie Mi
- Department of Gastroenterology, Shanxi Provincial People’s Hospital, Taiyuan 030012, Shanxi Province, China
| | - Rong Wang
- Department of Gastroenterology, The Fifth Hospital of Shanxi Medical University (Shanxi Provincial People’s Hospital), Taiyuan 030012, Shanxi Province, China
| |
Collapse
|
|
3
|
Miura H, Tanaka K, Umeda Y, Ikenoyama Y, Yukimoto H, Hamada Y, Yamada R, Tsuboi J, Nakamura M, Katsurahara M, Horiki N, Nakagawa H. Usefulness of magnifying endoscopy with acetic acid and narrow-band imaging for the diagnosis of duodenal neoplasms: proposal of a diagnostic algorithm. Surg Endosc 2022; 36:8086-8095. [PMID: 35449476 DOI: 10.1007/s00464-022-09239-1] [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: 08/12/2021] [Accepted: 04/02/2022] [Indexed: 02/07/2023]
Abstract
BACKGROUND This study aimed to clarify the features of superficial non-ampullary duodenal epithelial tumors (SNADETs) on magnifying endoscopy with narrow-band imaging (M-NBI) and magnifying endoscopy with acetic acid and narrow-band imaging (M-AANBI), and evaluate the efficacy of M-NBI/M-AANBI to distinguish high-grade adenomas or adenocarcinomas (HGA/AC) from low-grade adenomas (LGA). METHODS Clinicopathological data on 62 SNADETs in 58 patients who underwent preoperative M-NBI/M-AANBI and endoscopic resection were retrospectively reviewed. The pathological results were classified into two categories, LGA and HGA/AC. We evaluated microvascular patterns (MVPs) and microsurface patterns (MSPs) observed by M-NBI and MSPs observed by M-AANBI for characterizing LGA and HGA/AC. The kappa value was calculated to assess the interobserver and intraobserver agreements of evaluation of M-AANBI images. RESULTS Pathologically, 38 lesions (61.3%) were LGA and 24 lesions (38.7%) were HGA/AC. HGA/AC tended to have irregular MVP and/or MSP on M-NBI. M-NBI diagnostic performance to distinguish HGA/AC from LGA showed 62.5% sensitivity, 68.4% specificity, and 66.1% accuracy. SNADETs had irregular MSP on M-AANBI. Three irregularity grades (iG) of MSP were observed by M-AANBI as follows: iG1, mild; iG2, moderate; iG3, significant. HGA/AC lesions had a significantly higher rate of iG3 than LGA lesions (p < 0.001). The iG2 was associated with HGA/AC in elevated lesions and LGA in depressed lesions. The diagnostic performance of M-AANBI was as follows: 95.8% sensitivity, 97.4% specificity, and 96.8% accuracy. The diagnostic accuracy of M-AANBI was significantly higher than that of M-NBI (p < 0.001). The kappa value for interobserver agreement on the diagnosis and irregularity grading of M-AANBI images was 0.742 and 0.719, respectively. These data indicate substantial interobserver agreement. Based on the above-mentioned results, we developed a M-AANBI diagnostic algorithm for SNADETs. CONCLUSION The diagnostic algorithm for SNADETs using M-AANBI may be useful for differentiating between LGA and HGA/AC.
Collapse
Affiliation(s)
- Hiroshi Miura
- Department of Endoscopy, Mie University Hospital, 2-174 Edobashi, Tsu, Mie, 514-8507, Japan
- Department of Gastroenterology and Hepatology, Mie University Graduate School of Medicine, Tsu, Japan
| | - Kyosuke Tanaka
- Department of Endoscopy, Mie University Hospital, 2-174 Edobashi, Tsu, Mie, 514-8507, Japan.
- Department of Gastroenterology and Hepatology, Mie University Graduate School of Medicine, Tsu, Japan.
| | - Yuhei Umeda
- Department of Gastroenterology and Hepatology, Mie University Graduate School of Medicine, Tsu, Japan
| | - Yohei Ikenoyama
- Department of Gastroenterology and Hepatology, Mie University Graduate School of Medicine, Tsu, Japan
| | - Hiroki Yukimoto
- Department of Gastroenterology and Hepatology, Mie University Graduate School of Medicine, Tsu, Japan
| | - Yasuhiko Hamada
- Department of Gastroenterology and Hepatology, Mie University Graduate School of Medicine, Tsu, Japan
| | - Reiko Yamada
- Department of Gastroenterology and Hepatology, Mie University Graduate School of Medicine, Tsu, Japan
| | - Junya Tsuboi
- Department of Endoscopy, Mie University Hospital, 2-174 Edobashi, Tsu, Mie, 514-8507, Japan
| | - Misaki Nakamura
- Department of Gastroenterology and Hepatology, Mie University Graduate School of Medicine, Tsu, Japan
| | - Masaki Katsurahara
- Department of Endoscopy, Mie University Hospital, 2-174 Edobashi, Tsu, Mie, 514-8507, Japan
| | - Noriyuki Horiki
- Department of Gastroenterology and Hepatology, Mie University Graduate School of Medicine, Tsu, Japan
| | - Hayato Nakagawa
- Department of Endoscopy, Mie University Hospital, 2-174 Edobashi, Tsu, Mie, 514-8507, Japan
- Department of Gastroenterology and Hepatology, Mie University Graduate School of Medicine, Tsu, Japan
| |
Collapse
|
|
4
|
Shibagaki K, Itawaki A, Miyaoka Y, Kishimoto K, Takahashi Y, Kotani S, Mishiro T, Oshima N, Kawashima K, Ishimura N, Onuma H, Nagasaki M, Nagase M, Araki A, Kadota K, Kushima R, Ishihara S. Intestinal-type gastric dysplasia in Helicobacter pylori-naïve patients. Virchows Arch 2022; 480:783-792. [PMID: 34787713 DOI: 10.1007/s00428-021-03237-9] [Citation(s) in RCA: 5] [Impact Index Per Article: 1.7] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/23/2021] [Revised: 10/27/2021] [Accepted: 11/09/2021] [Indexed: 01/05/2023]
Abstract
Gastric dysplasia and gastric cancer in Helicobacter pylori (Hp)-naïve patients usually exhibit a gastric phenotype, reflecting gastric mucosa without intestinal metaplasia (IM). We showed that intestinal-type gastric dysplasia (IGD) rarely occurs in the Hp-naïve stomach. In the last 10 years, we treated 1760 gastric dysplasia and gastric cancer patients, with 3.6% (63/1760) being Hp-naïve. Among these, ten were diagnosed with 14 IGDs and enrolled in this retrospective analysis. All lesions were observed by white-light endoscopy (WLE) and narrow-band imaging with magnification endoscopy (NBIME). We analyzed their endoscopic and microscopic features and patient demographics. Five men and five women aged 64 ± 21 years were included. WLE showed the depressed lesions mimicking a benign raised erosion in the prepyloric compartment. Multiple growths were confirmed in 30% (3/10) of patients. NBIME showed a near-regular microstructure and capillaries in 50% (7/14) of lesions with a gastritis-like appearance. Histologically, background mucosa was non-atrophic pyloric gland tissue, but 40.0% of samples (4/10) contained sporadic IM. Most of the lesions (8/14) were low-grade dysplasia, and others had a high-grade component, with one progressing to intramucosal carcinoma. The neoplastic surface was widely covered with foveolar epithelium in 57.1% (8/14). Immunohistochemically, neoplastic cells expressed CDX2 in all patients (14/14), MUC2 and CD10 in 92.9% (13/14), MUC5AC in 14% (2/14), and no expression of MUC6, showing an intestinal phenotype. Ki-67 was overexpressed with a mean labeling index of 58.3 ± 38.5%, and p-53 was overexpressed in 92.9% (13/14), regardless of the dysplastic grade. The IGD rarely occurs in Hp-naïve patients with distinctive clinicopathologic characteristics.
Collapse
Affiliation(s)
- Kotaro Shibagaki
- Department of Endoscopy, Shimane University Hospital, Zip code 693-8501, 89-1 Enya, Izumo, Japan.
| | - Ayako Itawaki
- Department of Gastroenterology, Faculty of Medicine, Shimane University, Izumo, Japan
| | - Yoichi Miyaoka
- Department of Gastroenterology, Shimane Prefectural Central Hospital, Izumo, Japan
| | - Kenichi Kishimoto
- Department of Gastroenterology, Faculty of Medicine, Shimane University, Izumo, Japan
| | - Yusuke Takahashi
- Department of Gastroenterology, Faculty of Medicine, Shimane University, Izumo, Japan
| | - Satoshi Kotani
- Department of Gastroenterology, Faculty of Medicine, Shimane University, Izumo, Japan
| | - Tsuyoshi Mishiro
- Department of Gastroenterology, Faculty of Medicine, Shimane University, Izumo, Japan
| | - Naoki Oshima
- Department of Gastroenterology, Faculty of Medicine, Shimane University, Izumo, Japan
| | - Kousaku Kawashima
- Department of Gastroenterology, Faculty of Medicine, Shimane University, Izumo, Japan
| | - Norihisa Ishimura
- Department of Gastroenterology, Faculty of Medicine, Shimane University, Izumo, Japan
| | - Hideyuki Onuma
- Department of Pathology, Shimane Prefectural Central Hospital, Izumo, Japan
| | - Makoto Nagasaki
- Department of Pathology, National Hospital Organization Hamada Medical Center, Hamada, Japan
| | - Mamiko Nagase
- Department of Pathology, Faculty of Medicine, Shimane University, Izumo, Japan
| | - Asuka Araki
- Department of Pathology, Faculty of Medicine, Shimane University, Izumo, Japan
| | - Kyuichi Kadota
- Department of Pathology, Faculty of Medicine, Shimane University, Izumo, Japan
| | - Ryoji Kushima
- Department of Pathology, Shiga University of Medical Science, Otsu, Japan
| | - Shunji Ishihara
- Department of Gastroenterology, Faculty of Medicine, Shimane University, Izumo, Japan
| |
Collapse
|
|
5
|
Shibagaki K, Mishiro T, Fukuyama C, Takahashi Y, Itawaki A, Nonomura S, Yamashita N, Kotani S, Mikami H, Izumi D, Kawashima K, Ishimura N, Nagase M, Araki A, Ishikawa N, Maruyama R, Kushima R, Ishihara S. Sporadic foveolar-type gastric adenoma with a raspberry-like appearance in Helicobacter pylori-naïve patients. Virchows Arch 2021; 479:687-695. [PMID: 34043063 DOI: 10.1007/s00428-021-03124-3] [Citation(s) in RCA: 11] [Impact Index Per Article: 2.8] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/10/2021] [Revised: 05/12/2021] [Accepted: 05/14/2021] [Indexed: 02/07/2023]
Abstract
Sporadic foveolar-type gastric adenoma (FGA) has been described as an extremely rare polyp that is whitish and flatly elevated. However, we recently found that sporadic FGA with a raspberry-like appearance (FGA-RA) is not rare in Helicobacter pylori (H. pylori)-naïve gastric mucosa. We endoscopically or surgically treated 647 patients with gastric epithelial neoplasms in the last 5 years, with 7.7% (50/647) being H. pylori-naïve. Among these, 43 FGA-RAs were diagnosed based on histologic and endoscopic features in 34 patients, who were all enrolled in this retrospective study. All lesions were observed by white-light endoscopy (WLE) and narrow-band imaging with magnification endoscopy (NBIME). We subsequently analyzed their endoscopic and microscopic features and patient characteristics. The patients were 22 males and 12 females aged 57±23 years (mean±2SD). WLE showed raspberry-like small polyps mimicking gastric hyperplastic polyps in the oxyntic gastric compartment (body/fundus). Multiple growths were confirmed in 20.6% (7/34) of the patients. NBIME revealed irregularly shaped papillary/gyrus-like microstructures with abnormal capillaries. Histologically, all lesions were intraepithelial neoplasms, and most of lesions (62.8%, 27/43) exhibited low-grade dysplasia. Immunohistochemically, neoplastic cells featured strong and diffuse MUC5AC expression, negative or very low MUC6 expression, and negative MUC2/CD10 expression. They also showed Ki-67 hyperexpression with a mean labeling index of 59.4±48.7%. The coexistence of fundic gland polyps in the background mucosa was significantly higher in multiple FGA-RA cases than in solitary cases (100% vs. 55.5%, P< 0.05). FGA-RA is a newly suggested histologic variant of sporadic FGA whose occurrence is not rare in daily endoscopic practice.
Collapse
Affiliation(s)
- Kotaro Shibagaki
- Department of Endoscopy, Faculty of Medicine, Shimane University, 693-8501, 89-1 Enya, Izumo, Japan.
| | - Tsuyoshi Mishiro
- Department of Gastroenterology, Faculty of Medicine, Shimane University, Izumo, Japan
| | - Chika Fukuyama
- Department of Gastroenterology, Faculty of Medicine, Shimane University, Izumo, Japan
| | - Yusuke Takahashi
- Department of Gastroenterology, Faculty of Medicine, Shimane University, Izumo, Japan
| | - Ayako Itawaki
- Department of Gastroenterology, Faculty of Medicine, Shimane University, Izumo, Japan
| | - Saya Nonomura
- Department of Gastroenterology, Faculty of Medicine, Shimane University, Izumo, Japan
| | - Noritsugu Yamashita
- Department of Gastroenterology, Faculty of Medicine, Shimane University, Izumo, Japan
| | - Satoshi Kotani
- Department of Gastroenterology, Faculty of Medicine, Shimane University, Izumo, Japan
| | - Hironobu Mikami
- Department of Gastroenterology, Faculty of Medicine, Shimane University, Izumo, Japan
| | - Daisuke Izumi
- Department of Gastroenterology, Faculty of Medicine, Shimane University, Izumo, Japan
| | - Kousaku Kawashima
- Department of Gastroenterology, Faculty of Medicine, Shimane University, Izumo, Japan
| | - Norihisa Ishimura
- Department of Gastroenterology, Faculty of Medicine, Shimane University, Izumo, Japan
| | - Mamiko Nagase
- Department of Pathology, Faculty of Medicine, Shimane University, Izumo, Japan
| | - Asuka Araki
- Department of Pathology, Faculty of Medicine, Shimane University, Izumo, Japan
| | - Noriyoshi Ishikawa
- Department of Pathology, Faculty of Medicine, Shimane University, Izumo, Japan
| | - Riruke Maruyama
- Department of Pathology, Faculty of Medicine, Shimane University, Izumo, Japan
| | - Ryoji Kushima
- Department of Pathology, Shiga University of Medical Science, Otsu, Japan
| | - Shunji Ishihara
- Department of Gastroenterology, Faculty of Medicine, Shimane University, Izumo, Japan
| |
Collapse
|
|
6
|
Tang D, Wang L, Jiang J, Liu Y, Ni M, Fu Y, Guo H, Wang Z, An F, Zhang K, Hu Y, Zhan Q, Xu G, Zou X. A Novel Deep Learning System for Diagnosing Early Esophageal Squamous Cell Carcinoma: A Multicenter Diagnostic Study. Clin Transl Gastroenterol 2021; 12:e00393. [PMID: 34346911 PMCID: PMC8341371 DOI: 10.14309/ctg.0000000000000393] [Citation(s) in RCA: 24] [Impact Index Per Article: 6.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 02/04/2021] [Accepted: 07/12/2021] [Indexed: 12/29/2022] Open
Abstract
INTRODUCTION This study aims to construct a real-time deep convolutional neural networks (DCNNs) system to diagnose early esophageal squamous cell carcinoma (ESCC) with white light imaging endoscopy. METHODS A total of 4,002 images from 1,078 patients were used to train and cross-validate the DCNN model for diagnosing early ESCC. The performance of the model was further tested with independent internal and external validation data sets containing 1,033 images from 243 patients. The performance of the model was then compared with endoscopists. The accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and Cohen kappa coefficient were measured to assess performance. RESULTS The DCNN model had excellent performance in diagnosing early ESCC with a sensitivity of 0.979, a specificity of 0.886, a positive predictive value of 0.777, a negative predictive value of 0.991, and an area under curve of 0.954 in the internal validation data set. The model also depicted a tremendously generalized performance in 2 external data sets and exhibited superior performance compared with endoscopists. The performance of the endoscopists was markedly elevated after referring to the predictions of the DCNN model. An open-accessed website of the DCNN system was established to facilitate associated research. DISCUSSION A real-time DCNN system, which was constructed to diagnose early ESCC, showed good performance in validation data sets. However, more prospective validation is needed to understand its true clinical significance in the real world.
Collapse
Affiliation(s)
- Dehua Tang
- Department of Gastroenterology, Nanjing University Medical School Affiliated Drum Tower Hospital, Nanjing, Jiangsu, China
| | - Lei Wang
- Department of Gastroenterology, Nanjing University Medical School Affiliated Drum Tower Hospital, Nanjing, Jiangsu, China
| | - Jingwei Jiang
- Department of Gastroenterology, Nanjing University Medical School Affiliated Drum Tower Hospital, Nanjing, Jiangsu, China
| | - Yuting Liu
- Jiangsu Key Laboratory of Big Data Analysis Technology (B-DAT), Nanjing University of Information Science and Technology, Nanjing, China
| | - Muhan Ni
- Department of Gastroenterology, Nanjing University Medical School Affiliated Drum Tower Hospital, Nanjing, Jiangsu, China
| | - Yiwei Fu
- Department of Gastroenterology, Taizhou People's Hospital, Taizhou, Jiangsu, China
| | - Huimin Guo
- Department of Gastroenterology, Nanjing University Medical School Affiliated Drum Tower Hospital, Nanjing, Jiangsu, China
| | - Zhengwen Wang
- Jiangsu Key Laboratory of Big Data Analysis Technology (B-DAT), Nanjing University of Information Science and Technology, Nanjing, China
| | - Fangmei An
- Department of Gastroenterology, Wuxi People's Hospital Affiliated to Nanjing Medical University, Wuxi, Jiangsu, China
| | - Kaihua Zhang
- Jiangsu Key Laboratory of Big Data Analysis Technology (B-DAT), Nanjing University of Information Science and Technology, Nanjing, China
| | - Yanxing Hu
- Xiamen Innovision, Xiamen Software Park Phase III, Xiamen, Fujian, China
| | - Qiang Zhan
- Department of Gastroenterology, Wuxi People's Hospital Affiliated to Nanjing Medical University, Wuxi, Jiangsu, China
| | - Guifang Xu
- Department of Gastroenterology, Nanjing University Medical School Affiliated Drum Tower Hospital, Nanjing, Jiangsu, China
| | - Xiaoping Zou
- Department of Gastroenterology, Nanjing University Medical School Affiliated Drum Tower Hospital, Nanjing, Jiangsu, China
| |
Collapse
|
|
7
|
Lee SP, Lee J, Kae SH, Jang HJ, Koh DH, Jung JH, Byeon SJ. The role of linked color imaging in endoscopic diagnosis of Helicobacter pylori associated gastritis. Scand J Gastroenterol 2020; 55:1114-1120. [PMID: 32668999 DOI: 10.1080/00365521.2020.1794025] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.4] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 02/07/2023]
Abstract
OBJECTIVE Linked color imaging (LCI), a novel image-enhanced endoscopy, can make it easy to recognize differences in mucosal color. It may be helpful for diagnosing H. pylori associated gastritis and H. pylori infection status. We investigated whether LCI could improve the diagnostic accuracy of H. pylori associated gastritis. MATERIALS AND METHODS Upper endoscopy was performed for 100 patients using white light imaging (WLI) and LCI. During the exam, endoscopic video was recorded. It was then analyzed by four expert endoscopists. They reviewed these videos for endoscopic diagnosis of atrophic gastritis, metaplastic gastritis, nodular gastritis and H. pylori infection. Tissue biopsies with rapid urease test were done to confirm H. pylori infection status and intestinal metaplasia. RESULTS Kappa values for the inter-observer variability among the four endoscopists were fair to moderate under WLI and fair to good under LCI. Sensitivity, specificity, positive predictive value and negative predictive value for diagnosing H. pylori infection using WLI were 32.4%, 93.3%, 85.2% and 53.6%, respectively, while those for LCI were 57.4%, 91.3%, 88.7% and 64.3%, respectively. Total diagnostic accuracies for diagnosing H. pylori infection using WLI/LCI were 70.8%/78.8%. The accuracy and sensitivity of LCI for diagnosing H. pylori infection were significantly higher than those of WLI (p < .001 for both). However, there were no significant differences in the accuracy, sensitivity or specificity for diagnosing metaplastic gastritis between LCI and WLI. CONCLUSIONS LCI has better diagnostic accuracy for H. pylori infection status than WLI. Clinical trial registration number: KCT0003674.
Collapse
Affiliation(s)
- Sang Pyo Lee
- Department of Internal Medicine, Hallym University Dongtan Sacred Heart Hospital, Hallym University College of Medicine, Hwaseong, Korea
| | - Jin Lee
- Department of Internal Medicine, Hallym University Dongtan Sacred Heart Hospital, Hallym University College of Medicine, Hwaseong, Korea
| | - Sea Hyub Kae
- Department of Internal Medicine, Hallym University Dongtan Sacred Heart Hospital, Hallym University College of Medicine, Hwaseong, Korea
| | - Hyun Joo Jang
- Department of Internal Medicine, Hallym University Dongtan Sacred Heart Hospital, Hallym University College of Medicine, Hwaseong, Korea
| | - Dong Hee Koh
- Department of Internal Medicine, Hallym University Dongtan Sacred Heart Hospital, Hallym University College of Medicine, Hwaseong, Korea
| | - Jang Han Jung
- Department of Internal Medicine, Hallym University Dongtan Sacred Heart Hospital, Hallym University College of Medicine, Hwaseong, Korea
| | - Sun-Ju Byeon
- Department of Pathology, Hallym University Dongtan Sacred Heart Hospital, Hallym University College of Medicine, Hwaseong, Korea
| |
Collapse
|
|
8
|
Li L, Chen Y, Shen Z, Zhang X, Sang J, Ding Y, Yang X, Li J, Chen M, Jin C, Chen C, Yu C. Convolutional neural network for the diagnosis of early gastric cancer based on magnifying narrow band imaging. Gastric Cancer 2020; 23:126-132. [PMID: 31332619 PMCID: PMC6942561 DOI: 10.1007/s10120-019-00992-2] [Citation(s) in RCA: 136] [Impact Index Per Article: 27.2] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 06/06/2019] [Accepted: 07/12/2019] [Indexed: 02/07/2023]
Abstract
BACKGROUND Magnifying endoscopy with narrow band imaging (M-NBI) has been applied to examine early gastric cancer by observing microvascular architecture and microsurface structure of gastric mucosal lesions. However, the diagnostic efficacy of non-experts in differentiating early gastric cancer from non-cancerous lesions by M-NBI remained far from satisfactory. In this study, we developed a new system based on convolutional neural network (CNN) to analyze gastric mucosal lesions observed by M-NBI. METHODS A total of 386 images of non-cancerous lesions and 1702 images of early gastric cancer were collected to train and establish a CNN model (Inception-v3). Then a total of 341 endoscopic images (171 non-cancerous lesions and 170 early gastric cancer) were selected to evaluate the diagnostic capabilities of CNN and endoscopists. Primary outcome measures included diagnostic accuracy, sensitivity, specificity, positive predictive value, and negative predictive value. RESULTS The sensitivity, specificity, and accuracy of CNN system in the diagnosis of early gastric cancer were 91.18%, 90.64%, and 90.91%, respectively. No significant difference was spotted in the specificity and accuracy of diagnosis between CNN and experts. However, the diagnostic sensitivity of CNN was significantly higher than that of the experts. Furthermore, the diagnostic sensitivity, specificity and accuracy of CNN were significantly higher than those of the non-experts. CONCLUSIONS Our CNN system showed high accuracy, sensitivity and specificity in the diagnosis of early gastric cancer. It is anticipated that more progress will be made in optimization of the CNN diagnostic system and further development of artificial intelligence in the medical field.
Collapse
Affiliation(s)
- Lan Li
- Department of Gastroenterology, The First Affiliated Hospital, College of Medicine, Zhejiang University, 79 Qingchun Road, Hangzhou, 310003, China
| | - Yishu Chen
- Department of Gastroenterology, The First Affiliated Hospital, College of Medicine, Zhejiang University, 79 Qingchun Road, Hangzhou, 310003, China
| | - Zhe Shen
- Department of Gastroenterology, The First Affiliated Hospital, College of Medicine, Zhejiang University, 79 Qingchun Road, Hangzhou, 310003, China
| | - Xuequn Zhang
- Department of Gastroenterology, The First Affiliated Hospital, College of Medicine, Zhejiang University, 79 Qingchun Road, Hangzhou, 310003, China
| | - Jianzhong Sang
- Department of Gastroenterology, Yuyao People's Hospital, Yuyao, China
| | - Yong Ding
- Department of Gastroenterology, The Affiliated Hospital of School of Medicine of Ningbo University, Ningbo, China
| | - Xiaoyun Yang
- Department of Gastroenterology, Jinhua Hospital, Zhejiang University School of Medicine, Jinhua, China
| | - Jun Li
- Department of Pathology, The First Affiliated Hospital, College of Medicine, Zhejiang University, Hangzhou, China
| | - Ming Chen
- Hithink RoyalFlush Information Network Co., Ltd, Hangzhou, China
| | - Chaohui Jin
- Hithink RoyalFlush Information Network Co., Ltd, Hangzhou, China
| | - Chunlei Chen
- State Key Laboratory for Diagnosis and Treatment of Infectious Disease, Collaborative Innovation Center for Diagnosis and Treatment of Infectious Diseases, The First Affiliated Hospital, College of Medicine, Zhejiang University, Hangzhou, China
| | - Chaohui Yu
- Department of Gastroenterology, The First Affiliated Hospital, College of Medicine, Zhejiang University, 79 Qingchun Road, Hangzhou, 310003, China.
| |
Collapse
|
|
9
|
Kishino T, Oyama T, Funakawa K, Ishii E, Yamazato T, Shibagaki K, Miike T, Tanuma T, Kuwayama Y, Takeuchi M, Kitamura Y. Multicenter prospective study on the histological diagnosis of gastric cancer by narrow band imaging-magnified endoscopy with and without acetic acid. Endosc Int Open 2019; 7:E155-E163. [PMID: 30705947 PMCID: PMC6338541 DOI: 10.1055/a-0806-7275] [Citation(s) in RCA: 5] [Impact Index Per Article: 0.8] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 07/11/2018] [Accepted: 10/08/2018] [Indexed: 12/19/2022] Open
Abstract
Background and study aims The usefulness of endoscopy for diagnosing histological type remains unclear. This study aimed to examine the diagnostic accuracy of white light endoscopy (WLE), magnified endoscopy with narrow band imaging (NBI-ME), and NBI-ME with acetic acid enhancement (NBI-AA) for histological type of gastric cancer. Patients and methods Patients with depressed-type gastric cancers resected by endoscopic submucosal dissection were prospectively enrolled, and 221 cases were analyzed. Histological type was diagnosed by WLE, followed by NBI-ME and NBI-AA. Histological type was classified into differentiated adenocarcinoma and undifferentiated adenocarcinoma. Histological type was diagnosed based on lesion color in WLE, surface patterns (pit, villi, and unclear) and vascular irregularities in NBI-ME, and surface patterns in NBI-AA. Results Histological types of target areas were differentiated adenocarcinoma and undifferentiated adenocarcinoma in 206 and 15 cases, respectively. Diagnostic accuracy of WLE, NBI-ME, and NBI-AA for the histological type was 96.4 % (213/221), 96.8 % (214/221), and 95.5 % (211/221), respectively. No significant differences were observed among modalities. Positive predictive value based on endoscopic findings in NBI-ME was 98.0 % (149/152) for the villi pattern, 100 % (19/19) for the irregular pit pattern, 100 % (9/9) for the unclear surface pattern with a vascular network, 90.3 % (28/31) for the unclear surface pattern with mild vascular irregularity, and 88.9 % (8/9) for the unclear surface pattern with severe vascular irregularity. Conclusions NBI-ME and NBI-AA did not show any advantages over WLE for diagnostic accuracy. Villi pattern, irregular pit pattern, and vascular network may be useful for identifying differentiated adenocarcinoma.
Collapse
Affiliation(s)
- Takaaki Kishino
- Department of Endoscopy, Saku Central Hospital Advanced Care Center, Saku, Japan,Department of Gastroenterology, Nara City Hospital, Higashikidera-cho, Nara, Japan,Corresponding author Takaaki Kishino, MD Department of GastroenterologyNara City Hospital1-50-1 HigashikiderachoNara 630-8305Japan+81-742222478
| | - Tsuneo Oyama
- Department of Endoscopy, Saku Central Hospital Advanced Care Center, Saku, Japan
| | - Keita Funakawa
- Department of Gastroenterology, Kagoshima University School of Medical and Dental Sciences, Kagoshima, Japan
| | - Eiji Ishii
- Department of Gastroenterology, Kameda Medical Center, Kamogawa, Japan
| | - Tetsuro Yamazato
- Department of Gastroenterology, Tokyo Metropolitan Cancer Detection Center, Fuchu, Japan
| | - Kotaro Shibagaki
- Department of Gastroenterology, Tottori Municipal Hospital, Tottori, Japan
| | - Tadashi Miike
- Department of Gastroenterology, University of Miyazaki, Miyazaki, Japan
| | - Tokuma Tanuma
- Department of Gastroenterology, Teine Keijinkai Hospital, Sapporo, Japan
| | - Yasuharu Kuwayama
- Department of Gastroenterology, Tokushima Red Cross Hospital, Komatsushima, Japan
| | - Manabu Takeuchi
- Department of Gastroenterology, Niigata University Medical and Dental Hospital, Niigata, Japan
| | - Yoko Kitamura
- Department of Gastroenterology, Nara City Hospital, Higashikidera-cho, Nara, Japan
| |
Collapse
|
|
10
|
Takiyama H, Ozawa T, Ishihara S, Fujishiro M, Shichijo S, Nomura S, Miura M, Tada T. Automatic anatomical classification of esophagogastroduodenoscopy images using deep convolutional neural networks. Sci Rep 2018; 8:7497. [PMID: 29760397 PMCID: PMC5951793 DOI: 10.1038/s41598-018-25842-6] [Citation(s) in RCA: 79] [Impact Index Per Article: 11.3] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/07/2017] [Accepted: 04/30/2018] [Indexed: 12/14/2022] Open
Abstract
The use of convolutional neural networks (CNNs) has dramatically advanced our ability to recognize images with machine learning methods. We aimed to construct a CNN that could recognize the anatomical location of esophagogastroduodenoscopy (EGD) images in an appropriate manner. A CNN-based diagnostic program was constructed based on GoogLeNet architecture, and was trained with 27,335 EGD images that were categorized into four major anatomical locations (larynx, esophagus, stomach and duodenum) and three subsequent sub-classifications for stomach images (upper, middle, and lower regions). The performance of the CNN was evaluated in an independent validation set of 17,081 EGD images by drawing receiver operating characteristics (ROC) curves and calculating the area under the curves (AUCs). ROC curves showed high performance of the trained CNN to classify the anatomical location of EGD images with AUCs of 1.00 for larynx and esophagus images, and 0.99 for stomach and duodenum images. Furthermore, the trained CNN could recognize specific anatomical locations within the stomach, with AUCs of 0.99 for the upper, middle, and lower stomach. In conclusion, the trained CNN showed robust performance in its ability to recognize the anatomical location of EGD images, highlighting its significant potential for future application as a computer-aided EGD diagnostic system.
Collapse
Affiliation(s)
- Hirotoshi Takiyama
- Tada Tomohiro Institute of Gastroenterology and Proctology, Saitama, Japan
- Department of Surgical Oncology, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan
| | - Tsuyoshi Ozawa
- Tada Tomohiro Institute of Gastroenterology and Proctology, Saitama, Japan.
- Department of surgery, Sanno Hospital, International University of Health and Welfare, Tokyo, Japan.
| | - Soichiro Ishihara
- Tada Tomohiro Institute of Gastroenterology and Proctology, Saitama, Japan
- Department of surgery, Sanno Hospital, International University of Health and Welfare, Tokyo, Japan
| | - Mitsuhiro Fujishiro
- Department of Gastroenterology, Graduate School of Medicine, the University of Tokyo, Tokyo, Japan
| | - Satoki Shichijo
- Department of Gastrointestinal Oncology, Osaka International Cancer Institute, Osaka, Japan
| | - Shuhei Nomura
- Department of Global Health Policy, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan
- Department of Epidemiology and Biostatistics, School of Public Health, Imperial College London, London, United Kingdom
| | - Motoi Miura
- Tada Tomohiro Institute of Gastroenterology and Proctology, Saitama, Japan
- Graduate School of Public Health, Teikyo University, Tokyo, Japan
| | - Tomohiro Tada
- Tada Tomohiro Institute of Gastroenterology and Proctology, Saitama, Japan
- Department of Surgical Oncology, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan
| |
Collapse
|