Background/Objectives: Colonoscopies have some limitations that result in a miss rate detection of polyps. Microwave imaging has been demonstrated to detect colorectal polyps based on their dielectric properties in synthetic phantoms, ex vivo tissues and in vivo animal models. This study aims to evaluate, for the first time, the feasibility, safety and performance of microwave-based colonoscopy for diagnosis of polyps in real-time explorations in humans. Methods: This was a single-center, prospective, observational study. Patients referred for diagnostic colonoscopy were explored with a device with microwave antennas which was attached to the tip of a standard colonoscope. The primary outcomes were rate of cecal intubation, adverse events, mural injuries and performance metrics for the detection of polyps. Secondary outcomes were the following: patients' subjective feedback, procedural time and perception of difficulty according to the endoscopist. Results: Fifteen patients were enrolled. Cecal intubation rate was 100%, with a mean time of 12.7 ± 4.9 min (range 4-22). Use of the device did not affect the endoscopic image, and polypectomy was successfully performed in all cases. In on scale from zero (not difficult) to four (very difficult), the maneuverability during the insertion was considered ≤2 in the 86.7% (13/15) of colonoscopies. Only 16 incidents were reported in 14 patients: 11 (67%) superficial hematomas, 2 minor rectal bleedings, 1 anal fissure, 1 rhinorrhea and 1 headache. Most of the patients (94%) reported no discomfort or minimal discomfort before discharge (Gloucester score 1 and 2, respectively). In the six patients with 23 polyps used for the performance analysis, the sensitivity and specificity were 86.9% and 72.0%, respectively. Conclusions: microwave-based colonoscopy is safe and feasible and has the potential to detect polyps in real colonoscopies.

Despite enormous progress, advanced cancers are still one of the most serious medical problems in current society. Although various agents and therapeutic strategies with anticancer activity are known and used, they often fail to achieve satisfactory long-term patient outcomes and survival. Recently, immunotherapy has shown success in patients by harnessing important interactions between the immune system and cancer. However, many of these therapies lead to frequent side effects when administered systemically, prompting treatment modifications or discontinuation or, in severe cases, fatalities. New therapeutic approaches like intratumoral immunotherapy, characterized by reduced side effects, cost, and systemic toxicity, offer promising prospects for future applications in clinical oncology. In the context of locally advanced or metastatic cancer, combining diverse immunotherapeutic and other treatment strategies targeting multiple cancer hallmarks appears crucial. Such combination therapies hold promise for improving patient outcomes and survival and for promoting a sustained systemic response. This review aims to provide a current overview of immunotherapeutic approaches, specifically focusing on the intratumoral administration of drugs in patients with locally advanced and metastatic cancers. It also explores the integration of intratumoral administration with other modalities to maximize therapeutic response. Additionally, the review summarizes recent advances in intratumoral immunotherapy and discusses novel therapeutic approaches, outlining future directions in the field.

Colonoscopy remains the most commonly used colorectal cancer screening test in the United States. A substantial portion of the screening population value the high sensitivity of colonoscopy for precancerous colorectal lesions of all sizes, which allows it to be performed at 10 year intervals in average-risk persons with negative examinations. Emerging evidence supports the eventual endorsement of 15 year intervals for patients with normal examinations. Considerable evidence supports an impact of colonoscopy on colorectal cancer incidence and mortality, including a randomized controlled trial of colonoscopy vs. no screening, numerous case-control and cohort studies, an impact of fecal blood testing on cancer incidence, and the impact of one randomized controlled trial of flexible sigmoidoscopy on proximal colon cancer incidence. Colonoscopy is the gold standard for detection of colorectal precancerous lesions, and continues to evolve with regard to sensitivity for precancerous lesions and the effectiveness and safety of precancerous lesion resection. Gains in detection of precancerous lesions have followed from a robust movement to improve colonoscopy quality, and development of non-device techniques such as patient rotation during withdrawal, water exchange colonoscopy, and double examination of one or more colonic segments. Further, development of devices to improve mucosal exposure during withdrawal (e.g. Endocuff Vision, distal cap attachment, and Computer-Aided Quality), and devices that highlight flat lesions (e.g. chromoendoscopy, electronic chromoendoscopy, and Computer-Aided Detection) have created opportunities to achieve very high levels of detection and thereby increase the protective benefits of colonoscopy. Further, colonoscopy resection safety has improved via the widespread use of cold resection for lesions < 10 mm in size, as well as sessile serrated lesions of all sizes. Colonoscopy can be offered to patients as one of multiple options for screening, or as the test of choice for patients with the highest pre-screening probability of cancer and precancerous lesions, or as the first test offered followed by offers of other screening tests if colonoscopy is declined (sequential offers of screening). Sequential offers of screening result in overall adherence to screening similar to offering multiple options, but with a higher fraction of patients undergoing colonoscopy. Given the long-lasting protective effects of colonoscopy and its improving effectiveness and safety, colonoscopy remains a useful option for primary average-risk colorectal cancer screening.

Colorectal cancer (CRC) poses a significant global health threat, necessitating early detection. Traditional diagnostic tools like optical colonoscopy have limitations prompting our '5G-SUCCEEDS' initiative to explore a novel approach involving remote colon capsule endoscopy (CCE).

This prospective feasibility study was conducted at a single hospital in England. Between December 2022 and September 2023, we introduced a remote CCE service within the 5G-SUCCEEDS framework. We undertook a feasibility study of CCE in patients with low-risk/moderate-risk CRC stratified by faecal haemoglobin. Outcomes included carbon footprint analysis (outlined through three potential clinical pathways) and patient-reported outcomes through structured questionnaires and interviews.

Among 25 participants, 88% expressed satisfaction with remote CCE. 82% were willing to have remote CCE if clinically indicated in future. CCE findings included adenomatous polyps (58%), normal results (17%) and diverticulosis (21%), with no cancers identified in this pilot. Notably, we found that the carbon footprint associated with delivery of CCE at home (pathway 3) was lower compared with CCE delivered in a clinical setting (pathway 2). A fully optimised, automated scaled-up pathway would combine the delivery and collection of CCE equipment within a local area to reduce the carbon footprint of the travel element by 75%. Moreover, the conversion rate into a colonoscopy pathway is not static and clinicians acknowledge that this could be as low as 28%. Carbon footprint is more favourable for home-delivered CCE in the optimised scenario, while less so when considering the need for additional procedures (colonoscopy conversion).

The 5G-SUCCEEDS initiative highlights the feasibility and advantages of home-based diagnostics using CCE.

The ENDOANGEL (EN) computer-assisted detection technique has emerged as a promising tool for enhancing the detection rate of colorectal adenomas during colonoscopies. However, its efficacy in identifying missed adenomas during subsequent colonoscopies remains unclear. Thus, we herein aimed to compare the adenoma miss rate (AMR) between EN-assisted and standard colonoscopies. Data from patients who underwent a second colonoscopy (EN-assisted or standard) within 6 months between September 2022 and May 2023 were analyzed. The EN-assisted group exhibited a significantly higher AMR (24.3% vs 11.9%, P = .005) than the standard group. After adjusting for potential confounders, multivariable analysis revealed that the EN-assisted group had a better ability to detect missed adenomas than the standard group (odds ratio = 2.89; 95% confidence interval = 1.14-7.80, P = .029). These findings suggest that EN-assisted colonoscopy represents a valuable advancement in improving AMR compared with standard colonoscopy. The integration of EN-assisted colonoscopy into routine clinical practice may offer significant benefits to patients requiring hospital resection of lesions following adenoma detection during their first colonoscopy.

Adenomas per colonoscopy (APC) may be a better measure of colonoscopy quality than adenoma detection rate (ADR) because it credits endoscopists for each detected adenoma. There are few data examining the association between APC and postcolonoscopy colorectal cancer (PCCRC) incidence. We used data from the New Hampshire Colonoscopy Registry to examine APC and PCCRC risk.

We included New Hampshire Colonoscopy Registry patients with an index examination and at least 1 follow-up event, either a colonoscopy or a colorectal cancer (CRC) diagnosis. Our outcome was PCCRC defined as any CRC diagnosed ≥6 months after an index examination. The exposure variable was endoscopist-specific APC quintiles of .25, .40, .50, and .70. Cox regression was used to model the hazard of PCCRC on APC, controlled for age, sex, year of index examination, index findings, bowel preparation, and having more than 1 surveillance examination.

In 32,535 patients, a lower hazard for PCCRC (n = 178) was observed for higher APCs as compared to APCs of <.25 (reference): .25 to <.40: hazard ratio (HR), .35; 95% confidence interval (CI), .22-.56; .40 to <.50: HR, .31; 95% CI, .20-.49; .50 to <.70: HR, .20; 95% CI, .11-.36; and ≥.70: HR, .19; 95% CI, .09-.37. When examining endoscopists with an ADR of at least 25%, an APC of <.50 was associated with a significantly higher hazard than an APC of ≥.50 (HR, 1.65; 95% CI, 1.06-2.56). A large proportion of endoscopists-one-fifth (32 of 152; 21.1%)-had an ADR of ≥25% but an APC of <.50.

Our novel data demonstrating lower PCCRC risk in examinations performed by endoscopists with higher APCs suggest that APC could be a useful quality measure. Quality improvement programs may identify important deficiencies in endoscopist detection performance by measuring APC for endoscopists with an ADR of ≥25%.

Endocuff Vision (Olympus Europe, Hamburg, Germany) has been designed to enhance mucosal visualization, thereby improving detection of (pre-)malignant colorectal lesions. This multicenter, international, back-to-back, randomized colonoscopy trial compared the adenoma detection rate (ADR) and adenoma miss rate (AMR) between Endocuff Vision-assisted colonoscopy (EVC) and conventional colonoscopy (CC).

Patients aged 40 to 75 years referred for non-immunochemical fecal occult blood test-based screening, surveillance, or diagnostic colonoscopy were included at 10 hospitals and randomized into 4 groups: group 1, 2 × CC; group 2, CC followed by EVC; group 3, EVC followed by CC; and group 4, 2 × EVC. Primary outcomes included ADR and AMR.

A total of 717 patients were randomized, of whom 661 patients (92.2%) had 1 and 646 (90.1%) patients had 2 completed back-to-back colonoscopies. EVC did not significantly improve ADR compared to CC (41.1%; [95% confidence interval (CI), 36.1-46.3] vs 35.5% [95% CI, 30.7-40.6], respectively; P = .125), but EVC did reduce AMR by 11.7% (29.6% [95% CI, 23.6-36.5] vs 17.9% [95% CI, 12.5-23.5], respectively; P = .049). AMR of 2 × CC compared to 2 × EVC was also not significantly different (25.9% [95% CI, 19.3-33.9] vs 18.8% [95% CI, 13.9-24.8], respectively; P = .172). Only 3.7% of the polyps missed during the first procedures had advanced pathologic features. Factors affecting risk of missing adenomas were age (P = .002), Boston Bowel Preparation Scale (P = .008), and region where colonoscopy was performed (P < .001).

Our trial shows that EVC reduces the risk of missing adenomas but does not lead to a significantly improved ADR. Remarkably, 25% of adenomas are still missed during conventional colonoscopies, which is not different from miss rates reported 25 years ago; reassuringly, advanced features were only found in 3.7% of these missed lesions. (Clinical trial registration number: NCT03418948.).

Randomized controlled trials (RCTs) have reported that artificial intelligence (AI) improves endoscopic polyp detection. Different methodologies-namely, parallel and tandem designs-have been used to evaluate the efficacy of AI-assisted colonoscopy in RCTs. Systematic reviews and meta-analyses have reported a pooled effect that includes both study designs. However, it is unclear whether there are inconsistencies in the reported results of these 2 designs. Here, we aimed to determine whether study characteristics moderate between-trial differences in outcomes when evaluating the effectiveness of AI-assisted polyp detection.

A systematic search of Ovid MEDLINE, Embase, Cochrane Central, Web of Science, and IEEE Xplore was performed through March 1, 2023, for RCTs comparing AI-assisted colonoscopy with routine high-definition colonoscopy in polyp detection. The primary outcome of interest was the impact of study type on the adenoma detection rate (ADR). Secondary outcomes included the impact of the study type on adenomas per colonoscopy and withdrawal time, as well as the impact of geographic location, AI system, and endoscopist experience on ADR. Pooled event analysis was performed using a random-effects model.

Twenty-four RCTs involving 17,413 colonoscopies (AI assisted: 8680; non-AI assisted: 8733) were included. AI-assisted colonoscopy improved overall ADR (risk ratio [RR], 1.24; 95% confidence interval [CI], 1.17-1.31; I2 = 53%; P < .001). Tandem studies collectively demonstrated improved ADR in AI-aided colonoscopies (RR, 1.18; 95% CI, 1.08-1.30; I2 = 0%; P < .001), as did parallel studies (RR, 1.26; 95% CI, 1.17-1.35; I2 = 62%; P < .001), with no statistical subgroup difference between study design. Both tandem and parallel study designs revealed improvement in adenomas per colonoscopy in AI-aided colonoscopies, but this improvement was more marked among tandem studies (P < .001). AI assistance significantly increased withdrawal times for parallel (P = .002), but not tandem, studies. ADR improvement was more marked among studies conducted in Asia compared to Europe and North America in a subgroup analysis (P = .007). Type of AI system used or endoscopist experience did not affect overall improvement in ADR.

Either parallel or tandem study design can capture the improvement in ADR resulting from the use of AI-assisted polyp detection systems. Tandem studies powered to detect differences in endoscopic performance through paired comparison may be a resource-efficient method of evaluating new AI-assisted technologies.

The effect of computer-aided polyp detection (CADe) on adenoma detection rate (ADR) among endoscopists-in-training remains unknown.

We performed a single-blind, parallel-group, randomized controlled trial in Hong Kong between April 2021 and July 2022 (NCT04838951). Eligible subjects undergoing screening/surveillance/diagnostic colonoscopies were randomized 1:1 to receive colonoscopies with CADe (ENDO-AID[OIP-1]) or not (control) during withdrawal. Procedures were performed by endoscopists-in-training with <500 procedures and <3 years' experience. Randomization was stratified by patient age, sex, and endoscopist experience (beginner vs intermediate level, <200 vs 200-500 procedures). Image enhancement and distal attachment devices were disallowed. Subjects with incomplete colonoscopies or inadequate bowel preparation were excluded. Treatment allocation was blinded to outcome assessors. The primary outcome was ADR. Secondary outcomes were ADR for different adenoma sizes and locations, mean number of adenomas, and non-neoplastic resection rate.

A total of 386 and 380 subjects were randomized to CADe and control groups, respectively. The overall ADR was significantly higher in the CADe group than in the control group (57.5% vs 44.5%; adjusted relative risk, 1.41; 95% CI, 1.17-1.72; P < .001). The ADRs for <5 mm (40.4% vs 25.0%) and 5- to 10-mm adenomas (36.8% vs 29.2%) were higher in the CADe group. The ADRs were higher in the CADe group in both the right colon (42.0% vs 30.8%) and left colon (34.5% vs 27.6%), but there was no significant difference in advanced ADR. The ADRs were higher in the CADe group among beginner (60.0% vs 41.9%) and intermediate-level (56.5% vs 45.5%) endoscopists. Mean number of adenomas (1.48 vs 0.86) and non-neoplastic resection rate (52.1% vs 35.0%) were higher in the CADe group.

Among endoscopists-in-training, the use of CADe during colonoscopies was associated with increased overall ADR. (ClinicalTrials.gov, Number: NCT04838951).

Studies on the effect of computer-aided detection (CAD) in a daily clinical screening and surveillance colonoscopy population practice are scarce. The aim of this study was to evaluate a novel CAD system in a screening and surveillance colonoscopy population.

This multicentre, randomised, controlled trial was done in ten hospitals in Europe, the USA, and Israel by 31 endoscopists. Patients referred for non-immunochemical faecal occult blood test (iFOBT) screening or surveillance colonoscopy were included. Patients were randomomly assigned to CAD-assisted colonoscopy or conventional colonoscopy; a subset was further randomly assigned to undergo tandem colonoscopy: CAD followed by conventional colonoscopy or conventional colonoscopy followed by CAD. Primary objectives included adenoma per colonoscopy (APC) and adenoma per extraction (APE). Secondary objectives included adenoma miss rate (AMR) in the tandem colonoscopies. The study was registered at ClinicalTrials.gov, NCT04640792.

A total of 916 patients were included in the modified intention-to-treat analysis: 449 in the CAD group and 467 in the conventional colonoscopy group. APC was higher with CAD compared with conventional colonoscopy (0·70 vs 0·51, p=0·015; 314 adenomas per 449 colonoscopies vs 238 adenomas per 467 colonoscopies; poisson effect ratio 1·372 [95% CI 1·068-1·769]), while showing non-inferiority of APE compared with conventional colonoscopy (0·59 vs 0·66; p<0·001 for non-inferiority; 314 of 536 extractions vs 238 of 360 extractions). AMR in the 127 (61 with CAD first, 66 with conventional colonoscopy first) patients completing tandem colonoscopy was 19% (11 of 59 detected during the second pass) in the CAD first group and 36% (16 of 45 detected during the second pass) in the conventional colonoscopy first group (p=0·024).

CAD increased adenoma detection in non-iFOBT screening and surveillance colonoscopies and reduced adenoma miss rates compared with conventional colonoscopy, without an increase in the resection of non-adenomatous lesions.

Magentiq Eye.

With an increasing number of patients with gastrointestinal cancer, effective and accurate early diagnostic clinical tools are required provide better health care for patients with gastrointestinal cancer. Recent studies have shown that artificial intelligence (AI) plays an important role in the diagnosis and treatment of patients with gastrointestinal tumors, which not only improves the efficiency of early tumor screening, but also significantly improves the survival rate of patients after treatment. With the aid of efficient learning and judgment abilities of AI, endoscopists can improve the accuracy of diagnosis and treatment through endoscopy and avoid incorrect descriptions or judgments of gastrointestinal lesions. The present article provides an overview of the application status of various artificial intelligence in gastric and colorectal cancers in recent years, and the direction of future research and clinical practice is clarified from a clinical perspective to provide a comprehensive theoretical basis for AI as a promising diagnostic and therapeutic tool for gastrointestinal cancer.

Surveillance colonoscopy decreases colorectal cancer mortality; however, lesions are occasionally missed. Although an appropriate surveillance interval is indicated, variations may occur in the methods used, such as scope manipulation or observation. Therefore, individual endoscopists may miss certain areas. This study aimed to verify the effectiveness of performing repeat colonoscopies with a different endoscopist from the initial procedure.

We retrospectively reviewed a database of 8093 consecutive colonoscopies performed in the Omori Red Cross Hospital from January 1st 2018 to June 30th 2021. Data from repeat total colonoscopies performed within three months were collected to assess missed lesions. The patients were divided into two groups according to whether the two examinations were performed by different endoscopists (group D) or the same endoscopist (group S). The primary outcome in both groups was the missed lesion detection rate (MLDR).

Overall, 205 eligible patients were analyzed. In total, 102 and 103 patients were enrolled in groups D and S, respectively. The MLDR was significantly higher in group D (61.8% vs. 31.1%, P < 0.0001). Multivariate logistic regression analysis for the detection of missed lesions identified performance by the different endoscopists (odds ratio, 3.38; 95% CI, 1.81-6.30), and sufficient withdrawal time (> 6 min) (odds ratio, 3.10; 95% CI, 1.12-8.61) as significant variables.

Overall, our study showed a significant improvement in the detection of missed lesions when performed by different endoscopists. When performing repeat colonoscopy, it is desirable that a different endoscopist perform the second colonoscopy.

This study was approved by the Institutional Review Board of the Omori Red Cross Hospital on November 28, 2022 (approval number:22-43).

Colorectal cancer (CRC) is a leading cause of cancer. The detection of pre-malignant lesions by colonoscopy is associated with reduced CRC incidence and mortality. Narrow band imaging has shown promising but conflicting results for the detection of serrated lesions.

We performed a randomized clinical trial to compare the mean detection of serrated lesions and hyperplastic polyps ≥10 mm with NBI or high-definition white light (HD-WL) withdrawal. We also compared all sessile serrated lesions (SSLs), adenoma, and polyp prevalence and rates.

Overall, 782 patients were randomized (WL group 392 patients; NBI group 390 patients). The average number of serrated lesions and hyperplastic polyps ≥10 mm detected per colonoscopy (primary endpoint) was similar between the HD-WL and NBI group (0.118 vs. 0.156, p = 0.44). Likewise, the adenoma detection rate (55.2% vs. 53.2%, p = 0.58) and SSL detection rate (6.8% vs. 7.5%, p = 0.502) were not different between the two study groups. Withdrawal time was higher in the NBI group (10.88 vs. 9.47 min, p = 0.004), with a statistically nonsignificant higher total procedure time (20.97 vs. 19.30 min, p = 0.052).

The routine utilization of narrow band imaging does not improve the detection of serrated class lesions or any pre-malignant lesion and increases the withdrawal time.

To evaluate the effect of 3-dimensional (3D) imaging device on polyp and adenoma detection during colonoscopy.

In a single-blind, randomized controlled trial, participants aged 18-70 years who underwent diagnostic or screening colonoscopy were consecutively enrolled between August 2019 and May 2022. Each participant was randomized in a 1:1 ratio to undergo either 2-dimensional (2D-3D) colonoscopy or 3D-2D colonoscopy through computer-generated random numbers. Primary outcome included polyp detection rate (PDR) and adenoma detection rate (ADR), defined as the proportion of individuals with at least 1 polyp or adenoma detected during colonoscopy. The primary analysis was intention-to-treat.

Of 1,196 participants recruited, 571 in 2D-3D group and 583 in 3D-2D group were finally included after excluding those who met the exclusion criteria. The PDR between 2D and 3D groups was separately 39.6% and 40.5% during phase 1 (odds ratio [OR] = 0.96, 95% confidence interval [CI]: 0.76-1.22, P = 0.801), whereas PDR was significantly higher in 3D group (27.7%) than that of 2D group (19.9%) during phase 2, with a 1.54-fold increase (1.17-2.02, P = 0.002). Similarly, the ADR during phase 1 between 2D (24.7%) and 3D (23.8%) groups was not significant (OR = 1.05, 0.80-1.37, P = 0.788), while ADR was significantly higher in 3D group (13.8%) than that of 2D group (9.9%) during phase 2, with a 1.45-fold increase (1.01-2.08, P = 0.041). Further subgroup analysis confirmed significantly higher PDR and ADR of 3D group during phase 2, particularly in midlevel and junior endoscopists.

The 3D imaging device could improve overall PDR and ADR during colonoscopy, particularly in midlevel and junior endoscopists. Trial number: ChiCTR1900025000.

Colorectal cancer (CRC) is a rapidly escalating public health concern, which underlines the significance of its early detection and the need for the refinement of current screening methods. In this systematic review, we aimed to analyze the potential advantages and limitations of artificial intelligence (AI)-based computer-aided detection (CADe) systems as compared to routine colonoscopy. This review begins by shedding light on the global prevalence and mortality rates of CRC, highlighting the urgent need for effective screening techniques and early detection of this cancer type. It addresses the problems associated with undetected adenomas and polyps and the subsequent risk of interval CRC following colonoscopy. The incorporation of AI into diagnostics has been studied, specifically the use of CADe systems which are powered by deep learning. The review summarizes the findings from 13 randomized controlled trials (RCTs) (2019-2023), evaluating the impact of CADe on polyp and adenoma detection. The findings from the studies consistently show that CADe is superior to conventional colonoscopy procedures in terms of adenoma detection rate (ADR) and polyp detection rate (PDR), particularly with regard to small and flat lesions which are easily overlooked. The review acknowledges certain limitations of the included studies, such as potential performance bias and geographic limitations. The review ultimately concludes that AI-assisted colonoscopy can reduce missed lesion rates and improve CRC diagnosis. Collaboration between experts and clinicians is key for successful implementation. In summary, this review analyzes recent RCTs on AI-assisted colonoscopy for polyp and adenoma detection. It describes the likely benefits, limitations, and future implications of AI in enhancing colonoscopy procedures and lowering the incidence of CRC. More double-blinded trials and studies among diverse populations from different countries must be conducted to substantiate and expand upon the findings of this review.

Background and study aims Conventional endoscopic mucosal resection (C-EMR) is limited by low en-bloc resection rates, especially for large (> 20 mm) lesions. Underwater EMR (U-EMR) has emerged as an alternative for colorectal polyps and is being shown to improve en-bloc resection rates. We conducted a systematic review and meta-analysis comparing the two techniques. Methods Multiple databases were searched through November 2022 for randomized controlled trials (RCTs) comparing outcomes of U-EMR and C-EMR for colorectal polyps. Meta-analysis was performed to determine pooled proportions and relative risks (RRs) of R0 and en-bloc resection, polyp recurrence, resection time, and adverse events. Results Seven RCTs with 1458 patients (U-EMR: 739, C-EMR: 719) were included. The pooled rate of en-bloc resection was significantly higher with U-EMR vs C-EMR, 70.17% (confidence interval [CI] 46.68-86.34) vs 58.14% (CI 31.59-80.68), respectively, RR 1.21 (CI 1.01-1.44). R0 resection rates were higher with U-EMR vs C-EMR, 58.1% (CI 29.75-81.9) vs 44.6% (CI 17.4-75.4), RR 1.25 (CI 0.99-1.6). For large polyps (> 20 mm), en-bloc resection rates were comparable between the two techniques, RR 1.24 (CI 0.83-1.84). Resection times were comparable between U-EMR and C-EMR, standardized mean difference -1.21 min (CI -2.57 to -0.16). Overall pooled rates of perforation, and immediate and delayed bleeding were comparable between U-EMR and C-EMR. Pooled rate of polyp recurrence at surveillance colonoscopy was significantly lower with U-EMR than with C-EMR, RR 0.62 (CI 0.41-0.94). Conclusions Colorectal U-EMR results in higher en-bloc resection and lower recurrence rates when compared to C-EMR. Both techniques have comparable resection times and safety profiles.

Background and study aims  Incomplete resection of 4- to 20-mm colorectal polyps occur frequently (> 10 %), putting patients at risk for post-colonoscopy colorectal cancer. We hypothesized that routine use of wide-field cold snare resection with submucosal injection (CSP-SI) might reduce incomplete resection rates (IRRs). Patients and methods  Patients aged 45 to 80 years undergoing elective colonoscopies were enrolled in a prospective clinical study. All 4- to 20-mm non-pedunculated polyps were resected using CSP-SI. Post-polypectomy margin biopsies were obtained to determine IRRs through histopathology assessment. The primary outcome was IRR, defined as remnant polyp tissue found on margin biopsies. Secondary outcomes included technical success and complication rates. Results  A total of 429 patients (median age 65 years, 47.1 % female, adenoma detection rate 40 %) with 204 non-pedunculated colorectal polyps 4 to 20 mm removed using CSP-SI were included in the final analysis. CSP-SI was technical successful in 97.5 % (199/204) of cases (5 conversion to hot snare polypectomy). IRR for CSP-SI was 3.8 % (7/183) (95 % confidence interval [CI] 2.7 %-5.5 %). IRR was 1.6 % (2/129), 16 % (4/25), and 3.4 % (1/29) for adenomas, serrated lesions, and hyperplastic polyps respectively. IRR was 2.3 % (2/87), 6.3 % (4/64), 4.0 % (6/151), and 3.1 % (1/32) for polyps 4 to 5 mm, 6 to 9 mm, < 10 mm, and 10 to 20 mm, respectively. There were no CSP-SI-related serious adverse events. Conclusions  Use of CSP-SI results in lower IRRs compared to what has previously been reported in the literature for hot or cold snare polypectomy when not using wide-field cold snare resection with submucosal injection. CSP-SI showed an excellent safety and efficacy profile, however comparative studies to CSP without SI are required to confirm these results.

The aim was to investigate the adenoma detection rate (ADR) of endoscopists who have used full-spectrum endoscopy (Fuse) for 3 years and revert back to traditional forward-viewing endoscopes (R-TFV) at an ambulatory surgical center.

Traditional forward viewing (TFV) endoscopes have 1 camera and provide an angle of view of 140 to 170 degrees, whereas Fuse provides a 330 degrees view through the addition of 2 side cameras. It has previously been reported that Fuse increased the ADR by 5.4% when compared with previous rates using TFV. Fuse is no longer commercially available. The ADR of endoscopists who revert back to TFV is unknown.

We conducted a retrospective analysis of data examining the ADR from average risk screening colonoscopies at a 5-room ambulatory surgical center where endoscopists transitioned from TFV to Fuse in April 2014 and then reverted back to TFV in 2016. The primary outcome was ADR. Secondary outcomes were ADR for advanced and right-sided adenomas.

A total of 6110 procedures were reviewed. The ADR was 23.70% for TFV, 29.02% for Fuse and 28.88% for R-TFV. The ADR for advanced adenomas was 3.8% for TFV, 6.0% for Fuse and 7.3% for R-TFV. The ADR for right-sided adenomas was 13.0% for TFV, 16.7% for Fuse and 16.0% for R-TFV. The results for all 3 categories showed a statistical difference between TFV and Fuse as well as between TFV and R-TFV. There were no statistical differences between the ADR of Fuse compared with R-TFV.

During R-TFV, endoscopists are able to maintain their increased ability to detect adenomas. This would suggest that there was a change in behavior in endoscopists using Fuse that was durable.

The gastrointestinal (GI) tract can be affected by different diseases or lesions such as esophagitis, ulcers, hemorrhoids, and polyps, among others. Some of them can be precursors of cancer such as polyps. Endoscopy is the standard procedure for the detection of these lesions. The main drawback of this procedure is that the diagnosis depends on the expertise of the doctor. This means that some important findings may be missed. In recent years, this problem has been addressed by deep learning (DL) techniques. Endoscopic studies use digital images. The most widely used DL technique for image processing is the convolutional neural network (CNN) due to its high accuracy for modeling complex phenomena. There are different CNNs that are characterized by their architecture. In this article, four architectures are compared: AlexNet, DenseNet-201, Inception-v3, and ResNet-101. To determine which architecture best classifies GI tract lesions, a set of metrics; accuracy, precision, sensitivity, specificity, F1-score, and area under the curve (AUC) were used. These architectures were trained and tested on the HyperKvasir dataset. From this dataset, a total of 6,792 images corresponding to 10 findings were used. A transfer learning approach and a data augmentation technique were applied. The best performing architecture was DenseNet-201, whose results were: 97.11% of accuracy, 96.3% sensitivity, 99.67% specificity, and 95% AUC.

To determine the long-term outcomes after colorectal endoscopic submucosal dissection (ESD), we conducted a large, multicenter, prospective cohort trial with a 5-year observation period.

Between February 2013 and January 2015, we consecutively enrolled 1740 patients with 1814 colorectal epithelial neoplasms ≥20 mm who underwent ESD. Patients with noncurative resection (non-CR) lesions underwent additional radical surgery, as needed. After the initial treatment, intensive 5-year follow-up with planned multiple colonoscopies was conducted to identify metastatic and/or local recurrences. Primary outcomes were overall survival, disease-specific survival, and intestinal preservation rates. The rates of local recurrence and metachronous invasive cancer were evaluated as the secondary outcomes.

The 5-year overall survival, disease-specific survival, and intestinal preservation rates were 93.6%, 99.6%, and 88.6%, respectively. Patients with CR lesions had no metastatic occurrence, and patients with non-CR lesions had 4 metastatic occurrences. Kaplan-Meier curves revealed that overall survival and disease-specific survival rates were significantly higher in patients with CR lesions than in those with non-CR lesions (P > .001 and P = .009, respectively). Local recurrence occurred in only 8 lesions (0.5%), which were successfully resected by subsequent endoscopic treatment. Multiple logistic regression analyses revealed that piecemeal resection (hazard ratio, 8.19; 95% CI, 1.47-45.7; P = .02) and margin-positive resection (hazard ratio, 8.06; 95% CI, 1.76-37.0; P = .007) were significant independent predictors of local recurrence after colorectal ESD. Fifteen metachronous invasive cancers (1.0%) were identified during surveillance colonoscopy, most of which required surgical resection.

A favorable long-term prognosis indicates that ESD can be the standard treatment for large colorectal epithelial neoplasms.

UMIN000010136.

Background and study aims  A second examination of the right colon, either as a second forward view (SFV) or as retroflexion (RF) in the cecum, can increase adenoma detection rate (ADR) in the right colon. In this meta-analysis, we have evaluated the role of a second examination of the right colon in improving ADR. Methods  We reviewed several databases to identify randomized controlled trials that compared right colon SFV with no SFV, and RCTs that compared SFV with RF in the right colon, and reported data on ADR. Our outcomes of interest were ADR and polyp detection rate (PDR) with SFV vs no SFV, right colon and total withdrawal times, and additional ADR and PDR with SFV vs RF. For categorical variables, we calculated pooled risk ratios (RRs) with 95 % confidence intervals (CIs); for continuous variables, we calculated standardized mean difference (SMD) with 95 % CI. Data were analyzed using random effects model. Results  We included six studies with 3901 patients. Comparing SFV with no SFV, right colon ADR and PDR were significantly higher in the SFV group: ADR (RR [95 % CI] 1.39 [1.22,1.58]) and PDR (RR [95 % CI] 1.47 [1.30, 1.65]). We found no significant difference in right colon withdrawal time (SMD [95 % CI] 1.54 [-0.20,3.28]) or total withdrawal time (SMD (95 % CI) 0.37 [-0.39,1.13]) with and without SFV. We found no significant difference in additional ADR between SFV and RF. Conclusions  SFV of the right colon significantly increases right-sided and overall ADR.

Colorectal cancer is the second commonest cause of cancer death worldwide. Colonoscopy plays a key role in the control of colorectal cancer and, in that regard, maximizing detection (and removal) of pre-cancerous adenomas at colonoscopy is imperative. GI Genius™ (Medtronic Ltd) is a computer-aided detection system that integrates with existing endoscopy systems and improves adenoma detection during colonoscopy. COLO-DETECT aims to assess the clinical and cost effectiveness of GI Genius™ in UK routine colonoscopy practice.

Participants will be recruited from patients attending for colonoscopy at National Health Service sites in England, for clinical symptoms, surveillance or within the national Bowel Cancer Screening Programme. Randomization will involve a 1:1 allocation ratio (GI Genius™-assisted colonoscopy:standard colonoscopy) and will be stratified by age category (<60 years, 60-<74 years, ≥74 years), sex, hospital site and indication for colonoscopy. Demographic data, procedural data, histology and post-procedure patient experience and quality of life will be recorded. COLO-DETECT is designed and powered to detect clinically meaningful differences in mean adenomas per procedure and adenoma detection rate between GI Genius™-assisted colonoscopy and standard colonoscopy groups. The study will close when 1828 participants have had a complete colonoscopy. An economic evaluation will be conducted from the perspective of the National Health Service. A patient and public representative is contributing to all stages of the trial. Registered at ClinicalTrials.gov (NCT04723758) and ISRCTN (10451355). WHAT WILL THIS TRIAL ADD TO THE LITERATURE?: COLO-DETECT will be the first multi-centre randomized controlled trial evaluating GI Genius™ in real world colonoscopy practice and will, uniquely, evaluate both clinical and cost effectiveness.

Colonoscopy aims to early detect and remove precancerous colorectal polyps, thereby preventing development of colorectal cancer (CRC). Recently, computer-aided detection (CADe) systems have been developed to assist endoscopists in polyp detection during colonoscopy. The aim of this study was to investigate feasibility and safety of a novel CADe system during real-time colonoscopy in three European tertiary referral centers.

Ninety patients undergoing colonoscopy assisted by a real-time CADe system (DISCOVERY; Pentax Medical, Tokyo, Japan) were prospectively included. The CADe system was turned on only at withdrawal, and its output was displayed on secondary monitor. To study feasibility, inspection time, polyp detection rate (PDR), adenoma detection rate (ADR), sessile serrated lesion (SSL) detection rate (SDR), and the number of false positives were recorded. To study safety, (severe) adverse events ((S)AEs) were collected. Additionally, user friendliness was rated from 1 (worst) to 10 (best) by endoscopists.

Mean inspection time was 10.8 ± 4.3 min, while PDR was 55.6%, ADR 28.9%, and SDR 11.1%. The CADe system users estimated that < 20 false positives occurred in 81 colonoscopy procedures (90%). No (S)AEs related to the CADe system were observed during the 30-day follow-up period. User friendliness was rated as good, with a median score of 8/10.

Colonoscopy with this novel CADe system in a real-time setting was feasible and safe. Although PDR and SDR were high compared to previous studies with other CADe systems, future randomized controlled trials are needed to confirm these detection rates. The high SDR is of particular interest since interval CRC has been suggested to develop frequently through the serrated neoplasia pathway.

The study was registered in the Dutch Trial Register (reference number: NL8788).

Colonoscopy is a fundamental tool in colorectal cancer (CRC) prevention. Nevertheless, one-fourth of colorectal neoplasms are still missed during colonoscopy, potentially being the main reason for post-colonoscopy colorectal cancer (PCCRC). Adenoma detection rate (ADR) is currently known as the best quality indicator correlating with PCCRC incidence.

We performed a literature review in order to summarize evidences investigating key factors affecting ADR: endoscopists education and training, patient management, endoscopic techniques, improved navigation (exposition defect), and enhanced lesions recognition (vision defect) were considered.

'Traditional' factors, such as split dose bowel preparation, adequate withdrawal time, and right colon second view, held a significant impact on ADR. Several devices and technologies have been developed to promote high-quality colonoscopy, however artificial intelligence may be considered the most promising tool for ADR improvement, provided that endoscopists education and recording are guaranteed.

Mucosal exposure devices including distal attachments such as the cuff and cap have shown variable results in improving adenoma detection rate (ADR) compared with high-definition white light colonoscopy (HDWLE).

We performed a prospective, multicenter randomized controlled trial in patients undergoing screening or surveillance colonoscopy comparing HDWLE to 2 different types of distal attachments: cuff (CF) (Endocuff Vision) or cap (CP) (Reveal). The primary outcome was ADR. Secondary outcomes included adenomas per colonoscopy, advanced adenoma and sessile serrated lesion detection rate, right-sided ADR, withdrawal time, and adverse events. Continuous variables were compared using Student's t test and categorical variables were compared using chi-square or Fisher's exact test using statistical software Stata version16. A P value <.05 was considered significant.

A total of 1203 subjects were randomized to either HDWLE (n = 384; mean 62 years of age; 81.3% males), CF (n = 379; mean 62.7 years of age; 79.9% males) or CP (n = 379; mean age 62.1 years of age; 80.5% males). No significant differences were found among 3 groups for ADR (57.3%, 59.1%, and 55.7%; P = .6), adenomas per colonoscopy (1.4 ± 1.9, 1.6 ± 2.4, and 1.4 ± 2; P = .3), advanced adenoma (7.6%, 9.2%, and 8.2%; P = .7), sessile serrated lesion (6.8%, 6.3%, and 5.5%; P = .8), or right ADR (48.2%, 49.3%, and 46.2%; P = .7). The number of polyps per colonoscopy were significantly higher in the CF group compared with HDWLE and CP group (2.7 ± 3.4, 2.3 ± 2.5, and 2.2 ± 2.3; P = .013). In a multivariable model, after adjusting for age, sex, body mass index, withdrawal time, and Boston Bowel Preparation Scale score, there was no impact of device type on the primary outcome of ADR (P = .77). In screening patients, CF resulted in more neoplasms per colonoscopy (CF: 1.7 ± 2.6, HDWLE: 1.3 ± 1.7, and CP: 1.2 ± 1.8; P = .047) with a shorter withdrawal time.

Results from this multicenter randomized controlled trial do not show any significant benefit of using either distal attachment devices (CF or CP) over HDWLE, at least in high-detector endoscopists. The Endocuff may have an advantage in the screening population. (ClinicalTrials.gov, Number: NCT03952611).

Despite recent advances in endoscopic technology adenoma miss rate still is up to 20% contributing to interval cancers. Improved imaging modalities have been introduced to increase adenoma detection rate (ADR). Recently, narrow-band imaging (NBI) (Exera II series, Olympus Corporation) was not significantly better than high-definition white light colonoscopy (HD-WLC). An improved second generation of NBI (190-NBI) is characterized by better illumination of the bowel lumen and may be associated with a higher ADR.

We performed a prospective randomized study on patients referred to the Jena University Hospital for screening or surveillance colonoscopy between January 2015 and April 2017. Participating endoscopists were divided into 2 subgroups depending on their individual experience. Colonoscopy was performed by use of HD-WLC or 190-NBI upon withdrawal.

Five hundred fifty-three patients participated in the study. Eighty patients were excluded (insufficient bowel cleansing [n = 34], anticoagulation precluding polypectomy [n=15], partial colonic resection [n=9], other reasons [n = 22]). Mean age was 66.9 years (standard deviation 10.3 years), and 253 patients were male (53.5%). Bowel preparation and withdrawal time were not different. ADR among all subgroups was 39.4% using HD-WLC, but only 29.1% were using 190-NBI (P = .02). Number of polyps per patient was lower using 190-NBI than with HD-WLC (0.58 vs 0.86; P = .02). Subgroup analysis revealed that 190-NBI was inferior to HD-WLC only in unexperienced endoscopists.

In our stud,y ADR was lower by use of 190-NBI. These differences persisted only in unexperienced investigators. 190-NBI seems to be more challenging regarding ADR, requiring more intensive training prior to implementing this technology in daily clinical care.

ClinicalTrials.gov (identifier: NCT03081975).

The use of forceps for removal of nondiminutive polyps is associated with incomplete resection compared with snare polypectomy. However, few studies have characterized the frequency of forceps polypectomy for nondiminutive polyps or identified strategies to improve this practice. To address this gap, we estimated the prevalence and predictors of forceps polypectomy in clinical practice and examined the effectiveness of a multicomponent intervention to reduce inappropriate forceps polypectomy.

We retrospectively reviewed all colonoscopies with polypectomies performed at 2 U.S. health systems between October 1, 2017, and September 30, 2019. We used a mixed-effects logistic regression model to examine the effect of a multicomponent intervention, including provider education and a financial incentive, to reduce inappropriate forceps polypectomy, defined as use of forceps polypectomy for polyps ≥5 mm.

A total of 9968 colonoscopies with 25,534 polypectomies were performed by 42 gastroenterologists during the study period. Overall, 8.5% (n = 2176) of polyps were removed with inappropriate forceps polypectomy. Inappropriate forceps polypectomy significantly decreased after the intervention (odds ratio [OR], 0.34, 95% confidence interval [CI], 0.30-0.39), from 11.4% (n = 1539) to 5.3% (n = 637). Predictors of inappropriate forceps polypectomy included inadequate bowel prep (OR, 1.25; 95% CI, 1.06-1.47), polyps in the right colon (vs left: OR, 1.29; 95% CI, 1.09-1.51), and number of polyps removed (OR, 0.96; 95% CI, 0.94-0.97). Inappropriate forceps polypectomy also varied by gastroenterologist (median OR, 3.43). In a post hoc analysis, the proportion of polyps >2 mm removed with forceps decreased from 50.0% before the intervention to 43.0% after it (OR, 0.62; 95% CI, 0.58-0.68).

Inappropriate forceps polypectomy is common but modifiable. The proportion of nondiminutive polyps removed with forceps polypectomy should be considered as a quality measure.

Measuring the adenoma detection rate (ADR) is critical to providing quality care, however it is also challenging. We aimed to develop a tool using pre-existing electronic health record (EHR) functions to accurately and easily measure total ADR and to provide real-time feedback for endoscopists. We utilized the Epic EHR. With the help of an Epic analyst, using existing tools, we developed a method by which endoscopy staff could mark whether an adenoma was detected for a given colonoscopy. Using these responses and all colonoscopies performed by the endoscopist recorded in the EHR, ADR was calculated in a report and displayed to endoscopists within the EHR. One endoscopist piloted the tool, and results of the tool were validated against a manual chart review. Over the pilot period the endoscopist performed 145 colonoscopies, of which 78 had adenomas. The tool correctly identified 76/78 colonoscopies with an adenoma and 67/67 of colonoscopies with no adenomas (97.4% sensitivity, 100% specificity, 98% accuracy). There was no difference in ADR as determined by the tool compared to manual review (53.1% vs. 53.8%, p = 0.912). We successfully developed and pilot tested a tool to measure ADR using existing EHR functionality.

During the first wave of the COVID-19 pandemic in 2020, elective gastrointestinal endoscopy services were abbreviated for fear of viral transmission. However, urgent suspected colorectal cancer (CRC) referrals continued. Serendipitously, a national study suggested that a new faecal immunochemical test (FIT) might be helpful in triaging patients with colorectal alarm symptoms.

This was a single centre observational study of patients referred using NG12 criteria between March and August 2020. Patients were triaged to the urgent cancer pathway for FIT ≥ 10 μg/g and investigated using the latest National Health Service England guidance. Demographic data, method of investigations, cancer and polyp detection rates were compared to patients referred in the 6 months prior to the use of FIT as a triage tool.

In all, 1192 patients (median age 70) were referred using NG12 guidelines during the pandemic period, compared with 1592 patients (median age 72) in the previous 6 months. CRC detection was similar in both groups (n = 45, 2.8% vs. n = 38, 3.5%; P = 0.248). Two patients with a negative FIT (0.36%) had CRC. Using FIT as a triage tool resulted in a significant reduction in the use of endoscopy (n = 477, 43.6% vs. n = 1186, 74.5%; P > 0.001) with a significant increase in CT scanning (n = 696, 63.6% vs. n = 750, 47.1%; P < 0.001).

The use of FIT in NG12 patients triaged during the first wave of the COVID-19 pandemic reduced endoscopy but not CT scanning and did not compromise CRC detection rates. It is a safe method that aids in reducing the burden on services greatly. A negative FIT test does not absolutely exclude CRC.

Recent studies have shown that artificial intelligence-based computer-aided detection systems possess great potential in reducing the heterogeneous performance of doctors during endoscopy. However, most existing studies are based on high-quality static images available in open-source databases with relatively small data volumes, and, hence, are not applicable for routine clinical practice. This research aims to integrate multiple deep learning algorithms and develop a system (DeFrame) that can be used to accurately detect intestinal polyps in real time during clinical endoscopy.

A total of 681 colonoscopy videos were collected for retrospective analysis at Xiangya Hospital of Central South University from June 2019 to June 2020. To train the machine learning (ML)-based system, 6,833 images were extracted from 48 collected videos, and 1,544 images were collected from public datasets. The DeFrame system was further validated with two datasets, consisting of 24,486 images extracted from 176 collected videos and 12,283 images extracted from 259 collected videos. The remaining 198 collected full-length videos were used for the final test of the system. The measurement metrics were sensitivity and specificity in validation dataset 1, precision, recall and F1 score in validation dataset 2, and the overall performance when tested in the complete video perspective.

A sensitivity and specificity of 79.54 and 95.83%, respectively, was obtained for the DeFrame system for detecting intestinal polyps. The recall and precision of the system for polyp detection were determined to be 95.43 and 92.12%, respectively. When tested using full colonoscopy videos, the system achieved a recall of 100% and precision of 80.80%.

We have developed a fast, accurate, and reliable DeFrame system for detecting polyps, which, to some extent, is feasible for use in routine clinical practice.

There are limited data regarding the safety and efficacy of cold snare polypectomy (CSP) for large colorectal polyps. We evaluated factors affecting the clinical outcomes of CSP for polyps between 5 and 15 mm in size.

This was a prospective single-center observational study involving 1000 patients undergoing colonoscopy. Polyps (5-15 mm) were removed using CSP, and biopsies were taken from the resection margin. The primary outcome was the incomplete resection rate (IRR), and was determined by the presence of residual neoplasia on biopsy. Correlations between IRR and polyp size, morphology, histology, and resection time were assessed by generalized estimating equation model.

A total of 440 neoplastic polyps were removed from 261 patients. The overall IRR was 2.27%, 1.98% for small (5-9 mm) vs 3.45% for large (10-15 mm) polyps (P = .411). In univariate analysis, the IRR was more likely to be related to sessile serrated lesions (odds ratio [OR], 6.93; 95% confidence interval [CI], 1.88-25.45; P = .004), piecemeal resection (OR, 11.83; 95% CI, 1.20-116.49; P = .034), and prolonged resection time >60 seconds (OR, 7.56; 95% CI, 1.75-32.69; P = .007). In multivariable regression analysis, sessile serrated lesions (OR, 6.45; 95% CI, 1.48-28.03; P = .013) and resection time (OR, 7.39; 95% CI, 1.48-36.96; P = .015, respectively) were independent risk factors for IRR. Immediate bleeding was more frequent with resection of large polyps (6.90% vs 1.42%; P = .003). No recurrence was seen on follow-up colonoscopy in 37 cases with large polyps.

CSP is safe and effective for removal of colorectal polyps up to 15 mm in size, with a low IRR. (ClinicalTrials.gov; Number: NCT03647176).

Colorectal cancer is the third most common cancer worldwide. Because of the slow progression of the precancerous precursors, an efficient endoscopic surveillance strategy may be expected. It seems that around one-fourth of colorectal malignancies are still missed during colonoscopy. Several endoscopic technologies have been introduced, without radical changes. Interest in the development of artificial intelligence applications in the medical field has grown in the past decade. Artificial intelligence can help to highlight a specific region of interest that needs closer examination for the identification of polyps. The aim of this review is to report the first clinical experiences with the first US FDA-approved, real-time, deep-learning, computer-aided detection system (GI Genius™, Medtronic).

Colonoscopy is the reference standard procedure for the prevention and diagnosis of colorectal cancer, which is a leading cause of cancer-related deaths in Singapore. Artificial intelligence systems are automated, objective and reproducible. Artificial intelligence-assisted colonoscopy has recently been introduced into clinical practice as a clinical decision support tool. This review article provides a summary of the current published data and discusses ongoing research and current clinical applications of artificial intelligence-assisted colonoscopy.

Polyps are precursors to colorectal cancer, the third most common cancer in the United States. Large polyps, i.e.,, those with a size ≥ 20 mm, are more likely to harbor cancer. Colonic polyps can be removed through various techniques, with the goal to completely resect and prevent colorectal cancer; however, the management of large polyps can be relatively complex and challenging. Such polyps are generally more difficult to remove en bloc with conventional methods, and depending on level of expertise, may consequently be resected piecemeal, leading to an increased rate of incomplete removal and thus polyp recurrence. To effectively manage large polyps, endoscopists should be able to: (1) Evaluate the polyp for characteristics which predict high difficulty of resection or incomplete removal; (2) Determine the optimal resection technique (e.g., snare polypectomy, endoscopic mucosal resection, endoscopic submucosal dissection, etc.); and (3) Recognize when to refer to colleagues with greater expertise. This review covers important considerations in this regard for referring and receiving endoscopists and methods to best manage large colonic polyps.

The use of artificial intelligence represents an objective approach to increase endoscopist's adenoma detection rate (ADR) and limit interoperator variability. In this study, we evaluated a newly developed deep convolutional neural network (DCNN) for automated detection of colorectal polyps ex vivo as well as in a first in-human trial.

For training of the DCNN, 116 529 colonoscopy images from 278 patients with 788 different polyps were collected. A subset of 10 467 images containing 504 different polyps were manually annotated and treated as the gold standard. An independent set of 45 videos consisting of 15 534 single frames was used for ex vivo performance testing. In vivo real-time detection of colorectal polyps during routine colonoscopy by the DCNN was tested in 42 patients in a back-to-back approach.

When analyzing the test set of 15 534 single frames, the DCNN's sensitivity and specificity for polyp detection and localization within the frame was 90% and 80%, respectively, with an area under the curve of 0.92. In vivo, baseline polyp detection rate and ADR were 38% and 26% and significantly increased to 50% (P = 0.023) and 36% (P = 0.044), respectively, with the use of the DCNN. Of the 13 additionally with the DCNN detected lesions, the majority were diminutive and flat, among them three sessile serrated adenomas.

This newly developed DCNN enables highly sensitive automated detection of colorectal polyps both ex vivo and during first in-human clinical testing and could potentially increase the detection of colorectal polyps during colonoscopy.