Optical Diagnosis Of Diminutive Polyps Research Articles

Computer-aided diagnosis for the resect-and-discard strategy for colorectal polyps: a systematic review and meta-analysis

The resect-and-discard strategy allows endoscopists to replace post-polypectomy pathology with real-time prediction of polyp histology during colonoscopy (optical diagnosis). We aimed to investigate the benefits and harms of implementing computer-aided diagnosis (CADx) for polyp pathology into the resect-and-discard strategy. In this systematic review and meta-analysis, we searched MEDLINE, Embase, and Scopus from database inception to June 5, 2024, without language restrictions, for diagnostic accuracy studies that assessed the performance of real-time CADx systems, compared with histology, for the optical diagnosis of diminutive polyps (≤5 mm) in the entire colon. We synthesised data for three strategies: CADx-alone, CADx-unassisted, and CADx-assisted; when the endoscopist was involved in the optical diagnosis, we synthesised data exclusively from diagnoses for which confidence in the prediction was reported as high. The primary outcomes were the proportion of polyps that would have avoided pathological assessment (ie, the proportion optically diagnosed with high confidence; main benefit) and the proportion of polyps incorrectly predicted due to false positives and false negatives (main harm), directly compared between CADx-assisted and CADx-unassisted strategies. We used DerSimonian and Laird's random-effects model to calculate all outcomes. We used Higgins I2 to assess heterogeneity, the Grading of Recommendations, Assessment, Development, and Evaluation approach to rate certainty, and funnel plots and Egger's test to examine publication bias. This study is registered with PROSPERO, CRD42024508440. We found 1019 studies, of which 11 (7400 diminutive polyps, 3769 patients, and 185 endoscopists) were included in the final meta-analysis. Three studies (1817 patients and 4086 polyps [2148 neoplastic and 1938 non-neoplastic]) provided data to directly compare the primary outcome measures between the CADx-unassisted and CADx-assisted strategies. We found no significant difference between the CADx-assisted and CADx-unassisted strategies for the proportion of polyps that would have avoided pathological assessment (90% [88-93], 3653 [89·4%] of 4086 polyps diagnosed with high confidence vs 90% [95% CI 85-94], 3588 [87·8%] of 4086 polyps diagnosed with high confidence; risk ratio 1·01 [95% CI 0·99-1·04; I2=53·49%; low-certainty evidence; Egger's test p=0·18). The proportion of incorrectly predicted polyps was lower with the CADx-assisted strategy than with the CADx-unassisted strategy (12% [95% CI 7-17], 523 [14·3%] of 3653 polyps incorrectly predicted with a CADx-assisted strategy vs 13% [6-20], 582 [16·2%] of 3588 polyps incorrectly diagnosed with a CADx-unassisted strategy; risk ratio 0·88 [95% CI 0·79-0·98]; I2=0·00%; low-certainty evidence; Egger's test p=0·18). CADx did not produce benefit nor harm for the resect-and-discard strategy, questioning its value in clinical practice. Improving the accuracy and explainability of CADx is desired. European Commission (Horizon Europe), the Japan Society of Promotion of Science, and Associazione Italiana per la Ricerca sul Cancro.

Autonomous Artificial Intelligence vs Artificial Intelligence–Assisted Human Optical Diagnosis of Colorectal Polyps: A Randomized Controlled Trial

BackgroundArtificial intelligence (AI)-based optical diagnosis systems (CADx) have been developed to allow pathology prediction of colorectal polyps during colonoscopies. However, CADx systems have not yet been validated for autonomous performance. Therefore, we conducted a trial comparing Autonomous AI to AI assisted human (AI-H) optical diagnosis. MethodsWe performed a randomized non-inferiority trial of patients undergoing elective colonoscopies in one academic institution. Patients were randomized int: 1) Autonomous AI-based CADx optical diagnosis of diminutive polyps without human input; 2) endoscopists performed optical diagnosis of diminutive polyps after seeing the real-time CADx diagnosis. Primary outcome was accuracy in optical diagnosis in both arms using pathology as gold standard. Secondary outcomes included agreement with pathology for surveillance intervals. Results467 patients were randomized (238 patients/158 polyps in the Autonomous AI group; 229 patients/179 polyps in the AI-H group). Accuracy for optical diagnosis was 77.2% (95%Confidence Interval 69.7-84.7) in the Autonomous AI group and 72.1% (95%CI 65.5-78.6) in the AI-H group (p=0.86). For high confidence diagnoses, accuracy for optical diagnosis was 77.2% (95%CI 69.7-84.7) in the Autonomous AI group and 75.5% (95%CI 67.9-82.0) in the AI-H group. Autonomous AI had statistically significantly higher agreement with pathology-based surveillance intervals compared to AI-H (91.5% [95%CI 86.9-96.1] vs 82.1% [95%CI 76.5-87.7]; p=0.016). ConclusionsAutonomous AI-based optical diagnosis exhibits non-inferior accuracy to endoscopist-based diagnosis. Both Autonomous AI and AI-H exhibited relatively low accuracy for optical diagnosis, however Autonomous AI achieved higher agreement with pathology-based surveillance intervals.

Guidance for setting easy-to-adopt competence criteria for optical diagnosis of diminutive colorectal polyps: a simulation approach

One reason the optical diagnosis strategy for diminutive colorectal polyps has not yet been implemented is that the current competence criteria (Preservation and Incorporation of Valuable Endoscopic Innovation [PIVI] initiative) are difficult to use in daily practice. To provide guidance for setting alternative easy-to-adopt competence criteria, we determined the lowest proportion of diminutive polyps that should have a correct optical diagnosis to meet the PIVI. For this simulation study, we used datasets from 2 prospectively collected cohorts of patients who underwent colonoscopy in either a primary colonoscopy or fecal immunochemical test (FIT) screening setting. In the simulation approach, virtual endoscopists or computer-aided diagnosis systems performed optical diagnosis of diminutive polyps with a fixed diagnostic performance level (strategy) on all individuals in the cohort who had≥1 diminutive polyp. Strategies were defined by systematically varying the proportion of correct optical diagnoses for each polyp subtype (ie, adenomas, hyperplastic polyps, sessile serrated lesions). For each strategy, we determined whether PIVI-1 (≥90% agreement with U.S. or European Society for Gastrointestinal Endoscopy [ESGE] surveillance guidelines) and PIVI-2 (≥90% negative predictive value [NPV] for neoplastic lesions in the rectosigmoid) were met using Monte Carlo sampling with 1000 repetitions, with histology as reference. The level of overall diagnostic accuracy to achieve the PIVI differed significantly depending on the clinical setting and guidelines used. In the colonoscopy screening setting, all diagnostic strategies in which 92% of all diminutive polyps (regardless of histology) were diagnosed correctly led to 90% or more agreement with U.S. surveillance intervals (ie, PIVI-1). For all diagnostic strategies in which≥89% of all diminutive polyps were correctly diagnosed, at least 90% NPV was achieved (ie, PIVI-2). For the FIT screening setting, values were respectively≥77% and≥94%. When using ESGE guidelines, PIVI-1 was in both settings already met when 40% of all diminutive polyps were diagnosed correctly. In contrast to the fixed PIVI criteria, our simulation study shows that different thresholds for the proportion of correctly diagnosed diminutive polyps lead to different clinical consequences depending on guidelines and clinical setting. However, this target proportion of diminutive colorectal polyps correctly diagnosed with optical diagnosis represents easier-to-adopt competence criteria.

Optical diagnosis of diminutive polyps in the Dutch Bowel Cancer Screening Program: Are we ready to start?

Background and study aims Implementation of optical diagnosis of diminutive polyps may potentially increase the efficacy and cost-effectiveness of colonoscopies. To adopt such strategy in clinical practice, the Preservation and Incorporation of Valuable endoscopic Innovations (PIVI) thresholds provide the basis to be met: ≥ 90 % negative predictive value (NPV) for diagnosis of adenomatous histology and ≥ 90 % agreement on surveillance intervals. We evaluated this within the Dutch Bowel Cancer Screening Program (BCSP). Patients and methods Endoscopic and histological data were collected from participants of the national bowel cancer screening program with an unfavorable fecal immunochemical test referred for colonoscopy between February 2014 and August 2015 at four endoscopy centers. The “resect and discard” scenario was studied, resecting diminutive polyps without histological evaluation. Agreement between optical diagnosis and histological diagnosis was measured for surveillance intervals according to Dutch, European and American post-polypectomy surveillance guideline. Results Fifteen certified endoscopists participated in this study and included 3028 diminutive polyps. In 2,330 patients both optical and histological diagnosis were available. Optical diagnosis of diminutive polyps showed NPV of 84 % (95 % CI 80–87) for adenomatous histology in the rectosigmoid. Applying the ‘resect and discard’ strategy resulted in 90.6 %, 91.2 %, 90.9 % agreement on surveillance intervals for the Dutch, European and American guideline respectively. Conclusion Our data representing current clinical practice in the Dutch BCSP practice on optical diagnosis of diminutive polyps showed that accuracy of predicting histology remains challenging, and risk of incorrect optical diagnosis is still significant. Therefore, it is too early to safely implement these strategies.

Optical Diagnosis of Colorectal Polyps: Recent Developments.

Optical diagnosis of diminutive colorectal polyps has been recently proposed as an alternative to histopathologic diagnosis. Recent developments in imaging techniques, new classification systems, and the use of artificial intelligence have allowed for increased viability of optical diagnosis. This review provides an up-to-date overview of optical diagnosis recommendations, classifications, outcomes, and recent developments. There are currently seven major classification systems and three major society recommendations for quality benchmarks for optical diagnosis of diminutive polyps. The NICE classification has been extensively studied and meets quality benchmarks for most imaging techniques but does not allow for the diagnosis of sessile serrated polyps (SSPs). The SIMPLE classification has met quality benchmarks for NBI and i-Scan and allows for the diagnosis of SSPs. Other classification systems need to be further studied to validate effectiveness. Computer-assisted diagnosis of colorectal polyps is a very promising recent development with first studies showing that society-recommended quality benchmarks for real-time colonoscopies on patients are being met. Limitations include a non-negligible percentage of failure to diagnose, low specificity, and low number of real-time diagnostic studies. More research needs to be performed to further understand the value of artificial intelligence for optical polyp diagnosis. Optical diagnosis of diminutive colorectal polyps is currently a viable strategy for experienced endoscopists using validated classifications and imaging-enhanced endoscopy. Artificial intelligence-based diagnosis could make optical diagnosis widely applicable but is currently in its early developmental stage.

Self-Formation Assessed by Cumulative Summation Test Does Not Reach Recommended Thresholds for Optical Diagnosis of Colorectal Polyps ≤ 7mm.

Accurate optical diagnosis of diminutive polyps would allow implementing a resect and discard strategy. We evaluated the learning curve of a single training session followed by self-education in subjects with no endoscopic experience. Learning curves were evaluated in 38 subjects employing learning curve-cumulative summation (LC-CUSUM) tests, with each participant attending one training session regarding narrow band imaging and optical diagnosis and then individually assessing 100 lesions, receiving feedback after each diagnosis. Diagnostic accuracy was subsequently evaluated in 180 patients with lesions ≤ 7mm. Evaluators predicted each polyp's histology and recommended a surveillance interval. Determinants of accuracy were explored using regression analysis. According to the LC-CUSUM curve, 20 evaluators (52.6%) reached diagnostic competence after 57 lesions (IQR 55-76.5). During the diagnostic performance assessment, 11,666 diagnoses and 6840 follow-up recommendations were generated. Considering high confidence diagnoses, accuracy was 81.3% (80.5-82.1%), negative predictive value (NPV) for rectosigmoid adenomas 78.6% (76.4-80.6%), and sensitivity for adenomas 86.6% (85.8-87.4%). Two (5.3%) evaluators reached a ≥ 90% accuracy, 3 (7.9%) presented a NPV for rectosigmoid adenomas ≥ 90%, and 18 (47.4%) a sensitivity for adenomas ≥ 90%. Multivariable logistic regression showed high confidence and size ≥ 5mm as the strongest predictors of accuracy. Fifteen (39.5%) evaluators recommended a correct or reduced follow-up interval in over 90% of subjects. Self-formation after a single training session did not allow most evaluators to reach the required accuracy. LC-CUSUM tests did not identify competent evaluators. Despite these results, 86.7% of follow-up intervals would have been corrected or reduced.

Effects of Training and Feedback on Accuracy of Predicting Rectosigmoid Neoplastic Lesions and Selection of Surveillance Intervals by Endoscopists Performing Optical Diagnosis of Diminutive Polyps

Real-time differentiation of diminutive polyps (1-5 mm) during endoscopy could replace histopathology analysis. According to guidelines, implementation of optical diagnosis into routine practice would require it to identify rectosigmoid neoplastic lesions with a negative predictive value (NPV) of more than 90%, using histologic findings as a reference, and agreement with histology-based surveillance intervals for more than 90% of cases. We performed a prospective study with 39 endoscopists accredited to perform colonoscopies on participants with positive results from fecal immunochemical tests in the Bowel Cancer Screening Program at 13 centers in the Netherlands. Endoscopists were trained in optical diagnosis using a validated module (Workgroup serrAted polypS and Polyposis). After meeting predefined performance thresholds in the training program, the endoscopists started a 1-year program (continuation phase) in which they performed narrow bandimaging analyses during colonoscopies of participants in the screening program and predicted histological findings with confidence levels. The endoscopists were randomly assigned to groups that received feedback or no feedback on the accuracy of their predictions. Primary outcome measures were endoscopists' abilities to identify rectosigmoid neoplastic lesions (using histology as a reference) with NPVs of 90% or more, and selecting surveillance intervals that agreed with those determined by histology for at least 90% of cases. Of 39 endoscopists initially trained, 27 (69%) completed the training program. During the continuation phase, these 27 endoscopists performed 3144 colonoscopies in which 4504 diminutive polyps were removed. The endoscopists identified neoplastic lesions with a pooled NPV of 90.8% (95% confidence interval 88.6-92.6); their proposed surveillance intervals agreed with those determined by histologic analysis for 95.4% of cases (95% confidence interval 94.0-96.6). Findings did not differ between the group that did vs did not receive feedback. Sixteen endoscopists (59%) identified rectosigmoid neoplastic lesions with NPVs greater than 90% and selected surveillance intervals in agreement with those determined from histology for more than 90% of patients. In a prospective study following a validated training module, we found that a selected group of endoscopists identified rectosigmoid neoplastic lesions with pooled NPVs greater than 90% and accurately selected surveillance intervals for more than 90% of patients over the course of 1 year. Providing regular interim feedback on the accuracy of neoplastic lesion prediction and surveillance interval selection did not lead to differences in those endpoints. Monitoring is suggested, as individual performance varied. ClinicalTrials.gov no: NCT02516748; Netherland Trial Register: NTR4635.

Optical Diagnosis of Sessile Serrated Polyps: Bottleneck for the Optical Diagnosis Paradigm?

Optical diagnosis of diminutive (1 to 5 mm) polyps could result in a more cost-effective colonoscopy practice. Previous optical diagnosis studies did not incorporate the differentiation of sessile serrated polyps (SSPs). This study aimed to evaluate the impact of optical diagnosis of diminutive SSPs on the overall performance of endoscopic polyp differentiation in daily colonoscopy practice. Endoscopy data were prospectively collected between 2011 and 2014 in a colonoscopy center. Each endoscopist reported a real-time optical diagnosis (SSP, adenoma or hyperplastic polyp) for all lesions in a structured colonoscopy reporting system, using narrow band imaging at their discretion. Study outcomes were accuracy of optical diagnosis, surveillance interval agreement and negative predictive value for diminutive rectosigmoid neoplastic histology based on the optical diagnosis of diminutive polyps compared to histopathology. Of 2853 removed diminutive polyps, 202 (7.1%) were histologically proven SSPs. Optical diagnosis of diminutive SSPs was accurate in 24.4%. Diminutive SSPs determined 6.9% of postpolypectomy surveillance assignments. Inaccurate optical diagnosis of diminutive SSPs led to lower surveillance interval agreement (78.1% vs. 53.3%, P<0.01) and pooled negative predictive value per polyp (84.3% vs. 50.0%; P<0.01) in patients with diminutive SSPs when compared to patients without diminutive SSPs. Accurate endoscopic identification of diminutive SSPs improved from 0% in 2011 to 47% in 2014 (P=0.02). Endoscopic characterization of diminutive SSPs is difficult, impairing overall performance of optical diagnosis in patients with diminutive SSPs. Future optical diagnosis studies should use validated trainings and classification algorithms that include differentiation of SSPs.

Quality of optical diagnosis of diminutive polyps and associated factors

The aim of the study was to identify endoscopist-related and procedural factors that may be associated with the quality of optical diagnosis of diminutive polyps using narrow-band imaging (NBI). All subjects who participated in a randomized trial on cap-assisted colonoscopy were eligible for the current study. Optical polyp diagnosis was an a priori outcome of the initial trial. Ten participating endoscopists used NBI to assess all of the diagnosed polyps as adenomatous or non-adenomatous in real-time and provided a degree of diagnostic certainty. The main outcome measures were quality benchmarks of optical diagnosis (negative predictive value [NPV] for diminutive rectosigmoid adenomas, agreement with pathology-based surveillance interval) and assessment of endoscopist-related and procedural factors potentially associated with the quality of optical diagnosis. A total of 1650 polyps were found in 607 patients, with 1311 polyps (79 %) being diminutive, of which 672 (53 %) were adenomatous. The NPV of optical diagnosis for rectosigmoid adenomas was 95 %. The optical diagnosis-based surveillance interval agreed with the pathology-based recommendation in 93 % of patients. Prior experience with image-enhanced endoscopy had no effect on optical diagnosis. Low and high adenoma detectors were not different in achieving the quality benchmarks. Cap-assisted colonoscopy was not associated with quality of optical diagnosis. Quality metrics of optical diagnosis remained similar during the first and second half of the study period. High quality optical diagnosis of diminutive polyps can be achieved and sustained by endoscopists previously inexperienced in this practice with minimal training. None of the examined factors appear to affect the quality of optical diagnosis; particularly, endoscopists' adenoma detection was not associated with optical diagnosis.

110 Incorporating Sessile Serrated Polyps in Optical Diagnosis of Diminutive Polyps: What Are the Implications for the PIVI Thresholds?

110 Incorporating Sessile Serrated Polyps in Optical Diagnosis of Diminutive Polyps: What Are the Implications for the PIVI Thresholds?

Optical Diagnosis of Small Colorectal Polyp Histology with High-Definition Colonoscopy Using Narrow Band Imaging

Optical diagnosis of polyp histology can potentially result in enormous cost savings by way of the "resect and discard" strategy for diminutive polyps and the "do not resect" strategy for diminutive hyperplastic polyps in the distal colon. Narrow Band Imaging (NBI) highlights the surface mucosal and vascular pattern on polyps and has been shown to accurately characterize adenomatous and hyperplastic polyps by experts. However, the results have been a little discouraging amongst lesser experienced endoscopists. Studies have also shown that using the NBI diagnosis of diminutive polyp histology, experts can accurately define the future surveillance colonoscopy intervals. However nonexperts in academic or community setting have as yet failed to achieve the recommended thresholds. The subjectivity in assessment by endoscopists leads to the variable accuracy rates and can be circumvented by computer based automated tools. Although initial experience with a few computer based algorithms have shown accuracies comparable to experts, further refinement and validation will be required before these can be implemented in clinical practice. Incorporation of optical diagnosis of diminutive polyps into clinical practice is bound to face several hurdles. But the potential for enormous cost saving makes it an attractive strategy that can make colonoscopy more cost effective.