Quality indicators for capsule endoscopy and deep enteroscopy

Performance measures for small-bowel endoscopy: a European Society of Gastrointestinal Endoscopy (ESGE) Quality Improvement Initiative - Update 2025

Quality markers and patient experience are being implemented to ensure standardization of practice across gastrointestinal (GI) endoscopy procedures. The set benchmarks ensure high quality procedures are delivered and linked to measurable outcomes.There has been an increase in the demand for small-bowel endoscopy. In 2019, the European Society of Gastrointestinal Endoscopy (ESGE) embarked on setting performance measures for small-bowel endoscopy. This included major (key) and minor performance indicators for both small-bowel capsule endoscopy (SBCE) and device-assisted enteroscopy (DAE). These suggested quality indicators cover all procedure domains, from patient selection and preparation, to intraprocedural aspects such as pathology identification, appropriate management, the patient experience, and post-procedure complications. Since 2019, there has been an increase in published studies looking at different aspects of small-bowel endoscopy, including real-world data. This paper provides an update on the 2019 performance measures, considering the latest literature.

Assessing the Impact of a Structured Capsule Endoscopy Training Program Using a New Validated Assessment Tool

BACKGROUND AND AIM

We aimed to develop and validate a simple capsule endoscopy (CE) training assessment tool, the Capsule Endoscopy Training Assessment (CETA), and prospectively use it to analyze the learning progression achieved by participants in our CE training program.

METHODS

Over a 3-year period, all participants in our CE training program completed pre-training and post-training CETA, ranging between 0% and 100%, and encompassing theoretical questions and interpretation of segmented CE videos. We compared the mean differences in overall, theoretical, and practical pre-training and post-training CETA, and assessed the influence of previous endoscopic experience (upper gastrointestinal endoscopy [UGE], colonoscopy, device-assisted enteroscopy [DAE] and CE) using generalized linear models.

RESULTS

Fifty-seven participants were included. After training, there was a significant increase in participants' overall (mean difference, 26.3; 95% confidence interval [CI], 20.70 to 31.83), theoretical (mean difference, 27.2; 95% CI, 19.81 to 34.57), and practical (mean difference, 25.9; 95% CI, 20.09 to 31.63) CETA components. Compared to those without experience, participants with previous endoscopic experience demonstrated a smaller increase in overall CETA after training (UGE, rate ratio, 0.76; 95% CI, 0.63 to 0.91; colonoscopy (rate ratio, 0.80; 95% CI, 0.67 to 0.95; DAE (rate ratio, 0.84; 95% CI, 0.73 to 0.97; CE, rate ratio, 0.81; 95% CI, 0.72 to 0.92, respectively).

CONCLUSION

CETA is a valid and useful tool in assessing the learning progression achieved by participants following the CE training program. We demonstrated a significant improvement in participants' CETA after training, being the least experienced participants in endoscopic procedures who benefited the most from CE training.

Endoscopic Approaches for Managing Small Intestinal Disease

The endoscopic evaluation and management of small intestinal diseases continue to evolve and expand. The advent of small bowel wireless capsule endoscopy and deep enteroscopy with either a double- or single-balloon enteroscope now allows complete endoscopic visualization of the entire small intestine and enables access for endoscopic interventions such as biopsies or hemostasis for most of the small bowel. New endoscopic techniques are available to treat proximal malignant small bowel obstruction, including intraluminal stents and endoscopic gastrojejunal stents. Emerging technologies also aim to improve weight loss and diabetes management via small bowel endoscopic interventions.

From guidelines to clinical practice in care for ischaemic stroke patients: A systematic review and expert opinion

BACKGROUND AND PURPOSE

Guidelines help physicians to provide optimal care for stroke patients, but implementation is challenging due to the quantity of recommendations. Therefore a practical overview related to applicability of recommendations can be of assistance.

METHODS

A systematic review was performed on ischaemic stroke guidelines published in scientific journals, covering the whole acute care process for patients with ischaemic stroke. After data extraction, experts rated the recommendations on dimensions of applicability, that is, actionability, feasibility and validity, on a 9-point Likert scale. Agreement was defined as a score of ≥8 by ≥80% of the experts.

RESULTS

Eighteen articles were identified and 48 recommendations were ultimately extracted. Papers were included only if they described the whole acute care process for patients with ischaemic stroke. Data extraction and analysis revealed variation in terms of both content and comprehensiveness of this description. Experts reached agreement on 34 of 48 (70.8%) recommendations in the dimension actionability, for 16 (33.3%) in feasibility and for 15 (31.3%) in validity. Agreement on all three dimensions was reached for seven (14.6%) recommendations: use of a stroke unit, exclusion of intracerebral haemorrhage as differential diagnosis, administration of intravenous thrombolysis, performance of electrocardiography/cardiac evaluation, non-invasive vascular examination, deep venous thrombosis prophylaxis and administration of statins if needed.

DISCUSSION AND CONCLUSION

Substantial variation in agreement was revealed on the three dimensions of the applicability of recommendations. This overview can guide stroke physicians in improving the care process and removing barriers where implementation may be hampered by validity and feasibility.

Road map to small bowel endoscopy quality indicators

PURPOSE OF REVIEW

Quality indicators for upper and lower gastrointestinal endoscopy are well established and linked to patient outcomes. However, there is a perceived gap in the development and implementation of quality indicators for small bowel endoscopy. In this review, we aimed to discuss the development of quality indicators in small bowel endoscopy and their implementation in clinical practice.

RECENT FINDINGS

The proposed quality indicators for small bowel endoscopy focus on process measures, which mainly evaluate the procedural aspects, rather than the outcomes or the overall patient experience. These quality indicators have rarely been studied in clinical practice, leading to a limited understanding of their applicability and impact on patient outcomes and experience.

SUMMARY

Real-world studies evaluating the quality indicators of small bowel endoscopy are warranted to establish an evidence-based framework for their practical application and effectiveness. Linking these indicators to relevant patient outcomes is crucial for their broader acceptance and implementation.

Development and validation of a deep learning system for detection of small bowel pathologies in capsule endoscopy: a pilot study in a Singapore institution

INTRODUCTION

Deep learning models can assess the quality of images and discriminate among abnormalities in small bowel capsule endoscopy (CE), reducing fatigue and the time needed for diagnosis. They serve as a decision support system, partially automating the diagnosis process by providing probability predictions for abnormalities.

METHODS

We demonstrated the use of deep learning models in CE image analysis, specifically by piloting a bowel preparation model (BPM) and an abnormality detection model (ADM) to determine frame-level view quality and the presence of abnormal findings, respectively. We used convolutional neural network-based models pretrained on large-scale open-domain data to extract spatial features of CE images that were then used in a dense feed-forward neural network classifier. We then combined the open-source Kvasir-Capsule dataset (n = 43) and locally collected CE data (n = 29).

RESULTS

Model performance was compared using averaged five-fold and two-fold cross-validation for BPMs and ADMs, respectively. The best BPM model based on a pre-trained ResNet50 architecture had an area under the receiver operating characteristic and precision-recall curves of 0.969±0.008 and 0.843±0.041, respectively. The best ADM model, also based on ResNet50, had top-1 and top-2 accuracies of 84.03±0.051 and 94.78±0.028, respectively. The models could process approximately 200-250 images per second and showed good discrimination on time-critical abnormalities such as bleeding.

CONCLUSION

Our pilot models showed the potential to improve time to diagnosis in CE workflows. To our knowledge, our approach is unique to the Singapore context. The value of our work can be further evaluated in a pragmatic manner that is sensitive to existing clinician workflow and resource constraints.

Clean mucosal area detection of gastroenterologists versus artificial intelligence in small bowel capsule endoscopy

Studies comparing the detection of clean mucosal areas in capsule endoscopy (CE) using human judgment versus artificial intelligence (AI) are rare. This study statistically analyzed gastroenterologist judgments and AI results. Three hundred CE video clips (100 patients) were prepared. Five gastroenterologists classified the video clips into 3 groups (≥75% [high], 50%-75% [middle], and < 50% [low]) according to their subjective judgment of cleanliness. Visualization scores were calculated using an AI algorithm based on the predicted visible area, and the 5 gastroenterologists' judgments and AI results were compared. The 5 gastroenterologists evaluated CE clip video quality as "high" in 10.7% to 36.7% and as "low" in 28.7% to 60.3% and 29.7% of cases, respectively. The AI evaluated CE clip video quality as "high" in 27.7% and as "low" in 29.7% of cases. Repeated-measures analysis of variance (ANOVA) revealed significant differences in the 6 evaluation indicators (5 gastroenterologists and 1 AI) (P < .001). Among the 300 judgments, 90 (30%) were consistent with 5 gastroenterologists' judgments, and 82 (91.1%) agreed with the AI judgments. The "high" and "low" judgments of the gastroenterologists and AI agreed in 95.0% and 94.9% of cases, respectively. Bonferroni's multiple comparison test showed no significant difference between 3 gastroenterologists and AI (P = .0961, P = 1.0000, and P = .0676, respectively) but a significant difference between the other 2 with AI (P < .0001). When evaluating CE images for cleanliness, the judgments of 5 gastroenterologists were relatively diverse. The AI produced a relatively universal judgment that was consistent with the gastroenterologists' judgements.