Introduction

Capsule endoscopy (CE) is a minimally invasive and essential tool for assessing small bowel disorders, especially Crohn’s disease. Accurate detection of ulcers and erosions is vital for evaluating disease activity, guiding treatment, and monitoring therapeutic response. Although CE has revolutionized lesion detection, its manual interpretation remains time-consuming and subject to interobserver variability. Recent advances in artificial intelligence (AI), particularly convolutional neural networks (CNNs), offer promising solutions by enhancing diagnostic accuracy, reducing variability, and improving workflow efficiency. This study aimed to develop and validate a robust AI model capable of detecting and distinguishing small bowel ulcers and erosions across multiple CE platforms and clinical centers.

Aims & Methods

In a prospective, multicenter study conducted between January 2021 and April 2024, data were collected from centers in Portugal, Spain, and the United States. Two CE devices—PillCamSB3 and Olympus EC-10—were used to analyze 137 anonymized exams. The AI-assisted readings generated by a deep learning model were compared to standard-of-care (SoC) interpretations, with an expert consensus panel serving as the reference standard. Performance metrics included sensitivity, specificity, positive and negative predictive values (PPV, NPV), and area under the receiver operating characteristic curve (AUROC).

Results

Expert consensus identified ulcers and erosions in 56 patients (40.9%). Compared to the reference standard, SoC readings achieved 60.7% sensitivity, 98.8% specificity, 97.1% PPV, 78.4% NPV, and 83.2% accuracy. In contrast, the AI-assisted model achieved 94.6% sensitivity, 80.2% specificity, 76.8% PPV, 95.6% NPV, and 86.1% accuracy. The AI model was found to be both non-inferior (p < 0.001) and statistically superior (p < 0.001) to SoC. Notably, it identified 68 lesions, nearly double those detected by SoC (35), and demonstrated consistent performance across devices and centers. The mean AI-assisted reading time was 239 seconds (SD 138 seconds) per exam.

Conclusion

The AI-assisted model significantly outperformed conventional SoC in the detection of small bowel ulcers and erosions, achieving higher sensitivity and overall accuracy while substantially reducing reading time. This is the first multicenter validation of an interoperable AI solution for CE, capable of high diagnostic performance across different devices and clinical environments. These findings highlight the potential of AI to transform endoscopic evaluation and management in inflammatory bowel disease.

Introduction

Although endoscopic balloon dilation (EBD) remains the standard intervention for Crohn’s disease (CD)-associated strictures, its efficacy is limited by high recurrence and surgery rates. Retrospective analyses suggest endoscopic stricturotomy (ES) may offer superior outcomes, particularly in anastomotic strictures. However, no randomized trials have directly compared EBD and ES for short CD-related strictures.

Aims & Methods

This single-center, randomized controlled trial (August 2022–March 2025) enrolled adult CD patients (18–65 years) with symptomatic, predominantly fibrotic or mixed, short strictures (<3 cm), including both de novo and anastomotic types, accessible via standard endoscopy. Patients with multiple (>2) strictures, strictures beyond endoscopic reach, or prior stricturotomy were excluded. Participants were randomized to undergo either EBD or ES. The primary outcome was clinical recurrence at 1 year. Secondary outcomes included re-intervention, surgery, emergency visits, hospitalization, technical success, and adverse events.

Results

Out of 70 randomized patients, 69 were analyzed (EBD: 35; ES: 34). Baseline characteristics were comparable between groups. Technical success was similar (EBD: 91.4%, ES: 94.1%). On time to event analysis, clinical recurrence was significantly lower in the ES group (OR=0.343, 95% CI: 0.134–0.881; p=0.026). Re-intervention rates favored ES (EBD: 25.7%, ES: 8.8%; OR=0.337, 95% CI: 0.131–0.870; p=0.025). Surgery rates were numerically lower with ES (EBD: 11.4%, ES: 5.9%) but not statistically significant (p=0.581). Hospitalization was significantly more frequent in the EBD arm (OR=0.276, 95% CI: 0.091–0.840; p=0.023). ED visits also trended lower with ES, though not statistically significant (p=0.062) (see table). Adverse events occurred in 25.7% of EBD patients (bleeding n=3 [1 hemoclip], pain with admission n=3, micro-perforation n=1, post-procedure pain n=2) and 14.7% of ES patients (bleeding n=3 [1 self-limited, 1 coagrasper, 1 hemoclip], none required transfusion, suspected perforation n=1, mild pain n=1).

Conclusion

In this interim analysis, ES was associated with significantly lower clinical recurrence, re-intervention, and hospitalization compared to EBD for short CD-related strictures, with similar safety outcomes. These findings support ES as a potentially superior first-line endoscopic strategy over EBD in selected patients with Crohn’s disease strictures. (NCT05521867)

References

Lan N, Shen B. Endoscopic Stricturotomy Versus Balloon Dilation in the Treatment of Anastomotic Strictures in Crohn's Disease. Inflamm Bowel Dis. 2018 Mar 19;24(4):897-907. doi: 10.1093/ibd/izx085. PMID: 29546384.