Deep Learning and Capsule Endoscopy: Automatic Panendoscopic Detection of Protruding Lesions

Introduction Capsule endoscopy (CE) provides a minimally invasive exam modality for panendoscopic evaluation of the entire gastrointestinal (GI) tract. However, conventional reading methods can be time-consuming and error-prone. Protruding lesions are a relatively common entity that can be found with a variable incidence and different pathological significance throughout GI tract.

Aim The aim of this study was to develop and test a Convolutional Neural Network (CNN)-based algorithm for panendoscopic automatic detection of protruding lesions on CE exams.

Methods A multicentric retrospective study was conducted, based on 1245 CE exams. We used a total of 191455 frames, from 6 types of CE devices, of which 52717 had protruding lesions (polyps, epithelial tumors or subepithelial lesions) after triple validation. Data was divided into a training and testing set (90% vs 10%), using an exam-split design. During training stage, we performed a 5-fold cross validation. Our outcome measures were sensitivity, specificity, accuracy, positive predictive value (PPV), negative predictive value (NPV) and area under the conventional receiver operating characteristic curve (AUC-ROC) and the precision-recall curve (AUC-PR).

Results Regarding test set, the sensitivity were 99.7% and specificity 96.5%. The PPV and NPV were 96.8% and 99.3.0%, respectively. The global accuracy was 98.4%.

Conclusion This study aims to address the gap in AI-enhanced capsule panendoscopy by investigating the development of the first CNN for the detection of protruding lesions across the GI tract. AI‘s improvement of CE‘s diagnostic accuracy, along with the growing interest in minimally invasive procedures, may contribute to increasing access to this diagnostic tool.

No conflict of interest has been declared by the author(s).

Publication History

Article published online:
25 April 2025