Bowel Obstruction Detection and Localization on Abdominal CT with Deep Learning

2026-07-24

AI Summary

This study addresses the clinical challenge of automatically detecting intestinal obstruction and localizing its transition zone in abdominal CT scans. The authors propose a multitask deep learning framework that jointly performs lesion detection and precise transition zone localization. Innovatively, they introduce an intrinsically interpretable mechanism based on probabilistic selection masks to guide the model’s attention toward suspicious transitional regions, thereby enhancing decision transparency and clinical trustworthiness. Trained end-to-end on an internal dataset of 1,427 CT cases, the system achieves a detection accuracy of 93% and a Hit@10 score of 95% for transition zone localization, marking the first reliable automated method for identifying the transition zone in intestinal obstruction.

Abstract

Bowel obstruction is a common and potentially life-threatening gastrointestinal condition. In the face of rising diagnostic workloads, the automated diagnosis of bowel obstruction on CT scans supports radiologists by accelerating detection and improving patient outcomes. In this work, we propose a deep learning framework with a multi-task objective that jointly detects bowel obstruction and localizes its transition zone. Additionally, we extend the method with an inherently interpretable classification method that locates the suspected transition point within a slice. It does so by learning a probabilistic selection mask that faithfully bases the classifier's prediction solely on a small image region. The proposed method is evaluated on an internal dataset comprising 1,427 abdominal CTs. Here, the model achieves an obstruction detection test accuracy of 93% and a Hit@10 transition zone localization of 95%. As the first method to reliably localize the transition zone, this marks a significant step towards the automated identification of this critical clinical landmark.

Problem

• bowel obstruction
• abdominal CT
• transition zone localization
• automated diagnosis
• deep learning

Innovation

• multi-task learning
• transition zone localization
• interpretable deep learning
• probabilistic selection mask
• bowel obstruction detection