Related Experiment Video
Updated: Jun 29, 2025
07:33
A Mouse Model of Intestinal Partial Obstruction
Published on: March 5, 2018
21.2K
Unveiling new patterns: A surgical deep learning model for intestinal obstruction management
Ozan Can Tatar1,2, Mustafa Alper Akay3, Elif Tatar3
1Department of General Surgery, School of Medicine, Kocaeli University, Kocaeli, Turkey.
Summary
This study integrates deep learning (AI) with surgical expertise to improve decisions for intestinal obstruction. The developed AI model shows promise in differentiating surgical versus non-surgical management cases.
Area of Science:
- Medical Imaging Analysis
- Artificial Intelligence in Surgery
- Gastrointestinal Surgery
Background:
- Effective management of intestinal obstructions requires rapid and precise clinical decisions.
- Current diagnostic approaches may benefit from advanced computational tools.
Purpose of the Study:
- To develop and evaluate a deep learning model for enhanced decision-making in intestinal obstruction cases.
- To integrate artificial intelligence with surgical expertise for improved patient outcomes.
Main Methods:
- A deep learning model utilizing the YOLOv8 framework was created.
- The model was trained on 700 categorized images (operated vs. non-operated) with surgical outcomes as ground truth.
- Performance was assessed using standard evaluation metrics.
Main Results:
- The model achieved a sensitivity of 83.33% and specificity of 78.26% at a 0.5 confidence threshold.
- Precision was 81.7%, recall was 75.1%, and mean Average Precision (mAP@0.5) was 0.831.
- The model demonstrated strong performance in classifying management strategies.
Conclusions:
- The deep learning model shows significant potential in distinguishing between operative and nonoperative management for intestinal obstructions.
- Combining AI with surgical knowledge enhances clinical decision-making.
- This technology can aid surgeons in complex cases, fostering a synergy between AI and clinical judgment to advance patient care.
