Related Experiment Video
Updated: Jan 1, 2026
Multimodality Diagnosis of Mesenteric Ischemia
Published on: July 21, 2023
Performance of deep learning-based algorithm for detection of ileocolic intussusception on abdominal radiographs of
Sungwon Kim1, Haesung Yoon1, Mi-Jung Lee1
1Department of Radiology, Severance Hospital, Research Institute of Radiological Science, Center for Clinical Imaging Data Science, Yonsei University College of Medicine, 50-1 Yonsei-Ro, Seodaemun-Gu, Seoul, 03722, Korea.
Insights
A new deep learning algorithm shows promise in detecting ileocolic intussusception in young children using abdominal radiographs. This AI tool achieved higher sensitivity than radiologists, aiding in early diagnosis.
Area of Science:
- Medical Imaging
- Artificial Intelligence in Healthcare
- Pediatric Radiology
Background:
- Ileocolic intussusception is a common surgical emergency in young children.
- Abdominal radiography is frequently used for initial diagnosis, but interpretation can be challenging.
- Deep learning offers potential for improving diagnostic accuracy in medical imaging.
Purpose of the Study:
- To develop and evaluate a deep learning algorithm for detecting ileocolic intussusception on abdominal radiographs.
- To compare the diagnostic performance of the algorithm against human radiologists.
Main Methods:
- A YOLOv3-based deep learning algorithm was trained on abdominal radiographs from children (≤5 years old) with confirmed intussusception (based on ultrasonography).
- The algorithm was designed to identify the right abdominal region and diagnose intussusception.
- Performance was validated on a separate set of pediatric radiographs, comparing algorithm and radiologist diagnostic accuracy (sensitivity, specificity).
Main Results:
- The deep learning algorithm demonstrated significantly higher sensitivity (0.76) compared to radiologists (0.46) (p=0.013).
- Specificity was comparable between the algorithm (0.96) and radiologists (0.92) (p=0.32).
- The algorithm successfully identified intussusception in a validation cohort.
Conclusions:
- Deep learning algorithms, such as the YOLOv3-based model, can effectively aid in the screening of ileocolic intussusception using abdominal radiography in young children.
- This technology has the potential to improve early detection rates and support clinical decision-making in pediatric emergency settings.
Abstract:
The purpose of this study was to develop and test the performance of a deep learning-based algorithm to detect ileocolic intussusception using abdominal radiographs of young children. For the training set, children (≤5 years old) who underwent abdominal radiograph and ultrasonography (US) for suspicion of intussusception from March 2005 to December 2017 were retrospectively included and divided into control and intussusception groups according to the US results. A YOLOv3-based algorithm was developed to recognize the rectangular area of the right abdomen and to diagnose intussusception. For the validation set, children (≤5 years old) who underwent both radiograph and US from January to August 2018 with the suspicion of intussusception were included. Diagnostic performances of an algorithm and radiologists were compared. Total 681 children including 242 children in intussusception group were included in the training set and 75 children including 25 children in intussusception group were included in the validation set. The sensitivity of the algorithm was higher compared with that of the radiologists (0.76 vs. 0.46, p = 0.013), while specificity was not different between the algorithm and the radiologists (0.96 vs. 0.92, p = 0.32). Deep learning-based algorithm can aid screening of intussusception using abdominal radiography in young children.
