Related Experiment Video

Updated: Jul 12, 2026

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

A vision transformer deep learning model for assessing pediatric ileocolic intussusception severity using ultrasound

Jie Liu1,2,3, Yue Wang4, Danping Zeng5

  • 1Department of Pediatric Surgery, The First Affiliated Hospital of Wannan Medical University (Yijishan Hospital of Wannan Medical University), Wannan Medical University, Wuhu, China. liujie19842020@163.com.

NPJ Digital Medicine
|July 10, 2026
PubMed

Insights

A new Vision Transformer (ViT) deep learning model accurately predicts ileocolic intussusception reduction failure from ultrasound images. This AI tool surpasses human sonographers, improving clinical decisions and reducing perforation risks in children.

Area of Science:

  • Medical Imaging
  • Artificial Intelligence in Medicine
  • Pediatric Surgery

Background:

  • Ileocolic intussusception requires timely intervention to prevent complications like bowel perforation.
  • Predicting air-enema reduction failure is crucial but faces challenges due to inter-observer variability among sonographers.
  • Current diagnostic methods for intussusception severity can be subjective, impacting treatment decisions.

Purpose of the Study:

  • To develop and validate a Vision Transformer (ViT) deep learning system for predicting air-enema reduction failure in pediatric ileocolic intussusception.
  • To assess the performance of the ViT model against expert and junior sonographers in a prospective cohort.
  • To provide an objective and accurate tool to aid clinical decision-making in managing intussusception.

Main Methods:

  • A multicenter bidirectional cohort study involving 5602 children (4-60 months) with ileocolic intussusception.
  • Training a ViT model on 10,151 ultrasound images for binary classification of reduction success or failure.
  • External validation of the ViT model on a prospective cohort of 190 patients, compared with independent assessments by six sonographers.

Main Results:

  • The ViT model demonstrated high internal performance with an accuracy of 0.880 for failure and 0.970 for success.
  • In the prospective cohort, the ViT model achieved 93.7% overall accuracy.
  • The AI model's accuracy (93.7%) was significantly higher than senior (74.7%) and junior (60.7%) sonographers (p < 0.05).

Conclusions:

  • The Vision Transformer deep learning system offers an objective and highly accurate method for predicting ileocolic intussusception reduction failure.
  • This AI tool has the potential to significantly improve clinical decision-making, reduce treatment delays, and minimize the risk of bowel perforation in pediatric patients.
  • The study highlights the innovative application of ViT in assessing pediatric intussusception severity, paving the way for AI-assisted diagnostics in pediatric surgery.

Related Concept Videos

Imaging Studies III: Gastrointestinal Motility Studies and Virtual Colonoscopy01:26

Imaging Studies III: Gastrointestinal Motility Studies and Virtual Colonoscopy

This lesson explores three gastrointestinal imaging techniques: radionuclide testing, colonic transit studies, and virtual colonoscopy.
Radionuclide Testing
Radionuclide testing is a sophisticated medical technique for assessing gastrointestinal motility. It focuses on gastric emptying and colonic transit time. Radioactive markers track the movement of food through the digestive system, providing insights into gastrointestinal disorders.
In gastric emptying studies, a meal's liquid and solid...
Imaging Studies II: Ultrasonography01:24

Imaging Studies II: Ultrasonography

IntroductionUltrasonography, or renal ultrasound, is a noninvasive medical imaging technique that uses high-frequency sound waves to visualize the kidneys, ureters, bladder, and surrounding tissues.Indications for Urinary System UltrasonographyUrinary system ultrasonography is indicated in various clinical scenarios, such as:Kidney Stones (Urolithiasis): To detect and monitor the size and presence of kidney or urinary tract stones.Hydronephrosis: To assess the dilation of the renal pelvis and...
Ultrasound II: Endoscopic Ultrasound and FibroScan01:25

Ultrasound II: Endoscopic Ultrasound and FibroScan

Endoscopic Ultrasound (EUS) and FibroScan are valuable diagnostic tools in gastroenterology and hepatology, each with specific applications and techniques.
Endoscopic Ultrasound (EUS):
Imaging Studies I: Kidney, Ureter, and Bladder Studies01:28

Imaging Studies I: Kidney, Ureter, and Bladder Studies

Kidney, Ureter, and Bladder (KUB) StudiesKidney, Ureter, and Bladder (KUB) studies are standard diagnostic imaging procedures used to assess the anatomy of the urinary system. They are commonly utilized for patients experiencing abdominal pain or urinary symptoms. By using a simple X-ray of the abdomen, KUB studies can reveal structural and pathological abnormalities within the kidneys, ureters, and bladder. These studies are particularly valuable in diagnosing kidney stones, urinary...