Related Experiment Video
Updated: Jan 6, 2026
09:12
A Mouse Model of Chronic Liver Fibrosis for the Study of Biliary Atresia
Published on: February 3, 2023
3.2K
Development and validation of a minimally invasive diagnostic model for biliary atresia using artificial intelligence
Jing-Ying Jiang1, Rui Dong1, Ying-Hua Sun2
1Department of Pediatric Surgery, Shanghai Key Laboratory of Birth Defect, and Key Laboratory of Neonatal Disease, Children's Hospital of Fudan University, Ministry of Health, 399 Wan Yuan Rd, Shanghai 201102, China.
World Journal of Pediatrics : WJP
|November 11, 2025
Summary
An AI model combining ultrasound and serum matrix metalloproteinase-7 (MMP-7) shows high accuracy in diagnosing biliary atresia (BA). This AI approach offers a sensitive and specific tool for differentiating BA from other cholestatic diseases.
Area of Science:
- Medical Imaging
- Biomarkers
- Artificial Intelligence in Medicine
Background:
- Biliary atresia (BA) diagnosis is challenging, often requiring differentiation from other cholestatic diseases.
- Ultrasound imaging and serum matrix metalloproteinase-7 (MMP-7) are valuable diagnostic tools for BA.
- Accurate and early diagnosis of BA is crucial for effective treatment and improved patient outcomes.
Purpose of the Study:
- To evaluate the diagnostic accuracy of an artificial intelligence (AI) model integrating ultrasound and serum MMP-7 levels for biliary atresia (BA).
- To compare the performance of the combined AI model against individual ultrasound and MMP-7 tests in differentiating BA.
Main Methods:
- A multicenter diagnostic study involving six Chinese medical centers.
- Development of an AI algorithm using morphological operators on ultrasound images.
- Logistic regression modeling incorporating ultrasound features and serum MMP-7 levels.
- Validation of the AI model on both retrospective and prospective cohorts.
Main Results:
- The combined AI model achieved an area under the receiver-operating characteristic curve (AUROC) of 0.985, demonstrating superior diagnostic performance.
- The AI model showed high sensitivity (98.2%) and specificity (93.1%) in diagnosing BA.
- Individual models also showed strong performance, with AUROCs of 0.916 for MMP-7 and 0.945 for ultrasound.
Conclusions:
- An AI model integrating ultrasound and serum MMP-7 offers a highly sensitive and specific method for the differential diagnosis of biliary atresia.
- This AI-driven approach shows significant potential for improving the accuracy of BA diagnosis in clinical practice.
