Related Experiment Video
Updated: Jan 13, 2026
11:03
Robotic Left Hepatectomy using Indocyanine Green Fluorescence Imaging for an Intrahepatic Complex Biliary Cyst
Published on: June 24, 2022
6.1K
Artificial intelligence performance in image-based biliary atresia identification: a systematic review and
Anis Halimi1, Ahmed Msherghi2, Mohamedhen Vall Nounou3
1Faculty of Medicine, Badji Mokhtar University, Annaba, Algeria.
European Journal of Radiology
|January 7, 2026
Summary
Artificial Intelligence (AI) demonstrates strong performance in diagnosing biliary atresia (BA) using medical imaging. AI serves as a valuable assistive tool for clinicians, though further research is needed to confirm its widespread applicability.
Area of Science:
- Medical Imaging Analysis
- Artificial Intelligence in Healthcare
- Diagnostic Accuracy Studies
Background:
- Biliary atresia (BA) is a rare but serious condition requiring accurate diagnosis.
- Traditional imaging methods for BA detection have limitations.
- AI-based medical imaging offers potential for improved diagnostic accuracy in BA.
Purpose of the Study:
- To evaluate the diagnostic accuracy of Artificial Intelligence (AI) in medical imaging for detecting biliary atresia (BA).
- To synthesize existing evidence on AI's performance in BA diagnosis through a meta-analysis.
Main Methods:
- Adherence to PRISMA DTA guidelines and PROSPERO registration.
- Systematic search of major databases (PubMed, Web of Science, Embase, Scopus) for relevant studies.
- Calculation of pooled sensitivity, specificity, and AUC using R 4.4.2, with quality assessment via QUADAS-AI and QUADAS-2 tools.
Main Results:
- Nine studies involving 11,006 patients (2,357 BA cases) were analyzed.
- Patient-level AI analysis showed pooled sensitivity of 93.8% and specificity of 93.2% (AUC=0.94).
- Image-level analysis yielded pooled sensitivity of 86.9% and specificity of 94.3% (AUC=0.965).
Conclusions:
- AI demonstrates satisfactory performance for the imaging-based diagnosis of biliary atresia.
- AI can serve as an effective assistive tool to support clinical decision-making.
- Further high-quality research is necessary to validate the generalizability of AI diagnostic tools for BA.
