AI & softwareResearch findings
Vision Transformer-Based AI Achieves High Diagnostic Accuracy in Biliary Stricture Detection
Original published 2025.09.03 09:00 KSTAdded to Blackdeck 2026.10.18 23:37 KST
What has changed?
A vision transformer-based AI model demonstrated consistent diagnostic accuracy across different endoscopic systems
Key points
- ViT-based AI achieved 83.9% diagnostic accuracy in biliary stricture detection
- Performance was consistent across CHF and Spy endoscopic systems
- No statistically significant difference observed compared to expert endoscopists
Connect it to your work
Performance was comparable to conventional CNNs and expert endoscopists in the study
What remains unproven
This research note is based on an abstract; check the original for test conditions and application scope.
Originals used for this note
Links open the original publisher in a new tab.