AI & softwareResearch findings
Pilot Study on AI-Based Facial Image Diagnosis of Marfan Syndrome
Original published 2024.06.28 09:00 KSTAdded to Blackdeck 2026.11.03 04:19 KST
What has changed?
AI can distinguish Marfan from non-Marfan facial images from ordinary online photos
Key points
- Distinct facial features of Marfan syndrome suggest potential for AI-based diagnosis
- CNN-based model achieved 98.5% overall accuracy in distinguishing Marfan from non-Marfan faces
- F1 scores reached 97% for Marfan and 99% for non-Marfan facial patterns
Connect it to your work
AI successfully differentiates Marfan and non-Marfan facial features using common online photographs
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.