Researchcomputer visionmedical imagingendocrinology
Researchers Train AI To Detect Acromegaly
7.2
Kobe University researchers report in the Journal of Clinical Endocrinology and Metabolism (2026) that an AI model trained on more than 11,000 back-of-hand and clenched-fist photos from 725 Japanese participants can detect acromegaly. The model achieved a positive predictive value of 0.88 and negative predictive value of 0.93, outperforming endocrinologists on the same images, and could speed screening pending broader validation.
Key Points
- 1Used 11,000+ back-of-hand and clenched-fist images from 725 participants to train and validate AI
- 2Achieved high diagnostic performance with PPV 0.88 and NPV 0.93, outperforming specialist readers
- 3Enables non-specialist screening to accelerate case detection, but requires larger, more diverse validation
Scoring Rationale
Peer-reviewed clinical study demonstrates strong diagnostic accuracy, limited by single-country cohort and need for larger, diverse validation.
Sources
Public references used for this report.
Practice with real Health & Insurance data
90 SQL & Python problems · 15 industry datasets
Used by DS/ML engineers at top companies
250 free problems · No credit card
See all Health & Insurance problems
