AI Model Identifies Rare Endocrine Disorder from Hand Images

Acromegaly is a rare, intractable disease that typically begins in middle age and causes enlargement of the hands and feet, changes in facial appearance, and abnormal bone and organ growth. Because the condition is uncommon and progresses slowly, diagnosis is often delayed, exposing patients to preventable complications and reduced survival. Researchers now report a privacy-conscious artificial intelligence (AI) system that detects acromegaly using hand photographs rather than facial images. 

Endocrinologists at Kobe University (Kobe, Japan) developed a deep learning method that analyzes standardized photographs of the back of the hand and a clenched fist. The design deliberately avoids facial features and palm line patterns to reduce the chance of personal identification. The workflow reflects areas clinicians already inspect for disease‑related changes. The goal is to integrate a screening step that fits real‑world outpatient practice.

Investigators conducted a multicenter observational study to train and test the model. The dataset comprised more than 11,000 images collected from 725 patients at 15 medical institutions across Japan. Focusing on nonfacial hand views facilitated broad participation by addressing common privacy concerns among potential contributors.

In validation, the model achieved very high sensitivity and specificity for acromegaly recognition from hand images. In head-to-head comparisons, its performance surpassed that of experienced endocrinologists reviewing the same photographs. The researchers emphasize that the tool is designed to complement, not replace, clinical judgment, with the aim of supporting timely referral and reducing diagnostic oversight in routine care. The findings were published in The Journal of Clinical Endocrinology & Metabolism in 2026. 

“Frankly, I was surprised that the diagnostic accuracy reached such a high level using only photographs of the back of the hand and the clenched fist. What struck me as particularly significant was achieving this level of performance without facial features, which makes this approach a great deal more practical for disease screening,” said Yuka Ohmachi, graduate student at Kobe University.

“We believe that, by further developing this technology, it could lead to creating a medical infrastructure during comprehensive health check-ups to connect suspected cases of hand-related disorders to specialists. Furthermore, it could support non-specialist physicians in regional healthcare settings, thus contributing to a reduction of healthcare disparities there,” said Hidenori Fukuoka, endocrinologist at Kobe University.

Related Links
Kobe University