AI Detects Rare Disease From Hand Photos, Protects Privacy

A simple photo of your hand might soon catch a serious disease that doctors typically miss for a decade.

Researchers at Kobe University developed an AI system that identifies acromegaly, a rare hormone disorder, just by analyzing pictures of the back of the hand and a clenched fist. The technology achieves something remarkable: it's actually more accurate than experienced specialists looking at the same images.

Acromegaly happens when the body produces too much growth hormone, causing hands, feet and facial features to gradually enlarge over many years. Because changes develop so slowly, most patients wait up to ten years before getting diagnosed. Without treatment, the condition can shorten life expectancy by a decade.

"Because the condition progresses so slowly, and because it is a rare disease, it is not uncommon to take up to a decade for it to be diagnosed," says lead researcher Dr. Hidenori Fukuoka. Previous AI tools relied on facial photos for detection, but those raised serious privacy concerns that kept them out of real medical practice.

The Kobe team took a different approach. Graduate student Yuka Ohmachi explains they focused on hands because doctors already examine them during diagnosis, and acromegaly causes obvious changes there. The researchers deliberately avoided palm images since palm prints can identify individuals, choosing instead to photograph only the back of the hand and closed fist.

That privacy-first design helped recruit 725 patients from 15 hospitals across Japan who contributed over 11,000 images. The resulting AI model showed exceptional accuracy in the Journal of Clinical Endocrinology & Metabolism.

The Ripple Effect

The impact could reach far beyond acromegaly patients. Doctors in rural areas without specialist access could use the tool to screen patients during routine checkups, catching cases that might otherwise go unnoticed for years. The research team is already working to adapt their system for rheumatoid arthritis, anemia and other conditions visible in the hands.

"This result could be the entry point for expanding the potential of medical AI," says Ohmachi. She was surprised the technology reached such high accuracy without any facial features, making it far more practical for real-world screening.

The researchers stress their AI won't replace doctors. Instead, it will help physicians who see hundreds of patients identify the handful who need specialist referrals. Dr. Fukuoka believes the technology could create better medical infrastructure during health checkups, especially in underserved communities facing healthcare disparities.

Early detection could give thousands of people their healthy years back.