Researchers from Kobe University have developed an artificial intelligence (AI) system that can help detect acromegaly, a rare endocrine disorder, using simple photographs of the hand. The findings were published in March 2026 in a peer-reviewed journal.1
Acromegaly occurs due to prolonged excess secretion of growth hormone, usually from a pituitary tumor. The condition develops gradually and is often diagnosed late because early physical changes are subtle.
The research team collected and analyzed over 11,000 hand images from 725 individuals across 15 medical institutions in Japan. The dataset included both individuals diagnosed with acromegaly and those without the condition.
Using this dataset, the AI model was trained to identify patterns linked to the disorder. The system showed strong diagnostic performance, with a positive predictive value of 0.88 and a negative predictive value of 0.93, indicating its ability to correctly identify both affected and unaffected individuals.
According to the Science Daily, Kobe University endocrinologist Hidenori Fukuoka told,
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. With the progress of AI tools, there have been attempts to use photographs for early detection, but they have not been adopted in clinical practice.Hidenori Fukuoka, Endocrinologist, Kobe University
The AI system focuses on images of the back of the hand and a clenched fist. It evaluates subtle anatomical changes such as soft tissue thickening and structural variations that occur due to prolonged exposure to excess growth hormone.
Researchers avoided using facial images to reduce privacy concerns and instead selected hand photographs as a practical and less sensitive alternative. The method does not require specialized equipment and can be applied using standard digital images.
As reported in the Science Daily, Yuka Ohmachi, a graduate student at Kobe University, explained,
Trying to address this concern, we decided to focus on the hands, a body part we routinely examine alongside the face in clinical practice for diagnostic purposes, particularly because acromegaly often manifests changes in the hands.Yuka Ohmachi, Graduate Student, Division of Diabetes and Endocrinology, Kobe University
Current diagnostic pathways for acromegaly involve hormone testing and imaging studies, which may not be readily accessible in all healthcare settings. The AI-based screening tool could support early detection, especially in primary care or resource-limited environments.
The researchers suggest that this approach can serve as a preliminary screening method, prompting further evaluation in suspected cases rather than replacing established diagnostic procedures.
This study highlights the potential role of artificial intelligence in identifying subtle physical signs associated with endocrine diseases. Since many hormonal disorders present with gradual and non-specific symptoms, AI-based image analysis may help reduce diagnostic delays.
As reported in the Science Daily, the study lead, Hidenori Fukuoka said,
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.Hidenori Fukuoka, Lecturer (Diabetes & Endocrinology), Kobe University
The study demonstrates that AI analysis of hand photographs can assist in detecting acromegaly with high accuracy. While further validation is required before clinical implementation, the findings indicate a promising, non-invasive approach to early screening of rare hormonal disorders.
1. Ohmachi, Yuka, Mizuho Nishio, Ichiro Abe, Kunihisa Kobayashi, Tomoko Iida, Manabu Shirakawa, Yuichi Nagata, Kazuhito Takeuchi, Akira Taguchi, Yasuyuki Kinoshita, et al. 2026. “Automatic Acromegaly Detection Using Deep Learning on Hand Images: A Multicenter Observational Study.” The Journal of Clinical Endocrinology & Metabolism. https://doi.org/10.1210/clinem/dgag027.
(Rh/SS/MSM)