A privacy-first AI can diagnose a life-shortening hormone dysfunction—simply from a photograph of your hand.
Researchers at Kobe University have developed an artificial intelligence system that can identify a rare endocrine disorder by examining photos of the back of a person’s hand and their clenched fist. By avoiding facial images, the approach was designed with privacy in mind. The team believes this tool could help doctors refer patients to specialists more efficiently and help narrow gaps in access to care.
Acromegaly and Delayed Diagnosis
The condition, acromegaly, is an uncommon and difficult-to-treat disease that typically begins in middle age. It is caused by excess production of growth hormone, leading to enlarged hands and feet, noticeable facial changes, and abnormal growth of bones and internal organs. The disease develops gradually over many years. Without treatment, it can cause serious complications and shorten life expectancy by roughly 10 years.
“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 Kobe University endocrinologist Hidenori Fukuoka. He adds, “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.”
A Privacy Focused AI Approach
When reviewing earlier AI research, the team noted that many systems depend on facial photographs, which can raise privacy concerns. Yuka Ohmachi, a graduate student at Kobe University, explains, “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.”
To further protect patient identity, the researchers limited their dataset to images of the back of the hand and a clenched fist, deliberately excluding palm images that contain distinctive line patterns. This privacy-conscious strategy encouraged broad participation. A total of 725 patients from 15 medical centers across Japan contributed more than 11,000 images, which were used to train and test the AI system.
Study Results Show High Accuracy
The findings, published today (February 27) in the Journal of Clinical Endocrinology & Metabolism, show that the model achieved very high sensitivity and specificity in detecting acromegaly. In side-by-side comparisons using the same photographs, the AI system performed better than experienced endocrinologists.
“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,” says Ohmachi.
Expanding AI Screening to Other Conditions
The researchers plan to adapt their system to identify additional conditions that produce visible changes in the hands, including rheumatoid arthritis, anemia, and finger clubbing. Ohmachi says, “This result could be the entry point for expanding the potential of medical AI.”
Supporting Doctors and Reducing Healthcare Gaps
Doctors do not rely solely on hand images when making a diagnosis. They consider medical history, laboratory tests, and other clinical information. The Kobe University team views their AI tool as a way to support, not replace, medical professionals. In their paper, they write that it could “complement clinical expertise, reduce diagnostic oversight and enable earlier intervention.”
Study lead Fukuoka says: “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.”
Reference: “Automatic Acromegaly Detection Using Deep Learning on Hand Images: A Multicenter Observational Study” 27 February 2026, The Journal of Clinical Endocrinology & Metabolism.
DOI: 10.1210/clinem/dgag027
The research was funded by the Hyogo Foundation for Science Technology. Collaborators included researchers from Fukuoka University, Hyogo Medical University, Nagoya University, Hiroshima University, Toranomon Hospital, Nippon Medical School, Kagoshima University, Tottori University, Yamagata University, Okayama University, Hyogo Prefectural Kakogawa Medical Center, Hokkaido University, International University of Health and Welfare, Moriyama Memorial Hospital, and Konan Women’s University.
Never miss a breakthrough: Join the SciTechDaily newsletter.
Follow us on Google and Google News.
