Sometimes, the body leaves clues in plain sight.
Researchers at Kobe University say a new artificial intelligence system can identify a rare hormonal disorder simply by analyzing photos of the back of a person’s hand and a clenched fist.
The goal is to help doctors detect acromegaly, an uncommon endocrine disease that often goes unnoticed for years.
The condition is caused by excessive growth hormone and usually appears in middle age. Over time, it can lead to enlarged hands and feet, changes in facial features and abnormal growth of bones and organs.
Because those changes happen slowly, diagnosis can take a long time.
“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,” said Kobe University endocrinologist Hidenori Fukuoka.
If untreated, the disorder can lead to serious health problems and reduce life expectancy by about ten years.
Many experimental AI tools designed to detect diseases from photographs rely heavily on facial images. But researchers say that approach can raise privacy concerns for patients.
The Kobe team decided to take a different path.
Graduate student Yuka Ohmachi explained that the researchers focused on the hands, which doctors often examine when evaluating patients with suspected acromegaly.
“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,” Ohmachi said.
To further protect patient privacy, the scientists limited images to the back of the hand and a clenched fist. They deliberately avoided photographing palms because palm-line patterns can be unique enough to reveal someone’s identity.
The privacy-focused design helped researchers recruit a large group of participants.
In total, 725 patients from 15 medical institutions across Japan contributed more than 11,000 images used to train and test the system.
The results surprised even the scientists behind the project.
When researchers compared the AI’s performance with experienced endocrinologists who reviewed the same photos, the system showed higher diagnostic accuracy.
“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,” Ohmachi said.
“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.”
The findings were published in the Journal of Clinical Endocrinology & Metabolism.
Researchers emphasize that the technology is not meant to replace doctors.
Instead, it could act as a screening tool that flags potential cases earlier, helping patients reach specialists sooner.
The team also believes the same approach could be expanded to detect other conditions that show physical signs in the hands.
Possible future applications include identifying rheumatoid arthritis, anemia and finger clubbing, which can be linked to lung or heart disease.
Ohmachi says the technology could open the door to a broader role for artificial intelligence in medicine.
“This result could be the entry point for expanding the potential of medical AI,” she said.
Lead researcher Fukuoka believes tools like this could eventually become part of routine health checkups.
“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,” he said.
He added that it may also help doctors working in rural or underserved areas.
“Furthermore, it could support non-specialist physicians in regional healthcare settings, thus contributing to a reduction of healthcare disparities there.”
Sometimes, spotting a serious illness begins with something simple.
In this case, it may start with a photo of a hand.
