Scientists in Japan have developed a new artificial intelligence system that can help detect a rare hormone disease by looking only at photographs of a person’s hand.
The research was carried out by a team from Kobe University and was published in the Journal of Clinical Endocrinology & Metabolism.
The system focuses on identifying a condition called acromegaly. Acromegaly is a rare endocrine disease that usually appears in adults during middle age. It happens when the body produces too much growth hormone.
This extra hormone is usually caused by a small tumor in the pituitary gland, a tiny gland located at the base of the brain. The pituitary gland controls many hormones in the body, including the hormone that affects growth.
When too much growth hormone is released, the body’s tissues begin to grow abnormally. Over time, people with acromegaly may develop enlarged hands and feet, thicker facial features, and changes in the bones of the face and skull.
Internal organs may also grow larger than normal. Because these changes happen very slowly over many years, the disease can be difficult to recognize early.
Doctors say that many patients live with acromegaly for years before receiving the correct diagnosis. In some cases, it can take nearly ten years for the disease to be identified.
During that time, the condition can lead to serious health problems such as heart disease, diabetes, joint pain, and sleep disorders. If it is not treated, acromegaly can shorten life expectancy by about ten years.
The research team, led by endocrinologist Hidenori Fukuoka at Kobe University, wanted to find a faster and easier way to detect the disease.
Artificial intelligence has recently been used in medical research to analyze photographs of patients and detect visible signs of illness. Many earlier studies used facial images to train computer systems to recognize disease patterns.
However, using facial photos can raise concerns about privacy. Patients may not want their faces stored in medical databases or analyzed by computer systems. To avoid this problem, the Kobe University researchers decided to focus on another part of the body that doctors often examine during diagnosis: the hands.
Graduate student Yuka Ohmachi, who helped lead the research, explained that changes in the hands are a common sign of acromegaly. People with the condition often develop thicker fingers and larger hands.
Because doctors already look at the hands when examining patients for this disease, the researchers believed that hand images could be useful for training an AI system.
To further protect patient privacy, the scientists carefully limited the type of photographs used. They only used pictures showing the back of the hand and a clenched fist. They avoided using images of the palm because palm lines can be unique to each person and might reveal a person’s identity.
The research team collected a large set of images for the study. A total of 725 patients from 15 medical institutions across Japan participated in the project. Together they contributed more than 11,000 hand images. These photographs were used to train and test the artificial intelligence model.
After training the system, the scientists tested how well it could detect acromegaly. The AI program analyzed the hand images and looked for subtle patterns that might indicate the disease. The results showed that the system was highly accurate in identifying patients with acromegaly.
The researchers compared the AI system’s performance with that of experienced endocrinologists who examined the same images. Surprisingly, the artificial intelligence system performed even better than the specialists in identifying the condition based only on hand photographs.
Ohmachi said the team was surprised by how accurate the system became using such limited visual information. She noted that achieving high accuracy without using facial images makes the technology much more practical and acceptable for medical screening.
The researchers believe this technology could be useful in many healthcare settings. For example, it might help doctors during routine health check-ups identify patients who should be referred to hormone specialists. Early detection is important because treatment for acromegaly can prevent serious complications.
The AI tool could also be helpful in regions where access to specialists is limited. In some rural areas, patients may not easily reach endocrinologists who are trained to diagnose rare hormonal disorders. A simple screening tool using hand images could help general doctors identify possible cases and refer patients for further testing.
The scientists are also exploring whether similar AI systems could detect other diseases that affect the hands. Conditions such as rheumatoid arthritis, anemia, or finger clubbing can cause visible changes in hand shape and finger structure. Artificial intelligence may be able to recognize these patterns as well.
However, the researchers emphasize that the AI system is not meant to replace doctors. Diagnosing acromegaly still requires blood tests, hormone measurements, medical history, and imaging studies of the pituitary gland. The new tool is intended to support physicians by helping them notice possible cases earlier.
Looking at the findings overall, the study shows how artificial intelligence may assist doctors in detecting diseases that develop slowly and are easy to miss. The large number of images used in the research strengthens the reliability of the results. In addition, the strong privacy protections used in the study make the approach more acceptable for real-world use.
At the same time, the system still needs further testing before it can be widely used in healthcare. Future studies will need to confirm whether the technology works well with patients from different countries and ethnic backgrounds. Researchers will also need to test how the tool performs in everyday clinical settings.
Even with these limitations, the study highlights the growing potential of artificial intelligence in medicine. A simple photo of a hand, analyzed by advanced computer software, could one day help doctors detect rare diseases earlier and improve patient care.
If you care about medicine, please read studies that vitamin D could help lower the risk of autoimmune diseases, and drug for inflammation may stop spread of cancer.
For more health information, please see recent studies about which drug can harm your liver most, and results showing this drug can give your immune system a double boost against cancer.
Copyright © 2026 Knowridge Science Report. All rights reserved.
