OSAKA, Japan — Artificial intelligence can determine a person’s “real” age simply by looking at their chest, according to a new study. Specifically, researchers in Japan say AI can tell a patient’s age after examining chest X-rays, marking a significant leap ahead in medical imaging. The breakthrough paves the way for improved early detection of several potentially deadly diseases.
Researchers at Osaka Metropolitan University developed the advanced AI model to use chest radiographs to accurately estimate chronological age. If a disparity exists between the AI-estimated age and the patient’s actual age, this could be a warning sign of chronic disease.
The research team, led by graduate student Yasuhito Mitsuyama and Dr. Daiju Ueda, first constructed a deep learning-based AI model to estimate age using chest radiographs of healthy individuals. They then applied this model to radiographs of patients with known diseases to analyze the relationship between the AI-estimated age and each specific disease.
Acknowledging that AI models trained on a single dataset are prone to overfitting — when AI provides accurate results for data it trained on but not for new data — the researchers collected information from multiple institutions.
Between 2008 and 2021, they obtained more than 67,000 chest radiographs from over 36,000 volunteers. These volunteers underwent health check-ups at three facilities. The data from these radiographs were used for the development, training, internal, and external testing of the AI model for age estimation.
According to the findings, published in The Lancet Healthy Longevity, the AI model demonstrated a correlation coefficient of 0.95 between the estimated age and the patient’s chronological age. The Japanese team noted that, generally, a correlation coefficient of 0.9 or higher is considered to be “very strong.”
To further validate the usefulness of AI-estimated age using chest radiographs as a biomarker, the team compiled an additional 34,197 chest radiographs from the same number of patients with known diseases, sourced from two other institutions.
The results revealed that the difference between the AI-estimated age and the patient’s chronological age was positively correlated with several chronic diseases. These diseases include high blood pressure, chronic obstructive pulmonary disease (COPD), and hyperuricemia (elevated uric acid levels in the blood).
In simpler terms, the researchers explained that the greater the discrepancy between the AI-estimated age and the chronological age, the more likely it is that individuals have these diseases.
“Chronological age is one of the most critical factors in medicine,” Mitsuyama says in a media release. “Our results suggest that chest radiography-based apparent age may accurately reflect health conditions beyond chronological age. We aim to further develop this research and apply it to estimate the severity of chronic diseases, to predict life expectancy, and to forecast possible surgical complications.”
You might also be interested in:
- Stress really does age the body — but it’s possible to recover within days
- Eat better, live longer? Scientists uncover link between eating protein and aging
- Should governments stop AI? 70% want stricter laws clamping down on artificial intelligence
South West News Service writer Stephen Beech contributed to this report.