Some people appear much older than their chronological age. Japanese researchers have found that this could be linked to the presence of an illness that the individual is unaware of. They have developed an AI program that can accurately calculate a person’s age based on a chest X-ray.
Innovative AI Model for Age and Disease Research
While other previously established AI programs analyze X-rays to detect lung abnormalities, this model focuses on age assessment. Researchers then use this information to further predict potential diseases in the body. For instance, if the age calculated by the AI significantly exceeds the person’s actual age, it is likely that they have chronic conditions such as COPD, hypertension, or hyperuricemia (elevated uric acid levels).
This new AI model is likely the first program that allows scientists to establish a direct link between age and chronic diseases. According to the researchers, it could become an effective tool for early diagnosis, which is crucial for effective treatment.
As noted by one of the study’s authors, Yasuhito Mitsuyama, chronological age is one of the most important factors in medicine. The results indicate that the visible age based on chest X-rays can accurately reflect a person’s health status beyond their chronological age.
Testing the AI Age Estimator
The researchers input a significant amount of data into their program to develop an AI model capable of accurately estimating age based on chest X-rays. They collected 67,099 chest X-rays from over 36,000 individuals across three medical examination centers. All of these images were taken between 2008 and 2021 and pertained to healthy individuals.
In the next phase, the researchers tested their AI model using an additional 34,197 chest X-rays collected from two medical centers. This time, all the images belonged to individuals suffering from known diseases. When the AI model calculated age based on these X-rays, Mitsuyama’s team observed a remarkable positive correlation between the AI-estimated age and the likelihood of chronic diseases.
One Step Closer to New AI Predictions in Medicine
As reported by Interesting Engineering, the researchers noted that if the AI calculated a person’s age to be higher than their chronological age, they also tended to have persistent conditions such as high blood pressure or chronic bronchitis. In other words, the greater the discrepancy between the AI-estimated age and the chronological age, the higher the likelihood of having chronic diseases that the individual may not always suspect.
The researchers hope that with further studies and developments, they will be able to use their AI model for more complex tasks. For example, predicting a person’s lifespan, calculating their chances of survival from a specific illness, and forecasting the best treatment options for various diseases. In this way, the field of diagnostics using artificial intelligence could be significantly expanded.