The deep learning-based artificial intelligence (AI) model was trained on chest radiographs of healthy people from multiple institutions, then the model was applied to radiographs from those with known diseases to analyze the relationships between the AI-estimated age and each disease. For the development, training, and internal/external testing of the AI model for age estimation data was collected from 67,099 chest radiographs of 36,051 healthy people who underwent checkups at three facilities between 2008-2021. The model showed a correlation coefficient of 0.95 between the estimation and chronological age, typically a 0.9 or higher is considered to be very strong.
To validate the usefulness of the AI model using chest radiographs as a biomarker, an additional 34,197 chest radiographs were compiled from 34,197 people with known diseases visiting two other institutions. These additional results demonstrated that the difference between the AI estimation of age and the patient’s chronological age was positively correlated with a variety of chronic diseases like hypertension, COPD, and hyperuricemia. Simply put, the higher the AI estimation compared to the person’s chronological age the more likely that person was to have a chronic disease.
“Chronological age is one of the most critical factors in medicine,” stated Mr. Mitsuyama. “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.”
