A new AI model developed by Osaka Metropolitan University can accurately reveal a patient’s chronological age. It uses chest radiographs.
However, when there is a disparity, it can also indicate a link to chronic illness. These discoveries represent a breakthrough in medical imaging and open the door to better early disease identification and treatment.
Chest radiographs are commonly available and affordable, but more is needed to know about their potential as a biomarker of aging when combined with data from other institutions.
In this new study, scientists aimed to develop a biomarker of aging from chest radiography and examine the correlation between the biomarker and diseases.
They next applied the model to radiographs of individuals with known diseases to examine the association between AI-estimated age and every disease. Due to the risk of overfitting associated with AI trained on a single dataset, the scientists gathered data from other institutions.
For the biomarker modeling, scientists used healthy chest radiographs of 67,099 individuals from 36,051 healthy individuals between 2008 and 2021. These individuals underwent health check-ups at three facilities.
To evaluate the performance of the AI model in estimating ages, scientists calculated the correlation coefficient, mean square error, root mean square error, and mean absolute error. They also investigated the odds ratios (ORs) for various diseases. The developed model showed a correlation coefficient 0.95 between the AI-estimated age and chronological age. Generally, a correlation coefficient of 0.9 or higher is very strong.
An additional 34,197 chest radiographs from 34,197 patients with known disorders from two other institutions were collected to evaluate the usefulness of AI-estimated age using chest radiographs as a biomarker. The findings showed that several chronic conditions, including hypertension, hyperuricemia, and chronic obstructive pulmonary disease, were positively connected with the gap between AI-estimated age and the patient’s chronological age. In other words, people were more likely to develop these diseases the older their AI-estimated age was relative to their chronological age.
Graduate student Yasuhito Mitsuyama at the Graduate School of Medicine, OsakaMetropolitan University, said, “Chronological age is one of the most critical factors in medicine. Our results suggest that chest radiography-based apparent age may accurately reflect health conditions beyond chronological age. We aim to develop this research further and apply it to estimate the severity of chronic diseases, to predict life expectancy, and to forecast possible surgical complications.”
The AI-estimated age using chest radiographs strongly correlated with chronological age in the healthy cohorts. Furthermore, in cohorts of individuals with known diseases, the difference between estimated age and chronological age correlated with various chronic diseases. This biomarker might pave the way for enhanced risk stratification methodologies, individualized therapeutic interventions, and innovative early diagnostic and preventive approaches towards age-associated pathologies.”
