Artificial intelligence models can estimate age and age through chest X-rays | Artificial intelligence | Chest
Osaka Public University in Japan recently released a statement stating that its researchers have developed an advanced artificial intelligence model that can accurately estimate the age of filmmakers using chest X-rays. When there is a difference between the estimated results and the actual age, it indicates that it may be related to chronic diseases, and related research can help improve the detection and intervention of early diseases.
The research team first constructed a deep learning based artificial intelligence model to estimate the age of the filmmaker from chest X-rays of healthy individuals. Then, they used the artificial model to analyze chest X-rays of known disease patients to study the relationship between artificial intelligence estimation of age and various diseases.
Between 2008 and 2021, researchers obtained 67099 chest X-rays of healthy individuals from 36051 healthy individuals undergoing health examinations in three institutions. The analysis results show that the correlation coefficient between the estimated age of the artificial intelligence model and the actual age is 0.95.
In order to verify the effectiveness of artificial intelligence using chest X-rays to estimate age, researchers collected 34197 chest X-rays of known disease patients from two other institutions. Analysis shows that the difference between the estimated age of artificial intelligence and the actual age of patients is positively correlated with various chronic diseases such as hypertension, hyperuricemia, and chronic obstructive pulmonary disease. Compared to actual age, the higher the age estimated by artificial intelligence, the greater the likelihood of individuals suffering from these diseases.
The research paper has been published in the recently published journal The Lancet: Elderly Health.
Researchers suggest that artificial intelligence based on chest X-ray estimates can accurately reflect health status based on actual age. The next step of the research is to apply it to assess the severity of chronic diseases, predict life expectancy, and predict possible surgical complications.