When asked about a person's age, we usually look at their face to find any sign. A new study conducted by Japanese researchers suggests that an individual's true age can be determined using a chest X-ray.

Revealing a Person's True Age

In Osaka Metropolitan University (OMU), an advanced artificial intelligence (AI) model was developed by scientists to accurately estimate a person's chronological age. Their discovery can serve as a breakthrough in medical imaging and pave the way for improved detection of diseases.

Led by Yasuhito Mitsuyama and Dr. Daiju Ueda from the Department of Diagnostic and Interventional Radiology at OMU's Graduate School of Medicine, the research team first designed and constructed a deep learning-based AI model which can determine age from chest X-rays of healthy individuals. Then the model was applied to radiographs of patients with known diseases to evaluate the connection between AI-estimated age and each individual disease. The data were collected from multiple institutions since the AI trained on a single dataset is prone to bias.

Between 2008 and 2021, a total of 67,099 chest radiographs were obtained from 36,051 healthy individuals who underwent medical check-ups at three health institutions. These were used to develop, train, and test the AI model for age estimation. A correlation coefficient of 0.95 was shown by the developed model between the AI-estimated age and chronological age. This is considered as a good indicator since a correlation coefficient of 0.9 or higher is classified as very strong.

Additional 34,197 chest X-rays from two other institutions were compiled from 34,197 patients to validate the usefulness of AI-estimated age using radiographs as biomarker. It was revealed that the difference between AI-estimated age and the chronological age of the patient positively correlate with various chronic diseases like chronic obstructive pulmonary disease, hyperuricemia, and hypertension. This means that the higher the AI-estimated age is compared to the chronological age, the higher chance that an individual will develop these diseases.

As described by Mitsuyama, chronological age is one of the most significant factors in medicine. The result of their study suggests that the apparent age of a person based on chest radiography can accurately reflect their health conditions beyond chronological age.

In the future, the research team aims to improve their study and apply it in the estimation of the severity of chronic diseases. They are also hopeful in predicting life expectancy and in forecasting possible surgical complications.

READ ALSO: AI Tool Developed to Assess Lung Nodule Captured by CT Scans for Benign and Malignant Cancer


Radiography as Biomarker of Ageing

A biomarker refers to an objective measurement of the mechanisms that happen in a cell or an organism at a given time. Experts use biomarkers in understanding basic biological processes which can have practical medical and public health applications.

Radiography plays an important role as a biomarker in human age determination. Right after Wilhelm Roentgen discovered X-ray in 1895, radiology was applied in forensic sciences where the images are utilized in the process of age estimation.

In dentistry, dental radiography has been used as a simple and non-destructive method of age estimation since the 1982. The dental findings evaluated by radiography serve as an important source of information for forensic odontological age determination. Age assessment using dental X-rays involves the identification of mineralization stage of teeth shown on radiographic images and comparing them with the standard age in order to estimate the approximate age range.

 

 

RELATED ARTICLE: Explainable AI Accurately Labels X-Ray Images of Five Chest Pathologies Equivalent to Seven Human Experts


Check out more news and information on X-ray in Science Times.