In a groundbreaking study published in the Proceedings of the National Academy of Sciences, a team of Chinese scientists introduced an innovative approach to aging and disease prediction. Utilizing a multimodal image Transformer system, the researchers employed tongue, retinal, and facial images to estimate biological age, providing a novel method for predicting the risk of aging-related chronic diseases. Traditional markers for disease prediction have faced challenges due to the heterogeneity in tissues and organs, making identifying and standardizing such markers difficult. Biological age, based on functional and structural changes during aging, serves as a more reliable marker, and the study demonstrated the effectiveness of an AI model in estimating biological age using diverse image data.

Based on a Transformer architecture, the AI model incorporated information from retinal fundus, facial, and tongue images to predict biological age. Training and validation involved healthy participants, while testing the model utilized images from individuals with chronic diseases or known risk factors. The results indicated that the AI-based tool, using AgeDiff as a marker (the difference between chronological and predicted biological ages), outperformed other biological age prediction tools, offering a robust method to detect and predict age-related chronic diseases.

Read more:

#innovativeaging #diseaseprediction #multimodalAI #biologicalage #transformermodel #healthcareinnovation #precisionmedicine #chronicdisease #AIresearch #ageprogression