Using Biological Age to Detect and Mitigate Chronic Disease

By Olivia Ramirez

Dr. Vineet Raghu set out to pursue a career in medicine when he began attending the University of Pittsburgh. But, two years into his undergraduate studies, he took a computer science course and his plans changed. Raghu developed a passion for computational genomics, the process of using computational tools, like artificial intelligence (AI), to understand the relationship between genes and disease. Years later, his post-doctoral work focuses on AI in medical imaging, and specifically on detecting and preventing cardiovascular disease (CD) and cancer.  

In recent years, there has been an explosion of interest in the use of AI and machine learning to understand diseases. In 2017, Raghu’s teammate Dr. Michael Lu began a project that used chest X-rays to predict a person’s risk of death. That project was the basis for Raghu’s National Academy of Medicine (NAM) Healthy Longevity Catalyst Award-winning project, Deep Learning to Predict Biological Age and Longevity from Chest Radiographs.

Studying age isn’t a new concept. Researchers categorize age into two different groups. Chronological age is the number of years that a person has lived and is one of the biggest indicators of risk for chronic disease. Two people can have the same chronological age but are likely to have different levels of risk for developing diseases based on their lifestyles. That’s where researchers look to biological age, which better measures how lifestyle factors, environmental factors, and social determinants of health impact anatomy and physiology. So, how can we effectively evaluate an individual’s risk of chronic disease based on their unique lifestyle?

Example of images from chest X-ray and CNN on two patients.

Raghu and his team chose to study CD and lung cancer, among the deadliest diseases in the United States, contributing to half of all deaths. Raghu and his team propose using a convolutional neural network (CNN) to look at chest X-rays (CXRs) to estimate an individual’s biological age (they call chest X-ray age or CXR-Age), which they compare to chronological age to estimate the risk of CD and lung cancer. These images can reveal important information, such as deformities in the spine, heart size, and posture. When considered together, this provides insight into a person’s health.

To date, the team has demonstrated proof of concept by showing that their model can estimate biological aging. Their work has been recently published in JACC: Cardiovascular Imaging. In the next phase of research, the team’s goal is to show that the model can accurately predict risk for CD and lung cancer, and will compare those results to the detection rates of traditional prevention and screening guidelines for each disease. The new model could be used to complement current screening guidelines, by providing an easily assessable measure of risk. The NAM Catalyst Award will support the expansion of the team and work to improve the proof-of-concept model.

The ultimate goal of the project is to incorporate the model into a physician’s workflow and provide better outcomes for patients. In other words, when a chest X-ray is ordered during routine care, the patient’s biological age would be included in the data provided to physicians and help inform shared decision-making with the patient with regard to screening and prevention.

With his expertise in computer science, Raghu continues to learn the medical and biological implications of his work as the project progresses. He is looking forward to the opportunity to meet others at the competition’s Innovator Summit to gain a more in-depth understanding of how the tools he’s creating can be translated into clinical settings. Raghu said, “I’m building these tools but don’t necessarily know how to use them. I’m excited to meet other Catalyst Award winners to learn how to use these tools to solve real problems.”

The growing population of people over 65 is one of the most important challenges we are facing, according to Raghu. Finding ways to keep people healthy as they are living longer is an important task, and he believes that AI has the potential to help address general chronic disease processes that occur as a result of aging.


This article is part of a series of profiles on the 2020 winners of the National Academy of Medicine’s Healthy Longevity Catalyst Awards — a component of the Healthy Longevity Global Competition, a multiyear, multimillion-dollar international competition seeking breakthrough innovations to improve physical, mental, and social well-being for people as they age The current application cycle has been extended and will accept applications through April 25, 2021 Apply today! Learn more about the award, the winners’ research, and ideas for promoting healthy aging here.


Sign up for updates