Researchers have developed a deep learning model that estimates chronic stress burden by measuring adrenal gland volume on standard CT scans, introducing what they describe as the first imaging-based biomarker for chronic stress. The metric, called the Adrenal Volume Index, is linked to cortisol exposure, perceived stress, overall physiological stress load and long-term cardiovascular risk, according to findings to be presented at the Radiological Society of North America's annual meeting.
Chronic stress can profoundly affect health, contributing to problems such as anxiety, trouble sleeping, muscle pain, high blood pressure, a less effective immune system, and major conditions including heart disease, depression and obesity, according to the American Psychological Association.
A new study led by Elena Ghotbi, M.D., a postdoctoral research fellow at Johns Hopkins University School of Medicine in Baltimore, Maryland, proposes a way to visualize the long-term impact of stress using CT scans that patients already receive for other reasons.
According to a report from the Radiological Society of North America (RSNA), Ghotbi and colleagues trained a deep learning artificial intelligence tool to automatically measure adrenal gland size on routine chest CT images. Each year, tens of millions of chest CT scans are performed in the United States alone.
From these measurements, the team derived a metric they call the Adrenal Volume Index (AVI), defined as adrenal volume in cubic centimeters divided by height squared in meters (cm³/m²). The researchers describe AVI as an imaging marker that reflects chronic stress burden, in contrast to single time-point cortisol tests that capture hormone levels only at the moment of sampling.
The study used data from the Multi-Ethnic Study of Atherosclerosis (MESA) and included adults who had undergone chest CT imaging as well as detailed stress-related assessments. RSNA's summary reports that the team linked AI-derived AVI to cortisol measurements, allostatic load (a composite measure that can include factors such as body mass index, blood pressure and glucose levels) and psychosocial indicators like perceived stress and depression scores.
Higher AVI values were associated with greater overall cortisol exposure, higher peak cortisol levels and increased allostatic load. Participants who reported higher levels of perceived stress had higher AVI than those reporting lower stress. The researchers also found that AVI was related to a higher left ventricular mass index, a measure of heart structure, and that for every 1 cm³/m² increase in AVI, the risk of heart failure and death rose over follow-up of up to 10 years, according to the RSNA summary.
"Our approach leverages widely available imaging data and opens the door to large-scale evaluations of the biological impact of chronic stress across a range of conditions using existing chest CT scans," Ghotbi said in remarks released by RSNA.
Senior author Shadpour Demehri, M.D., a professor of radiology at Johns Hopkins, said the technique allows clinicians to visualize the long-term burden of stress inside the body using a scan that many patients already undergo as part of routine care.
In the RSNA report, co-author Teresa E. Seeman, Ph.D., professor of epidemiology at UCLA, said the work is especially notable because it links a routinely obtained imaging feature—adrenal volume—with validated biological and psychological measures of stress and demonstrates that it independently predicts a major clinical outcome.
The researchers say this imaging biomarker could potentially refine cardiovascular risk assessment and preventive strategies without additional radiation exposure or extra testing, and that it may be relevant across a range of stress-related diseases that commonly affect middle-aged and older adults.
Other contributors listed in the RSNA summary include Roham Hadidchi, Seyedhouman Seyedekrami, Quincy A. Hathaway, M.D., Ph.D., Michael Bancks, Nikhil Subhas, Matthew J. Budoff, M.D., David A. Bluemke, M.D., Ph.D., R. Graham Barr and Joao A.C. Lima, M.D.
