AI helps experts 'see' stress on routine CT scans
Using artificial intelligence and routine CT scans, researchers say they have identified the first ever imaging biomarker of stress.
For their work, the team focused on the adrenal glands. Located on top of the kidneys, the glands play a significant role in stress due to the hormones they produce, such as cortisol, epinephrine and norepinephrine, which regulate the body’s response to stressful situations and events.
It has been well established that chronic stress can affect an individual both physically and psychologically but measuring its impact objectively has remained a challenge. Recently, experts developed a deep learning model they believe can help achieve this.
The model measures adrenal gland volume using routine chest CT scans. Larger volume is indicative of chronic or severe stress and could give experts an idea of how vulnerable patients are to the associated risks of long-term tension.
“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,” lead author Elena Ghotbi, MD, a postdoctoral research fellow at the Johns Hopkins University School of Medicine in Baltimore, said in a news announcement on the findings. “This AI-driven biomarker has the potential to enhance cardiovascular risk stratification and guide preventive care without additional testing or radiation.”
The team retrospectively applied the algorithm to the CT scans of 2,842 participants from the Multi-Ethnic Study of Atherosclerosis to segment and calculate their adrenal gland volume. That study required participants undergo imaging in addition to a slew of other assessments, including stress questionnaires, cortisol measures and markers of allostatic load. This gave researchers ample subjective and objective data to compare.
An Adrenal Volume Index for each participant was calculated by dividing the glands’ volume by its height. This measurement was compared to individuals’ salivary cortisol levels (collected via swab eight times per day over two days) and allostatic load, which was based on body mass index, creatinine, hemoglobin, albumin, glucose, white blood count, heart rate and blood pressure.
The algorithm’s AVI calculations were found to have significant associations with increased cortisol levels, allostatic load and psychosocial stress measured via stress questionnaires. Those with the highest perceived stress also had the highest AVI measurements; each 1 cm³/m² increase in AVI was linked to greater risk of heart failure and other adverse cardiac events in the future.
“With up to 10-year follow-up data on our participants, we were able to correlate AI-derived AVI with clinically meaningful and relevant outcomes,” Ghotbi said. “This is the very first imaging marker of chronic stress that has been validated and shown to have an independent impact on a cardiovascular outcome, namely, heart failure.”
“For over three decades, we’ve known that chronic stress can wear down the body across multiple systems,” added Teresa E. Seeman, PhD, study co-author, professor of epidemiology at UCLA and a researcher in stress and health. “What makes this work so exciting is that it links a routinely obtained imaging feature, adrenal volume, with validated biological and psychological measures of stress and shows that it independently predicts a major clinical outcome. It’s a true step forward in operationalizing the cumulative impact of stress on health.”
The full research is set to be presented during this year’s annual Radiological Society of North America (RSNA) meeting, set to begin Sunday, Nov. 30, in Chicago.
In addition to her background in journalism, Hannah also has patient-facing experience in clinical settings, having spent more than 12 years working as a registered rad tech. She began covering the medical imaging industry for Innovate Healthcare in 2021.