Stressed out? AI can detect first biological marker of chronic stress by looking at medical scans

Artificial intelligence (AI) can now tell whether someone may have chronic stress by looking at routine medical scans, according to a new study.

Using a deep learning AI model to analyse routine chest CT scans, scientists can now identify a first-of-its-kind biological marker of chronic stress.

Chronic stress can contribute to the development of major illnesses, such as heart disease, depression, and obesity. Yet until now, medical professionals lacked an objective, scalable way to measure its impact.

The research, to be presented next week at the meeting of the Radiological Society of North America (RSNA), collected data from almost 3,000 patients, which included computed tomography (CT) scans of their adrenal glands, stress surveys, cortisol levels, and other health data that can indicate long-term stress.

The researchers then used an AI deep learning model that they had developed, which can measure the size of the patient’s adrenal glands from their scans. That data was compared to other stress measures the team took of their patients, such as body mass index (BMI), blood pressure, and heart rate.

Adrenal glands are responsible for producing and regulating hormones that control metabolism, the immune system, blood pressure, and stress responses. The researchers call them a “biological barometer” for measuring stress in the body.

The patients who reported stress before the study had a higher adrenal volume, along with higher levels of cortisol and risks of heart failure.

“For the first time, we can ‘see’ the long-term burden of stress inside the body, using a scan that patients already get every day in hospitals across the country,” radiology professor and report co-author Shadpour Demehri said in a statement.

Demehri added that there “hasn’t been a way to measure and quantify the cumulative effects of chronic stress,” aside from patient questionnaires or markers such as chronic inflammation or cortisol spikes, which are “very cumbersome,” to get.

The researchers said their AI model can be used to identify “a variety of diseases” that are associated with stress in older adults.

The study has not yet been published in a peer-reviewed journal.