A new study suggests artificial intelligence can detect chronic stress by examining routine medical scans. Researchers say their model identifies a biological marker of long-term stress using standard chest CT images.
Scientists analysed scans from almost 3,000 patients and focused on the adrenal glands, which regulate stress hormones. They compared adrenal size measured by AI with other indicators such as cortisol levels, blood pressure, and body mass index. Patients who reported stress showed larger adrenal glands and higher risks linked to heart failure.
The research will be presented next week at the Radiological Society of North America meeting. Experts believe the findings could reshape how doctors understand the hidden impact of chronic stress.
Until now, stress has been measured through surveys or demanding laboratory tests, limiting accurate assessment. The new approach uses scans that many patients already undergo, making detection easier and more accessible for hospitals. Researchers describe the glands as a “biological barometer” that reflects how the body handles long-lasting pressure.
Chronic stress is known to contribute to heart disease, depression, and obesity, affecting millions worldwide. The study’s authors argue that early identification may help prevent these conditions before symptoms become severe. The AI system may also help uncover other illnesses linked to stress, especially in older adults.
Radiology professor Shadpour Demehri said the technology allows clinicians to visualise the burden of stress directly. He noted that previous methods were limited, offering only partial insights into the body’s response.
The study has not yet been published in a peer-reviewed journal, but researchers expect growing interest as stress levels rise globally. They say future work will explore how the technology might guide treatment or support public health decisions.
