AI Unveils a Hidden Stress Signal in Routine CT Scans
A groundbreaking study has harnessed the power of artificial intelligence to uncover a novel biomarker of chronic stress, offering a glimpse into the body's hidden stress burden. The research, presented at the Radiological Society of North America (RSNA) annual meeting, introduces a deep learning model that can identify adrenal gland size on standard medical images, a significant advancement in understanding the long-term effects of stress.
Chronic stress, a pervasive issue, extends beyond mood, impacting both physical and mental health. It's linked to anxiety, sleep disturbances, muscle pain, high blood pressure, and a weakened immune system, according to the American Psychological Association. Moreover, ongoing stress has been associated with severe conditions like heart disease, depression, and obesity.
AI Measures Adrenal Glands on Routine CT Scans
The study's lead researcher, Dr. Elena Ghotbi, developed a deep learning tool to calculate adrenal gland size from existing CT scans. This innovation is significant because it utilizes widely available imaging data, enabling large-scale evaluations of chronic stress's biological impact.
Each year, millions of chest CT scans are performed in the U.S. alone, providing a vast resource for stress assessment. Dr. Ghotbi emphasizes the potential of this AI-driven biomarker to enhance cardiovascular risk assessment and guide preventive care without additional testing or radiation.
Visualizing the Stress Burden in the Body
Senior author Dr. Shadpour Demehri highlights the commonality of chronic stress, experienced daily by many adults. The study introduces a groundbreaking approach to measuring and quantifying the cumulative effects of chronic stress, a challenge previously addressed only through questionnaires and surrogate markers.
Unlike single cortisol tests, which measure stress at a specific moment, adrenal gland size serves as a long-term gauge of chronic stress. This novel biomarker offers a more comprehensive view of stress's impact on the body.
Large Multi-Ethnic Cohort Links Imaging, Hormones, and Stress Load
The research analyzed data from 2,842 participants in the Multi-Ethnic Study of Atherosclerosis, a comprehensive study combining chest CT imaging, validated stress questionnaires, cortisol measurements, and allostatic load indicators. This unique cohort, integrating imaging, biochemical data, and psychosocial assessments, was instrumental in creating an imaging-based marker of chronic stress.
The investigators applied their deep learning model to CT scans, automatically outlining and measuring adrenal gland volume. They introduced the Adrenal Volume Index (AVI), calculated as adrenal volume (cm³) divided by height² (m²). Salivary cortisol measurements and allostatic load calculations provided insights into hormonal patterns and the cumulative physiological and psychological effects of stress.
AVI Tracks Stress, Hormones, and Heart Risk
The team's findings revealed a strong correlation between AVI, cortisol levels, allostatic load, and psychosocial stress indicators. Higher AVI values were associated with increased cortisol exposure, peak cortisol levels, and allostatic load. People reporting high perceived stress had elevated AVI compared to those with low stress.
AVI was also linked to a higher left ventricular mass index, a heart structure measure. For every 1 cm³/m² increase in AVI, the risk of heart failure and death rose. Dr. Ghotbi emphasized the clinical relevance of these findings, correlating AI-derived AVI with heart failure outcomes over a 10-year follow-up.
A New Approach to Quantifying Stress's Impact
Co-author Dr. Teresa E. Seeman, a pioneering stress and health researcher, highlights the excitement of linking a routine imaging feature with validated biological and psychological stress measures. This approach predicts a major clinical outcome, marking a significant step in understanding stress's cumulative impact on health.
Dr. Demehri explains that this simple imaging measure, combined with well-established stress and disease markers, offers a practical method for measuring chronic stress in clinical practice. The biomarker's accessibility from widely performed CT scans and its physiological relevance to chronic stress make it a valuable tool for future research and clinical applications.
The researchers suggest that this imaging biomarker could be applied to various stress-related diseases affecting middle-aged and older adults, opening new avenues for stress management and cardiovascular health.
