AI technology now unveils chronic stress through routine CT scans, a significant breakthrough for healthcare.

Story Highlights

• AI detects the first imaging biomarker of chronic stress using CT scans.
• Findings link adrenal gland volume to stress indicators like cortisol.
• Potential to revolutionize stress-related healthcare without additional tests.

AI Innovation in Stress Detection

Researchers at Johns Hopkins University have pioneered a deep learning model that identifies chronic stress by analyzing adrenal gland volume on routine chest CT scans. The model leverages existing data, avoiding additional radiation or procedures. This approach offers a new biomarker for stress, validated against cortisol levels and other indicators, marking a substantial advance in integrating AI with healthcare.

AI finds a hidden stress signal inside routine CT scans https://t.co/pyHgKLSnZH

— Zicutake USA Comment (@Zicutake) December 14, 2025

Linking Adrenal Volume to Health Risks

The study, presented at the Radiological Society of North America’s 2025 Annual Meeting, involved data from 2,842 participants of the Multi-Ethnic Study of Atherosclerosis (MESA). Findings show larger adrenal volume correlates with higher cortisol levels and increased allostatic load, both markers of chronic stress. This volume also aligns with a greater risk of future heart failure and mortality, emphasizing the biomarker’s clinical significance.

This AI-driven approach could significantly enhance cardiovascular risk stratification, integrating stress burden into routine cardiology assessments. By visualizing the physiological impacts of stress, healthcare providers can gain insights into patients’ conditions without invasive testing.

Watch:

Implications for Healthcare and Beyond

The adoption of this technology could usher in a new era of preventative healthcare, focusing on stress management as a cornerstone of disease prevention. The AI model’s ability to detect stress without additional testing could transform how stress-related conditions are diagnosed and managed, potentially reducing healthcare costs and improving patient outcomes.

However, challenges such as data privacy and the risk of algorithmic bias must be addressed to ensure equitable and ethical implementation. The MESA cohort’s diversity supports generalizability, but further evaluation is needed across various clinical settings to avoid biases.

24 7 confidential AI help for stress and burnout.

Sources:

AI Uncovers Imaging Biomarker of Chronic Stress in Routine CT Scans
AI Uncovers Hidden Stress Damage in the Body
RSNA Press Release
AI Helps Experts ‘See’ Stress on Routine CT Scans