A deep learning model can predict cardiovascular risk in women using routine mammograms, performing comparably to traditional risk calculators.
Key Details
- 1Developed by The George Institute for Global Health in collaboration with UNSW and University of Sydney.
- 2Uses only mammographic features and age—no need for additional clinical data.
- 3Validated using data from over 49,000 routine mammograms linked to hospital and death records.
- 4Performs comparably to traditional cardiovascular risk models requiring more extensive data.
- 5Potential for integration into existing high-participation breast screening programs globally.
Why It Matters
This approach could allow opportunistic cardiovascular risk screening for women already undergoing mammograms, potentially improving early identification and prevention of heart disease. It demonstrates innovative use of imaging AI beyond traditional cancer detection, with public health implications for women's care.
Source
EurekAlert
Related News
•EurekAlert
AI-Assisted Tracking Reveals Stem Cell Therapy Promotes Stroke Recovery in Mice
Mice with ischemic stroke receiving human neural stem cell transplants showed robust brain recovery, measured using AI-based limb tracking.
•EurekAlert
Moffitt Develops AI Model to Predict Urgent Care in Lung Cancer Patients
Moffitt Cancer Center researchers created machine learning models that use patient-reported outcomes and wearable data to predict urgent care visits for non-small cell lung cancer patients.
•EurekAlert
AI Predicts Keratoconus Progression Using OCT Scans
Researchers have developed an AI model that accurately predicts which keratoconus patients require treatment by analyzing OCT eye scans and clinical data.