Artificial Intelligence pinpoints diabetes susceptibility through assessment of fat surrounding the heart
New AI Tool Developed to Measure Fat Around the Heart, Predict Diabetes Risk
A team from Queen Mary University of London has developed an AI tool that can accurately determine the amount of fat around the heart from MRI scan images. The tool, developed as part of the CAP-AI programme led by Barts Life Sciences, has been tested on over 45,000 people's heart MRI scans, including participants in the UK Biobank.
The research paper titled 'Automated quality-controlled cardiovascular magnetic resonance pericardial fat quantification using a convolutional neural network in the UK Biobank' was published in Frontiers in Cardiovascular Medicine. The authors of the paper include Andrew Bard from the Department of Radiology at the University of California, San Francisco, and a team from Queen Mary University of London.
The AI tool has been designed to measure the fat around the heart automatically and quickly, in under three seconds. It includes an in-built method for calculating its own results' uncertainty, ensuring accurate and reliable measurements.
Researchers found that a larger amount of fat around the heart is associated with significantly greater odds of diabetes, independent of a person's age, sex, and body mass index. This discovery could have important implications for patient care, as the distribution of fat in the body, particularly around the heart, can influence a person's risk of developing various diseases, including diabetes, atrial fibrillation, and coronary artery disease.
The funding for CAP-AI comes from the European Regional Development Fund and Barts Charity. The project is led by Capital Enterprise in partnership with Barts Health NHS Trust, Digital Catapult, and The Alan Turing Institute.
The research paper is now available for reading, and can be accessed through the following link: https://www.frontiersin.org/articles/10.3389/fcvm.2021.677574
While the paper does not mention any new AI tool or its clinical utility, the novel AI tool has high utility for future research and, if clinical utility is demonstrated, may be applied in clinical practice to improve patient care.
