A new method using facial thermal imaging and AI has potential in predicting coronary artery disease more accurately than traditional methods, as per research in BMJ Health & Care Informatics. This non-invasive technique could enhance diagnostic workflows in clinical practice pending further confirmation.
Facial Thermal Imaging and AI Predict Coronary Artery Disease
A new non-invasive method using facial thermal imaging combined with artificial intelligence (AI) has shown promise in accurately predicting the presence of coronary artery disease (CAD), according to research published in BMJ Health & Care Informatics. This method could potentially improve diagnosis workflows in clinical practice pending further validation.
Current diagnostic guidelines for coronary heart disease include risk assessments, ECGs, angiograms, and blood tests. These methods, while effective, can be invasive and time-consuming. Thermal imaging, which detects infrared radiation to highlight temperature distributions on the skin, can identify abnormal blood circulation and inflammation. When combined with AI, this technique can process complex information to enhance diagnostic accuracy.
In a study involving 460 individuals with suspected heart disease, with an average age of 58 and 27.5% women, thermal images of participants’ faces were captured before confirmatory tests. The developed AI model was able to predict CAD with 13% greater accuracy than traditional methods. Key predictors included temperature differences between the left and right side of the face, maximal facial temperature, and average facial temperature, with the left jaw region’s temperature being the strongest indicator.
Although the study’s sample size was relatively small and conducted at a single center, researchers believe this method could streamline clinical workflows and facilitate mass prescreening in the future. Further research with larger, more diverse populations is recommended to validate these findings.
Elevated Troponin Levels Linked to Increased Death Risk
A study conducted by the University of Southampton and published in Heart links high levels of the cardiac protein troponin to an increased risk of death from any cause over the next few years, even without known cardiovascular disease. Tracking 20,000 hospital patients of average age 61, the research found a 76% higher risk of death associated with elevated troponin levels.
High cardiac troponin levels are typically used to diagnose heart attacks. However, in this study, the heightened death risk also included non-cardiovascular causes, most notably cancer. After excluding deaths within 30 days, the risk association persisted, indicating a broader prognostic role for troponin levels. The findings suggest a potential for the troponin blood test as a general predictor of medium-term mortality.
Professor Nick Curzen from the University of Southampton emphasized that troponin levels could serve as a snapshot indicator of health beyond their current use in diagnosing acute heart attacks.
