Reviewed by Lexie CornerJun 4 2024
Researchers from Tsinghua University predicted the presence of coronary artery disease by combining facial thermal imaging and artificial intelligence (AI). The study has been published in the open-access journal BMJ Health & Care Informatics.
According to the researchers, this non-invasive real-time methodology is more effective than traditional methods. It could be used in clinical practice to enhance the accuracy of diagnosis and workflow, pending validation with a larger and more ethnically diverse group of patients).
The current criteria for diagnosing coronary heart disease rely on probability assessments of risk factors, which are not always highly reliable or widely applicable. While they can be supplemented with other diagnostics, including blood testing, angiography, and ECG readings, they are frequently intrusive and time-consuming.
Thermal imaging is a non-invasive method of obtaining surface temperature distribution and fluctuations by detecting the infrared radiation produced by the object. Since it can detect aberrant blood circulation and inflammation from skin temperature patterns, it has emerged as a promising tool for disease evaluation.
The ability of AI to collect, interpret, and integrate complicated data may improve the precision and efficacy of diagnostics using thermal imaging.
Therefore, the researchers set out to investigate if it would be possible to reliably predict the existence of coronary artery disease in 460 patients with suspected heart disease using thermal imaging and AI, all without the need for invasive or time-consuming procedures. Of them, 126 (27.5 %) were women, with an average age of 58.
Before confirmatory examinations, thermal images of their faces were taken to create and verify an AI-assisted imaging model for diagnosing coronary artery disease.
Thermal images of patients’ faces were used to develop and validate an AI-assisted imaging model for diagnosing coronary artery disease.
The thermal imaging plus AI technique predicted coronary artery disease 13 % better than the pre-test risk assessment, which included conventional risk factors and clinical signs and symptoms.
The overall left-right temperature difference of the face, along with the maximum and average facial temperatures, was the most important of the three predictive thermal indicators.
The greatest predictive factor was the average temperature of the left jaw area, followed by the right eye region’s temperature range and the left-right temperature differential of the left temple regions.
This method also successfully detected traditional coronary artery disease risk factors, such as high cholesterol, male sex, smoking, excess weight (BMI), fasting blood glucose, and inflammatory indicators.
The researchers acknowledge the study’s extremely small sample size and the fact that it was conducted at only one center (all research participants were referred for follow-up testing for possible cardiac conditions).
The researchers noted, “The feasibility of [thermal imaging] based [coronary artery disease] prediction suggests potential future applications and research opportunities. As a biophysiological-based health assessment modality, [it] provides disease-relevant Information beyond traditional clinical measures that could enhance [atherosclerotic cardiovascular disease] and related chronic condition assessment.”
They added, “The non-contact, real-time nature of [it] allows for instant disease assessment at the point of care, which could streamline clinical workflows and save time for important physician-patient decision-making. In addition, it has the potential to enable mass prescreening. Our developed [thermal imaging] prediction models, based on advanced [machine learning] technology, have exhibited promising potential compared with the current conventional clinical tools.”
“Further investigations incorporating larger sample sizes and diverse patient populations are needed to validate the external validity and generalisability of the current findings,” the researchers concluded.
Journal Reference:
Kung, M., et al. (2024) Prediction of coronary artery disease based on facial temperature information captured by non-contact infrared thermography. BMJ Health & Care Informatics. doi:10.1136/bmjhci-2023-100942