Regression model for predicting core body temperature in infrared thermal mass screening

• DOI: 10.1016/j.ipemt.2022.100006
• Publication Date: 2022-07-15T14:46:01Z
• Authors: Chayabhan Limpaba...; Frances Sophie Wo...; Rui Li; Zion Tse
• Keywords: Core body temperature; Medicine (General); 03 medical and health sciences; R5-920; 0302 clinical medicine; Infrared thermography; COVID-19; Regression analysis; Article; 3. Good health

With fever being one of the most prominent symptoms COVID-19, implementation screening has become commonplace around world to help mitigate spread virus. Non-contact methods temperature screening, such as infrared (IR) forehead thermometers and thermal cameras, benefit by minimizing infection risk. However, IR measurements may not be reliably correlated with actual core body temperatures. This study proposed a trained model prediction using IR-measured facial feature temperatures predict comparable an FDA-approved product. The reference were measured commercially available monitoring system. Optimal inputs training models selected correlation between predicted temperature. Five regression tested during study. linear showed lowest minimum-root-mean-square error (RSME) compared temple nose region interest (ROI) identified optimal inputs. suggests that data could provide comparatively accurate for rapid mass potential COVID cases model. Using modeling, non-contact measurement SpotOn system mean SD ± 0.285 °C MAE 0.240 °C.