Could we use digital infrared medical thermography to distinguish SARS-СоV-2 viral from community acquired pneumonia?

Abstract

Objectives: This study set out to determine the possibility to differ SARS-СоV-2 viral pneumonia from community acquired pneumonia (CAP) by mean of digital infrared medical thermography. Material, methods: Totally 380 patients were evaluated: with diagnosis 'U07.1COVID-19, virus identified. SARS-CoV-2 pneumonia' (group 1) — 208 pts, with community acquired viral or bacterial pneumonia (group 2) — 172 pts. Thermographic assessment was performed using TBC300-med infrared camera (manufactured by LLC 'STC SILAR', Saint Petersburg, Russia, image format 384 x 288, thermal resolution 30 mK). To store and analyze thermographic images cloud application 'TVision' (LLC 'Dignosys', Russia) was used. Quadrangle in shape regions of interest (ROI) were automatically created on the front and back of the thorax roughly in the projection of the upper lobe (ULP) and the lower lobe (LLP) of the lungs. Mean temperature was measured and thermal gradients- ΔТ° (side-to-side temperature difference) - were calculated between symmetric ROI both on front and back of the thorax. Values obtained were taken as modulus, the greatest value used for analysis. Results: Mean ΔТ° value were significantly higher in SARS-CoV-2 pneumonia compared with CAP group (0,56 ± 0,35 ºC and 0,32 ± 0,14ºC respectively; P < 0,05). The ΔТ° value > 0,51 °C is associated with SARS-CoV-2 pneumonia with sensitivity 71 % and specificity 62,8 %, area under the ROC curve 0.739; P < 0,05). Conclusions. Infrared thermography proved to be a sensitive method for detection of temperature differences between both types of pneumonia and may contribute to preliminary diagnosis.