Quality assurance for a multi-centre thermography study

In this study we describe the quality assurance programme developed to evaluate the performance of six thermal imaging devices in the VALIDS study, a project set up by a United Kingdom (UK) group to investigate the reliability of thermography for the assessment of Raynaud's phenomenon secondary to systemic sclerosis. At each of the six centres the start-up drift, accuracy and repeatability of the thermal cameras was assessed by imaging an Isotech 988 blackbody source. All of the thermal cameras, except one (FLIR A35SC) had stable measurements within a 30 minute warm-up period. All of the thermal cameras, except the FLIR A35SC had accurate measurements within ±2°C of the blackbody temperature. Repeatability, as demonstrated by the calculation of the within-subject standard deviation, was less than 0.5°C for 5 of the 6 cameras; the exception being the FLIR Agema Thermovision 570 thermal camera. In conclusion the results confirmed that the thermal imaging device with less exacting accuracy specifications (FLIR A35SC) did not match the performance of the higher-specification devices, and it is questionable if it is fit for purpose when being used clinically. All the other devices demonstrated acceptable accuracy and repeatability for clinical use. These quality assurance methods, when employed along with rigorous patient preparation, image capture and analysis protocols, provide essential confidence in valid and reliable temperature measurements across multiple clinical centres.