Comparison of Three Thermal Cameras with Canine Hip Area Thermographic Images

Abstract

The objective of this study was to compare the method of thermography by using three different resolution thermal cameras and basic software for thermographic images, separating the two persons taking the thermographic images (thermographers) from the three persons interpreting the thermographic images (interpreters). This was accomplished by studying the repeatability between thermographers and interpreters. Forty-nine client-owned dogs of 26 breeds were enrolled in the study. The thermal cameras used were of different resolutions—80 × 80, 180 × 180 and 320 × 240 pixels. Two trained thermographers took thermographic images of the hip area in all dogs using all three cameras. A total of six thermographic images per dog were taken. The thermographic images were analysed using appropriate computer software, the FLIR QuickReport 2.1. Three trained interpreters independently evaluated the mean temperatures of hip joint areas of the six thermographic images of each dog. The repeatability between thermographers was >0.975 with the two higher-resolution cameras and 0.927 with the lowest resolution camera. The repeatability between interpreters was >0.97 with each camera. Thus the between-interpreter variation was small. The repeatability between thermographers and interpreters was considered high enough to encourage further studies with thermographic imaging in dogs.