Abstract
OBJECTIVES: To assess changes in facial skin temperature over time, to identify sources of variation related to skin temperature and to evaluate interobserver reproducibility in measurements of the thermograms. METHODS: 62 volunteers (32 females, 30 males, mean age 23.4, range 19.5-29.5 years) underwent thermography of the face (left and right side lateral images) on four occasions with approximately 2 months between each session. Three observers recorded the images and marked regions of interest (ROIs) in each image using dedicated software. Smoking, exercise habits and use of oral contraceptives were recorded. RESULTS: A significant difference between sessions (≤1 °C, p<0.001) and between observers (≤0.11 °C, p<0.001) was identified. The difference between sides was not significant (≤0.07 °C, p=0.7). None of the interactions between side, session and observer were significant. Smoking, exercise habits and oral contraceptive intake were not significant impact factors when included as covariates in the analysis (p>0.1). ROI temperature was significantly higher in males than in females (0.7 °C, p<0.001). A mixed model analysis of variance showed that observer had little impact on the expected standard deviation, whereas session and subject had a greater impact. CONCLUSIONS: Face temperature is symmetrical and varies over time. The non-significant difference between sides is highly reproducible, even between observers.