Brown adipose tissue (BAT) is a unique organ in existence in mammals. It can induce non-shivering thermogenesis to control body temperature and energy balance through the expression of uncoupling protein 1 (UCP1). In our study, we aimed to evaluate the prevalence of BAT, as detected by fluorine 18-fluorodeoxyglucose (18F-FDG) positron emission tomography combined computer tomography (PET/CT), in a Hong Kong Chinese population. We also assessed the influence of age and sex to BAT in Hong Kong Chinese population. We also determined the factors associated with it, in particular, its relationship with overweight and other metabolic disorders such as diabetes mellitus. We analyzed 1765 consecutive 18F-FDG PET-CT scans of 1442 Chinese for the presence of BAT. Comparison of the variables between positive and negative BAT scans was performed using Student’s t-test. The association between maximum value of standardized uptake value (SUVmax) and variables were explored by Spearman correlation. The predictors of observed BAT were analyzed by multiple logistic regression to determine the significant predictors of positive BAT. The relationship between the monthly numbers of subjects with BAT and the respective mean monthly outdoor temperature was evaluated by Pearson’s correlation co-efficient. P < 0.05 was considered to be statistically significant. Brown adipose tissue was detected in 66 out of 1442 subjects (4.6%). BAT was significantly more commonly found in younger (43.7±13.5 years old vs. 61.4±14.2 years old, P<0.001) and female (59% vs. 46%, P<0.05) subjects. BAT also existed more frequently in subjects with lower body mass index (BMI) (21.2±3.1 kg/m2 vs. 22.4±3.7 kg/m2, P<0.01) and lower blood glucose level (5.9±0.9 mmol/L vs. 6.4±1.6 mmol/L, P<0.01). Also, BAT was detected only in subjects with no history of diabetes meallitus (DM) (0 vs. 10%, P<0.01). Moreover, lower outdoor temperature (21.6±4.6。C vs. 23.4±4.7。C, P<0.005) resulted in higher prevalence of detected BAT. In the multiple logistic regression test, age and mean monthly temperatures were the significant independent predictors of the presence of BAT (P< 0.001 and P=0.001). Age was also significantly correlated to SUVmax (P< 0.001). The monthly prevalence of positive BAT correlated negatively with mean monthly temperature by Pearson’s correlation (r = -0.79; P<0.01). To summarize, BAT was more commonly found in young, female subjects with lower BMI and blood glucose levels, and non-diabetes subjects. Age was the most important factor associated with the prevalence of BAT in humans. Lower outdoor temperature in winter can increase the prevalence of BAT even in Hong Kong’s sub-tropical climates. Also, there was an association of BAT with normal BMI (<=23) and lower blood sugar levels supporting the notion that BAT may potentially be a therapeutic target for obesity and diabetes. / published_or_final_version / Diagnostic Radiology / Master / Master of Philosophy