Wearable devices have racial bias issues in health monitoring, which are mainly reflected in the following aspects: - Data accuracy differences - There are differences in the physical characteristics and physiological parameter ranges of different racial groups. For example, the skin color of dark-skinned people will affect the accuracy of photoplethysmography (PPG) sensors in monitoring indicators such as heart rate and blood oxygen saturation. When detecting the same physiological state, wearable devices may produce greater measurement errors on dark-skinned individuals, resulting in inaccurate data, which in turn affects health assessments based on these data. - Algorithm bias - The algorithms in wearable devices are trained based on data from a specific population, which is often based on a certain race or region. When applied to other races, the algorithm may not accurately identify and interpret the unique physiological characteristics and health patterns of different racial groups, resulting in bias. - Health risk assessment bias - Health risk assessment models based on wearable device data may be inaccurate due to racial differences. For example, some cardiovascular disease risk assessment models have different predictive abilities in different races. They have higher predictive accuracy for white people, but may overestimate or underestimate the risk for other races such as black people, resulting in inaccurate health intervention measures for different racial groups.