Wearable computing is about data collected from body-worn sensors. Accelerometry data, in raw format, are high-frequency triaxial accelerations which are proxies of intensity and direction of movements, while, in processed format, are often minuted-by-minute activity counts, measurements of average activity intensity in each minute. Accelerometry data provide objective and detailed measurements of physical activity and have been widely used in observational studies and clinical trials.
Physical activity is an important biomarker of human aging. One interesting question is to quantify the circadin rhythms of physical activity (activity counts) and its relation with age. I designed a computationally fast bivariate smoother, the sandwich smoother, to smooth functional data and obtained the results displayed in the left figure. In the two heatmaps, bule corresponds to no or little activity, green and yellow correspond to light intensity activity, and red corresponds to moderate and vigorous activity intensities. Blue areas overlap with the known resting periods of people; for example, 11pm to 6am for 40 year old people. Red areas overlap with the working and activity hours of the day; for example 8am to 8pm for 40 year old people and 8am to 2pm for 80 year old people. For more details, please refer to my to-appear-in-biostatistics paper Quantifying the life-time circadian rhythm of physical activity: a covariate-dependent functional approach .