This paper describes ongoing work in analyzing sensor data logged in the homes of seniors. A visualization of motion sensor data in the form of a density map which uses different colors to show levels of activity was introduced previously. For evaluating changes in activity level and periodicity of life style, we introduce a dissimilarity measurement of density maps based on co-occurrence matrices. The dissimilarity between two density maps is captured using texture features for automatically determining changes in activity patterns. A case study is included to illustrate how the density map can be used to track activity patterns over time.