The analysis of astrocyte morphology is performed in multiple steps. First, astrocytes are detected from IF images stained for GFAP and DAPI using a deep-learning model. Then, for each cell, the soma and processes are segmented and separated. Many morphological features, such as the soma area, the process thickness, and the number of branching points, are then extracted. Each astrocyte is then classified as having a normal or hypertrophic morphology based on the morphological metrics using a machine-learning model, and a morphology score is calculated. Finally, metrics such as the density of astrocytes of a specific morphology or the average morphology score, are aggregated by ROI, mice, and group for statistical comparison.