The results of mice segmentation based on dynamic contrast enhanced micro-CT images. The data derived from seven female Balb/c mice with body weights ranging from 14 to 18 g. For each mouse, five sets of data matrix were collected, representing three manual segmentation results named M1, M2, and M3, automatic segmentation results from the support vector machine classifier named SVM, and the random forest classifier named RF, respectively. The matrix size is 200*200*450, containing 8-bit integeral data. The values rangeing from 1 to 6 with a step of 1 denote the heart, liver, spleen, lung, kidney and bone, respectively.The manual segmentation of the bone was only contained in M1.