To evaluate the high-grade breast cancer morphological complexity on mammogram. We conducted a retrospective study using an open source data got from figshare repository. These anonymized data were collected and used for a study approved by the institutional review board. Cranio-Caudal and Medio-lateral mammograms and their tumor segmented images from 66 patients subdivided in two groups high histological grade (n=23) low-grade (low and intermediate, n=41). From breast cancer image segmentation, we extracted fractal dimension using Fraclac, plugin of ImageJ software based on box-counting method. For our analysis we used comparatively the fractal dimension from cranio-caudal (CC) and medio-lateral (MLO) images. We summarized the fractal dimension of our cohort using boxplot and performed the Wilcoxon non-parametric statistic for fractal dimension comparison of two groups (High-grade and low-grade). There was not difference between CC (mean ± std= 1.1583±0.067) andmLO (mean ± std =1.1551±0.055) breast cancer fractal dimension. For the high-grade differentiation, CC andmLO images fractal dimension were contributed respectively at a little difference but without statistically difference (P value=0.438 and 0.435). High-grade fractal dimensions mean were respectively 1.142±0.044 and 1.144±0.075 for CC andmLO images against 1.166±0.050 and 1.160±0.057 for low-grade. It had been recorded a lower mean value of fractal dimension for high-grade breast cancer without statistically significant. This finding shows that the high-grade breast cancer tends to have a regular shape.