Mammogram Segmentation Using Neuro-fuzzy for Breast Cancer Auto-detection

A CAD system for the purpose of detecting and identifying cancerous tumors in mammograms is introduced in this study. Using this modern computer system segmentation is set to work based upon the combination of two issues_ fuzzy logic and neuro fuzzy to detect cancerous tumors. To train this segmenter, 70 mammograms containing tumors have been utilized. To, further, evaluate the efficiency of the system 30 more mammograms, not used in the training period, were added later. By applying this method the specificity index is raised to %95 and the sensitivity index to %98. Considering these measures the true algorithm capability is ensured.

mammography, neural networks, fuzzy-logic, neuro fuzzy systems, breast cancer, segmentation.