Breast abnormality detection in mammograms using fuzzy inference system

One of the leading diseases in women is breast cancer. The detection in an earlier stage is done by indicating the presence of microcalcification or mass. We develop two detection systems that can help a radiologist to detect microcalcifications and masses in mammograms. In particular, we utilize Mamdani inference system with four features, i.e., B-descriptor, D-descriptor, average intensity inside boundary, and intensity difference between inside and outside boundary in microcalcification detection system. In mass detection with Mamdani inference system, there are 3 features used, i.e., intensity of the center, average intensity and maxmin average intensity. We found that both systems yield good results, i.e. 78.07% correct classification with 20 false positives in microcalcification detection system and 98.33% correct classification with 4 false positives in mass detection system.