%0 Journal Article %T Automated Clustering of Cancer Cells Using Fuzzy C Means with Repulsions in Ultrasound Images %A N. Alamelumangai %A J. Devishree %J Journal of Artificial Intelligence %D 2012 %I Asian Network for Scientific Information %X In the report provided by World Health Organization (WHO), breast cancer is one of the highly deadliest cancers occurred in middle-aged women. Accurate diagnosis and prediction are essential to decrease the high death rate. In the past few years, breast ultrasound images have turn out to be an optional for mammography to help distinguish benign from malignant lesions. Its advantages safety and cost-effectiveness as discussed by various authors have turned ultrasound method into an increasingly significant function in the estimate of breast lesions. Like ultrasound exams in general, breast ultrasound exams are comparatively low-priced and do not utilize X-rays or other kinds of probably dangerous radiation. As a result, breast inspection with the help of ultrasound method has turn out to be a main option to mammography. This study provides a new technique in the field of computer science for diagnosis of breast tumors on ultrasound images. Some preprocessing steps are performed in the obtained ultrasound image in to enhance it for better diagnosis. Finally, the cancerous cells are distinguished by clustering them. This study uses Modified Fuzzy Possibilistic C-Means technique with Repulsion factor for the purpose of clustering the cancerous cells. The proposed computer-aided diagnosis systems are evaluated using the real time breast ultrasound images. The accuracy of predicting the cancer regions in the breast is higher when compared to the conventional technique. Also, the standard deviation resulted for the proposed approach is lesser than conventional technique. The experimental results show that the proposed technique results in better detection of cancer regions as it produces higher accuracy of detecting the cancer regions. %K repulsion %K fuzzy possibilistic c-means %K Ultrasound image %K clustering %U http://docsdrive.com/pdfs/ansinet/jai/2012/14-25.pdf