AN IMPROVED GA-MILSVM CLASSIFICATION APPROACH FOR DIAGNOSIS OF BREAST LESIONS FROM STAIN IMAGES

Cancer cells spread to more distant parts of the body through the lymphatic system or bloodstream. Not all tumors are cancerous. Benign tumors do not grow uncontrollably, do not invade neighboring tissues, and do not spread throughout the body. Intraductal Carcinoma is a noninvasive condition in which abnormal cells are found in the lining of a breast duct. The abnormal cells have not spread outside the duct to other tissues in the breast. In some cases, Intraductal Carcinoma may become invasive cancer and spread to other tissues, although it is not known at this time how to predict which lesions will become invasive. Intraductal cancer is the most common type of breast cancer in women. Memory Intraductal includes 3-types of cancer: Usual Ductal Hyperplasia (UDH), Atypical Ductal Hyperplasia (ADH), and Ductal Carcinoma in Situ (DCIS). So the system of detecting the breast microscopic tissue of UDH, ADH, DCIS is proposed. The current standard of care is to perform percutaneous needle biopsies for diagnosis of palpable and image-detected breast abnormalities. UDH is considered benign and patients diagnosed UDH undergo routine follow-up, whereas ADH and DCIS are considered actionable and patients diagnosed with these two subtypes get additional surgical procedures. The system classify the tissue based on the quantitative feature derived from the images. The statistical features are obtained. The approach makes use of preprocessing, Cell region segmentation, Individual cell segmentation, Feature extraction technique and MILSVM classifier for the detection of cancer and optimized using Genetic Algorithm An overall accuracy of 87.9% precision is obtained using GA and 4.5% recall are achieved on the entire test data. The test accuracy of 82.6% precision and 3.5% recall are obtained using MILSVM. When compared with MILSVM, GA has a great potential in improving diagnostic accuracy and reproducibility.