BREAST CANCER PROGNOSIS PREDICTION USING NOVEL SHRUNK KERNEL KNN METHOD WITH MLO AND CC FEATURES
Dr. V. Sridevi · 2023
The most continual malignancy found in women is breast cancer. The good news is that if detected early, it is one of the most treatable forms of cancer. High-dimensional data results in large number of computation redundancy but also advances diagnostic techniques. As a result significant information must be extracted and the feature dimension must be reduced for good prediction and a precise treatment decision. However, past studies for diagnosing breast cancer have relied mostly on labelled data that is difficult to get. To solve this problem, two different sorts of perspectives, such as CC and MLO are employed to improve diagnostic effectiveness. This chapter comprises segmentation, feature extraction and classification of images. The two perspectives from a mammography picture are segmented using the adaptive K-means clustering approach. The Gabor filter is used in conjunction with the traditional k-means clustering method during the feature extraction stage to extract the features of the CC and MLO perspectives. The mammography image is finally classified into benign and malignant using a unique Shrunk Kernel K-Nearest Neighbor (SKKNN) classifier.