DIAGNOSIS OF MICRO CALCIFICATIONS IN DIGITAL MAMMOGRAM USING DEEP PROGNOSTIC LEARNING CLASSIFIER
Journal of Critical Reviews · 2020
Breast cancer is the second most common cause of cancer death in women worldwide. The primary prevention is complicated in the early stages of the disease. However, some of the common features of the disease, such as the appearance of mammograms and micro calcifications, are early signs of breast cancer. The Image processing based the diagnosis and detection of Micro calcification in the mammogram images. Since the mammogram images are difficult to identify the micro calcifications, because the scanned images had lot of noises, so the separation of the Micro calcification and the noise is the challenging process in the image processing. In this proposed work a multilevel wavelet decomposition filtering of the work proposed here takes the detection of images at an early stage of the cancer process and is used for noise suppression and enhancement in digital mammography images. During the second stage, the breast area is separated from the pre-processed image. Suboptimal clustering algorithm is used to identify and divide it into negative and positive suspicious areas. This process eliminates unnecessary areas of interest. The Positive suspicious mass containing Micro calcification pixels are taken to the training optimal clustering optimal clustering segmentation network. After distinguishing of micro calcifications in the breast regions the feature like entropy, mean, standard deviation, variance and skewness and the gained values are utilized for classification. The proposed Deep Prognostic Learning Classifier (DPLC) was used to classify the suspicious mass in the mammogram using the above feature. Also the precision, recall and the F-measure values are utilized for analysing the performance of the proposed DPLC method. Finally, the False Positive (FP) and True Positive (TP) rates for the Micro calcification were computed to compare the performance of the trained networks.