Breast Cancer Detection by Feature Extraction and Classification Using Deep Learning Architectures

Abstract

In the world, breast cancer is regarded as one of the leading causes of death for females between ages of 20 and 59. Machine learning is the method that is utilised in research the most frequently. There have been a lot of earlier machine learning-based studies.This research propose novel technique in breast cancer detection based on feature extraction and classification by deep learning techniques. here the input data is taken as breast cancer dataset and processed for noise removal and smoothening. In order to improve the accuracy of categorising microcalcifications as benign, malignant, or normal, textural features from the processed mammography picture have been retrieved using kernel independent component analysis.Utilizing optimization techniques, the tumour portion in the breast region is excised. Here, U-net convolutional learning (U-NetCL), which eliminates human labour, is suggested for diagnosing breast cancer. The U-NetCL framework is designed for effectively extracting features.This specifically created method recognises cancerous areas in mammography (MG) pictures and quickly categorises those areas as normal or abnormal.