Mammography Image Classification and Detection by Bi-LSTM with Residual Network using XG-Boost Approach

Abstract

CNNs play an essential role in the process of deep learning approaches because of their ability to categorize and classify images. In the same case, CNN is also successful in identifying mammogram-based breast cancer photos. It is helpful in the process of automatically classifying and recognizing pictures. Despite this, there are a lot of challenges that need to be overcome in order to reduce the amount of noise in mammography images. Some of these challenges include patch categorization and detection, as well as the extraction of useful characteristics for certain tumors. It is essential to recognize the significance of feature engineering in this sector of the economy. Because LSTM has a very little amount of memory, the goal of this study is to find ways to extract valuable features by making use of Attention layers and to reduce error by using the BI-LSTM approach.