Application of Artificial Intelligence using Mammograms to Identify Breast Cancer
Sudeshna Rath, Kalyanbrata Giri, Satya Ranjan Dash, Paola Barra, Azian Azamimi Abdullah · 2024
An affordable and precise way to identify breast cancer is through Mammography. It has significantly decreased mortality by detecting malignancies early and is essential for early detection. It has also increased survival rates. However, disparities in mammography access are still mostly driven by socioeconomic factors. Ongoing research aims to address these problems, improve screening methods, and raise the precision and usability of mammography. Additionally, radiologists are being assisted by computer-aided detection (CAD) and artificial intelligence (AI), which are being used to improve diagnostic accuracy. The goal of this research is to identify the best AI algorithms for breast cancer detection. We employ a range of preprocessing methods in this study, first resizing the images to see which method cleans up our data the best. After segmenting our images to locate the region of interest, we extract features using a local binary pattern. Numerous classification models have been employed, such as SVM, KNN, Random Forest, and Decision Tree. Moreover, CNN and RCN have been used for the same objective. Furthermore, it was shown that, generally speaking, neural networks outperformed classification.