Efficient breast cancer classification using LS-SVM and dimensionality reduction

Breast cancer is the main cause of cancer-related deaths among women worldwide. Several diagnostic methods, including mammography, ultrasound, and biopsy, are used to discover breast tumors, since early identification and diagnosis are crucial for improved treatment results. This work aims to identify and classify breast cancer using logistic regression with kernel-based principal component analysis (KPCA) dimensionality reduction and least square support vector machine (LS-SVM) classification (LR-KPCA-LS-SVM). Earlier categorization schemes for breast cancer were plagued by inaccuracies in diagnosis and inefficiency. The suggested LR-KPCA-LS-SVM solves these problems by lowering complexity, concentrating on key characteristics, and integrating different approaches and algorithms to improve accuracy. The LR-KPCA-LS-SVM has accuracy rates of 95% and 96%, SVM has accuracy rates of 85% and 83%, KNN has accuracy rates of 89% and 88%, and CNN has accuracy rates of 80% and 88% when applied over WBCD and WDBC datasets.