A Hybrid Method for Breast Cancer Classification Utilizing Feature Fusion

Abstract

Breast cancer is a prevalent form of malignancy, particularly among women. Estimates indicate that one in nine women receives a diagnosis of breast cancer. The insufficiency of adequate facilities is causing delays in breast cancer diagnosis, hence elevating the prospective mortality rate. A variety of statistical techniques and machine learning algorithms are frequently utilized in research to enhance the accuracy of breast cancer detection. Machine learning (ML) has yielded significant outcomes for physicians, and the healthcare sector is employing ML-based models for the detection of breast cancer in women. This facilitates the analysis of healthcare data and employs conventional computer-aided detection (CAD) to evaluate breast cancer. Machine learning has been integrated into clinical practice, enabling physicians to assess the ML model for early breast cancer detection. This research utilizes various machine learning methods to categorize cancer as malignant or benign. MLP signifies a more efficient and accurate methodology for breast cancer categorization. The Matthews correlation coefficient for the MLP is 0.89%, whereas the accuracy score for the random forest is 0.94%.