Divulging Patterns: An Analytical Review for Machine Learning Methodologies for Breast Cancer Detection.

Area of Science:

• Oncology
• Computer Science
• Medical Imaging

Background:

• Breast cancer is a significant global health concern, with early detection being crucial for effective management.
• Accurate classification of breast tumors (benign vs. malignant) is vital for timely diagnosis and treatment planning.
• Machine learning (ML) shows promise in enhancing breast cancer diagnostic accuracy.

Purpose of the Study:

• To survey emerging machine learning (ML) approaches for breast cancer diagnosis.
• To analyze classification techniques based on image segmentation and feature selection.
• To identify research gaps and propose solutions for improved clinical adoption of ML tools.

Main Methods:

• Evaluation of various ML models, including Support Vector Machines (SVM), Convolutional Neural Networks (CNNs), and ensemble methods.
• Analysis of diverse datasets such as Wisconsin Diagnostic Breast Cancer (WDBC), Wisconsin Breast Cancer Original (WBCD), Wisconsin Prognostic Breast Cancer (WPBC), and BreakHis.
• Investigation of classification performance influenced by segmentation and feature selection strategies.

Main Results:

• Demonstration of the impact of different datasets and ML models on diagnostic tool performance and accuracy.
• Identification of key research gaps including dataset bias, limited generalizability, and interpretation challenges.
• Highlighting the potential of hybrid methodologies, cross-dataset validation, and explainable AI.

Conclusions:

• ML offers powerful tools for breast cancer diagnosis, but challenges remain.
• Addressing dataset bias, generalizability, and interpretability is essential for clinical integration.
• Hybrid approaches, robust validation, and explainable AI are key to advancing ML in breast cancer detection.