Breast Cancer Detection and Classification using Traditional Computer Vision Techniques: A Comprehensive Review

Area of Science:

• Medical Imaging
• Artificial Intelligence in Medicine
• Oncology

Background:

• Breast cancer remains a leading cause of mortality among women globally.
• Timely diagnosis in the initial stages significantly improves patient survival rates.
• Existing diagnostic methods require enhancement for accurate and early detection.

Purpose of the Study:

• To present image processing, machine learning, and deep learning techniques for breast cancer diagnosis.
• To analyze the performance of various Computer-Aided Diagnosis (CAD) system phases.
• To highlight the importance of advanced methods for early and accurate breast cancer detection.

Main Methods:

• Review and analysis of image processing techniques for breast cancer detection.
• Exploration of machine learning and deep learning algorithms in CAD systems.
• Detailed examination of CAD system phases: preprocessing, segmentation, feature extraction, and classification.

Main Results:

• Various techniques show promise in detecting breast masses, microcalcifications, and malignant cells.
• Performance measurements of different methods are reported, indicating areas for improvement.
• Segmentation and classification remain challenging phases for accurate diagnosis.

Conclusions:

• Improving CAD system techniques is essential for enhancing breast cancer survival rates.
• Advanced tools and methodologies are critically needed for accurate breast cancer diagnosis and classification.
• Early detection through improved diagnostic systems offers the best chance for patient survival.