Methods Used in Computer-Aided Diagnosis for Breast Cancer Detection Using Mammograms: A Review

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Oncology

Background:

• Breast cancer diagnosis relies heavily on mammography, creating a demand for specialized expertise.
• Computer-Aided Diagnosis (CAD) systems offer a potential solution to augment breast cancer detection and diagnosis.
• The increasing incidence of breast cancer necessitates efficient and accurate diagnostic tools.

Purpose of the Study:

• To review recent advancements in CAD systems for mammogram-based breast cancer detection.
• To provide an overview of methodologies employed in CAD system development, from preprocessing to classification.
• To assess the current performance and future potential of CAD systems in clinical practice.

Main Methods:

• Review of existing literature on CAD systems for breast cancer detection.
• Analysis of preprocessing, enhancement, and classification techniques used in CAD systems.
• Evaluation of current performance metrics and limitations of CAD systems.

Main Results:

• Current CAD systems demonstrate encouraging performance but are not yet suitable for standalone clinical use.
• Advancements in deep learning, data augmentation, and computational power are crucial for improving CAD systems.
• CAD systems currently function as a supplementary tool, providing a second opinion in clinical settings.

Conclusions:

• Further enhancement of CAD system performance is essential for their integration as primary diagnostic tools.
• Exploiting novel pattern recognition methods and increased computational power can significantly improve CAD accuracy.
• CAD systems are valuable adjuncts to radiologists, improving the efficiency and potentially the accuracy of breast cancer diagnosis.