Breast Mass Detection and Classification Using Machine Learning Approaches on Two-Dimensional Mammogram: A Review.

Area of Science:

• Medical Imaging
• Artificial Intelligence in Healthcare
• Oncology

Background:

• Breast cancer is a significant global health concern, particularly for women, with breast masses common in women aged 20-60.
• Mammography is a key imaging tool for breast abnormality detection, but interpretation is time-consuming and requires expert radiologists.
• Accurate classification of breast masses (e.g., fibroadenoma, cysts, benign, malignant) is vital for timely and effective disease management.

Purpose of the Study:

• To review machine learning (ML) applications in digital mammogram analysis for breast mass detection and classification.
• To assess the effectiveness of various ML approaches in identifying normal, benign, and malignant breast masses.
• To highlight the advantages and limitations of current ML techniques in breast imaging.

Main Methods:

• Review of studies applying diverse machine learning algorithms to digital mammograms.
• Analysis of ML techniques focused on the identification and classification of breast masses.
• Evaluation of ML-based decision support systems for radiologists.

Main Results:

• Machine learning and AI show promise in automating the identification and classification of breast masses from mammograms.
• Various ML approaches have been investigated for their ability to distinguish between normal, benign, and malignant findings.
• The reviewed studies indicate potential for improved accuracy and efficiency in breast disorder diagnosis.

Conclusions:

• ML and AI offer powerful tools to enhance breast cancer screening and diagnosis through mammography analysis.
• Further research into ML techniques can optimize decision support systems, aiding radiologists and improving patient outcomes.
• Addressing the limitations of current ML methods is crucial for advancing breast health and reducing mortality rates.