Bio-Imaging-Based Machine Learning Algorithm for Breast Cancer Detection

Area of Science:

• Medical Imaging
• Machine Learning
• Oncology

Background:

• Breast cancer is a leading cause of mortality in women globally.
• Early and accurate diagnosis is crucial for improving survival rates.
• Computer-aided detection (CAD) systems assist radiologists in distinguishing between normal and abnormal breast tissue.

Purpose of the Study:

• To evaluate machine learning techniques for breast cancer recurrence rate estimation.
• To develop an accurate computer-aided detection system for breast cancer classification.
• To minimize false positive and false negative rates in breast cancer diagnosis.

Main Methods:

• Overview of imaging modalities: ultrasound, histography, and mammography.
• Application of machine learning algorithms: Support Vector Machine (SVM), Logistic Regression (LR), and K-Nearest Neighbor (KNN).
• Data preprocessing, including noise reduction and transformation, with a 60/40 train-test split.

Main Results:

• Achieved a highest accuracy of 97.7% in breast cancer classification.
• Reported a low false positive rate (FPR) of 0.01 and false negative rate (FNR) of 0.03.
• Obtained an Area Under the ROC Curve (AUC) score of 0.99, indicating high model performance.

Conclusions:

• The proposed machine learning model demonstrates high accuracy and effectiveness in classifying breast tumors.
• The system successfully overcomes limitations of previous research in breast cancer detection.
• Future research directions include classification and segmentation challenges in breast cancer detection.