Breast cancer detection employing stacked ensemble model with convolutional features

Area of Science:

• Medical Imaging and Diagnostics
• Machine Learning in Healthcare
• Oncology

Background:

• Breast cancer remains a leading cause of mortality in women, particularly in developing nations.
• Early and accurate diagnosis is crucial for effective treatment and improved patient survival rates.
• Current automated diagnostic methods, while promising, often lack the desired accuracy.

Purpose of the Study:

• To develop a highly accurate ensemble model for automated breast cancer detection.
• To enhance diagnostic accuracy by integrating optimized feature extraction with machine learning classifiers.
• To improve upon existing state-of-the-art models for breast cancer diagnosis.

Main Methods:

• An ensemble model combining Random Forest and Support Vector Classifier was developed.
• Automatic feature extraction was performed using an optimized Convolutional Neural Network (CNN).
• The model's performance was evaluated using the Wisconsin dataset, comparing original and CNN-based features.

Main Results:

• CNN-based features significantly outperformed original features in breast cancer detection.
• The proposed ensemble model achieved an exceptional accuracy of 99.99%.
• Comparative analysis demonstrated the superior performance of the proposed model over existing methods.

Conclusions:

• The proposed ensemble model offers a highly accurate solution for automated breast cancer detection.
• Optimized feature extraction via CNN is key to achieving superior diagnostic performance.
• This approach holds significant potential for improving early detection and patient outcomes in breast cancer care.