Enhancing Automated Breast Cancer Detection: A CNN-Driven Method for Multi-Modal Imaging Techniques.

Area of Science:

• Medical Imaging Analysis
• Artificial Intelligence in Oncology
• Computational Pathology

Background:

• Breast cancer remains a leading cause of mortality in women globally.
• Accurate and efficient diagnostic methods are crucial for effective treatment.
• Current diagnostic approaches necessitate improvement in speed and accuracy.

Purpose of the Study:

• To develop an automated diagnostic framework for breast cancer detection.
• To utilize convolutional neural networks (CNNs) for analyzing diverse imaging modalities.
• To enhance the accuracy and efficiency of early breast cancer diagnosis.

Main Methods:

• Training and evaluating a CNN model on benchmark datasets: mammography (DDSM, MIAS, INbreast), ultrasound, MRI, and histopathology (BreaKHis).
• Implementing standardized preprocessing techniques across all imaging data.
• Comparing the CNN model's performance against state-of-the-art methods.

Main Results:

• Achieved high classification accuracy: 99.2% (DDSM), 98.97% (MIAS), 99.43% (INbreast), 98.00% (Ultrasound), 98.43% (MRI), and 86.42% (BreaKHis).
• Demonstrated superior performance compared to many existing diagnostic techniques.
• Confirmed the robustness and reliability of the developed CNN framework.

Conclusions:

• Introduced a reliable and scalable automated diagnostic system for breast cancer.
• The framework supports radiologists in achieving earlier and more accurate detection.
• High accuracy and adaptability across imaging modalities suggest clinical integration potential.