Applying Deep Learning for Breast Cancer Detection in Radiology

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Oncology

Background:

• Breast cancer is a leading cause of death among women globally.
• Early detection significantly improves patient outcomes and survival rates.
• Deep learning (DL) shows promise in advancing medical imaging analysis.

Purpose of the Study:

• To review the application of deep learning methods in breast cancer screening workflows.
• To summarize DL techniques, data considerations, and various imaging modalities used.
• To explore the integration of artificial intelligence (AI) in diagnostic breast imaging.

Main Methods:

• Literature review of deep learning applications in breast cancer screening.
• Summarization of different breast imaging techniques: mammography, thermography, ultrasound, and MRI.
• Analysis of data availability and DL model performance in diagnostic tasks.

Main Results:

• Deep learning models demonstrate potential in improving the accuracy and efficiency of breast cancer detection across various imaging modalities.
• The review identifies key DL architectures and their suitability for specific screening tasks.
• Data availability and quality are critical factors influencing DL model generalizability.

Conclusions:

• AI, particularly deep learning, offers significant opportunities to enhance breast cancer screening protocols.
• Challenges remain in clinical implementation, including data privacy, regulatory approval, and model interpretability.
• Further research is needed to optimize DL algorithms and facilitate seamless integration into routine clinical practice.