Related Experiment Video
Updated: Jul 12, 2025
07:15
Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
Published on: August 16, 2020
6.8K
Recent advancements in machine learning and deep learning-based breast cancer detection using mammograms.
Adyasha Sahu1, Pradeep Kumar Das2, Sukadev Meher1
1Department of Electronics and Communication Engineering, National Institute of Technology, Rourkela, Odisha, 769008, India.
Summary
This study reviews artificial intelligence (AI) methods for breast cancer detection using mammograms. Deep learning models, particularly transfer learning, show superior performance for accurate diagnosis and treatment planning.
Area of Science:
- Radiology
- Medical Imaging
- Artificial Intelligence
Background:
- Mammography is a crucial screening tool for breast cancer, aiding in early diagnosis and treatment.
- Accurate breast cancer detection from mammograms is vital for patient survival and effective treatment planning.
- Existing surveys on mammogram-based breast cancer detection methods are limited.
Purpose of the Study:
- To provide a comprehensive overview of recent advancements in machine learning (ML) and deep learning (DL) for mammogram-based breast cancer detection.
- To categorize existing techniques for breast cancer detection using mammograms.
- To highlight challenges and future research directions in AI-driven mammography.
Main Methods:
- A structured framework was developed to categorize various mammogram-based breast cancer detection techniques.
- Publicly available mammogram databases and performance metrics were identified and discussed.
- Recent literature on ML and DL applications in mammography was systematically reviewed.
Main Results:
- Most studies classify tumors into two categories (normal-abnormal or malignant-benign) rather than three.
- Deep learning (DL) models extract more significant features compared to traditional hand-crafted features.
- Transfer learning techniques demonstrate superior performance, especially with limited datasets, outperforming classical DL approaches.
Conclusions:
- Artificial intelligence (AI), particularly ML and DL, offers significant advancements in breast cancer detection from mammograms.
- Transfer learning is a promising approach for improving diagnostic accuracy, especially in data-scarce scenarios.
- Identifying current challenges and future research avenues will guide further progress in AI-assisted mammography.
