Related Experiment Video
Updated: Jan 1, 2026
13:44
Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
Published on: August 30, 2013
43.5K
Classification of Mammogram Images Using Multiscale all Convolutional Neural Network (MA-CNN)
S Akila Agnes1, J Anitha1, S Immanuel Alex Pandian2
1Department of Computer Science and Engineering, Karunya Institute of Technology and Sciences, Coimbatore, India.
Journal of Medical Systems
|December 16, 2019
Summary
This study introduces a Multiscale All Convolutional Neural Network (MA-CNN) for early breast cancer detection. The MA-CNN model accurately classifies mammograms, improving diagnostic accuracy and aiding radiologists in identifying normal, malignant, and benign tumors.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Oncology
Background:
- Breast cancer is a leading cause of cancer death in women globally.
- Early diagnosis significantly improves survival rates and treatment efficacy.
- Digital mammography is a key tool for early breast cancer detection.
Purpose of the Study:
- To develop an effective deep learning model for automated breast cancer diagnosis.
- To assist radiologists in accurately classifying mammogram images.
- To improve the accuracy and efficiency of breast cancer detection using artificial intelligence.
Main Methods:
- Development of a Multiscale All Convolutional Neural Network (MA-CNN).
- Utilizing convolutional neural networks for feature extraction and classification.
- Training and validation on the mini-MIAS mammographic dataset.
Main Results:
- The MA-CNN model achieved high accuracy in classifying mammograms.
- The system automatically categorized images into normal, malignant, and benign classes.
- Achieved an overall sensitivity of 96% and an Area Under the Curve (AUC) of 0.99.
Conclusions:
- MA-CNN is a powerful tool for effective breast cancer diagnosis.
- The multiscale approach enhances classification accuracy without compromising speed.
- This AI-driven method aids in early and accurate detection of breast cancer.
