Related Experiment Video
Updated: Jun 24, 2025
Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
Published on: August 30, 2013
Recurrent attention U-Net for segmentation and quantification of breast arterial calcifications on synthesized 2D
Manar AlJabri1,2, Manal Alghamdi1, Fernando Collado-Mesa3
1Department of Computer Science and Artificial Intelligence, Umm Al-Qura University, Makkah, Makkah, Saudi Arabia.
Insights
This study introduces a deep learning model to detect breast arterial calcifications (BAC) on mammograms. The AI tool shows high accuracy, aiding radiologists in identifying these calcifications and their potential cardiovascular links.
Area of Science:
- Radiology
- Artificial Intelligence
- Cardiovascular Health
Background:
- Breast arterial calcifications (BAC) are common mammogram findings, generally considered benign.
- Emerging evidence links BAC to cardiovascular disease, a leading cause of mortality in women.
Purpose of the Study:
- To develop and evaluate a deep learning model for detecting and quantifying BAC in synthesized 2D mammograms.
- To assist radiologists in identifying BAC and potentially assessing cardiovascular risk.
Main Methods:
- A recurrent attention U-Net model was designed, incorporating recurrent mechanisms and attention modules.
- The model utilizes skip connections, similar to U-Net architectures, for enhanced feature processing.
- Evaluation was performed on a dataset of 2,000 synthesized 2D mammogram images.
Main Results:
- The deep learning model achieved an overall accuracy of 99.8861%.
- Sensitivity, F-1 score, and Jaccard coefficient were reported as 69.6107%, 66.5758%, and 59.5498%, respectively.
- The model demonstrated promising performance compared to existing related models.
Conclusions:
- The developed deep learning method shows potential for accurate BAC detection and quantification in mammography.
- This AI tool could support radiologists in identifying BAC, contributing to cardiovascular risk assessment.
- Further validation is warranted to integrate this technology into clinical practice.
Abstract:
Breast arterial calcifications (BAC) are a type of calcification commonly observed on mammograms and are generally considered benign and not associated with breast cancer. However, there is accumulating observational evidence of an association between BAC and cardiovascular disease, the leading cause of death in women. We present a deep learning method that could assist radiologists in detecting and quantifying BAC in synthesized 2D mammograms. We present a recurrent attention U-Net model consisting of encoder and decoder modules that include multiple blocks that each use a recurrent mechanism, a recurrent mechanism, and an attention module between them. The model also includes a skip connection between the encoder and the decoder, similar to a U-shaped network. The attention module was used to enhance the capture of long-range dependencies and enable the network to effectively classify BAC from the background, whereas the recurrent blocks ensured better feature representation. The model was evaluated using a dataset containing 2,000 synthesized 2D mammogram images. We obtained 99.8861% overall accuracy, 69.6107% sensitivity, 66.5758% F-1 score, and 59.5498% Jaccard coefficient, respectively. The presented model achieved promising performance compared with related models.
