Related Experiment Video
Updated: Jun 16, 2025
15:48
Tracking the Mammary Architectural Features and Detecting Breast Cancer with Magnetic Resonance Diffusion Tensor Imaging
Published on: December 15, 2014
22.5K
AI-Based screening for thoracic aortic aneurysms in routine breast MRI
Dimitrios Bounias1,2, Tobit Führes3, Luise Brock3
1German Cancer Research Center (DKFZ) Heidelberg, Division of Medical Image Computing, Im Neuenheimer Feld 280, Heidelberg, Germany.
Nature Communications
|June 12, 2025
Summary
An artificial neural network (ANN) pipeline can detect thoracic aortic aneurysms on breast MRI scans, improving detection rates by 3.5-fold. This technology offers simultaneous screening for breast cancer and aortic aneurysms, potentially improving patient outcomes.
Area of Science:
- Radiology
- Artificial Intelligence
- Cardiovascular Imaging
Background:
- Thoracic aortic aneurysms (TAAs) have a worse prognosis in women.
- Magnetic resonance imaging (MRI) is increasingly used for breast cancer screening.
- Simultaneous detection of TAAs during breast MRI could improve early diagnosis.
Purpose of the Study:
- To validate a fully-automated artificial neural network (ANN) pipeline for detecting TAAs on breast MRI examinations.
- To assess the ANN's performance against ground-truth segmentations, clinical reports, and expert readers.
- To evaluate the potential of integrated screening for breast cancer and TAAs.
Main Methods:
- Retrospective validation of an ANN pipeline on 5057 breast MRI scans from public and in-house datasets.
- Benchmarking ANN performance using 3D-ground-truth segmentations, clinical reports, and a multireader panel.
- Assessing technical robustness across different MRI vendors and field strengths.
Main Results:
- The ANN pipeline demonstrated high technical robustness with dice/clDice scores of 0.88-0.91/0.97-0.99.
- ANN-assisted detection of TAAs was 3.5-fold higher compared to routine clinical readings.
- A higher odds ratio (OR=2.29) for TAAs was observed in women with a history of breast cancer.
Conclusions:
- A fully-automated ANN pipeline effectively detects thoracic aortic aneurysms on breast MRI scans.
- This technology significantly enhances TAA detection rates, offering potential for improved early diagnosis and patient outcomes.
- Integrated simultaneous assessment for breast cancer and TAAs may provide additional benefits, particularly for women with a history of breast cancer.
