Related Experiment Video
Updated: Sep 24, 2025
Human Fetal Blood Flow Quantification with Magnetic Resonance Imaging and Motion Compensation
Published on: January 7, 2021
Automatic Time-Resolved Cardiovascular Segmentation of 4D Flow MRI Using Deep Learning
Mariana Bustamante1,2, Federica Viola1, Jan Engvall1,3
1Division of Diagnostics and Specialist Medicine, Department of Health, Medicine and Caring Sciences, Linköping University, Linköping, Sweden.
Insights
This study introduces a deep learning method for segmenting cardiac structures in 4D flow MRI, significantly improving accuracy and efficiency for clinical applications.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Cardiovascular Research
Background:
- 4D flow MRI segmentation of the heart is complex due to motion and low contrast.
- Manual segmentation is time-consuming and challenging.
Purpose of the Study:
- To develop and validate a deep learning (DL) model for automated segmentation of cardiac chambers and great vessels in 4D flow MRI.
- To assess the DL model's performance against manual segmentation.
Main Methods:
- A 3D U-net based neural network was trained on 4D flow MRI data from 205 subjects.
- The model segmented four cardiac chambers, aorta, and pulmonary artery.
- Segmentation accuracy was evaluated using Dice score, Hausdorff distance, and other metrics. Volumetric parameters were compared using Bland-Altman analysis.
Main Results:
- The DL model achieved a mean Dice score of 0.908 ± 0.023.
- High accuracy was observed for Hausdorff distance (1.253 ± 0.293 mm) and average surface distance (0.466 ± 0.136 mm).
- Bland-Altman analysis demonstrated good agreement for end-diastolic and end-systolic volumes between DL and manual segmentations.
Conclusions:
- Deep learning-based segmentation of 4D flow MRI is accurate and efficient.
- This automated approach can expedite cardiac assessment and enhance the clinical utility of 4D flow MRI.
Background:
Segmenting the whole heart over the cardiac cycle in 4D flow MRI is a challenging and time-consuming process, as there is considerable motion and limited contrast between blood and tissue.
Purpose:
To develop and evaluate a deep learning-based segmentation method to automatically segment the cardiac chambers and great thoracic vessels from 4D flow MRI.
Study Type:
Retrospective.
Subjects:
A total of 205 subjects, including 40 healthy volunteers and 165 patients with a variety of cardiac disorders were included. Data were randomly divided into training (n = 144), validation (n = 20), and testing (n = 41) sets.
Field Strength/Sequence:
A 3 T/time-resolved velocity encoded 3D gradient echo sequence (4D flow MRI).
Assessment:
A 3D neural network based on the U-net architecture was trained to segment the four cardiac chambers, aorta, and pulmonary artery. The segmentations generated were compared to manually corrected atlas-based segmentations. End-diastolic (ED) and end-systolic (ES) volumes of the four cardiac chambers were calculated for both segmentations.
Statistical Tests:
Dice score, Hausdorff distance, average surface distance, sensitivity, precision, and miss rate were used to measure segmentation accuracy. Bland-Altman analysis was used to evaluate agreement between volumetric parameters.
Results:
The following evaluation metrics were computed: mean Dice score (0.908 ± 0.023) (mean ± SD), Hausdorff distance (1.253 ± 0.293 mm), average surface distance (0.466 ± 0.136 mm), sensitivity (0.907 ± 0.032), precision (0.913 ± 0.028), and miss rate (0.093 ± 0.032). Bland-Altman analyses showed good agreement between volumetric parameters for all chambers. Limits of agreement as percentage of mean chamber volume (LoA%), left ventricular: 9.3%, 13.5%, left atrial: 12.4%, 16.9%, right ventricular: 9.9%, 15.6%, and right atrial: 18.7%, 14.4%; for ED and ES, respectively.
Data Conclusion:
The addition of this technique to the 4D flow MRI assessment pipeline could expedite and improve the utility of this type of acquisition in the clinical setting.
Evidence Level:
4 TECHNICAL EFFICACY: Stage 1.
