Automated intracranial vessel segmentation of 4D flow MRI data in patients with atherosclerotic stenosis using a
Patrick Winter1,2, Haben Berhane2, Jackson E Moore2
1Department of Medical Physics, Faculty of Mathematics and Natural Sciences, University of Greifswald, Greifswald, Germany.
Insights
This study introduces a deep learning method for automated segmentation of intracranial vessels in 4D flow MRI scans, improving accuracy and speed for diagnosing intracranial atherosclerotic disease (ICAD). The automated approach matches human observer performance, enhancing reproducibility in clinical practice.
Area of Science:
- Medical Imaging
- Artificial Intelligence in Medicine
- Neuroscience
Background:
- Intracranial 4D flow MRI is crucial for quantifying hemodynamics in patients with intracranial atherosclerotic disease (ICAD).
- Manual vessel segmentation is time-consuming and prone to user variability, hindering reproducible and robust quantitative assessments, especially in stenosed vessels.
Purpose of the Study:
- To develop an accurate, fully automated deep learning-based segmentation method for stenosed intracranial vessels using 4D flow MRI data.
- To improve the reproducibility, robustness, and efficiency of hemodynamic analysis in ICAD patients.
Main Methods:
- A 3D U-Net deep learning model was trained using 154 dual-VENC 4D flow MRI scans (68 ICAD patients, 86 controls) with manual segmentations as ground truth.
- Automated segmentations were evaluated against manual segmentations from two independent observers using Dice scores, Hausdorff distance, and average symmetrical surface distance.
- Flow parameters and cross-sectional areas were compared, and stenosis assessments were validated against black blood vessel wall imaging (VWI).
Main Results:
- The automated segmentation achieved performance comparable to independent observers, with Dice scores of 0.85±0.03 (CoW) and 0.86±0.06 (sinuses) in controls, and 0.85±0.04 (CoW) and 0.82±0.07 (sinuses) in patients.
- Automated segmentation time was significantly reduced to 2.2±1.0 seconds per scan, compared to manual methods.
- Cross-sectional lumen areas in stenosed vessels showed very good agreement (ICC: 0.93) with VWI, despite a consistent overestimation bias.
Conclusions:
- Deep learning enables accurate and fully automated segmentation of stenosed intracranial vessels from 4D flow MRI data.
- The developed method enhances reproducibility and significantly accelerates data analysis for ICAD patients.
- Future work will focus on expanding the dataset to improve performance and generalization across various intracranial vascular pathologies.
Introduction:
Intracranial 4D flow MRI enables quantitative assessment of hemodynamics in patients with intracranial atherosclerotic disease (ICAD). However, quantitative assessments are still challenging due to the time-consuming vessel segmentation, especially in the presence of stenoses, which can often result in user variability. To improve the reproducibility and robustness as well as to accelerate data analysis, we developed an accurate, fully automated segmentation for stenosed intracranial vessels using deep learning.
Methods:
154 dual-VENC 4D flow MRI scans (68 ICAD patients with stenosis, 86 healthy controls) were retrospectively selected. Manual segmentations were used as ground truth for training. For automated segmentation, deep learning was performed using a 3D U-Net. 20 randomly selected cases (10 controls, 10 patients) were separated and solely used for testing. Cross-sectional areas and flow parameters were determined in the Circle of Willis (CoW) and the sinuses. Furthermore, the flow conservation error was calculated. For statistical comparisons, Dice scores (DS), Hausdorff distance (HD), average symmetrical surface distance (ASSD), Bland-Altman analyses, and interclass correlations were computed using the manual segmentations from two independent observers as reference. Finally, three stenosis cases were analyzed in more detail by comparing the 4D flow-based segmentations with segmentations from black blood vessel wall imaging (VWI).
Results:
Training of the network took approximately 10 h and the average automated segmentation time was 2.2 ± 1.0 s. No significant differences in segmentation performance relative to two independent observers were observed. For the controls, mean DS was 0.85 ± 0.03 for the CoW and 0.86 ± 0.06 for the sinuses. Mean HD was 7.2 ± 1.5 mm (CoW) and 6.6 ± 3.7 mm (sinuses). Mean ASSD was 0.15 ± 0.04 mm (CoW) and 0.22 ± 0.17 mm (sinuses). For the patients, the mean DS was 0.85 ± 0.04 (CoW) and 0.82 ± 0.07 (sinuses), the HD was 8.4 ± 3.1 mm (CoW) and 5.7 ± 1.9 mm (sinuses) and the mean ASSD was 0.22 ± 0.10 mm (CoW) and 0.22 ± 0.11 mm (sinuses). Small bias and limits of agreement were observed in both cohorts for the flow parameters. The assessment of the cross-sectional lumen areas in stenosed vessels revealed very good agreement (ICC: 0.93) with the VWI segmentation but a consistent overestimation (bias ± LOA: 28.1 ± 13.9%).
Discussion:
Deep learning was successfully applied for fully automated segmentation of stenosed intracranial vasculatures using 4D flow MRI data. The statistical analysis of segmentation and flow metrics demonstrated very good agreement between the CNN and manual segmentation and good performance in stenosed vessels. To further improve the performance and generalization, more ICAD segmentations as well as other intracranial vascular pathologies will be considered in the future.
