What does this research mean for the field?

A deep learning convolutional neural network can perform fully automated, fast, and highly accurate 3D aortic segmentation from 4D-flow MRI, achieving reproducibility comparable to manual analysis while drastically reducing processing time. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

March 13, 2020Open Access

Fully automated 3D aortic segmentation of 4D flow MRI for hemodynamic analysis using deep learning

Structured PICO

Does a deep learning convolutional neural network provide accurate and fast 3D aortic segmentation from 4D-flow MRI compared to manual segmentation in subjects with and without aortic pathology?

Population

1,018 subjects with aortic 4D-flow MRI (528 with bicuspid aortic valve, 376 with tricuspid aortic valve and aortic dilation, 114 healthy controls)

Intervention

Convolutional neural network (deep learning) for automated 3D aortic segmentation from 4D-flow MRI data

Comparator

Manual segmentations (ground truth)

Outcome

Segmentation performance assessed by Dice scores, Hausdorff distance, and average symmetrical surface distance; agreement in aortic flow, peak velocity, and lumen dimensionssurrogate

Deep learning enables fast, automated, and highly accurate 3D aortic segmentation from 4D-flow MRI, significantly reducing analysis time compared to manual methods.

Abstract

PURPOSE: To generate fully automated and fast 4D-flow MRI-based 3D segmentations of the aorta using deep learning for reproducible quantification of aortic flow, peak velocity, and dimensions. METHODS: A total of 1018 subjects with aortic 4D-flow MRI (528 with bicuspid aortic valve, 376 with tricuspid aortic valve and aortic dilation, 114 healthy controls) comprised the data set. A convolutional neural network was trained to generate 3D aortic segmentations from 4D-flow data. Manual segmentations served as the ground truth (N = 499 training, N = 101 validation, N = 418 testing). Dice scores, Hausdorff distance, and average symmetrical surface distance were calculated to assess performance. Aortic flow, peak velocity, and lumen dimensions were quantified at the ascending, arch, and descending aorta and compared using Bland-Altman analysis. Interobserver variability of manual analysis was assessed on a subset of 40. RESULTS: Convolutional neural network segmentation required 0.438 ± 0.355 seconds versus 630 ± 254 seconds for manual analysis and demonstrated excellent performance with a median Dice score of 0.951 (0.930-0.966), Hausdorff distance of 2.80 (2.13-4.35), and average symmetrical surface distance of 0.176 (0.119-0.290). Excellent agreement was found for flow, peak velocity, and dimensions with low bias and limits of agreement less than 10% difference versus manual analysis. For aortic volume, limits of agreement were moderate within 16.3%. Interobserver variability (median Dice score: 0.950; Hausdorff distance: 2.45; and average symmetrical surface distance: 0.145) and convolutional neural network-based analysis (median Dice score: 0.953-0.959; Hausdorff distance: 2.24-2.91; and average symmetrical surface distance: 0.145-1.98 to observers) demonstrated similar reproducibility. CONCLUSIONS: Deep learning enabled fast and automated 3D aortic segmentation from 4D-flow MRI, demonstrating its potential for efficient clinical workflows. Future studies should investigate its utility for other vasculature and multivendor applications.

Bookmark

View Full Paper