Multi-site CNN-based segmentation for pediatric 4D flow MRI maintained the performance of single-site CNNs for geometrical similarity in the aorta (Dice score 0.916 vs. 0.915, P=0.55).
Observational (n=174)
Yes
Does multi-site CNN segmentation perform equivalently to single-site CNN segmentation for 4D flow MRI in pediatric patients with and without congenital heart disease?
Multi-site CNN-based segmentation and blood flow measurement are feasible for pediatric 4D flow MRI and maintain the performance of single-site CNNs.
Background Automated segmentation using convolutional neural networks (CNNs) have been developed using four‐dimensional (4D) flow magnetic resonance imaging (MRI). To broaden usability for congenital heart disease (CHD), training with multi‐institution data is necessary. However, the performance impact of heterogeneous multi‐site and multi‐vendor data on CNNs is unclear. Purpose To investigate multi‐site CNN segmentation of 4D flow MRI for pediatric blood flow measurement. Study Type Retrospective. Population A total of 174 subjects across two sites (female: 46%; N = 38 healthy controls, N = 136 CHD patients). Participants from site 1 (N = 100), site 2 (N = 74), and both sites (N = 174) were divided into subgroups to conduct 10‐fold cross validation (10% for testing, 90% for training). Field Strength/Sequence 3 T/1.5 T; retrospectively gated gradient recalled echo‐based 4D flow MRI. Assessment Accuracy of the 3D CNN segmentations trained on data from single site (single‐site CNNs) and data across both sites (multi‐site CNN) were evaluated by geometrical similarity (Dice score, human segmentation as ground truth) and net flow quantification at the ascending aorta (Qs), main pulmonary artery (Qp), and their balance (Qp/Qs), between human observers, single‐site and multi‐site CNNs. Statistical Tests Kruskal–Wallis test, Wilcoxon rank‐sum test, and Bland–Altman analysis. A P‐ value <0.05 was considered statistically significant. Results No difference existed between single‐site and multi‐site CNNs for geometrical similarity in the aorta by Dice score (site 1: 0.916 vs. 0.915, P = 0.55; site 2: 0.906 vs. 0.904, P = 0.69) and for the pulmonary arteries (site 1: 0.894 vs. 0.895, P = 0.64; site 2: 0.870 vs. 0.869, P = 0.96). Qs site‐1 medians were 51.0–51.3 mL/cycle ( P = 0.81) and site‐2 medians were 66.7–69.4 mL/cycle ( P = 0.84). Qp site‐1 medians were 46.8–48.0 mL/cycle ( P = 0.97) and site‐2 medians were 76.0–77.4 mL/cycle ( P = 0.98). Qp/Qs site‐1 medians were 0.87–0.88 ( P = 0.97) and site‐2 medians were 1.01–1.03 ( P = 0.43). Bland–Altman analysis for flow quantification found equivalent performance. Data Conclusion Multi‐site CNN‐based segmentation and blood flow measurement are feasible for pediatric 4D flow MRI and maintain performance of single‐site CNNs. Level of Evidence 3 Technical Efficacy Stage 2
Fujiwara et al. (Thu,) conducted a observational in Congenital heart disease (n=174). Multi-site CNN segmentation vs. Single-site CNN segmentation was evaluated on Geometrical similarity (Dice score) and net flow quantification. Multi-site CNN-based segmentation for pediatric 4D flow MRI maintained the performance of single-site CNNs for geometrical similarity in the aorta (Dice score 0.916 vs. 0.915, P=0.55).