基于全自动化多站点、多厂商、多标签密集 U-Net 的儿科 4DFlowMRI 主动脉和肺动脉分割。
Segmentation of the Aorta and Pulmonary Arteries Based on 4D Flow MRI in the Pediatric Setting Using Fully Automated Multi-Site, Multi-Vendor, and Multi-Label Dense U-Net.
机构信息
Department of Radiology, Section of Pediatric Radiology, Children's Hospital Colorado, University of Colorado Anschutz Medical Campus, Aurora, Colorado, USA.
Department of Biomedical Engineering, Northwestern University, Chicago, Illinois, USA.
出版信息
J Magn Reson Imaging. 2022 Jun;55(6):1666-1680. doi: 10.1002/jmri.27995. Epub 2021 Nov 18.
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.
背景
基于卷积神经网络（CNN）的自动分割技术已应用于四维（4D）血流磁共振成像（MRI）。为了拓宽在先天性心脏病（CHD）中的应用范围，需要使用多机构数据进行训练。然而，不同机构和不同供应商的数据对 CNN 的性能影响尚不清楚。
目的
研究用于儿科血流测量的 4D 流 MRI 的多中心 CNN 分割。
研究类型
回顾性。
人群
共有两个中心的 174 名受试者（女性：46%；N=38 名健康对照，N=136 名 CHD 患者）。来自中心 1（N=100）、中心 2（N=74）和两个中心（N=174）的参与者被分为亚组进行 10 折交叉验证（10%用于测试，90%用于训练）。
磁场强度/序列：3T/1.5T；回顾性门控梯度回波 4D 流 MRI。
评估
评估在单中心（单中心 CNN）和多中心数据（多中心 CNN）上训练的 3D CNN 分割的准确性，使用几何相似性（Dice 评分，以人体分割作为金标准）和升主动脉（Qs）、主肺动脉（Qp）及其平衡（Qp/Qs）的净流量定量，在人体观察者、单中心和多中心 CNN 之间进行评估。
统计学检验
Kruskal-Wallis 检验、Wilcoxon 秩和检验和 Bland-Altman 分析。P 值<0.05 被认为具有统计学意义。
结果
在主动脉的 Dice 评分的几何相似性方面，单中心和多中心 CNN 之间没有差异（中心 1：0.916 与 0.915，P=0.55；中心 2：0.906 与 0.904，P=0.69），对于肺动脉也是如此（中心 1：0.894 与 0.895，P=0.64；中心 2：0.870 与 0.869，P=0.96）。中心 1 的 Qs 中位数为 51.0-51.3mL/循环（P=0.81），中心 2 的中位数为 66.7-69.4mL/循环（P=0.84）。中心 1 的 Qp 中位数为 46.8-48.0mL/循环（P=0.97），中心 2 的中位数为 76.0-77.4mL/循环（P=0.98）。中心 1 的 Qp/Qs 中位数为 0.87-0.88（P=0.97），中心 2 的中位数为 1.01-1.03（P=0.43）。流量定量的 Bland-Altman 分析发现性能相当。
数据结论
基于多中心 CNN 的分割和血流测量对于儿科 4D 流 MRI 是可行的，并且保持了单中心 CNN 的性能。
证据水平
3 级技术功效：阶段 2。
Zotero 插件