基于深度学习的钆塞酸二钠增强肝脏MRI上运动相关伪影的自动检测与分级

Deep Learning-Based Automatic Detection and Grading of Motion-Related Artifacts on Gadoxetic Acid-Enhanced Liver MRI.

作者信息

Park Taeyong, Kim Dong Wook, Choi Sang Hyun, Khang Seungwoo, Huh Jimi, Hong Seung Baek, Lee Tae Young, Ko Yousun, Kim Kyung Won, Lee Seung Soo

机构信息

From the Department of Radiology and Research Institute of Radiology, Asan Medical Center, University of Ulsan College of Medicine.

School of Computer Science and Engineering, Soongsil University, Seoul.

出版信息

Invest Radiol. 2023 Feb 1;58(2):166-172. doi: 10.1097/RLI.0000000000000914. Epub 2022 Aug 31.

DOI:10.1097/RLI.0000000000000914

PMID:36070544

Abstract

OBJECTIVES

The aim of this study was to develop and validate a deep learning-based algorithm (DLA) for automatic detection and grading of motion-related artifacts on arterial phase liver magnetic resonance imaging (MRI).

MATERIALS AND METHODS

Multistep DLA for detection and grading of motion-related artifacts, based on the modified ResNet-101 and U-net, were trained using 336 arterial phase images of gadoxetic acid-enhanced liver MRI examinations obtained in 2017 (training dataset; mean age, 68.6 years [range, 18-95]; 254 men). Motion-related artifacts were evaluated in 4 different MRI slices using a 3-tier grading system. In the validation dataset, 313 images from the same institution obtained in 2018 (internal validation dataset; mean age, 67.2 years [range, 21-87]; 228 men) and 329 from 3 different institutions (external validation dataset; mean age, 64.0 years [range, 23-90]; 214 men) were included, and the per-slice and per-examination performances for the detection of motion-related artifacts were evaluated.

RESULTS

The per-slice sensitivity and specificity of the DLA for detecting grade 3 motion-related artifacts were 91.5% (97/106) and 96.8% (1134/1172) in the internal validation dataset and 93.3% (265/284) and 91.6% (948/1035) in the external validation dataset. The per-examination sensitivity and specificity were 92.0% (23/25) and 99.7% (287/288) in the internal validation dataset and 90.0% (72/80) and 96.0% (239/249) in the external validation dataset, respectively. The processing time of the DLA for automatic grading of motion-related artifacts was from 4.11 to 4.22 seconds per MRI examination.

CONCLUSIONS

The DLA enabled automatic and instant detection and grading of motion-related artifacts on arterial phase gadoxetic acid-enhanced liver MRI.

摘要

目的

本研究旨在开发并验证一种基于深度学习的算法（DLA），用于在动脉期肝脏磁共振成像（MRI）上自动检测与运动相关的伪影并进行分级。

材料与方法

基于改良的ResNet-101和U-net的用于检测与运动相关伪影并进行分级的多步骤DLA，使用2017年获得的336例钆塞酸增强肝脏MRI检查的动脉期图像进行训练（训练数据集；平均年龄68.6岁[范围18 - 95岁]；254名男性）。使用三级分级系统在4个不同的MRI切片中评估与运动相关的伪影。在验证数据集中，纳入了2018年来自同一机构的313幅图像（内部验证数据集；平均年龄67.2岁[范围21 - 87岁]；228名男性）以及来自3个不同机构的329幅图像（外部验证数据集；平均年龄64.0岁[范围23 - 90岁]；214名男性），并评估了检测与运动相关伪影的每切片和每次检查的性能。

结果

在内部验证数据集中，DLA检测3级与运动相关伪影的每切片灵敏度和特异性分别为91.5%（97/106）和96.8%（1134/1172），在外部验证数据集中分别为93.3%（265/284）和91.6%（948/1035）。在内部验证数据集中，每次检查的灵敏度和特异性分别为92.0%（23/25）和99.7%（287/288），在外部验证数据集中分别为90.0%（72/80）和96.0%（239/249）。DLA对与运动相关伪影进行自动分级的处理时间为每次MRI检查4.11至4.22秒。