Automated MR Image Prescription of the Liver Using Deep Learning: Development, Evaluation, and Prospective Implementation
Automated MR Image Prescription of the Liver Using Deep Learning: Development, Evaluation, and Prospective Implementation
复制标题
DOI：
10.1002/jmri.28564
复制
发表时间：
2022-12-30
影响因子：
4.4
通讯作者：
Hernando, Diego
中科院分区：
文献类型：
--
作者：
Geng, Ruiqi;Buelo, Collin J.;Hernando, Diego
背景：对于肝脏的全自动图像处方，以实现高效、可再现的MRI，存在未满足的需求。目的：开发和评价基于人工智能（AI）的肝脏图像处方。研究类型：前瞻性。人口：共有570名女性/469名男性患者（年龄：56岁和PLUSMN; 17岁），其中72%/8%/20%被随机分配用于培训/验证/测试;两名女性/四名男性健康志愿者（年龄：31 +/- 6岁）。场强/序列：1.5 T、3.0 T;自旋回波、梯度回波、bSSFP。考核：回顾性检索了来自连续临床肝脏MRI检查的共计1039个三平面定位器采集（26，929个切片），并由6名放射科医生进行了注释。定位器图像和手动注释用于训练对象检测卷积神经网络（YOLOv 3），以跨定位器图像方向检测多个对象类别（肝脏、躯干和手臂），并输出相应的2D边界框。基于这些边界框获得标准方向上的全肝图像处方。通过计算手动和自动标记之间的交集（IoU），在测试数据集上评估2D检测性能。通过测量每个维度的边界不匹配和AI处方覆盖的手动体积百分比来计算3D处方准确度。自动处方在3 T MR系统上实现，并在健康志愿者身上进行前瞻性评价。统计检验：进行配对t检验（阈值= 0.05），以评估经训练的网络之间性能差异的显著性。结果如下：在208个测试数据集中，所提出的全网络方法与手动注释具有良好的一致性，所有七个类别的中位数IoU > 0.91（四分位距< 0.09）。自动3D处方是准确的，对于99.5%的测试数据集，3D轴向处方的上级/下部尺寸的偏移< 2.3 cm，与放射科医师的阅片员间再现性相当。对于患者的3D轴向处方，全网络的性能明显优于微小网络的上级性能。在前瞻性研究中，自动处方在单次激发快速自旋回波、梯度回波和平衡稳态自由旋进序列中表现良好。数据结论：基于AI的自动肝脏图像处方在研究的患者、病理和场强中表现出了良好的性能。
Background: There is an unmet need for fully automated image prescription of the liver to enable efficient, reproducible MRI. Purpose: To develop and evaluate artificial intelligence (AI)-based liver image prescription. Study Type: Prospective. Population: A total of 570 female/469 male patients (age: 56 & PLUSMN; 17 years) with 72%/8%/20% assigned randomly for training/validation/testing; two female/four male healthy volunteers (age: 31 +/- 6 years). Field Strength/Sequence: 1.5 T, 3.0 T; spin echo, gradient echo, bSSFP. Assessment: A total of 1039 three-plane localizer acquisitions (26,929 slices) from consecutive clinical liver MRI examinations were retrieved retrospectively and annotated by six radiologists. The localizer images and manual annotations were used to train an object-detection convolutional neural network (YOLOv3) to detect multiple object classes (liver, torso, and arms) across localizer image orientations and to output corresponding 2D bounding boxes. Whole-liver image prescription in standard orientations was obtained based on these bounding boxes. 2D detection performance was evaluated on test datasets by calculating intersection over union (IoU) between manual and automated labeling. 3D prescription accuracy was calculated by measuring the boundary mismatch in each dimension and percentage of manual volume covered by AI prescription. The automated prescription was implemented on a 3 T MR system and evaluated prospectively on healthy volunteers. Statistical Tests: Paired t-tests (threshold = 0.05) were conducted to evaluate significance of performance difference between trained networks. Results: In 208 testing datasets, the proposed method with full network had excellent agreement with manual annotations, with median IoU > 0.91 (interquartile range < 0.09) across all seven classes. The automated 3D prescription was accurate, with shifts < 2.3 cm in superior/inferior dimension for 3D axial prescription for 99.5% of test datasets, comparable to radiologists' interreader reproducibility. The full network had significantly superior performance than the tiny network for 3D axial prescription in patients. Automated prescription performed well across single-shot fast spin-echo, gradient-echo, and balanced steady-state free-precession sequences in the prospective study. Data Conclusion: AI-based automated liver image prescription demonstrated promising performance across the patients, pathologies, and field strengths studied.