Automated magnetic resonance image segmentation of the anterior cruciate ligament.
Automated magnetic resonance image segmentation of the anterior cruciate ligament.
复制标题
前十字韧带的自动磁共振图像分割。
DOI：
10.1002/jor.24926
复制
发表时间：
2021-04
期刊：
影响因子：
--
通讯作者：
Fleming BC
中科院分区：
文献类型：
--
作者：
Flannery SW;Kiapour AM;Edgar DJ;Murray MM;Fleming BC
这项工作的目的是开发一种前十字韧带的自动分割方法，该方法能够促进临床和研究环境中韧带的定量评估。改进的 U-Net 全卷积网络模型在 246 张完整前十字韧带稳态磁共振图像中的相长干涉进行了训练、验证和测试。相对于经验丰富（> 5 年）的“地面实况”分割器，在图像集上评估了两个领域的总体模型性能：解剖相似性和从自动分割获得的定量测量（即信号强度和体积）的准确性。为了建立相对于手动分割的模型可靠性，由地面真实分割器和两个附加分割器（A：6 个月，B：2 年经验）对成像数据的子集进行重新分割，并相对于地面真实情况评估其性能。最终模型在解剖性能指标上得分很高（Dice 系数=.84，精度=.82，灵敏度=.85）。自动分割的中值信号强度和体积与真实值没有显着差异（分别为 0.3% 差异，p=.9；2.3% 差异，p=.08）。将模型结果与独立分段器进行比较时，模型预测显示出更高的中值 Dice 系数（A=.73，p=.001；B=.77，p=NS）和敏感性（A=.68，p=.001；B=.72，p=.003）。该模型在所有测量上都表现得同样好，可以通过地面实况分割器重新测试分割。从自动分割模型中提取的定量测量与手动分割没有什么不同，这使得它们能够在定量 MRI 管道中使用来评估前十字韧带。
The objective of this work was to develop an automated segmentation method for the anterior cruciate ligament that is capable of facilitating quantitative assessments of ligament in clinical and research settings. A modified U-Net fully convolutional network model was trained, validated, and tested on 246 Constructive Interference in Steady State magnetic resonance images of intact anterior cruciate ligaments. Overall model performance was assessed on the image set relative to an experienced (>5 years) “ground truth” segmenter in two domains: anatomical similarity and the accuracy of quantitative measurements (i.e. signal intensity and volume) obtained from the automated segmentation. To establish model reliability relative to manual segmentation, a subset of the imaging data was re-segmented by the ground truth segmenter and two additional segmenters (A: 6 months, B: 2 years of experience), with their performance evaluated relative to the ground truth. The final model scored well on anatomical performance metrics (Dice coefficient=.84, precision=.82, sensitivity=.85). The median signal intensities and volumes of the automated segmentations were not significantly different from ground truth (0.3% difference, p=.9; 2.3% difference, p=.08, respectively). When the model results were compared to the independent segmenters, the model predictions demonstrated greater median Dice coefficient (A=.73, p=.001; B=.77, p=NS) and sensitivity (A=.68, p=.001; B=.72, p=.003). The model performed equivalently well to re-test segmentation by the ground truth segmenter on all measures. The quantitative measures extracted from the automated segmentation model did not differ from those of manual segmentation, enabling their use in quantitative MRI pipelines to evaluate the anterior cruciate ligament.
登录
查看更多内容
前交叉韧带的结构和解剖学恢复与手术后1年的软骨损伤有关：愈合韧带的特性会影响软骨损伤。
影响因子：
2.6
作者：
Kiapour AM;Fleming BC;Murray MM
通讯作者：
Murray MM
影响因子：
7
作者：
Koff, M. F.;Shah, P.;Pownder, S.;Romero, B.;Williams, R.;Gilbert, S.;Maher, S.;Fortier, L. A.;Rodeo, S. A.;Potter, H. G.
通讯作者：
Potter, H. G.
影响因子：
2.8
作者：
Biercevicz, Alison M.;Proffen, Benedikt L.;Murray, Martha M.;Walsh, Edward G.;Fleming, Braden C.
通讯作者：
Fleming, Braden C.
影响因子：
19.7
作者：
Pauli, Chantal;Bae, Won C.;Du, Jiang
通讯作者：
Du, Jiang
影响因子：
3.9
作者：
Ahn, Chunsoo;Bui, Toan Duc;Park, Hyunjin
通讯作者：
Park, Hyunjin