Automated detection of axonal damage along white matter tracts in acute severe traumatic brain injury.
Automated detection of axonal damage along white matter tracts in acute severe traumatic brain injury.
复制标题
DOI：
10.1016/j.nicl.2022.103294
复制
发表时间：
2023
影响因子：
4.2
通讯作者：
Edlow, Brian L.
中科院分区：
文献类型：
--
作者：
Maffei, Chiara;Gilmore, Natalie;Snider, Samuel B.;Foulkes, Andrea S.;Bodien, Yelena G.;Yendiki, Anastasia;Edlow, Brian L.
我们提出了一种自动纤维束成像方法，用于个体化TAI检测。我们测试了自动纤维束成像在急性严重TBI患者中的可行性。我们调查了与意识行为测量的关联。TRACULA管道成功重建了93%的患者管道。管道准确区分患者和对照组（AUC：0.91）需要个性化评估白色物质完整性的新技术来检测创伤性轴突损伤（TAI）并预测急性重度创伤性脑损伤（TBI）危重患者的结局。弥散磁共振纤维束成像具有量化体内白色物质微观结构的潜力，并已被用于表征TBI后的特定纤维束变化。然而，纤维束成像在临床环境中并不常规用于评估TAI的程度，部分原因是局灶性病变降低了自动化方法的鲁棒性。在这里，我们提出了一个管道，结合自动纤维束成像重建的40个白色的物质tracts与沿道扩散指标的多变量分析，以评估存在TAI急性严重TBI的个别患者。我们使用Mahalanobis距离来识别18例急性严重TBI患者和33例健康受试者的异常白色物质束。在可获得FreeSurfer解剖分割的所有患者（18例患者中的17例）中，包括13例局灶性病变，自动化管道成功重建了平均37.5 ± 2.1个白色物质束，无需手动干预。平均2.5 ± 2.1个束导致部分重建或重建失败，需要在目视检查时重新初始化。管道在所有患者中检测到至少一个异常管道（平均值：9.1 ± 7.9），并准确区分患者和对照组（AUC：0.91）。异常神经束的数量和神经解剖位置因患者和意识水平而异。胼胝体的运动前区、颞区和顶叶部分是最常见的受损部位（分别为10例、9例和8例），这与之前的TAI组织病理学研究一致。TAI指标与同期意识行为指标无关。总之，我们提供了原理证明证据，证明自动纤维束成像管道具有检测和量化急性重度TBI个体患者TAI的转化潜力。
We proposed an automated tractography approach for individualized TAI detection. We tested the feasibility of automated tractography in patients with acute severe TBI. We investigated associations with behavioral measures of consciousness. The TRACULA pipeline successfully reconstructed 93% of the tracts across patients. The pipeline accurately discriminated between patients and controls (AUC: 0.91) New techniques for individualized assessment of white matter integrity are needed to detect traumatic axonal injury (TAI) and predict outcomes in critically ill patients with acute severe traumatic brain injury (TBI). Diffusion MRI tractography has the potential to quantify white matter microstructure in vivo and has been used to characterize tract-specific changes following TBI. However, tractography is not routinely used in the clinical setting to assess the extent of TAI, in part because focal lesions reduce the robustness of automated methods. Here, we propose a pipeline that combines automated tractography reconstructions of 40 white matter tracts with multivariate analysis of along-tract diffusion metrics to assess the presence of TAI in individual patients with acute severe TBI. We used the Mahalanobis distance to identify abnormal white matter tracts in each of 18 patients with acute severe TBI as compared to 33 healthy subjects. In all patients for which a FreeSurfer anatomical segmentation could be obtained (17 of 18 patients), including 13 with focal lesions, the automated pipeline successfully reconstructed a mean of 37.5 ± 2.1 white matter tracts without the need for manual intervention. A mean of 2.5 ± 2.1 tracts resulted in partial or failed reconstructions and needed to be reinitialized upon visual inspection. The pipeline detected at least one abnormal tract in all patients (mean: 9.1 ± 7.9) and accurately discriminated between patients and controls (AUC: 0.91). The number and neuroanatomic location of abnormal tracts varied across patients and levels of consciousness. The premotor, temporal, and parietal sections of the corpus callosum were the most commonly damaged tracts (in 10, 9, and 8 patients, respectively), consistent with prior histopathological studies of TAI. TAI measures were not associated with concurrent behavioral measures of consciousness. In summary, we provide proof-of-principle evidence that an automated tractography pipeline has translational potential to detect and quantify TAI in individual patients with acute severe TBI.
登录
查看更多内容
DOI：
10.7759/cureus.1723
发表时间：
2017-09-28
期刊：
Cureus
影响因子：
--
作者：
Ordóñez-Rubiano EG;Johnson J;Enciso-Olivera CO;Marín-Muñoz JH;Cortes-Lozano W;Baquero-Herrera PE;Ordóñez-Mora EG;Cifuentes-Lobelo HA
通讯作者：
Cifuentes-Lobelo HA
影响因子：
4.2
作者：
Bodien, Yelena G.;Barra, Alice;Edlow, Brian L.
通讯作者：
Edlow, Brian L.
影响因子：
14.5
作者：
Edlow, Brian L.;Chatelle, Camille;Wu, Ona
通讯作者：
Wu, Ona
影响因子：
5.7
作者：
Dale, AM;Fischl, B;Sereno, MI
通讯作者：
Sereno, MI
影响因子：
3.3
作者：
Benner, Thomas;van der Kouwe, Andre J. W.;Sorensen, A. Gregory
通讯作者：
Sorensen, A. Gregory