Impact of Aortic Arch Anatomy on Technical Performance and Clinical Outcomes in Patients with Acute Ischemic Stroke.

BACKGROUND AND PURPOSE: Arterial access is a technical consideration of mechanical thrombectomy that may affect procedural time, but few studies exist detailing the relationship of anatomy to procedural times and patient outcomes. We sought to investigate the respective impact of aortic arch and carotid artery anatomy on endovascular procedural times in patients with large-vessel occlusion. MATERIALS AND METHODS: We retrospectively reviewed imaging and medical records of 207 patients from 2 academic institutions who underwent mechanical thrombectomy for anterior circulation large-vessel occlusion from January 2015 to July 2018. Preintervention CTAs were assessed to measure features of the aortic arch and ipsilateral great vessel anatomy. These included the cranial-to-caudal distance from the origin of the innominate artery to the top of the aortic arch and the takeoff angle of the respective great vessel from the arch. mRS scores were calculated from rehabilitation and other outpatient documentation. We performed bootstrap, stepwise regressions to model groin puncture to reperfusion time and binary mRS outcomes (good outcome, mRS ≤ 2). RESULTS: From our linear regression for groin puncture to reperfusion time, we found a significant association of the great vessel takeoff angle (P = .002) and caudal distance from the origin of the innominate artery to the top of the aortic arch (P = .05). Regression analysis for the binary mRS revealed a significant association with groin puncture to reperfusion time (P < .001). CONCLUSIONS: These results demonstrate that patients with larger takeoff angles and extreme aortic arches have an association with longer procedural times as approached from transfemoral access routes.

Convolutional Neural Network-Based Automated Segmentation of the Spinal Cord and Contusion Injury: Deep Learning Biomarker Correlates of Motor Impairment in Acute Spinal Cord Injury.

BACKGROUND AND PURPOSE: Our aim was to use 2D convolutional neural networks for automatic segmentation of the spinal cord and traumatic contusion injury from axial T2-weighted MR imaging in a cohort of patients with acute spinal cord injury. MATERIALS AND METHODS: Forty-seven patients who underwent 3T MR imaging within 24 hours of spinal cord injury were included. We developed an image-analysis pipeline integrating 2D convolutional neural networks for whole spinal cord and intramedullary spinal cord lesion segmentation. Linear mixed modeling was used to compare test segmentation results between our spinal cord injury convolutional neural network (Brain and Spinal Cord Injury Center segmentation) and current state-of-the-art methods. Volumes of segmented lesions were then used in a linear regression analysis to determine associations with motor scores. RESULTS: Compared with manual labeling, the average test set Dice coefficient for the Brain and Spinal Cord Injury Center segmentation model was 0.93 for spinal cord segmentation versus 0.80 for PropSeg and 0.90 for DeepSeg (both components of the Spinal Cord Toolbox). Linear mixed modeling showed a significant difference between Brain and Spinal Cord Injury Center segmentation compared with PropSeg (P < .001) and DeepSeg (P < .05). Brain and Spinal Cord Injury Center segmentation showed significantly better adaptability to damaged areas compared with PropSeg (P < .001) and DeepSeg (P < .02). The contusion injury volumes based on automated segmentation were significantly associated with motor scores at admission (P = .002) and discharge (P = .009). CONCLUSIONS: Brain and Spinal Cord Injury Center segmentation of the spinal cord compares favorably with available segmentation tools in a population with acute spinal cord injury. Volumes of injury derived from automated lesion segmentation with Brain and Spinal Cord Injury Center segmentation correlate with measures of motor impairment in the acute phase. Targeted convolutional neural network training in acute spinal cord injury enhances algorithm performance for this patient population and provides clinically relevant metrics of cord injury.

MRI Atlas-Based Measurement of Spinal Cord Injury Predicts Outcome in Acute Flaccid Myelitis.

BACKGROUND AND PURPOSE: Recent advances in spinal cord imaging analysis have led to the development of a robust anatomic template and atlas incorporated into an open-source platform referred to as the Spinal Cord Toolbox. Using the Spinal Cord Toolbox, we sought to correlate measures of GM, WM, and cross-sectional area pathology on T2 MR imaging with motor disability in patients with acute flaccid myelitis. MATERIALS AND METHODS: Spinal cord imaging for 9 patients with acute flaccid myelitis was analyzed by using the Spinal Cord Toolbox. A semiautomated pipeline using the Spinal Cord Toolbox measured lesion involvement in GM, WM, and total spinal cord cross-sectional area. Proportions of GM, WM, and cross-sectional area affected by T2 hyperintensity were calculated across 3 ROIs: 1) center axial section of lesion; 2) full lesion segment; and 3) full cord atlas volume. Spearman rank order correlation was calculated to compare MR metrics with clinical measures of disability. RESULTS: Proportion of GM metrics at the center axial section significantly correlated with measures of motor impairment upon admission (r [9] = -0.78; P = .014) and at 3-month follow-up (r [9] = -0.66; P = .05). Further, proportion of GM extracted across the full lesion segment significantly correlated with initial motor impairment (r [9] = -0.74, P = .024). No significant correlation was found for proportion of WM or proportion of cross-sectional area with clinical disability. CONCLUSIONS: Atlas-based measures of proportion of GM T2 signal abnormality measured on a single axial MR imaging section and across the full lesion segment correlate with motor impairment and outcome in patients with acute flaccid myelitis. This is the first atlas-based study to correlate clinical outcomes with segmented measures of T2 signal abnormality in the spinal cord.