Search

Aim: Vertebral compression fractures (VCF) occur most commonly at the thoracolumbar junction (TLJ). Balloon kyphoplasty (BKP) is an effective method of bone remodeling in these cases. In this study, we evaluate the parameters that affect bone retropulsion and restoration in TLJ VCF without neurological deficits., Material and Methods: Thirty-one of Frankel E and AO A3-4 type VCFs fractures at the TLJ, with bone retropulsion into the spinal canal, from 2017 to 2020, were evaluated retrospectively. Data was gathered on patient demographics and medical histories. Measurements of anterior vertebral heights, posterior vertebral heights, local kyphotic angles, spinal cord area, and bone retropulsion into the spinal canal (BRC) were evaluated preoperatively, early postoperatively, and late postoperatively., Results: In those patients who underwent early surgery ( 4 weeks postfracture), a significantly greater increase in anterior vertebral heights was seen between early postoperative and preoperative measurements than in patients who underwent late surgery ( 4 weeks postfracture) (p = 0.016). At the six-month follow-up, a significantly greater decrease in local kyphotic angle measurements was seen in patients over 65 years of age than those under 65 (p = 0.023). Comparison of local kyphotic angles between sexes revealed a significant decrease in measurements at follow-up in female patients (p = 0.029). Both early postoperative and late postoperative local kyphotic angle measurements of patients with a body mass index (BMI) ≥25 were significantly lower than those of patients with a BMI 25 (p = 0.012)., Conclusion: The restoration of vertebral angles and heights with the maximum level of BKP can effectively reduce BRC.

Purpose: Rapid detection and vascular territorial classification of stroke enable the determination of the most appropriate treatment. In this study, we aimed to investigate the performance of convolutional neural network (CNN) models in the detection and vascular territorial classification of stroke on diffusion-weighted images (DWI)., Methods: DWI of 421 cases (271 acute ischemic stroke patients and 150 cases without any ischemia findings on DWI) obtained between January 2017 to April 2020 were reviewed. We created two custom datasets. A stroke detection dataset was created with 1800 slices (900 S and 900 normal) consisting of 1400 for training, 200 for validation, 200 for test. A vascular territorial type dataset was created with 1717 slices (883 middle cerebral artery stroke, 416 posterior circulatory stroke, and 418 watershed stroke) consisting of 1117 slices for training, 300 for validation, 300 for test. A transfer learning approach based on MobileNetV2 and EfficientNet-B0 CNN architecture was used. The performance of the models was evaluated., Results: Modified MobileNetV2 and EfficientNet-B0 models achieved 96% (κ: 0.92) and 93% (κ: 0.86) accuracy in stroke detection, respectively. In vascular territorial classification of stroke as middle cerebral artery, posterior circulation, or watershed infarction, an accuracy of 93% (κ: 0.895) was achieved with modified MobileNetV2 model and 87% (κ: 0.805) with modified EfficientNet-B0 CNN model., Conclusion: Transfer learning approach with custom top CNN models achieve sufficiently high performance for both the detection of ischemic stroke and the classification of its vascular territorial type on DWI., (Copyright © 2021 Elsevier B.V. All rights reserved.)