1Department of Neurosurgery, Johns Hopkins University School of Medicine, Baltimore.
2Malone Center for Engineering in Healthcare, Johns Hopkins University, Baltimore; and.
Neurosurg Focus. 2022 Apr;52(4):E5. doi: 10.3171/2022.1.FOCUS21745.
Damage to the thoracolumbar spine can confer significant morbidity and mortality. The Thoracolumbar Injury Classification and Severity Score (TLICS) is used to categorize injuries and determine patients at risk of spinal instability for whom surgical intervention is warranted. However, calculating this score can constitute a bottleneck in triaging and treating patients, as it relies on multiple imaging studies and a neurological examination. Therefore, the authors sought to develop and validate a deep learning model that can automatically categorize vertebral morphology and determine posterior ligamentous complex (PLC) integrity, two critical features of TLICS, using only CT scans.
All patients who underwent neurosurgical consultation for traumatic spine injury or degenerative pathology resulting in spine injury at a single tertiary center from January 2018 to December 2019 were retrospectively evaluated for inclusion. The morphology of injury and integrity of the PLC were categorized on CT scans. A state-of-the-art object detection region-based convolutional neural network (R-CNN), Faster R-CNN, was leveraged to predict both vertebral locations and the corresponding TLICS. The network was trained with patient CT scans, manually labeled vertebral bounding boxes, TLICS morphology, and PLC annotations, thus allowing the model to output the location of vertebrae, categorize their morphology, and determine the status of PLC integrity.
A total of 111 patients were included (mean ± SD age 62 ± 20 years) with a total of 129 separate injury classifications. Vertebral localization and PLC integrity classification achieved Dice scores of 0.92 and 0.88, respectively. Binary classification between noninjured and injured morphological scores demonstrated 95.1% accuracy. TLICS morphology accuracy, the true positive rate, and positive injury mismatch classification rate were 86.3%, 76.2%, and 22.7%, respectively. Classification accuracy between no injury and suspected PLC injury was 86.8%, while true positive, false negative, and false positive rates were 90.0%, 10.0%, and 21.8%, respectively.
In this study, the authors demonstrate a novel deep learning method to automatically predict injury morphology and PLC disruption with high accuracy. This model may streamline and improve diagnostic decision support for patients with thoracolumbar spinal trauma.
胸腰椎损伤可导致严重的发病率和死亡率。胸腰椎损伤分类及严重程度评分（TLICS）用于对损伤进行分类，并确定存在脊柱不稳定风险需要手术干预的患者。然而，计算该评分可能会成为分诊和治疗患者的瓶颈，因为它依赖于多个影像学研究和神经学检查。因此，作者试图开发和验证一种深度学习模型，该模型仅使用 CT 扫描即可自动对椎体形态进行分类，并确定后纵韧带复合体（PLC）的完整性，这是 TLICS 的两个关键特征。
回顾性评估了 2018 年 1 月至 2019 年 12 月在一家三级中心因创伤性脊柱损伤或退行性病变导致脊柱损伤而接受神经外科咨询的所有患者，以纳入研究。在 CT 扫描上对损伤的形态和 PLC 的完整性进行分类。利用基于区域的卷积神经网络（R-CNN）的先进目标检测 Faster R-CNN 来预测椎体位置和相应的 TLICS。该网络使用患者 CT 扫描、手动标记的椎体边界框、TLICS 形态和 PLC 注释进行训练，从而使模型能够输出椎体位置、对其形态进行分类并确定 PLC 完整性状态。
共纳入 111 例患者（平均年龄 62 ± 20 岁），共 129 例单独损伤分类。椎体定位和 PLC 完整性分类的 Dice 评分分别为 0.92 和 0.88。非损伤和损伤形态评分之间的二进制分类显示准确率为 95.1%。TLICS 形态准确性、真阳性率和阳性损伤错配分类率分别为 86.3%、76.2%和 22.7%。无损伤与疑似 PLC 损伤之间的分类准确率为 86.8%，而真阳性、假阴性和假阳性率分别为 90.0%、10.0%和 21.8%。
在这项研究中，作者展示了一种新的深度学习方法，可以非常准确地自动预测损伤形态和 PLC 中断。该模型可以简化并改进胸腰椎脊柱创伤患者的诊断决策支持。
