Department of Spinal Surgery, The Fifth Affiliated Hospital of Sun Yat-Sen University, Zhuhai, China.
Department of Radiology, The Fifth Affiliated Hospital of Sun Yat-Sen University, Zhuhai, China.
Front Endocrinol (Lausanne). 2022 Jun 6;13:890371. doi: 10.3389/fendo.2022.890371. eCollection 2022.
Accurate severity grading of lumbar spine disease by magnetic resonance images (MRIs) plays an important role in selecting appropriate treatment for the disease. However, interpreting these complex MRIs is a repetitive and time-consuming workload for clinicians, especially radiologists. Here, we aim to develop a multi-task classification model based on artificial intelligence for automated grading of lumbar disc herniation (LDH), lumbar central canal stenosis (LCCS) and lumbar nerve roots compression (LNRC) at lumbar axial MRIs.
Total 15254 lumbar axial T2W MRIs as the internal dataset obtained from the Fifth Affiliated Hospital of Sun Yat-sen University from January 2015 to May 2019 and 1273 axial T2W MRIs as the external test dataset obtained from the Third Affiliated Hospital of Southern Medical University from June 2016 to December 2017 were analyzed in this retrospective study. Two clinicians annotated and graded all MRIs using the three international classification systems. In agreement, these results served as the reference standard; In disagreement, outcomes were adjudicated by an expert surgeon to establish the reference standard. The internal dataset was randomly split into an internal training set (70%), validation set (15%) and test set (15%). The multi-task classification model based on ResNet-50 consists of a backbone network for feature extraction and three fully-connected (FC) networks for classification and performs the classification tasks of LDH, LCCS, and LNRC at lumbar MRIs. Precision, accuracy, sensitivity, specificity, F1 scores, confusion matrices, receiver-operating characteristics and interrater agreement (Gwet k) were utilized to assess the model's performance on the internal test dataset and external test datasets.
A total of 1115 patients, including 1015 patients from the internal dataset and 100 patients from the external test dataset [mean age, 49 years ± 15 (standard deviation); 543 women], were evaluated in this study. The overall accuracies of grading for LDH, LCCS and LNRC were 84.17% (74.16%), 86.99% (79.65%) and 81.21% (74.16%) respectively on the internal (external) test dataset. Internal and external testing of three spinal diseases showed substantial to the almost perfect agreement (k, 0.67 - 0.85) for the multi-task classification model.
The multi-task classification model has achieved promising performance in the automated grading of LDH, LCCS and LNRC at lumbar axial T2W MRIs.
磁共振成像（MRI）对腰椎疾病进行准确的严重程度分级在为该疾病选择合适的治疗方法方面发挥着重要作用。然而，对于临床医生来说，尤其是放射科医生来说，解读这些复杂的 MRI 是一项重复且耗时的工作。在这里，我们旨在开发一种基于人工智能的多任务分类模型，用于对腰椎轴向 MRI 进行腰椎间盘突出症（LDH）、腰椎中央管狭窄症（LCCS）和腰椎神经根受压症（LNRC）的自动分级。
本回顾性研究共分析了中山大学附属第五医院 2015 年 1 月至 2019 年 5 月期间获得的 15254 例腰椎轴向 T2W MRI 作为内部数据集和南方医科大学第三附属医院 2016 年 6 月至 2017 年 12 月期间获得的 1273 例轴向 T2W MRI 作为外部测试数据集。两位临床医生使用三种国际分类系统对所有 MRI 进行了注释和分级。一致的结果作为参考标准；不一致的结果由一位专家外科医生进行裁决，以建立参考标准。内部数据集被随机分为内部训练集（70%）、验证集（15%）和测试集（15%）。基于 ResNet-50 的多任务分类模型由一个用于特征提取的骨干网络和三个用于分类的全连接（FC）网络组成，用于对腰椎 MRI 进行 LDH、LCCS 和 LNRC 的分类任务。在内部测试数据集和外部测试数据集上，使用精度、准确度、敏感度、特异性、F1 分数、混淆矩阵、接收者操作特征和评分者间一致性（Gwet k）来评估模型的性能。
共有 1115 名患者（内部数据集 1015 名患者，外部测试数据集 100 名患者[平均年龄 49 岁±15（标准差）；543 名女性]）参与了本研究。在内部（外部）测试数据集上，LDH、LCCS 和 LNRC 分级的总体准确率分别为 84.17%（74.16%）、86.99%（79.65%）和 81.21%（74.16%）。三种脊柱疾病的内部和外部测试对于多任务分类模型具有高度一致到几乎完美的一致性（k 值，0.67-0.85）。
多任务分类模型在腰椎轴向 T2W MRI 上对 LDH、LCCS 和 LNRC 的自动分级中取得了有希望的性能。
