Chen Yi, He Zhong, Yang Kenneth Guangpu, Qin Xiaodong, Lau Adam Yiu-Chung, Liu Zhen, Lu Neng, Cheng Jack Chun-Yiu, Lee Wayne Yuk-Wai, Chui Elvis Chun-Sing, Qiu Yong, Liu Xiaoli, Chen Xipu, Zhu Zezhang
Division of Spine Surgery, Department of Orthopedic Surgery, Affiliated Hospital of Medical School, Nanjing University, Nanjing Drum Tower Hospital, Nanjing, China.
Joint Scoliosis Research Center of The Chinese University of Hong Kong and Nanjing University, Nanjing, China.
Global Spine J. 2025 Jun 11:21925682251349999. doi: 10.1177/21925682251349999.
Study DesignRetrospective diagnostic study.ObjectivesTo develop a fine-grained classification model based on deep learning using X-ray images, to screen for scoliosis, and further to screen for atypical scoliosis patterns associated with Chiari Malformation type I (CMS).MethodsA total of 508 pairs of coronal and sagittal X-ray images from patients with CMS, adolescent idiopathic scoliosis (AIS), and normal controls (NC) were processed through construction of the ResNet-50 model, including the development of ResNet-50 Coronal, ResNet-50 Sagittal, ResNet-50 Dual, ResNet-50 Concat, and ResNet-50 Bilinear models. Evaluation metrics calculated included accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for both the scoliosis diagnosis system and the CMS diagnosis system, along with the generation of receiver operating characteristic (ROC) curves and heatmaps for CMS diagnosis.ResultsThe classification results for the scoliosis diagnosis system showed that the ResNet-50 Coronal model had the best overall performance. For the CMS diagnosis system, the ResNet-50 Coronal and ResNet-50 Dual models demonstrated optimal performance. Specifically, the ResNet-50 Dual model reached the diagnostic level of senior spine surgeons, and the ResNet-50 Coronal model even surpassed senior surgeons in specificity and PPV. The CMS heatmaps revealed that major classification weights were concentrated on features such as atypical curve types, significant lateral shift of scoliotic segments, longer affected segments, and severe trunk tilt.ConclusionsThe fine-grained classification model based on the ResNet-50 network can accurately screen for atypical scoliosis patterns associated with CMS, highlighting the importance of radiographic features such as atypical curve types in model classification.
研究设计
回顾性诊断研究。
目的
基于深度学习利用X线图像开发一种细粒度分类模型，用于筛查脊柱侧弯，并进一步筛查与I型Chiari畸形（CMS）相关的非典型脊柱侧弯模式。
方法
通过构建ResNet-50模型，对来自CMS患者、青少年特发性脊柱侧弯（AIS）患者和正常对照（NC）的总共508对冠状位和矢状位X线图像进行处理，包括开发ResNet-50冠状位模型、ResNet-50矢状位模型、ResNet-50双模型、ResNet-50拼接模型和ResNet-50双线性模型。计算的评估指标包括脊柱侧弯诊断系统和CMS诊断系统的准确性、敏感性、特异性、阳性预测值（PPV）和阴性预测值（NPV），以及生成CMS诊断的受试者操作特征（ROC）曲线和热图。
结果
脊柱侧弯诊断系统的分类结果表明，ResNet-50冠状位模型具有最佳的整体性能。对于CMS诊断系统，ResNet-50冠状位模型和ResNet-50双模型表现出最佳性能。具体而言，ResNet-50双模型达到了资深脊柱外科医生的诊断水平，而ResNet-50冠状位模型在特异性和PPV方面甚至超过了资深外科医生。CMS热图显示，主要分类权重集中在非典型曲线类型、脊柱侧弯节段明显侧移、受累节段较长和严重躯干倾斜等特征上。
结论
基于ResNet-50网络的细粒度分类模型能够准确筛查与CMS相关的非典型脊柱侧弯模式，突出了非典型曲线类型等影像学特征在模型分类中的重要性。
