中国组织工程研究

中国组织工程研究 ›› 2026, Vol. 30 ›› Issue (21): 5369-5375.doi: 10.12307/2026.776

• 骨与关节生物力学 bone and joint biomechanics • 上一篇    下一篇

颈椎失稳深度学习预测模型的构建及验证

路广琦，孙馨悦，韩  雪，刘亚坤，马明明，毛瀚泽，周帅琪，梁  龙，李  靖，胡家铭，朱立国，于  杰，庄明辉   

  1. 中国中医科学院望京医院，北京市  100102
  • 接受日期:2025-08-26 出版日期:2026-07-28 发布日期:2026-03-03
  • 通讯作者: 于杰，博士，主任医师，中国中医科学院望京医院，北京市 100102 庄明辉，博士，主治医师，中国中医科学院望京医院，北京市 100102
  • 作者简介:路广琦，男，1997年生，山东省济南市人，汉族，中国中医科学院研究生院在读博士，医师，主要从事中西医结合治疗脊柱疾病的临床与基础研究。
  • 基金资助:
    中国中医科学院望京医院高水平中医医院建设项目中医药临床循证研究专项(WJYY-XZKT-2023-01)，项目负责人：于杰；国家自然科学基金面上项目(82274560)，项目负责人：于杰；国家自然科学基金青年项目(82405438)，项目负责人：庄明辉；中国中医科学院望京医院自主选题专项课题(WJYY-ZZXT-2023-24)，项目负责人：庄明辉

Construction and validation of a deep learning prediction model for cervical instability

Lu Guangqi, Sun Xinyue, Han Xue, Liu Yakun, Ma Mingming, Mao Hanze, Zhou Shuaiqi, Liang Long, Li Jing, Hu Jiaming, Zhu Liguo, Yu Jie, Zhuang Minghui   

  1. Wangjing Hospital, Chinese Academy of Traditional Chinese Medicine, Beijing 100102, China  
  • Accepted:2025-08-26 Online:2026-07-28 Published:2026-03-03
  • Contact: Yu Jie, MD, Chief physician, Wangjing Hospital, Chinese Academy of Traditional Chinese Medicine, Beijing 100102, China Zhuang Minghui, MD, Attending physician, Wangjing Hospital, Chinese Academy of Traditional Chinese Medicine, Beijing 100102, Chin
  • About author:Lu Guangqi, MD candidate, Physician, Wangjing Hospital, Chinese Academy of Traditional Chinese Medicine, Beijing 100102, China
  • Supported by:
    Special Project on Clinical Evidence-Based Research on Traditional Chinese Medicine in the High-Level Traditional Chinese Medicine Hospital Construction Project of Wangjing Hospital of Chinese Academy of Traditional Chinese Medicine, No. WJYY-XZKT-2023-01 (to YJ); National Natural Science Foundation of China (General Program), No. 82274560 (to YJ); National Natural Science Foundation of China (Youth Program), No. 82405438 (to ZMH); Special Project on Independently Selected Topics of Wangjing Hospital of Chinese Academy of Traditional Chinese Medicine, No. WJYY-ZZXT-2023-24 (to ZMH)

PDF

20

摘要：

文题释义：

颈椎失稳：是由于多种原因造成颈椎的结构功能退化，导致患者在生理状态下椎体节段过度或异常活动，并由此引发颈肩痛、头痛头晕等一系列的临床症状。
深度学习：是机器学习的一个子领域，其核心是通过具有多级非线性变换的层级化架构(称为深度神经网络)对数据进行表征学习。

摘要
背景：早期预测颈椎失稳对颈椎病的防治具有重要意义，深度学习技术可为颈椎失稳的智能预测提供有力支持。
目的：建立基于颈椎核磁图像的颈椎失稳深度学习模型，实现颈椎失稳的早期智能预测。
方法：通过中国中医科学院望京医院脊柱科门诊及社会招募，选择18-45岁的中青年人群，包括颈椎失稳患者和健康对照者。所有受试者均接受颈椎核磁检查，并在横断面图像上手动标注5个关键解剖结构：椎间盘、关节突、椎前肌、颈后部深层和浅层肌群，而后基于原始图像和勾画的感兴趣区域，采用深度学习算法构建颈椎失稳的预测模型，并对模型的预测性能进行评价和验证。

结果与结论：①共纳入308例中青年受试者，包括196例颈椎失稳受试者和112例健康受试者；根据纳入时间不同，受试者数据被分别分配至模型训练集和测试集；②模型在训练集中的受试者工作特征曲线下面积、F1-分数、精确率和召回率分别为0.97，0.98，0.98和0.97，在测试集中分别为0.97，0.95，1.00和0.90；③结果表明，基于颈椎核磁图像构建颈椎失稳的深度学习模型能够对颈椎失稳进行早期智能化预测，且具有较高的预测性能。



中国组织工程研究杂志出版内容重点：人工关节；骨植入物；脊柱；骨折；内固定；数字化骨科；组织工程

关键词: 中青年, 颈椎失稳, 颈椎核磁, 深度学习, 预测模型, 受试者工作特征曲线下面积

Abstract: BACKGROUND: Early prediction of cervical instability is crucial for the prevention and treatment of cervical spondylosis, and deep learning technology can provide robust support for intelligent prediction of cervical instability. 
OBJECTIVE: To develop a deep learning model of cervical instability based on cervical magnetic resonance imaging for early intelligent prediction of cervical instability. 
METHODS: This study recruited young and middle-aged participants (18-45 years), including both cervical instability patients and healthy controls, through the Spine Department Outpatient Clinic of Wangjing Hospital, China Academy of Chinese Medical Sciences, as well as community-based recruitment. All participants underwent cervical magnetic resonance imaging examinations. On the axial magnetic resonance imaging images, five key anatomical structures were manually annotated: intervertebral disc, facet, prevertebral muscle, deep muscle group in the back of the neck, and superficial muscle group in the back of the neck. A deep learning algorithm was then employed to develop a predictive model for cervical instability, utilizing both the original images and the delineated regions of interest. Finally, the model's predictive performance was systematically evaluated and validated. 
RESULTS AND CONCLUSION: (1) The study included a total of 308 young and middle-aged participants, comprising 196 individuals with cervical instability and 112 healthy controls. Based on enrollment time, the subjects' data were allocated to either the model training set or the test set. (2) The model demonstrated high predictive performance, with an area under the curve values of 0.97, an F1-score of 0.98, a precision of 0.98, and a recall of 0.97 in the training set. In the test set, these metrics were 0.97, 0.95, 1.00, and 0.90, respectively. (3) The results indicate that the deep learning model based on cervical magnetic resonance imaging images can effectively enable early intelligent prediction of cervical instability, exhibiting strong diagnostic accuracy.

Key words: young and middle-aged adults, cervical instability, cervical magnetic resonance imaging, deep learning, prediction model, the area under the receiver operating characteristic curve

中图分类号: