文献类型：期刊文献

英文题名：Research on a deep learning-based model for measurement of X-ray imaging parameters of atlantoaxial joint

作者：Wu, Yuhua[1,2];Zheng, Yuwen[3];Zhu, Jinping[4];Chen, Xiaofei[5,6];Dong, Fuwen[5,6];He, Linyang[7];Zhu, Jinyang[7];Cheng, Guohua[7];Wang, Ping[8];Zhou, Sheng[8]

第一作者：吴玉泓;Wu, Yuhua

通信作者：Wu, YH[1];Wu, YH[2];Zhou, S[3]

机构：[1]Gansu Univ Tradit Chinese Med, Lanzhou, Peoples R China;[2]Xian Hosp Tradit Chinese Med, Xian, Peoples R China;[3]Zigong First Peoples Hosp, Zigong, Peoples R China;[4]Xian Daxing Hosp, Xian, Peoples R China;[5]Gansu Prov Hosp TCM, Lanzhou, Peoples R China;[6]Gansu Univ Tradit Chinese Med, Affiliated Hosp 1, Lanzhou, Peoples R China;[7]Hangzhou Jianpei Technol Co Ltd, Hangzhou, Peoples R China;[8]Gansu Prov Hosp, Lanzhou, Peoples R China

第一机构：甘肃中医药大学

通信机构：[1]corresponding author), Gansu Univ Tradit Chinese Med, Lanzhou, Peoples R China;[2]corresponding author), Xian Hosp Tradit Chinese Med, Xian, Peoples R China;[3]corresponding author), Gansu Prov Hosp, Lanzhou, Peoples R China.|[10735]甘肃中医药大学;

年份：2025

外文期刊名：EUROPEAN SPINE JOURNAL

收录：;Scopus(收录号：2-s2.0-105010499951);WOS:【SCI-EXPANDED(收录号:WOS:001527688400001)】；

基金：We sincerely appreciate Hangzhou Jianpei Technology Company Ltd. for providing us with the excellent technical support needed to establish the algorithm.

语种：英文

外文关键词：Deep learning; Atlantoaxial subluxation; Cervical open-mouth view radiographs; Imaging parameters; Automatic measurement

摘要：PurposeTo construct a deep learning-based SCNet model, in order to automatically measure X-ray imaging parameters related to atlantoaxial subluxation (AAS) in cervical open-mouth view radiographs, and the accuracy and reliability of the model were evaluated.MethodsA total of 1973 cervical open-mouth view radiographs were collected from picture archiving and communication system (PACS) of two hospitals(Hospitals A and B). Among them, 365 images of Hospital A were randomly selected as the internal test dataset for evaluating the model's performance, and the remaining 1364 images of Hospital A were used as the training dataset and validation dataset for constructing the model and tuning the model hyperparameters, respectively. The 244 images of Hospital B were used as an external test dataset to evaluate the robustness and generalizability of our model. The model identified and marked landmarks in the images for the parameters of the lateral atlanto-dental space (LADS), atlas lateral mass inclination (ALI), lateral mass width (LW), axis spinous process deviation distance (ASDD). The measured results of landmarks on the internal test dataset and external test dataset were compared with the mean values of manual measurement by three radiologists as the reference standard. Percentage of correct key-points (PCK), intra-class correlation coefficient (ICC), mean absolute error (MAE), Pearson correlation coefficient (r), mean square error (MSE), root mean square error (RMSE) and Bland-Altman plot were used to evaluate the performance of the SCNet model.Results(1) Within the 2 mm distance threshold, the PCK of the SCNet model predicted landmarks in internal test dataset images was 98.6-99.7%, and the PCK in the external test dataset images was 98-100%. (2) In the internal test dataset, for the parameters LADS, ALI, LW, and ASDD, there were strong correlation and consistency between the SCNet model predictions and the manual measurements (ICC = 0.80-0.96, r = 0.86-0.96, MAE = 0.47-2.39 mm/degrees, MSE = 0.38-8.55 mm2/degrees 2, RMSE = 0.62-2.92 mm/degrees). (3) The same four parameters also showed strong correlation and consistency between SCNet and manual measurements in the external test dataset (ICC = 0.81-0.91, r = 0.82-0.91, MAE = 0.46-2.29 mm/degrees, MSE = 0.29-8.23mm2/degrees 2, RMSE = 0.54-2.87 mm/degrees).ConclusionThe SCNet model constructed based on deep learning algorithm in this study can accurately identify atlantoaxial vertebral landmarks in cervical open-mouth view radiographs and automatically measure the AAS-related imaging parameters. Furthermore, the independent external test set demonstrates that the model exhibits a certain degree of robustness and generalization capability under meet radiographic standards.