Yilihamu Elzat Elham-Yilizati, Shang Jun, Su Zhi-Hai, Yang Jin-Tao, Zhao Kun, Zhong Hai, Feng Shi-Qing
Orthopedic Research Center of Shandong University & Advanced Medical Research Institute, Shandong University, Jinan, 250000, China.
Qilu Hospital of Shandong University, Shandong University, Jinan, 250000, China.
Radiol Med. 2025 Mar 24. doi: 10.1007/s11547-025-01996-y.
Application of a deep learning model visualization plugin for rapid and accurate automatic quantification and classification of lumbar disc herniation (LDH) types on axial T2-weighted MRIs.
Retrospective analysis of 2500 patients, with the training set comprising data from 2120 patients (25,554 images), an internal test set covering data from 80 patients (784 images), and an external test set including data from 300 patients (3285 images). To enhance implementation, this study categorized normal and bulging discs as a grade without significant abnormalities, defining the region and severity grades of LDH based on the relationship between the disc and the spinal canal. The automated detection training and validation process employed the YOLOv8 object detection model for target area localization, the YOLOv8-seg segmentation model for disc recognition, and the YOLOv8-pose keypoint detection model for positioning. Finally, the stability of the detection results was verified using metrics such as Intersection over Union (IoU), mean error (ME), precision (P), F1 score (F1), Kappa coefficient (kappa), and 95% confidence interval (95%CI).
The segmentation model achieved an mAP50:95 of 98.12% and an IoU of 98.36% in the training set, while the keypoint detection model achieved an mAP50:95 of 93.58% with a mean error (ME) of 0.208 mm. For the internal and external test sets, the segmentation model's IoU was 97.58 and 97.49%, respectively, while the keypoint model's ME was 0.219 mm and 0.221 mm, respectively. In the quantification validation of the extent of LDH, P, F1, and kappa were measured. For LDH classification (18 categories), the internal and external test sets showed P = 81.21% and 74.50%, F1 = 81.26% and 74.42%, and kappa = 0.75 (95%CI 0.68, 0.82, p = 0.00) and 0.69 (95%CI 0.65, 0.73, p = 0.00), respectively. For the severity grades of LDH (four categories), the internal and external test sets showed P = 92.51% and 90.07%, F1 = 92.36% and 89.66%, and kappa = 0.88 (95%CI 0.80, 0.96, p = 0.00) and 0.85 (95%CI 0.81, 0.89, p = 0.00), respectively. For the regions of LDH (eight categories), the internal and external test sets showed P = 83.34% and 77.87%, F1 = 83.85% and 78.21%, and kappa = 0.77 (95%CI 0.70, 0.85, p = 0.00) and 0.71 (95%CI 0.67, 0.75, p = 0.00), respectively.
The automated aided diagnostic model achieved high performance in detecting and classifying LDH and demonstrated substantial consistency with expert classification.
应用深度学习模型可视化插件，对腰椎间盘突出症（LDH）在轴向T2加权磁共振成像（MRI）上进行快速准确的自动定量分析和类型分类。
对2500例患者进行回顾性分析，训练集包括2120例患者的数据（25554幅图像），内部测试集涵盖80例患者的数据（784幅图像），外部测试集包括300例患者的数据（3285幅图像）。为便于实施，本研究将正常椎间盘和膨出椎间盘归为无明显异常的等级，根据椎间盘与椎管的关系定义LDH的区域和严重程度等级。自动检测训练和验证过程采用YOLOv8目标检测模型进行目标区域定位，YOLOv8-seg分割模型进行椎间盘识别，YOLOv8-pose关键点检测模型进行定位。最后，使用交并比（IoU）、平均误差（ME）、精度（P）、F1分数（F1）、kappa系数（kappa）和95%置信区间（95%CI）等指标验证检测结果的稳定性。
分割模型在训练集中的mAP50:95为98.12%，IoU为98.36%，而关键点检测模型的mAP50:95为93.58%，平均误差（ME）为0.208毫米。对于内部和外部测试集，分割模型的IoU分别为97.58%和97.49%，而关键点模型的ME分别为0.219毫米和0.221毫米。在LDH范围的定量验证中，测量了P、F1和kappa。对于LDH分类（18类），内部和外部测试集的P分别为81.21%和74.50%，F1分别为81.26%和74.42%，kappa分别为0.75（95%CI 0.68, 0.82, p = 0.00）和0.69（95%CI 0.65, 0.73, p = 0.00）。对于LDH的严重程度等级（4类），内部和外部测试集的P分别为92.51%和90.07%，F1分别为92.36%和89.66%，kappa分别为0.88（95%CI 0.80, 0.96, p = 0.00）和0.85（95%CI 0.81, 0.89, p = 0.00）。对于LDH的区域（8类），内部和外部测试集的P分别为83.34%和77.87%，F1分别为83.85%和78.21%，kappa分别为0.77（95%CI 0.70, 0.85, p = 0.00）和0.71（95%CI 0.67, 0.75, p = 0.00）。
自动辅助诊断模型在LDH的检测和分类方面具有高性能，与专家分类具有高度一致性。
