Scholarlink

Purpose <p>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.</p> Methods <p>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).</p> Results <p>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&#xa0;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&#xa0;mm and 0.221&#xa0;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 <i>P</i> = 81.21% and 74.50%, F1 = 81.26% and 74.42%, and kappa = 0.75 (95%CI 0.68, 0.82, <i>p</i> = 0.00) and 0.69 (95%CI 0.65, 0.73, <i>p</i> = 0.00), respectively. For the severity grades of LDH (four categories), the internal and external test sets showed <i>P</i> = 92.51% and 90.07%, F1 = 92.36% and 89.66%, and kappa = 0.88 (95%CI 0.80, 0.96, <i>p</i> = 0.00) and 0.85 (95%CI 0.81, 0.89, <i>p</i> = 0.00), respectively. For the regions of LDH (eight categories), the internal and external test sets showed <i>P</i> = 83.34% and 77.87%, F1 = 83.85% and 78.21%, and kappa = 0.77 (95%CI 0.70, 0.85, <i>p</i> = 0.00) and 0.71 (95%CI 0.67, 0.75, <i>p</i> = 0.00), respectively.</p> Conclusion <p>The automated aided diagnostic model achieved high performance in detecting and classifying LDH and demonstrated substantial consistency with expert classification.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Quantification and classification of lumbar disc herniation on axial magnetic resonance images using deep learning models

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年3月24日

查看全文
  • Elzat Elham-Yilizati Yilihamu,
  • Jun Shang,
  • Zhi-Hai Su,
  • Jin-Tao Yang,
  • Kun Zhao,
  • Hai Zhong,
  • Shi-Qing Feng

摘要

Purpose

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.

Methods

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).

Results

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.

Conclusion

The automated aided diagnostic model achieved high performance in detecting and classifying LDH and demonstrated substantial consistency with expert classification.