Deep-Learning-Based Computer-Aided Grading of Cervical Spinal Stenosis Yields Strong Performance and Clinical Value

Strategic Insights -

Research from the Fifth Affiliated Hospital of Sun Yat-sen University in Zhuhai, People's Republic of China, has made significant strides in the non-invasive grading of cervical spinal stenosis using deep learning convolutional neural network algorithms. The study, which utilized a dataset of 954 patients with cervical spine magnetic resonance imaging (MRI) data, demonstrated the potential of computer-aided diagnostic support in accurately assessing the severity of spinal stenosis. The findings, published in the journal Bioengineering, show that the EfficientNet_B5 model achieved a five-fold cross-validated accuracy of 79.45% and near-perfect agreement with clinician grading on the test set.

Key Takeaways:

  • The study aimed to apply different deep learning convolutional neural network algorithms to assess the grading of cervical spinal stenosis and evaluate their consistency with clinician grading results as well as clinical manifestations of patients.
  • The researchers retrospectively enrolled 954 patients with cervical spine MRI data and medical records from the Fifth Affiliated Hospital of Sun Yat Sen University.
  • The study adopted the Kang grading method for sagittal MR images of the cervical spine and the spinal cord compression ratio for horizontal MR images of the cervical spine for cervical spinal canal stenosis grading.
  • The EfficientNet_B5 model achieved a five-fold cross-validated accuracy of 79.45% and near-perfect agreement with clinician grading on the test set (k=0.848, 0.822), surpassing resident-clinician consistency (k=0.732, 0.702).
  • The model-derived compression ratio (0.45 ± 0.07) did not differ significantly from manual measurements (0.46 ± 0.07).
  • Correlation analysis showed moderate associations between model outputs and clinical symptoms: EfficientNet_B5 grades (r=0.526) were comparable to clinician assessments (r=0.517, 0.503) and higher than those of residents (r=0.457, 0.448).
  • CNN models demonstrate strong performance in the objective, consistent, and efficient grading of cervical spinal stenosis severity, offering potential clinical value in automated diagnostic support.

Statistics:

  • 954 patients were retrospectively enrolled in the study with cervical spine MRI data and medical records from the Fifth Affiliated Hospital of Sun Yat Sen University.
  • The EfficientNet_B5 model achieved a five-fold cross-validated accuracy of 79.45%.
  • The model-derived compression ratio (0.45 ± 0.07) did not differ significantly from manual measurements (0.46 ± 0.07).
  • Correlation analysis showed moderate associations between model outputs and clinical symptoms: EfficientNet_B5 grades (r=0.526) were comparable to clinician assessments (r=0.517, 0.503) and higher than those of residents (r=0.457, 0.448).

Sources:

  • Deep-Learning-Based Computer-Aided Grading of Cervical Spinal Stenosis from MR Images: Accuracy and Clinical Alignment. Bioengineering, 2025, 12(6):604. (Bioengineering - http://www.mdpi.com/journal/bioengineering)
  • NewsRx. New Findings from Fifth Affiliated Hospital of Sun Yat-sen University Describe Advances in Spinal Stenosis (Deep-Learning-Based Computer-Aided Grading of Cervical Spinal Stenosis from MR Images: Accuracy and Clinical Alignment). Health & Medicine Week. July 11, 2025; p 3089.