Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 Jun 28;12(7):709.
doi: 10.3390/bioengineering12070709.

Deep Learning Spinal Cord Segmentation Based on B0 Reference for Diffusion Tensor Imaging Analysis in Cervical Spondylotic Myelopathy

Affiliations

Deep Learning Spinal Cord Segmentation Based on B0 Reference for Diffusion Tensor Imaging Analysis in Cervical Spondylotic Myelopathy

Shuoheng Yang et al. Bioengineering (Basel). .

Abstract

Diffusion Tensor Imaging (DTI) is a crucial imaging technique for accurately assessing pathological changes in Cervical Spondylotic Myelopathy (CSM). However, the segmentation of spinal cord DTI images primarily relies on manual methods, which are labor-intensive and heavily dependent on the subjective experience of clinicians, and existing research on DTI automatic segmentation cannot fully satisfy clinical requirements. Thus, this poses significant challenges for DTI-assisted diagnostic decision-making. This study aimed to deliver AI-driven segmentation for spinal cord DTI. To achieve this goal, a comparison experiment of candidate input features was conducted, with the preliminary results confirming the effectiveness of applying a diffusion-free image (B0 image) for DTI segmentation. Furthermore, a deep-learning-based model, named SCS-Net (Spinal Cord Segmentation Network), was proposed accordingly. The model applies a classical U-shaped architecture with a lightweight feature extraction module, which can effectively alleviate the training data scarcity problem. The proposed method supports eight-region spinal cord segmentation, i.e., the lateral, dorsal, ventral, and gray matter areas on the left and right sides. To evaluate this method, 89 CSM patients from a single center were collected. The model demonstrated satisfactory accuracy for both general segmentation metrics (precision, recall, and Dice coefficient) and a DTI-specific feature index. In particular, the proposed model's error rate for the DTI-specific feature index was evaluated as 5.32%, 10.14%, 7.37%, and 5.70% on the left side, and 4.60%, 9.60%, 8.74%, and 6.27% on the right side of the spinal cord, respectively, affirming the model's consistent performance for radiological rationality. In conclusion, the proposed AI-driven segmentation model significantly reduces the dependence on DTI manual interpretation, providing a feasible solution that can improve potential diagnostic outcomes for patients.

Keywords: cervical spondylotic myelopathy; deep learning; diffusion tensor imaging; medical image segmentation.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflicts of interest.

Figures

Figure A1
Figure A1
The loss of SCS-Net during the training process.
Figure 1
Figure 1
Spinal cord anatomical-level ROI visualization, in the segmentation experiment, the ROI definitions distinguished between the anatomical structures on the left and right sides, including the left lateral column (L_LC), left dorsal column (L_DC), left ventral column (L_VC), left gray matter(L_GM), right lateral column (R_LC), right dorsal column (R_DC), right ventral column (R_VC), and right gray matter (R_GM).
Figure 2
Figure 2
The structure of the proposed segmentation model SCS-Net.
Figure 3
Figure 3
The structure of the feature extraction block of SCS-Net.
Figure 4
Figure 4
The design of the proposed segmentation pipeline, major steps including image registration, segmentation model configuration, and result visualization.
Figure 5
Figure 5
The Dice performance of the different input features for segmentation, ROIs including left lateral column (L_LC), left dorsal column (L_DC), left ventral column (L_VC), left gray matter (L_GM), right lateral column (R_LC), right dorsal column (R_DC), right ventral column (R_VC), right gray matter (R_GM), and Mean result.
Figure 6
Figure 6
The Dice results of segmentation model performance under DTI B0 images.
Figure 7
Figure 7
The segmentation results predicted on the sample DTI slice.
Figure 8
Figure 8
The Visualization of segmentation results. Three samples has been demonstrated. (a) sample input DTI slice, (b) visualization of labeled ground truth, and (c) visualization of segmentation prediction.

References

    1. Al-Shaari H., Fulford J., Heales C. Diffusion tensor imaging within the healthy cervical spinal cord: Within-participants reliability and measurement error. Magn. Reson. Imaging. 2024;109:56–66. doi: 10.1016/j.mri.2024.03.005. - DOI - PubMed
    1. Beaulieu C. The basis of anisotropic water diffusion in the nervous system–a technical review. NMR Biomed. Int. J. Devoted Dev. Appl. Magn. Reson. In Vivo. 2002;15:435–455. doi: 10.1002/nbm.782. - DOI - PubMed
    1. Kara B., Celik A., Karadereler S., Ulusoy L., Ganiyusufoglu K., Onat L., Mutlu A., Ornek I., Sirvanci M., Hamzaoglu A. The role of DTI in early detection of cervical spondylotic myelopathy: A preliminary study with 3-T MRI. Neuroradiology. 2011;53:609–616. doi: 10.1007/s00234-011-0844-4. - DOI - PubMed
    1. Shabani S., Kaushal M., Budde M.D., Wang M.C., Kurpad S.N. Diffusion tensor imaging in cervical spondylotic myelopathy: A review. J. Neurosurg. Spine. 2020;33:65–72. - PubMed
    1. Jin R., Luk K.D., Cheung J.P.Y., Hu Y. Prognosis of cervical myelopathy based on diffusion tensor imaging with artificial intelligence methods. NMR Biomed. 2019;32:e4114. doi: 10.1002/nbm.4114. - DOI - PubMed

LinkOut - more resources