Deep Learning-Based Automated Magnetic Resonance Image Segmentation of the Lumbar Structure and Its Adjacent Structures at the L4/5 Level

Abstract

(1) Background: This study aims to develop a deep learning model based on a 3D Deeplab V3+ network to automatically segment multiple structures from magnetic resonance (MR) images at the L4/5 level. (2) Methods: After data preprocessing, the modified 3D Deeplab V3+ network of the deep learning model was used for the automatic segmentation of multiple structures from MR images at the L4/5 level. We performed five-fold cross-validation to evaluate the performance of the deep learning model. Subsequently, the Dice Similarity Coefficient (DSC), precision, and recall were also used to assess the deep learning model’s performance. Pearson’s correlation coefficient analysis and the Wilcoxon signed-rank test were employed to compare the morphometric measurements of 3D reconstruction models generated by manual and automatic segmentation. (3) Results: The deep learning model obtained an overall average DSC of 0.886, an average precision of 0.899, and an average recall of 0.881 on the test sets. Furthermore, all morphometry-related measurements of 3D reconstruction models revealed no significant difference between ground truth and automatic segmentation. Strong linear relationships and correlations were also obtained in the morphometry-related measurements of 3D reconstruction models between ground truth and automated segmentation. (4) Conclusions: We found it feasible to perform automated segmentation of multiple structures from MR images, which would facilitate lumbar surgical evaluation by establishing 3D reconstruction models at the L4/5 level.

One-hole split endoscope versus unilateral biportal endoscopy for lumbar spinal stenosis: a retrospective propensity score study

“…All imaging findings were measured three times by three independent investigators and were averaged. In addition, we used previously reported computer vision and mathematical modelling methods to automatically measure the imaging data [ 13 – 15 ], and compared them with the manual measurement methods described above to verify the reliability of the data.

Fig.

…”

Accurate Intervertebral Disc Segmentation Approach Based on Deep Learning

“…Several deep learning models have emerged in the scientific literature, focusing on the segmentation of MRI images depicting intervertebral discs [20][21][22][23][24][25][26][27][28][29]. For instance, Wang et al [20] introduced a convolutional architecture based on the 3D U-Net, designed for the segmentation of 66 intervertebral discs within multimodal MRI images.…”

“…Their results show that the segmentation masks and associated metrics exhibited high similarity between human-and computer-generated methods, with Dice coefficients of 0.77. Wang et al [26] proposed a deep learning model based on a 3D Deeplab V3+ network to automatically segment multiple structures from MRI images at the L4/5 level. The deep learning model obtained an average precision of 89.9% from a total of 50 participants who had undergone a 3T MRI with T2-3D-space sequences.…”