Deep learning-driven MRI trigeminal nerve segmentation with SEVB-net

Abstract

Purpose: Trigeminal neuralgia (TN) poses significant challenges in its diagnosis and treatment due to its extreme pain. Magnetic resonance imaging (MRI) plays a crucial role in diagnosing TN and understanding its pathogenesis. Manual delineation of the trigeminal nerve in volumetric images is time-consuming and subjective. This study introduces a Squeeze and Excitation with BottleNeck V-Net (SEVB-Net), a novel approach for the automatic segmentation of the trigeminal nerve in three-dimensional T2 MRI volumes.

Methods: We enrolled 88 patients with trigeminal neuralgia and 99 healthy volunteers, dividing them into training and testing groups. The SEVB-Net was designed for end-to-end training, taking three-dimensional T2 images as input and producing a segmentation volume of the same size. We assessed the performance of the basic V-Net, nnUNet, and SEVB-Net models by calculating the Dice similarity coefficient (DSC), sensitivity, precision, and network complexity. Additionally, we used the Mann-Whitney U test to compare the time required for manual segmentation and automatic segmentation with manual modification.

Results: In the testing group, the experimental results demonstrated that the proposed method achieved state-of-the-art performance. SEVB-Net combined with the ωDoubleLoss loss function achieved a DSC ranging from 0.6070 to 0.7923. SEVB-Net combined with the ωDoubleLoss method and nnUNet combined with the DoubleLoss method, achieved DSC, sensitivity, and precision values exceeding 0.7. However, SEVB-Net significantly reduced the number of parameters (2.20 M), memory consumption (11.41 MB), and model size (17.02 MB), resulting in improved computation and forward time compared with nnUNet. The difference in average time between manual segmentation and automatic segmentation with manual modification for both radiologists was statistically significant (p < 0.001).

Conclusion: The experimental results demonstrate that the proposed method can automatically segment the root and three main branches of the trigeminal nerve in three-dimensional T2 images. SEVB-Net, compared with the basic V-Net model, showed improved segmentation performance and achieved a level similar to nnUNet. The segmentation volumes of both SEVB-Net and nnUNet aligned with expert annotations but SEVB-Net displayed a more lightweight feature.

Keywords: automatic segmentation; deep learning; magnetic resonance imaging; trigeminal nerve; trigeminal neuralgia.

Figure 1

MRI imaging of the trigeminal nerve. (A) A three-dimensional T2WI-CUBE fs reconstruction image displays the ocular nerve (white arrow) and maxillary nerve (red arrow). (B) A three-dimensional T2WI-CUBE fs reconstruction image shows the branches of the mandibular nerve, inferior alveolar nerve (green arrows), and lingual nerve (yellow arrows). (C) A three-dimensional T2WI-CUBE fs reconstruction image provides a comprehensive view of the mandibular branch of the trigeminal nerve, including the trigeminal nerve itself (orange arrows), Meckel cavity (purple arrows), and mandibular nerve (blue arrows).

Figure 2

Pre-processing steps applied to each MRI case.

Figure 3

Schematic representation of our network architecture.

Figure 4

An overview of the multi-resolution network.

Figure 5

An example of morphological processing. Case 1: morphological processing of the original manual segmentation. Case 2: morphological processing after cropping and resampling of manual segmentation.

Figure 6

Convergence curve of SEVB-Net combined with the ωDoubleLoss method.

Figure 7

Comparative analysis of the Dice similarity coefficient (DSC), sensitivity, and precision of the three experiments on the testing set.

Figure 8

Visualization of the segmentation results.

Figure 9

Comparison of the boundaries of segmentation results.