The following is a summary of “Enhanced choroid plexus segmentation with 3D UX-Net and its association with disease progression in multiple sclerosis,” published in the June 2024 issue of Neurology by Wang et al.
A study is being conducted to determine the link between the choroid plexus (CP) and multiple sclerosis (MS) with deep learning (DL) for fast and consistent CP volume measurement.
Researchers conducted a retrospective study to establish a dependable DL model for automatic CP segmentation and assess its past clinical value in patients with MS.
They trained and validated the 3D UX-Net model (3D U-Net for comparison) using T1-weighted MRI data from 216 patients with relapsing-remitting MS (RRMS) and 75 healthy subjects. Dataset1b had 53 patients with RRMS, including baseline and 2-year follow-up scans, while dataset2 had 58 patients with RRMS from multiple centers. The study assessed segmentation performance using the Dice coefficient and correlated CP volume from automatic and manual segmentation with MS clinical outcomes, evaluating disability with the Expanded Disability Status Scale (EDSS) and cognitive function with the Symbol Digit Modalities Test (SDMT).
The results showed that the 3D UX-Net model achieved Dice coefficients of 0.875 ± 0.030 and 0.870 ± 0.044 for CP segmentation on dataset1b and dataset2, respectively, outperforming 3D U-Net’s scores of 0.809 ± 0.098 and 0.601 ± 0.226. The CP volumes segmented by the 3D UX-Net model aligned consistently with clinical outcomes compared to manual segmentation. In dataset1b, both manual and automatic segmentation revealed a significant positive correlation between normalized CP volume (nCPV) and EDSS scores at baseline (manual: r = 0.285, P=0.045; automatic: r = 0.287, P=0.044) and a negative correlation with SDMT scores (manual: r = -0.331, P=0.020; automatic: r = -0.329, P=0.021). In dataset2, similar correlations were found with EDSS scores (manual: r = 0.337, P=0.021; automatic: r = 0.346, P=0.017). Meanwhile, in dataset1b, both manual and automatic segmentation showed a significant increase in nCPV from baseline to follow-up (P<0.05). The rise in nCPV was more pronounced in patients with worsened disability compared to stable patients (manual: P=0.023; automatic: P=0.018). Patients receiving disease-modifying therapy (DMT) exhibited a significantly lower increase in nCPV compared to patients untreated (manual: P=0.004; automatic: P=0.004).
Investigators concluded that DL could effectively segment the CP, suggesting its potential as a tool for MS research. However, further studies are needed to validate its role as a clinical biomarker.
Source: msard-journal.com/article/S2211-0348(24)00327-4/abstract
The post 3D UX-Net Enhanced CP Segmentation and Its Association with MS Disease Progression first appeared on Physician's Weekly.