Search

Objectives: To implement a machine learning model using radiomic features extracted from quantitative susceptibility mapping (QSM) in discriminating multiple sclerosis (MS) from neuromyelitis optica spectrum disorder (NMOSD). Materials and Methods: Forty-seven patients with MS (mean age = 40.00 ± 13.72 years) and 36 patients with NMOSD (mean age = 42.14 ± 12.34 years) who underwent enhanced gradient-echo T2*-weighted angiography (ESWAN) sequence in 3.0-T MRI were included between April 2017 and October 2019. QSM images were reconstructed from ESWAN, and QSM-derived radiomic features were obtained from seven regions of interest (ROIs), including bilateral putamen, globus pallidus, head of the caudate nucleus, thalamus, substantia nigra, red nucleus, and dentate nucleus. A machine learning model (logistic regression) was applied to classify MS and NMOSD, which combined radiomic signatures and demographic information to assess the classification accuracy using the area under the receiver operating characteristic (ROC) curve (AUC). Results: The radiomics-only models showed better discrimination performance in almost all deep gray matter (DGM) regions than the demographic information-only model, with the highest AUC in DN of 0.902 (95% CI: 0.840–0.955). Moreover, the hybrid model combining radiomic signatures and demographic information showed the highest discrimination performance which achieved the AUC of 0.927 (95% CI: 0.871–0.984) with fivefold cross-validation. Conclusion: The hybrid model based on QSM and powered with machine learning has the potential to discriminate MS from NMOSD. [ABSTRACT FROM AUTHOR]

• This is a combined cross-sectional and longitudinal study reported about the difference in microstructural integrity between patients with NMOSD and MS. • In this study, we report the detail about the extent of white matter integrity damage, and we found the extend of white matter integrity damage was differ between NMOSD and MS at cross-sectional result. • The pattern of longitudinal white matter microstructure damage after an average time interval of approximately one year was different between NMOSD and MS, and the result showed that the progressive change in brain microstructural properties is observed in patients with MS but may not in patients with NMOSD. Neuromyelitis optica spectrum disorder (NMOSD) may sometimes be misdiagnosed as multiple sclerosis (MS) because both disorders have similar clinical presentations and commonly show white matter damage in the brain. Diffusion tensor imaging (DTI) is an advanced MRI technique to assess the microstructural organization of white matter and provides greater pathological specificity than conventional MRI. In the present combined cross-sectional and longitudinal study, the novel DTI technique of Track-Based Spatial Statistics (TBSS) was used to investigate the difference of DTI parameter abnormalities between NMOSD and MS. A total of 42 patients with NMOSD, 51 patients with MS and 56 health controls (HC) were recruited and of these 14 patients with NMOSD and 13 patients with MS were also studied at follow-up after an average interval of approximately one year. Measurements of fractional anisotropy (FA), mean diffusion (MD), axial diffusivity (AD) and radial diffusivity (RD) were compared at baseline and follow-up in patients with NMOSD and MS. Significant reduction in FA, increase in MD, AD and RD were observed in patients with MS (p < 0.05) and reduced FA was shown in NMOSD (p < 0.05) compared to HC, with all the effects, together with lesion load on T1WI and T2WI, being greater in patients with MS than in patients with NMOSD (p < 0.05). There was no significant difference in the time interval to follow-up in patients with MS (1.37 years) and NMOSD (1.25 years) (p > 0.05), during which there were significant changes in EDSS score between baseline and follow-up in NMOSD and MS patients (p < 0.05). There was a significantly reduced FA, and increased MD and RD in patients with MS (p < 0.05), but no significant changes in patients with NMOSD (p > 0.05). Both MS and NMOSD have microstructure damage in white matter, while the progressive change in brain microstructural properties is observed in patients with MS but may not in patients with NMOSD in a short-term follow-up. [ABSTRACT FROM AUTHOR]