Objectives
Spinocerebellar ataxia type 3 is a disorder within the brain network. However, the relationship between the brain network and disease severity is still unclear. This study aims to investigate changes in the white matter (WM) structural motor network, both in preclinical and ataxic stages, and its relationship with disease severity.
Methods
For this study, 20 ataxic, 20 preclinical SCA3 patients, and 20 healthy controls were recruited and received MRI scans. Disease severity was quantified using the SARA and ICARS scores. The WM motor structural network was created using probabilistic fiber tracking and was analyzed using graph theory and network‐based statistics at global, nodal, and edge levels. In addition, the correlations between network topological measures and disease duration or clinical scores were analyzed.
Results
Preclinical patients showed increasing assortativity of the motor network, altered subnetwork including 12 edges of 11 nodes, and 5 brain regions presenting reduced nodal strength. In ataxic patients assortativity of the motor network also increased, but global efficiency, global strength, and transitivity decreased. Ataxic patients showed a wider altered subnetwork and a higher number of reduced nodal strengths. A negative correlation between the transitivity of the motor network and SARA and ICARS scores was observed in ataxic patients.
Interpretation
Changes to the WM motor network in SCA3 start before ataxia onset, and WM motor network involvement increases with disease progression. Global network topological measures of the WM motor network appear to be a promising image biomarker for disease severity. This study provides new insights into the pathophysiology of disease in SCA3/MJD.
The analysis of structural covariance has emerged as a powerful tool to explore the morphometric correlations among broadly distributed brain regions. However, little is known about the interactions between the damaged primary motor cortex (M1) and other brain regions in stroke patients with motor deficits. This study is aimed at investigating the structural covariance pattern of the ipsilesional M1 in chronic subcortical stroke patients with motor deficits. High-resolution T1-weighted brain images were acquired from 58 chronic subcortical stroke patients with motor deficits (29 with left-sided lesions and 29 with right-sided lesions) and 50 healthy controls. Structural covariance patterns were identified by a seed-based structural covariance method based on gray matter (GM) volume. Group comparisons between stroke patients (left-sided or right-sided groups) and healthy controls were determined by a permutation test. The association between alterations in the regional GM volume and motor recovery after stroke was investigated by a multivariate regression approach. Structural covariance analysis revealed an extensive increase in the structural interactions between the ipsilesional M1 and other brain regions in stroke patients, involving not only motor-related brain regions but also non-motor-related brain regions. We also identified a slightly different pattern of structural covariance between the left-sided stroke group and the right-sided stroke group, thus indicating a lesion-side effect of cortical reorganization after stroke. Moreover, alterations in the GM volume of structural covariance brain regions were significantly correlated to the motor function scores in stroke patients. These findings indicated that the structural covariance patterns of the ipsilesional M1 in chronic subcortical stroke patients were induced by motor-related plasticity. Our findings may help us to better understand the neurobiological mechanisms of motor impairment and recovery in patients with subcortical stroke from different perspectives.
ObjectiveTo develop a machine learning (ML)-based classifier for discriminating between low-grade (ISUP I-II) and high-grade (ISUP III-IV) clear cell renal cell carcinomas (ccRCCs) using MRI textures.Materials and MethodsWe retrospectively evaluated a total of 99 patients (with 61 low-grade and 38 high-grade ccRCCs), who were randomly divided into a training set (n = 70) and a validation set (n = 29). Regions of interest (ROIs) of all tumors were manually drawn three times by a radiologist at the maximum lesion level of the cross-sectional CMP sequence images. The quantitative texture analysis software, MaZda, was used to extract texture features, including histograms, co-occurrence matrixes, run-length matrixes, gradient models, and autoregressive models. Reproducibility of the texture features was assessed with the intra-class correlation coefficient (ICC). Features were chosen based on their importance coefficients in a random forest model, while the multi-layer perceptron algorithm was used to build a classifier on the training set, which was later evaluated with the validation set.ResultsThe ICCs of 257 texture features were equal to or higher than 0.80 (0.828–0.998. Six features, namely Kurtosis, 135dr_RLNonUni, Horzl_GLevNonU, 135dr_GLevNonU, S(4,4)Entropy, and S(0,5)SumEntrp, were chosen to develop the multi-layer perceptron classifier. A three-layer perceptron model, which has 229 nodes in the hidden layer, was trained on the training set. The accuracy of the model was 95.7% with the training set and 86.2% with the validation set. The areas under the receiver operating curves were 0.997 and 0.758 for the training and validation sets, respectively.ConclusionsA machine learning-based grading model was developed that can aid in the clinical diagnosis of clear cell renal cell carcinoma using MRI images.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.