Support Vector Machine Classification of Obsessive-Compulsive Disorder Based on Whole-Brain Volumetry and Diffusion Tensor Imaging.
ABSTRACT: Magnetic resonance imaging (MRI) methods have been used to detect cerebral anatomical distinction between obsessive-compulsive disorder (OCD) patients and healthy controls (HC). Machine learning approach allows for the possibility of discriminating patients on the individual level. However, few studies have used this automatic technique based on multiple modalities to identify potential biomarkers of OCD. High-resolution structural MRI and diffusion tensor imaging (DTI) data were acquired from 48 OCD patients and 45 well-matched HC. Gray matter volume (GMV), white matter volume (WMV), fractional anisotropy (FA), and mean diffusivity (MD) were extracted as four features were examined using support vector machine (SVM). Ten brain regions of each feature contributed most to the classification were also estimated. Using different algorithms, the classifier achieved accuracies of 72.08, 61.29, 80.65, and 77.42% for GMV, WMV, FA, and MD, respectively. The most discriminative gray matter regions that contributed to the classification were mainly distributed in the orbitofronto-striatal "affective" circuit, the dorsolateral, prefronto-striatal "executive" circuit and the cerebellum. For WMV feature and the two feature sets of DTI, the shared regions contributed the most to the discrimination mainly included the uncinate fasciculus, the cingulum in the hippocampus, corticospinal tract, as well as cerebellar peduncle. Based on whole-brain volumetry and DTI images, SVM algorithm revealed high accuracies for distinguishing OCD patients from healthy subjects at the individual level. Computer-aided method is capable of providing accurate diagnostic information and might provide a new perspective for clinical diagnosis of OCD.
Project description:Structural abnormalities in schizophrenia (SZ) patients have been well documented with structural magnetic resonance imaging (MRI) data using voxel-based morphometry (VBM) and region of interest (ROI) analyses. However, these analyses can only detect group-wise differences and thus, have a poor predictive value for individuals. In the present study, we applied a machine learning method that combined support vector machine (SVM) with recursive feature elimination (RFE) to discriminate SZ patients from normal controls (NCs) using their structural MRI data. We first employed both VBM and ROI analyses to compare gray matter volume (GMV) and white matter volume (WMV) between 41 SZ patients and 42 age- and sex-matched NCs. The method of SVM combined with RFE was used to discriminate SZ patients from NCs using significant between-group differences in both GMV and WMV as input features. We found that SZ patients showed GM and WM abnormalities in several brain structures primarily involved in the emotion, memory, and visual systems. An SVM with a RFE classifier using the significant structural abnormalities identified by the VBM analysis as input features achieved the best performance (an accuracy of 88.4%, a sensitivity of 91.9%, and a specificity of 84.4%) in the discriminative analyses of SZ patients. These results suggested that distinct neuroanatomical profiles associated with SZ patients might provide a potential biomarker for disease diagnosis, and machine-learning methods can reveal neurobiological mechanisms in psychiatric diseases.
Project description:INTRODUCTION:Studies using voxel-based morphometry report variable and inconsistent abnormalities of gray matter volume (GMV) and white matter volume (WMV) in brains of preterm-born adolescents (PBA). In such circumstances a meta-analysis can help identify the most prominent and consistent abnormalities. METHOD:We identified 9 eligible studies by systematic search of the literature up to October 2017. We used Seed-based d Mapping to analyze GMV and WMV alterations between PBA and healthy controls. RESULTS:In the GMV meta-analysis, PBA compared to healthy controls showed: increased GMV in left cuneus cortex, left superior frontal gyrus, and right anterior cingulate cortex; decreased GMV in bilateral inferior temporal gyrus (ITG), left superior frontal gyrus, and right caudate nucleus. In the WMV meta-analysis, PBA showed: increased WMV in right fusiform gyrus and precuneus; decreased WMV in bilateral ITG, and right inferior frontal gyrus. In meta-regression analysis, the percentage of male PBA negatively correlated with decreased GMV of bilateral ITG. INTERPRETATION:PBA show widespread GMV and WMV alterations in the default mode network, visual recognition network, and salience network. These changes may be causally relevant to socialization difficulties and cognitive impairments. The meta-regression results perhaps reveal the structural underpinning of the cognition-related sex differences in PBA.
Project description:Growth hormone (GH) and its anabolic mediator, insulin-like growth factor-1 (IGF-1), have a critical role in the central nervous system. However, their detailed roles in the adult human brain are not clear. In this study, structural MRIs of 48 patients with GH-secreting pituitary adenoma (GH-PA), 48 sex- and age-matched clinical Non-Functional pituitary adenoma patients (NonFun-PA) and healthy controls (HCs) were assessed using voxel-based morphometry (VBM) and region-based morphometry (RBM). Correlation analyses helped determine the relationships between serum hormone levels and brain structure. The whole-brain gray matter volume (GMV) and white matter volume (WMV) significantly increased at the expense of cerebrospinal fluid volume (CSFV) in GH-PA (Bonferroni corrected, p<0.01). The increase in GMV and reduction in CSFV were significantly correlated with serum GH/IGF-1 levels (p<0.05). VBM showed significant correlations of the GMV/WMV alteration pattern between GH-PA vs HCs and GH-PA vs NonFun-PA and widespread bilateral clusters of significantly increased GMV and WMV in GH-PA (pFDR<0.05). RBM showed obviously increased GMV/WMV in 54 of 68 brain regions (p<0.05) in GH-PA compared to HCs. Our results provide imaging evidence that serum GH/IGF-1 contributes to brain growth, which may be a potential treatment option for neurodegenerative disorders and brain injury in humans.
Project description:We examined if cerebral volume reduction occurs very early during the course of systemic lupus erythematosus (SLE), and observed prospectively whether gray (GMV) and white matter volumes (WMV) of the brain would improve with lowered SLE disease activity. T1-weighted MRI brain images were obtained from 14 healthy controls (HC) and 14 newly-diagnosed SLE patients within 5 months of diagnosis (S1) and after achieving low disease activity (S2). Whole brain voxel-based morphometry was used to detect differences in the GMV and WMV between SLE patients and HC and those between SLE patients at S1 and S2. SLE patients were found to have lower GMV than HC in the middle cingulate cortex, middle frontal gyrus and right supplementary motor area, and lower WMV in the superior longitudinal fasciculus, cingulum cingulate gyrus and inferior fronto-occipital fasciculus at both S1 and S2. Whole-brain voxel-wise analysis revealed increased GMV chiefly in the prefrontal regions at S2 compared to S1 in SLE patients. The GMV increase in the left superior frontal gyrus was significantly associated with lowered SLE disease activity. In conclusion, GMV and WMV reduced very early in SLE patients. Reduction of SLE disease activity was accompanied by region-specific GMV improvement in the prefrontal regions.
Project description:The aim of the study was to find structural brain changes in systemic lupus erythematosus patients without major neuropsychiatric manifestations [non-neuropsychiatric systemic lupus erythematosus (non-NPSLE)] using quantitative magnetic resonance imaging (MRI) and possible associations with clinical characteristics. 89 non-NPSLE patients with normal conventional MRI and 84 healthy controls (HCs) were recruited. The whole brain gray matter volume (GMV) and white matter volume (WMV) were calculated for each individual. We found obvious GMV and WMV reduction in the systemic lupus erythematosus (SLE) group compared with HCs. Female patients showed significant reduction of GMV and WMV compared with male patients. Patients treated with immunosuppressive agents (ISA) showed less WMV reduction than those without. Cognitive impairment was the most common subclinical neuropsychiatric manifestation and had a prevalence of 46.1%. Association between WMV reduction with cognitive impairment was found. Thus, we concluded that structural brain atrophy could happen even before occurrence of obvious neuropsychiatric signs and symptoms and was associated with subclinical symptoms such as cognitive impairment. ISA treatment might have a protective effect on the brain atrophy. Early treatment might prevent the progressive damage to the brain. More studies are needed to fully understand the complicated underlying mechanisms of brain atrophy in SLE.
Project description:Longitudinal neuroimaging investigation of fragile X syndrome (FXS), the most common cause of inherited intellectual disability and autism, provides an opportunity to study the influence of a specific genetic factor on neurodevelopment in the living human brain. We examined voxel-wise gray and white matter volumes (GMV, WMV) over a 2-year period in 1- to 3-year-old boys with FXS (n = 41) and compared these findings to age- and developmentally matched controls (n = 28). We found enlarged GMV in the caudate, thalamus, and fusiform gyri and reduced GMV in the cerebellar vermis in FXS at both timepoints, suggesting early, possibly prenatal, genetically mediated alterations in neurodevelopment. In contrast, regions in which initial GMV was similar, followed by an altered growth trajectory leading to increased size in FXS, such as the orbital gyri, basal forebrain, and thalamus, suggests delayed or otherwise disrupted synaptic pruning occurring postnatally. WMV of striatal-prefrontal regions was greater in FXS compared with controls, and group differences became more exaggerated over time, indicating the possibility that such WM abnormalities are the result of primary FMRP-deficiency-related axonal pathology, as opposed to secondary connectional dysregulation between morphologically atypical brain structures. Our results indicate that structural abnormalities of different brain regions in FXS evolve differently over time reflecting time-dependent effects of FMRP deficiency and provide insight into their neuropathologic underpinnings. The creation of an early and accurate human brain phenotype for FXS in humans will significantly improve our capability to detect whether new disease-specific treatments can "rescue" the FXS phenotype in affected individuals.
Project description:To compare the sample size requirements for a neuroprotection trial with change in cerebral gray matter volume (GMV), white matter volume (WMV) or whole brain parenchymal volume (BPV) as outcome measures in patients with relapsing-remitting multiple sclerosis (RRMS).Two datasets with longitudinal MRI measures of untreated patients with RRMS (n = 116 and n = 26) and one dataset of treated patients with RRMS (n = 109) were investigated. In each dataset, normalised GMV, normalised WMV and normalised BPV were analysed using a random intercepts and slopes model to estimate the variance components and per cent change. The required sample size to observe a 33%, 50% and 90% reduction in the per cent change was calculated for each dataset using both a constant per cent change for each measurement and the estimated per cent change for each dataset.The per cent change was greatest in GMV but all variance components were smallest in BPV. Using the estimated per cent change, the sample size required in the untreated cohorts was similar for GMV and BPV, and both were lower than WMV. In the treated cohort, the sample size for GMV was the smallest of all measures. Including additional scans reduced the sample size but increasing the length of the trial and clustering scans led to greater reductions.Cerebral GMV may be a viable outcome measure for clinical trials investigating neuroprotection in RRMS patients, especially considering that the treatment effect may be larger on GMV compared with BPV. However, GMV was somewhat limited by increased variability versus BPV.
Project description:Evidence shows that there are reductions in gray matter volume (GMV) and changes in long association white matter fibres within the left insula-temporoparietal junction (TPJ) during the early stages of psychotic disorders but less is known about short association fibres (sAFs). In this study we sought to characterise the changes in sAFs and associated volumetric changes of the left insula-TPJ during the early stages of psychosis. Magnetic resonance imaging was obtained from a sample of young people with psychosis (n = 42) and healthy controls (n = 45), and cortical parcellations of the left insula-TPJ were used as seeding masks to reconstruct 13 sAFs. Compared to healthy counterparts, the psychosis group showed significant reductions in fractional anisotropy (FA) in the sAFs connecting the superior (STG) and middle temporal gyri (MTG) and as well as reduced GMV within the inferior temporal gyrus and increased white matter volume (WMV) within Heschl's gyrus (HG). Furthermore, adolescent-onset psychosis subjects (onset 18 year or earlier) showed FA reductions in the STG-HG sAF when compared to adult-onset subjects, but this was not associated with changes in GMV nor WMV of the STG or HG. These findings suggest that during the early stages of psychosis, changes in sAFs and associated cortical GMV and WMV appear to occur independently, however age of onset of a psychotic syndrome/disorder influences the pattern of neuroanatomical abnormalities.
Project description:Although, acquired amusia is a common deficit following stroke, relatively little is still known about its precise neural basis, let alone to its recovery. Recently, we performed a voxel-based lesion-symptom mapping (VLSM) and morphometry (VBM) study which revealed a right lateralized lesion pattern, and longitudinal gray matter volume (GMV) and white matter volume (WMV) changes that were specifically associated with acquired amusia after stroke. In the present study, using a larger sample of stroke patients (N = 90), we aimed to replicate and extend the previous structural findings as well as to determine the lesion patterns and volumetric changes associated with amusia recovery. Structural MRIs were acquired at acute and 6-month post-stroke stages. Music perception was behaviorally assessed at acute and 3-month post-stroke stages using the Scale and Rhythm subtests of the Montreal Battery of Evaluation of Amusia (MBEA). Using these scores, the patients were classified as non-amusic, recovered amusic, and non-recovered amusic. The results of the acute stage VLSM analyses and the longitudinal VBM analyses converged to show that more severe and persistent (non-recovered) amusia was associated with an extensive pattern of lesions and GMV/WMV decrease in right temporal, frontal, parietal, striatal, and limbic areas. In contrast, less severe and transient (recovered) amusia was linked to lesions specifically in left inferior frontal gyrus as well as to a GMV decrease in right parietal areas. Separate continuous analyses of MBEA Scale and Rhythm scores showed extensively overlapping lesion pattern in right temporal, frontal, and subcortical structures as well as in the right insula. Interestingly, the recovered pitch amusia was related to smaller GMV decreases in the temporoparietal junction whereas the recovered rhythm amusia was associated to smaller GMV decreases in the inferior temporal pole. Overall, the results provide a more comprehensive picture of the lesions and longitudinal structural changes associated with different recovery trajectories of acquired amusia.
Project description:OBJECTIVE: Metachromatic leukodystrophy (MLD) is an inherited lysosomal disorder due to a deficiency in arylsulfatase A with progressive demyelination and neurological decline. This retrospective MRI study investigated the extent of cortical involvement at time of diagnosis, and clinical correlates to both conventional and regional volumetric measures of brain involvement. METHODS: 3D-T1-weighted MRI scans were used to determine cortical thickness and surface-based cerebral cortical gray matter (GM) and cerebral white matter (WM) volume (GMV and WMV), WM lesions, thalamus, and cerebellum. MRI-MLD severity scores were obtained from FLAIR images. Associations between clinical and imaging data were examined using correlation coefficients. RESULTS: Twenty patients with MLD (mean age 13.7 years, range 2-35) and 20 controls (mean age 13.9 years, range 2-40) were included. Compared with control subjects, late-infantile, and juvenile patients (n = 14) had significantly diminished cerebral cortical GMV and thalamus volume (P < 0.05), but did not differ in WMV and cortical thickness. Adult patients (n = 6) showed significantly reduced GMV, WMV and cortical thickness (all P < 0.05). Regional analysis showed statistically significant cortical thinning in the cingulate gyrus and most pronounced thinning with age in the frontal lobe of MLD patients. Intelligence quotient (IQ) correlated with MRI-MLD scores (r = -0.87, P < 0.001). INTERPRETATION: Significant cerebral cortical GMV loss is already present in early stages of MLD. IQ correlates with WM severity scores and lesion volume, but not with volumetric measures. In adult presentations, there is more pronounced global atrophy with GMV and WMV loss and accelerated cortical thinning, most prominently in the cingulate gyrus and frontal lobes.