There are growing numbers of studies using machine learning approaches to characterize patterns of anatomical difference discernible from neuroimaging data. The high-dimensionality of image data ...often raises a concern that feature selection is needed to obtain optimal accuracy. Among previous studies, mostly using fixed sample sizes, some show greater predictive accuracies with feature selection, whereas others do not. In this study, we compared four common feature selection methods. 1) Pre-selected region of interests (ROIs) that are based on prior knowledge. 2) Univariate t-test filtering. 3) Recursive feature elimination (RFE), and 4) t-test filtering constrained by ROIs. The predictive accuracies achieved from different sample sizes, with and without feature selection, were compared statistically. To demonstrate the effect, we used grey matter segmented from the T1-weighted anatomical scans collected by the Alzheimer's disease Neuroimaging Initiative (ADNI) as the input features to a linear support vector machine classifier. The objective was to characterize the patterns of difference between Alzheimer's disease (AD) patients and cognitively normal subjects, and also to characterize the difference between mild cognitive impairment (MCI) patients and normal subjects. In addition, we also compared the classification accuracies between MCI patients who converted to AD and MCI patients who did not convert within the period of 12months. Predictive accuracies from two data-driven feature selection methods (t-test filtering and RFE) were no better than those achieved using whole brain data. We showed that we could achieve the most accurate characterizations by using prior knowledge of where to expect neurodegeneration (hippocampus and parahippocampal gyrus). Therefore, feature selection does improve the classification accuracies, but it depends on the method adopted. In general, larger sample sizes yielded higher accuracies with less advantage obtained by using knowledge from the existing literature.
Recent research on Alzheimer's disease (AD) has shown that the decline of cognitive and memory functions is accompanied by a disrupted neuronal connectivity characterized by white matter (WM) ...degeneration. However, changes in the topological organization of WM structural network in AD remain largely unknown. Here, we used diffusion tensor image tractography to construct the human brain WM networks of 25 AD patients and 30 age- and sex-matched healthy controls, followed by a graph theoretical analysis. We found that both AD patients and controls had a small-world topology in WM network, suggesting an optimal balance between structurally segregated and integrative organization. More important, the AD patients exhibited increased shortest path length and decreased global efficiency in WM network compared with controls, implying abnormal topological organization. Furthermore, we showed that the WM network contained highly connected hub regions that were predominately located in the precuneus, cingulate cortex, and dorsolateral prefrontal cortex, which was consistent with the previous diffusion-MRI studies. Specifically, AD patients were found to have reduced nodal efficiency predominantly located in the frontal regions. Finally, we showed that the alterations of various network properties were significantly correlated with the behavior performances. Together, the present study demonstrated for the first time that the Alzheimer's brain was associated with disrupted topological organization in the large-scale WM structural networks, thus providing the structural evidence for abnormalities of systematic integrity in this disease. This work could also have implications for understanding how the abnormalities of structural connectivity in AD underlie behavioral deficits in the patients.
Structural covariance assesses similarities in gray matter between brain regions and can be applied to study networks of the brain. In this study, we explored correlations between structural ...covariance networks (SCNs) and cognitive impairment in Parkinson's disease patients. 101 PD patients and 58 age- and sex-matched healthy controls were enrolled in the study. For each participant, comprehensive neuropsychological testing using the Wechsler Adult Intelligence Scale-III and Cognitive Ability Screening Instrument were conducted. Structural brain MR images were acquired using a 3.0T whole body GE Signa MRI system. T1 structural images were preprocessed and analyzed using Statistical Parametric Mapping software (SPM12) running on Matlab R2016a for voxel-based morphometric analysis and SCN analysis. PD patients with normal cognition received follow-up neuropsychological testing at 1-year interval. Cognitive impairment in PD is associated with degeneration of the amygdala/hippocampus SCN. PD patients with dementia exhibited increased covariance over the prefrontal cortex compared to PD patients with normal cognition (PDN). PDN patients who had developed cognitive impairment at follow-up exhibited decreased gray matter volume of the amygdala/hippocampus SCN in the initial MRI. Our results support a neural network-based mechanism for cognitive impairment in PD patients. SCN analysis may reveal vulnerable networks that can be used to early predict cognitive decline in PD patients.
Background. Parkinson’s disease (PD) is a common neurodegenerative disease associated with accumulation of misfolding proteins and increased neuroinflammation, which may further impair the glymphatic ...system. The purpose of this study was to utilize diffusion tensor image analysis along the perivascular space (DTI-ALPS) to evaluate glymphatic system activity and its relationship with systemic oxidative stress status in PD patients. Methods. Magnetic resonance imaging and neuropsychological tests were conducted on 25 PD patients with normal cognition (PDN), 25 PD patients with mild cognitive impairment (PD-MCI), 38 PD patients with dementia (PDD), and 47 normal controls (NC). Oxidative stress status was assessed by plasma DNA level. Differences in ALPS-index among the subgroups were assessed and further correlated with cognitive functions and plasma DNA levels. Results. The PD-MCI and PDD groups showed significantly lower ALPS-index compared to normal controls. The ALPS-index was inversely correlated with plasma nuclear DNA, mitochondrial DNA levels, and cognitive scores. Conclusions. Lower diffusivity along the perivascular space, represented by lower ALPS-index, indicates impairment of the glymphatic system in PD patients. The correlation between elevated plasma nuclear DNA levels and lower ALPS-index supports the notion that PD patients may exhibit increased oxidative stress associated with glymphatic system microstructural alterations.
Love hurts: An fMRI study Cheng, Yawei; Chen, Chenyi; Lin, Ching-Po ...
NeuroImage (Orlando, Fla.),
June 2010, 2010-Jun, 2010-06-00, 20100601, Volume:
51, Issue:
2
Journal Article
Peer reviewed
Being in a close relationship is essential to human existence. Such closeness can be described as including other in the self and be underpinned on social attachment system, which evolved from a ...redirection of nociceptive mechanisms. To what extent does imagining a loved-one differs from imagining an unfamiliar individual being in painful situations? In this functional MRI study, participants were exposed to animated stimuli depicting hands or feet in painful and non-painful situations, and instructed to imagine these scenarios from three different perspectives: self, loved-one and stranger after being primed with their respective photographs. In line with previous studies, the three perspectives were associated with activation of the neural network involved in pain processing. Specifically, adopting the perspective of a loved-one increased activity in the anterior cingulate cortex and insula, whereas imagining a stranger induced a signal increase in the right temporo-parietal junction (TPJ) and superior frontal gyrus. The closer the participants' relationships were with their partner, the greater the deactivation in the right TPJ. A negative effective connectivity between the right TPJ and the insula, and a positive one with the superior frontal gyrus were found when participants imagined the perspective of a stranger. These results demonstrate that intimacy affects the bottom-up information processing involved in empathy, as indicated by greater overlap between neural representations of the self and the other.
Abstract
The aging process is accompanied by changes in the brain’s cortex at many levels. There is growing interest in summarizing these complex brain-aging profiles into a single, quantitative ...index that could serve as a biomarker both for characterizing individual brain health and for identifying neurodegenerative and neuropsychiatric diseases. Using a large-scale structural covariance network (SCN)-based framework with machine learning algorithms, we demonstrate this framework’s ability to predict individual brain age in a large sample of middle-to-late age adults, and highlight its clinical specificity for several disease populations from a network perspective. A proposed estimator with 40 SCNs could predict individual brain age, balancing between model complexity and prediction accuracy. Notably, we found that the most significant SCN for predicting brain age included the caudate nucleus, putamen, hippocampus, amygdala, and cerebellar regions. Furthermore, our data indicate a larger brain age disparity in patients with schizophrenia and Alzheimer’s disease than in healthy controls, while this metric did not differ significantly in patients with major depressive disorder. These findings provide empirical evidence supporting the estimation of brain age from a brain network perspective, and demonstrate the clinical feasibility of evaluating neurological diseases hypothesized to be associated with accelerated brain aging.
The present study aimed to determine whether a recently proposed cerebral small vessel disease (CSVD) classification scheme could differentiate the 5-year all-cause mortality in middle-to-old aged ...asymptomatic CSVD. Stroke-free and non-demented participants recruited from the community-based I-Lan Longitudinal Aging Study underwent baseline brain magnetic resonance imaging (MRI) between 2011 and 2014 and were followed-up between 2018 and 2019. The study population was classified into control (non-CSVD) and CSVD type 1-4 groups based on MRI markers. We determined the association with mortality using Cox regression models, adjusting for the age, sex, and vascular risk factors. A total of 735 participants were included. During a mean follow-up of 5.7 years, 62 (8.4%) died. There were 335 CSVD type 1 (57.9 ± 5.9 years), 249 type 2 (65.6 ± 8.1 years), 52 type 3 (67.8 ± 9.2 years), and 38 type 4 (64.3 ± 9.0 years). Among the four CSVD types, CSVD type 4 individuals had significantly higher all-cause mortality (adjusted hazard ratio = 5.0, 95% confidence interval 1.6-15.3) compared to controls. This novel MRI-based CSVD classification scheme was able to identify individuals at risk of mortality at an asymptomatic, early stage of disease and might be applied for future community-based health research and policy.