Brain connectivity networks underlying resting heart rate variability in acute ischemic stroke.

Acute strokes can affect heart rate variability (HRV), the mechanisms how are not well understood. We included 42 acute stroke patients (2-7 days after ischemic stroke, mean age 66 years, 16 women). For analysis of HRV, 20 matched controls (mean age 60.7, 10 women) were recruited. HRV was assessed at rest, in a supine position and individual breathing rhythmus for 5 min. The coefficient of variation (VC), the root mean square of successive differences (RMSSD), the powers of low (LF, 0.04-0.14 Hz) and high (HF, 0.15-0.50 Hz) frequency bands were extracted. HRV parameters were z-transformed related to age- and sex-matched normal subjects. Z-values < -1 indicate reduced HRV. Acute stroke lesions were marked on diffusion-weighted images employing MRIcroN and co-registered to a T1-weighted structural volume-dataset. Using independent component analysis (ICA), stroke lesions were related to HRV. Subsequently, we used the ICA-derived lesion pattern as a seed and estimated the connectivity between these brain regions and seven common functional networks, which were obtained from 50 age-matched healthy subjects (mean age 68.9, 27 women). Especially, LF and VC were frequently reduced in patients. ICA revealed one covarying lesion pattern for LF and one similar for VC, predominantly affecting the right hemisphere. Activity in brain areas corresponding to these lesions mainly impact on limbic (r = 0.55 ± 0.08) and salience ventral attention networks (0.61 ± 0.10) in the group with reduced LF power (z-score < -1), but on control and default mode networks in the group with physiological LF power (z-score > -1). No different connectivity could be found for the respective VC groups. Our results suggest that HRV alteration after acute stroke might be due to affecting resting-state brain networks.

Identifying and characterizing different stages toward Alzheimer's disease using ordered core features and machine learning.

Based on the joint HCPMMP parcellation method we developed before, which divides the cortical brain into 360 regions, the concept of ordered core features (OCF) is first proposed to reveal the functional brain connectivity relationship among different cohorts of Alzheimer's disease (AD), late mild cognitive impairment (LMCI), early mild cognitive impairment (EMCI) and healthy controls (HC). A set of core network features that change significantly under the specifically progressive relationship were extracted and used as supervised machine learning classifiers. The network nodes in this set mainly locate in the frontal lobe and insular, forming a narrow band, which are responsible for cognitive impairment as suggested by previous finding. By using these features, the accuracy ranged from 86.0% to 95.5% in binary classification between any pair of cohorts, higher than 70.1%-91.0% when using all network features. In multi-group classification, the average accuracy was 75% or 78% for HC, EMCI, LMCI or EMCI, LMCI, AD against baseline of 33%, and 53.3% for HC, EMCI, LMCI and AD against baseline of 25%. In addition, the recognition rate was lower when combining EMCI and LMCI patients into one group of mild cognitive impairment (MCI) for classification, suggesting that there exists a big difference between early and late MCI patients. This finding supports the EMCI/LMCI inclusion criteria introduced by ADNI based on neuropsychological assessments.

Brain functional connectivity analysis based on multi-graph fusion.

In this paper, we propose a framework for functional connectivity network (FCN) analysis, which conducts the brain disease diagnosis on the resting state functional magnetic resonance imaging (rs-fMRI) data, aiming at reducing the influence of the noise, the inter-subject variability, and the heterogeneity across subjects. To this end, our proposed framework investigates a multi-graph fusion method to explore both the common and the complementary information between two FCNs, i.e., a fully-connected FCN and a 1 nearest neighbor (1NN) FCN, whereas previous methods only focus on conducting FCN analysis from a single FCN. Specifically, our framework first conducts the graph fusion to produce the representation of the rs-fMRI data with high discriminative ability, and then employs the L1SVM to jointly conduct brain region selection and disease diagnosis. We further evaluate the effectiveness of the proposed framework on various data sets of the neuro-diseases, i.e., Fronto-Temporal Dementia (FTD), Obsessive-Compulsive Disorder (OCD), and Alzheimers Disease (AD). The experimental results demonstrate that the proposed framework achieves the best diagnosis performance via selecting reasonable brain regions for the classification tasks, compared to state-of-the-art FCN analysis methods.