Figure 3. Group differences between healthy controls and the PSD schizoaffective subtype in gray matter cortical areas in a) perfusion fraction (PF), b) free water (FW) and (c) fractional anisotropy of tissue (FAt) measures. Green indicates ROIs with significant differences at q < .05 FDR BH corrected, blue indicates ROIs with differences noted at p < .05.

Figure 4. Significant Pearson’s correlations between IVIM metrics (PF, FW and FAt) and the duration of psychosis in the PSD schizoaffective and schizophrenia subtypes in gray and white matter ROIs. All plots reached significance at the FDR BH q < .05 correction-level.

Figure 1. Three major steps in learning VAE-MMD model. 1. For training we input both ABIDE I with labels and ABIDE II training datasets into VAE-MMD model. The original t-SNE figure and domain adapted t-SNE figure (both shown in Fig.2) are generated in the beginning and last iteration of this process. The total loss is constructed by SSL loss, reconstruction loss and MMD loss. 2.For validation, ABIDE I and ABIDE II validation dataset are used to fine-tuning the hyperparameters α and β. 3. For testing, ABIDE I & ABIDE II test dataset are used in testing model to evaluate the model’s performance

Figure 3. Three-way classification result (in terms of % accuracy) in different scenarios with/without domain adaptation (DA) and statistical ComBat harmonization.

Figure 1. Overview of data processing and analysis pipeline. Group-ICA was used to calculate and extract the average signal time series from validated brain parcellations. Following data reduction were performed by a rCCA. The subtypes of depression were defined by k-means clustering based on combination of neuroimaging and clinical data. Finally, group comparisons of ICNs and clinical symptoms between each subgroup and HC were implemented a one-way ANOVA analysis, respectively. PLSR analysis was employed to assess the association between these features in each subgroup.

Figure 4. Statistically significant FNC differences between MDD, MDD-subtypes and HCs. (A) A heatmap and a circular plot showing were significantly abnormal FNC in MDD relative to HCs. (B) A heatmap showing the significant results of ANOVA in FNC among MDD-subtypes and HCs. (C) Three circular plots showing significant FNC patterns between subtype1, subtype2 and HCs using post-hoc analyses. Statistical significance for all comparisons were thresholded at p < 0.05, FDR corrected for cluster level and p < 0.001, uncorrected for voxel level.

Figure 1: Spectral fit of one subject during the Rest period (baseline). Metabolites included in the fitting template were: alanine, aspartate, choline, creatine, GABA, glucose, glutamate, glutamine, glutathione, glycine, lactate, myo-inositol, N-acetyl aspartate, N-acetyl aspartyl glutamate, phosphorylethanolamine, scyllo-inositol, and taurine. Macromolecules were not included due to the long echo time used.

Figure 2: Glutathione (GSH) dynamics. Plots of (a) mean GSH concentration [mM], (b) mean GSH concentration difference [mM] relative to baseline (‘Rest’), and (c) mean GSH percentage change (%) normalized to baseline (‘Rest’). Solid blue, solid red, light blue, and pink lines represent mean HC, mean FES, individual HC, and individual FES values, respectively. Asterisks above time points indicate significant difference between group.

Figure 2.

Figure 1.

Figure 2: TD vs LF-ASD. Fixels are colour-coded by family-wise error corrected p-values and overlaid on the cohort FOD template. Upper and middle rows: Fixels with a significant (P<0.05) lower fibre density (FD) in the LF-ASD subgroup than TD group, illustrated at two slice locations per plane. Bottom row: Fixels with a significant (P<0.05) lower fibre density and cross-section (FDC) in the LF-ASD subgroup than TD group.

Figure 3: Low IQ ASD vs minimally verbal LF-ASD. Fixels are colour-coded by family-wise error corrected p-values and overlaid on the cohort FOD template. Upper and middle rows: Fixels with a significant (P<0.05) lower fibre density (FD) in the minimally verbal ASD than the low IQ-only subgroup, illustrated at two slice locations per plane. Bottom row: Fixels with a significant (P<0.05) lower fibre density and cross-section (FDC) in the minimally verbal ASD than the low IQ-only subgroup.

Figure 2. Regional changes in CBF (left), ATT (middle), and aCBV (right). A) Mean CBF in the frontal, parietal, and temporal lobes. B) Mean ATT in frontal and parietal lobes. C) Mean aCBV in the bilateral insula. Dashed lines represent individual participants. Insets illustrate regions of interest.

Figure 1. Representative CBF (left), ATT (middle), and aCBV (right) maps from a single participant in the nitrous oxide treatment arm at baseline (top row) and post-treatment (bottom row).

Fig. 1. Cohen’s d of regional subcortical volume in two clusters of treated patients and the group of never-treated patients compared with those in healthy controls. HC, healthy controls.

Fig. 2. Cohen’s d of cognitive function in two clusters of treated patients compared with those in healthy controls. ASIP, attention and speed of information processing; EF, executive functioning; HC, healthy controls; MS, motor speed; VF, verbal fluency; VM, verbal memory; WM, working memory.

Figure 3.| FMRI results of the face-matching task. A) Region of interest analysis: Post-hoc tests showed significant differences between conditions in right amygdala reactivity for children at 8 weeks (DT) (mean with 95% CI). B) Exploratory whole-brain analysis: Increased reactivity in the Superior-Frontal-Gyrus and Paracingulate-Cortex in methylphenidate treated children from DT to PT and decreased reactivity in the Lateral-Occipital-Cortex in placebo-treated adults from BL to PT.

Figure 1.| Timeline of the ePOD-MPH RCT. A 16-week double-blind, randomized, placebo-controlled, multicenter trial with methylphenidate and a blinded endpoint evaluation in stimulant treatment-naive patients with ADHD. We measured fMRI activity on a face-matching task at three times points (baseline (BL), eight weeks during treatment (DT), and one week after the trial (post-treatment (PT)). Furthermore, clinical measures of anxiety, depression, and emotional lability were assessed.

Fig 1. Global properties of whole brain in obsessive-compulsive disorder (OCD) and healthy control (HC) groups. The bars showed significant between-group differences in the area under the curves (AUC) of global metrics at slow2, slow3, slow4, slow5 and reference frequency bands (0.01-0.08Hz). (a) γ, normalized clustering coefficient; (b) λ, normalized characteristic path length; (c) σ, normalized small-world; (d) Global efficiency; (e) Modularity, Q. * p<0.05, with t test. referB, reference frequency band between 0.01 and 0.08 Hz.

Fig 3. The significantly different nodal properties at slow3 band between the OCD patients and healthy controls. (p<0.05, FDR corrected). (a) Degree centrality; (b) Betweenness centrality; (c) Nodal clustering coefficient; (d) Nodal shortest path. Warm colors indicate increase in OCD grooup, cool colors indicate decrease.

Fig. 1 The NBS result showed the disrupted subnetwork in the TSC patients with intellectual disability compared with who have normal intelligence (ID < normal).

Fig. 2 The NBS result showed the disrupted subnetwork in the TSC patients with intractable seizure of intractable compared with who are none / controlled (active + refractory < none + remission).

Figure 4. Reconstructed typical multi-parametric MR images and GluCEST maps of the control (a) and FS (b) rats. ADC, apparent diffusion coefficient; CBF, cerebral blood flow; FS, forced swimming test group; GluCEST, glutamate-weighted chemical exchange saturation transfer

Figure 2. LCModel fitted results of the ¹H-MRS spectra in the hippocampal region acquired from representative FS and control rats (a and b). The bar chart with data points shows the mean glutamate concentration, and vertical lines on each of the bars represent the standard deviation of the mean values (FS, forced swimming test group; ppm, part per million; VOI, volume of interest; green color: control group; red color: FS group; significance levels: ***p < 0.001).

Figure 2. Neuromelanin Contrast Ratio (NMcr) in treatment resistant (TR) and responders.

Figure 1. Example of the segmentation procedures on an individual scan. A)Individual image in MNI space B)Individual image in MNI space with standardized mask, red represents the substantia nigra (SN). The SN mask was drawn on a standardized average image, and then placed on the individual data. The Crus Cerebri (CC) (not shown here) was analyzed in a similar manner. C)Raw image D)Raw image with manually drawn masks, red represents the SN. The blue dots represents the CC.

Figure 3. Functional network connectivity correlation matrix of four transient states in early childhood, late childhood and early adolescence group of ADHD and TDC.

Figure 4. Temporal characteristics (from left to right: fraction time, dwell time, number of transitions) differences between each age group of ADHD and TDC.

Figure 1 Seed region of functional connectivity, MRS voxel location, and representative spectra of semi-LASER and MEGA-PRESS (6).

Figure 4 Correlations between the MPFC-DLPFC resting-state connectivity and MPFC GABA and Glu concentrations respectively in schizophrenia and healthy control groups.

Figure 2. Group differences in non-directional dMRI metrics: a) MD, b) FA, c) ODI, d) NDI, and e) MK in the sub-cortical white matter. Results are presented by projecting WM areas of difference onto the corresponding sub-cortical WM surface. T-tests were conducted between HC and PSD and PSD subtypes: schizoaffective (SZA), schizophrenia (SZ) and bipolar with psychotic features (BPF). Group differences that reached significance at the FDR BH q < .05 correction-level are shown in pink, while group differences that reached significance at p < .05 trend-level are shown in purple.

Figure 3. Representative significant Pearson’s correlations between dMRI metrics: 1) NODDI NDI, ODI and ODIs, 2) DTI FA, RD and MD, and 3) DKI RK and MK and several tests of verbal learning and memory (HVLT), processing speed (symbol coding) and working memory (spatial span, symbol span) in both PSD and HC groups. Tests that reached FDR BH q < .05 correction-level are noted with an asterisk (*).

Figure 1: Significantly lower NODDI NDI found in ASD compared to TDC (p<0.05, corrected). Lower NDI were observed in areas of the genu of the corpus callosum, superior longitudinal fasciculus, and uncinate fasciculus.

Figure 2: Significant age by group interactions of qT1 between ASD and TDC subjects overlaid on the mean qT1 map. Significant interactions were localized in the in body and splenium of the corpus callosum.

Fig. 3. HAM-D score is inversely correlated with ACC GHS level.

Fig. 1. Placement of a 20×20×30 mm³ voxel at left dorsal/rostral ACC.

Table 1. Demographic and clinical characteristics of AN-FES and HCs.

Table 2. Demographic and clinical characteristics of schizophrenia patient subgroups at baseline and after one-year follow-up.

Fig. 1 Flow chart of Meta-analysis of resting-state functional imaging and VBM studies of patients with AN

Fig. 2 Meta-analyses results regarding a) resting-state functional activity difference between AN and HCs, b) GMV difference between AN and HCs, c) conjunction of resting-state functional activity differences and GMV differences. Areas with decreased resting-state functional activity value or GMV value are displayed in blue, and areas with increased resting-state functional activity value or GMV value are displayed in red. The color bar indicates the maximum and minimum SDM-Z values.

Fig. 3. Evolution of total Cho level in patients and healthy controls during 4 visits over 26 weeks.

Fig. 1. Placement of 20×20×30 mm³ voxel at left dorsal/rostral ACC.

Group differences in global topological properties among healthy control subjects and patients with major depressive disorder either with or without a history of suicidality.

Regions with significant alterations in nodal efficiency among three groups (A) and associations between clinical measures and network properties (B).

Fig. 1. A. ROI for SCN in bilateral anterior hypothalamus. B. Negative FC with SCN in the HC subjects. C. Positive FC with SCN in the HCs (p < 0.005, extent threshold = 5 voxels). The brain structures labeled 1 to 15 are listed in Table 2.

Table 2. Regions with negative (blue) and positive (red) functional connectivity with the SCN in HCs (voxel ≥ 5, p < 0.005). Note: 2-sample t-test were used for the FCs’ comparisons between HC group and CID group, ＊significant difference between two groups (p<0.05).

Fig. 2: Bar chart with the standard deviation for each RS-fMRI measure within the right executive control network (RECN, right side) and the left executive control network (LECN, left side) between a healthy subject and a depressed patient. Right and left view of the masks of the RECN and LECN used in the analysis are shown above the corresponding bar chart.

Fig. 1: Mean RS-fMRI measures of healthy subjects (right column) and depressed patients (left column). The fMRI measures degree centrality (DC, top row), regional homogeneity (ReHo, second row), amplitude of low frequency fluctuations (ALFF, third row) and fractional ALFF (bottom row) are shown in right, and left views.

Figure 1.Flow chart of data processing and analysis. A) fMRI data used in this research. B) We extracted time courses from each ROI of the anatomical automatic labeling atlas-90 (AAL-90) mask and calculated the temporal variability of each region. C) Statistical analysis was performed, including grouping comparison of temporal variability of RS, NRS and HC, and Pearson correlation analysis to reveal the relationship between temporal variability and ΔPANSS.

Figure 2.The relationship between the response to ECT (△PANSS) and the temporal variability in brain regions of discriminant differences in SZ. A) The temporal variability of IFGtriang.R in RS is negatively correlated with the response to ECT. The temporal variability of TPOsup.R B) and MTG.R C) in NRS is positively correlated with the response to ECT.

Figure 2: MRI measurements of gray matter changes induced by tDCS: A voxelwise 1-way ANOVA revealed significant (p < 0.05, FDR corrected) differences between the Sham, Active-Conventional and Active-HD groups in the left dorsolateral prefrontal cortex (DLPFC) and the posterior cingulate cortex (PCC). Post-hoc t-tests revealed significant (p<0.05) gray matter increases with the active-HD montage (relative to Sham) in both regions, but only in the left DLPFC with the active-conventional montage.

Figure 1: Study design: Fifty-nine moderately depressed participants were randomized into Active/Sham X HD/Conventional groups using a double-blind, parallel design (n=20/19/20 in Active-HD/Active-Conventional/Sham groups respectively). Each participant received tDCS treatment from the High definition (HD) or Conventional montages, according to the allocation. Treatments were administered using a double-blind device (for details about tDCS treatments, see abstract text).

Figure 3. Brain networks dysfunction patterns of posttraumatic stress disorder. DMN, default mode network; VAN, ventral attention network; DAN, dorsal attention network; SMN, somatomotor network; LN, limbic network; FPN, frontoparietal network

Figure 2. The results of the meta-analysis of altered resting-state functional connectivity with the default mode network (DMN) and limbic network in the PTSD group compared with the groups of trauma-exposed controls (TEC) or nonexposed controls (NEC).

ReHo difference brain map of three groups

Correlation between different brain regions and IES-R return scores(IES-R:Impact of Event Scale-Revised )

The comparison of the brain tissue and myelin volume between ADHD group and control group with the measurements acquired by SyMRI (*: p<0.05).

Figure 2: NODDI and DTI group differences. Color bar indicates level of significance of voxels from group difference model. A. Significant voxels and neuroanatomical location of FICVF group differences. B. Significant voxels and neuroanatomical location of MD group differences. C. Significant voxels and neuroanatomical location of RD group differences. D. Significant voxels and neuroanatomical location of AD group differences.

Figure 4: DTI age by group interactions. Significant voxels displayed on the neuroanatomical maps, color bar indicates level of significance (B, D, F, H). A. Scatter plot representing mean FA values of significant voxels of age-by-group interaction. C. Scatter plot representing MD values of significant voxels of age-by-group interaction. E. Scatter plot representing mean RD values of significant voxels of age-by-group interaction. G. Scatter plot representing mean AD values of significant voxels of age-by-group interaction.

Power spectra under TAF task and resting state of OCD (red) and HC (black). During resting state, intrinsic neural activity power within DMN had significant difference at Bin1 and Bin2. Salience network showed significant difference at Bin1 and Bin3 during TAF task. (Bin1: 0-0.05 Hz, Bin2: 0.05-0.1 Hz, Bin3: 0.1-0.15 Hz, Bin4: 0.15-0.2 Hz, Bin5: 0.2-0.25Hz)

Independent components of each group under the TAF task. Spatial maps were thresholded at p < 0.05 using one sample t-test, corrected for family wise error (FWE). (a) Default Mode Network (DMN), (b) Central Executive Network (CEN), and (c) Salience Network (SN).

Figure 2. Significant differences in fixel-based metrics between HIV-infected and HIV-uninfected individuals. (top) Fiber Density (FD), (middle) Fiber bundle Cross-section (FC), and (bottom) Fiber density and cross-section (FDC). Significant fixels (family-wise error corrected p < 0.05) were mapped to streamlines and thresholded at p < 0.05.

Figure 4. Scatterplots showing FDC as a function of Overall Z-score (A), and as a function of Executive z-score (B). Only regions with significant correlations are shown. Solid lines are linear fits and shaded areas are for 95% confidence interval. FDC: Fiber density and cross-section, IC: internal capsule.

Figure 3. NBS results. (a) Compared with the HC group, the SA group demonstrated significantly stronger subnetwork connections in the frontal and parietal lobe. (b) Compared with the HC group, the D group demonstrated significantly stronger subnetwork connections in the parietal lobe (p < 0.05).

Figure 2. Topological parameters of graph theoretical analysis. Topological parameters including (a) global efficiency, (b) normalized shortest path length (λ), (c) characteristic path length, (d) normalized clustering coefficient (γ), (e) modularity and (f) small-worldness index among three groups (SA, D and HC).

Figure 1: lesion progression in three patients projected on WM segmentations (left) and corresponding slices of ADC maps (right). Colors indicate the age of the lesion. Only four time points were available for patient one and does therefore not show lesion age of five years. Patient two and three had five available time points.

Figure 2: averaged ADC values within newly detected WML as a function of lesion age. WMLs that were present at the first time point were not included because the age of these lesions cannot be determined. Error bars denote standard deviations. Linear regression of all individual newly detected lesion areas after each year showed a significant correlation between lesion age and ADC.

Axial slice from an HIV-1C positive subject illustrating the different parametric maps obtained with obtained with FWE-DTI: free-water volume fraction (f_FW), fractional anisotropy (FWE-FA), mean diffusivity (FWE-MD), axial diffusivity (FWE-AD), and radial diffusivity (FWE-RD). A b0 image is included for anatomical reference.

Free-water volume fraction (f_FW) measured at 20 gray matter ROIs for the HIV-1C and control groups.

Axial slices from the brain of an HE subject showing b0 images (top row) and the corresponding the f_FW maps (bottom row).

Voxelwise z-score maps comparing f_FW values from the brain of an HE subjects with the mean values of the control group. The z-scores are overlaid on a template T1 image in MNI space.

Figure 1. Axial slices of an individual participant with DMD and a healthy control. The first row (A-C) show T1-weighted images of a HC at baseline (A), follow-up (B) and the corresponding image of changes at the brain edges superimposed on a interpolated halfway point between baseline and follow-up MR images (edge-motion image). (C) The bottom row shows T1-weighted images of a participant with DMD, with baseline (D), follow-up (E) and edge-motion image. The brain edge movement image shows local reductions (in blue-green) and increases (red-yellow).

Figure 2. Left top to bottom: results of the paired t-test between baseline (Time1) and follow-up (Time2) within groups; DMD in blue, HC in black. Intracranial volume (ICV), total brain volume (TBV), grey matter volume (GMV), white matter volume (WMV) and cerebrospinal fluid volume (CSF) with average increase or average decrease over time. Results were considered significant at Bonferroni corrected p≤0.005. Right top to bottom: results of the longitudinal mixed model analysis between groups corrected for age. DMD in blue, HC in black.

Fig.1 ROI selection of gray nucleus. (A)The HCN、PUT、GP、THA are shown front and back. (B)Showing the RN and SN. (C) Showing the DN. (D) Showing the GF.

Fig.2 Comparison of gray nucleus MSV between T2DM and CON group. ★P < 0.05.

Figure 3: Child-rated SCT symptoms (controlling for age, sex, and parent-rated ADHD inattention symptoms) shows decreased functional connectivity between the attention seed region and the: left cuneus cortex, left isthmus – cingulate cortex, left lingual gyrus, left pericalcarine cortex, left precuneus cortex, right cuneus cortex, right isthmus – cingulate cortex, right lingual gyrus, right parahippocampal gyrus, right pericalcarine cortex, right precuneus cortex.

Figure 2: Teacher-rated SCT (controlling for age, sex, and teacher-rated ADHD inattention symptoms) shows increased functional connectivity between the attention seed region and the: right cuneus cortex, right lateral occipital cortex, fight pericalcarine cortex, right superior parietal cortex, and decreased functional connectivity between the: left cuneus cortex, left isthmus – cingulate cortex, left lingual gyrus, left pericalcarine cortex, Left precuneus cortex.

Figure 1: The top panels show the areas where the prenatal methadone exposed group has lower FC and the bottom panels show and the areas where they have lower FDC values. All the results are overlaid to the WM FOD template for reference. Axial and coronal views follow radiological convention.

Table 1: Infant characteristics.

Figure 2. Images of CVSD-D patient (A-1, T1WI 3D MPRAGE, A-2, DKI-Dmean map) and CVSD-ND patient (B-1, T1WI 3D MPRAGE, B-2, DKI-Dmean map). Compared with non-depression CVDS patient, the Dmean map of CVSD-D patient have more warm pixels in region of ACC. White arrow indicate left ACC.

Table 1. A comparison of DKI derived parameters in ACC between two groups.

Note: Values are presented as mean ± SD or median (interquartile range). Abbreviations: Dmean, mean diffusivity; KFA, kurtosis fractional anisotropy; Kmean, mean kurtosis.

Fig. 1 Bar chart comparing brain iron content measurement results across children in the same age group

Fig. 2 Bar chart comparing brain iron content measurement results across children in the same age and sex group

Table 1 demographic characteristics of the cohorts

Table 2 Difference of IVIM Parameters between the patients with acute CO poisoning and healthy controls

Figure 2(A) ROC curves of the non-NPSLE classification on different classifiers (B) Histograms of AUC on different classifiers. SVM, support vector machine, LDA, linear discriminant analysis, LR, logistic regression.

Figure 1. Brain structures with grey matter volume decreased in non-NPSLE compared to healthy controls

Comparison of iron deposition between patients with cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) and healthy controls. Increased iron deposition of caudate, putamen, and pallium in patients with CADASIL compared with healthy controls. ns indicates nonsignificant. *A significant correlation (P<0.05).

Correlations between white matter microstructure and putamen iron deposition. FA: fractional anisotropy; MD: mean diffusivity; RD: radial diffusivity

The images show the mean FCS in RA (A), HC (B) and the difference between RA and HC (C). The color bar at the bottom of each picture represents the FCS value for each group.

The correlation between FCS and neuropsychological scores (A, MMSE; B, MoCA) in RA patients. The color bar represents the ! value.

Fig 1. Comparison of APTw values of left and right hemispheres in different regions of limbic system.

Table 2. Normal APTw signals in limbic system

Figure 1. Data derived from a healthy control. The rows show data from the arm, cubital tunnel and forearm. The columns contain T2-weighted scans, and corresponding maps of normalised quantitative anisotropy (nQA), mean diffusivity (MD) and the principal eigenvector (V1) with the colours red, green and blue representing diffusion in x, y and z directions, and the intensity scaled by QA.

Figure 2. Scatter plot with linear fit (and 95% CI) showing the relationship between Fractional Anisotropy of the ulnar nerve in volunteers and patients, at different positions within the upper limb.

Figure 3. The NBS analysis results. A disrupted subnetwork was found in the SI group compared with the NS group (NS > SI).

Figure 1. (a) Two-sample t-test results of mfALFF between the SI and the NS groups (NS > SI, color bar represents t-score). (b) Lower mfALFF of the left cuneus and (c) higher mfALFF of the right middle temporal pole gyrus were found in the SI group compared with the NS group. (d) Two-sample t-test results of mReHo between the SI and the NS groups (NS > SI; color bar represents t-score). (e) Lower mReHo of the right cuneus and (f) higher mReHo of the left middle temporal gyrus (MTG) were found in the SI group compared with the NS group.