MVPANI: A Toolkit With Friendly Graphical User Interface for Multivariate Pattern Analysis of Neuroimaging Data

Functional Connectivity-Based Searchlight Multivariate Pattern Analysis for Discriminating Schizophrenia Patients and Predicting Clinical Variables

BACKGROUND

Schizophrenia, a multifaceted psychiatric disorder characterized by functional dysconnectivity, poses significant challenges in clinical practice. This study explores the potential of functional connectivity (FC)-based searchlight multivariate pattern analysis (CBS-MVPA) to discriminate between schizophrenia patients and healthy controls while also predicting clinical variables.

STUDY DESIGN

We enrolled 112 schizophrenia patients and 119 demographically matched healthy controls. Resting-state functional magnetic resonance imaging data were collected, and whole-brain FC subnetworks were constructed. Additionally, clinical assessments and cognitive evaluations yielded a dataset comprising 36 clinical variables. Finally, CBS-MVPA was utilized to identify subnetworks capable of effectively distinguishing between the patient and control groups and predicting clinical scores.

STUDY RESULTS

The CBS-MVPA approach identified 63 brain subnetworks exhibiting significantly high classification accuracies, ranging from 62.2% to 75.6%, in distinguishing individuals with schizophrenia from healthy controls. Among them, 5 specific subnetworks centered on the dorsolateral superior frontal gyrus, orbital part of inferior frontal gyrus, superior occipital gyrus, hippocampus, and parahippocampal gyrus showed predictive capabilities for clinical variables within the schizophrenia cohort.

CONCLUSION

This study highlights the potential of CBS-MVPA as a valuable tool for localizing the information related to schizophrenia in terms of brain network abnormalities and capturing the relationship between these abnormalities and clinical variables, and thus, deepens our understanding of the neurological mechanisms of schizophrenia.

Voxel- and tensor-based morphometry with machine learning techniques identifying characteristic brain impairment in patients with cervical spondylotic myelopathy

Aim

The diagnosis of cervical spondylotic myelopathy (CSM) relies on several methods, including x-rays, computed tomography, and magnetic resonance imaging (MRI). Although MRI is the most useful diagnostic tool, strategies to improve the precise and independent diagnosis of CSM using novel MRI imaging techniques are urgently needed. This study aimed to explore potential brain biomarkers to improve the precise diagnosis of CSM through the combination of voxel-based morphometry (VBM) and tensor-based morphometry (TBM) with machine learning techniques.

Methods

In this retrospective study, 57 patients with CSM and 57 healthy controls (HCs) were enrolled. The structural changes in the gray matter volume and white matter volume were determined by VBM. Gray and white matter deformations were measured by TBM. The support vector machine (SVM) was used for the classification of CSM patients from HCs based on the structural features of VBM and TBM.

Results

CSM patients exhibited characteristic structural abnormalities in the sensorimotor, visual, cognitive, and subcortical regions, as well as in the anterior corona radiata and the corpus callosum [P < 0.05, false discovery rate (FDR) corrected]. A multivariate pattern classification analysis revealed that VBM and TBM could successfully identify CSM patients and HCs [classification accuracy: 81.58%, area under the curve (AUC): 0.85; P < 0.005, Bonferroni corrected] through characteristic gray matter and white matter impairments.

Conclusion

CSM may cause widespread and remote impairments in brain structures. This study provided a valuable reference for developing novel diagnostic strategies to identify CSM.

NeuroDecodeR: a package for neural decoding in R

Neural decoding is a powerful method to analyze neural activity. However, the code needed to run a decoding analysis can be complex, which can present a barrier to using the method. In this paper we introduce a package that makes it easy to perform decoding analyses in the R programing language. We describe how the package is designed in a modular fashion which allows researchers to easily implement a range of different analyses. We also discuss how to format data to be able to use the package, and we give two examples of how to use the package to analyze real data. We believe that this package, combined with the rich data analysis ecosystem in R, will make it significantly easier for researchers to create reproducible decoding analyses, which should help increase the pace of neuroscience discoveries.

Optimization of structural connectomes and scaled patterns of structural-functional decoupling in Parkinson's disease

Parkinson's disease (PD) is manifested with disrupted topology of the structural connection network (SCN) and the functional connection network (FCN). However, the SCN and its interactions with the FCN remain to be further investigated. This multimodality study attempted to precisely characterize the SCN using diffusion kurtosis imaging (DKI) and further identify the neuropathological pattern of SCN-FCN decoupling, underscoring the neurodegeneration of PD. Diffusion-weighted imaging and resting-state functional imaging were available for network constructions among sixty-nine patients with PD and seventy demographically matched healthy control (HC) participants. The classification performance and topological prosperities of both the SCN and the FCN were analyzed, followed by quantification of the SCN-FCN couplings across scales. The SCN constructed by kurtosis metrics achieved optimal classification performance (area under the curve 0.89, accuracy 80.55 %, sensitivity 78.40 %, and specificity 80.65 %). Along with diverse alterations of structural and functional network topology, the PD group exhibited decoupling across scales including: reduced global coupling; increased nodal coupling within the sensorimotor network (SMN) and subcortical network (SN); higher intramodular coupling within the SMN and SN and lower intramodular coupling of the default mode network (DMN); decreased coupling between the modules of DMN-fronto-parietal network and DMN-visual network, but increased coupling between the SMN-SN module. Several associations between the coupling coefficient and topological properties of the SCN, as well as between network values and clinical scores, were observed. These findings validated the clinical implementation of DKI for structural network construction with better differentiation ability and characterized the SCN-FCN decoupling as supplementary insight into the pathological process underlying PD.

Neuroimaging profiles of the negative affective network predict anxiety severity in patients with chronic insomnia disorder: A machine learning study

BACKGROUND

Sleep is instrumental in safeguarding emotional well-being. While the susceptibility to both insomnia and anxiety has been demonstrated to involve intricate brain systems, the neuroimaging profile of chronic insomnia disorder with comorbid anxiety symptoms (CID-A) remains unexplored. Employing machine learning methodologies, this study aims to elucidate the distinct neural substrates underlying CID-A and to investigate whether these cerebral markers can prognosticate anxiety symptoms in patients with insomnia.

METHODS

Functional magnetic resonance imaging (fMRI) data were procured from a relatively large cohort (dataset 1) comprised of 47 CID-A patients, 49 CID patients without anxiety (CID-NA), and 48 good sleeper controls (GSC). Aberrant cerebral functional alterations were assessed through functional connectivity strength (FCS) and resting-state functional connectivity (rsFC). Subsequently, Support Vector Regression (SVR) models were constructed to predict anxiety symptoms in CID patients based on neuroimaging features, which were validated utilizing an external cohort (dataset 2).

RESULTS

In comparison to CID-NA and GSC subjects, CID-A patients exhibited heightened FCS in the right dorsomedial prefrontal cortex (DMPFC), a central hub within the negative affective network. Moreover, the SVR models revealed that DMPFC-related rsFC/FCS features could be employed to predict anxiety symptoms in two independent cohorts of CID patients.

LIMITATION

Modifications in brain functionality might vary across insomnia subtypes.

CONCLUSION

The present findings suggest a potential negative affective network model for the neuropathophysiology of CID accompanied by anxiety. Importantly, the negative affective network pattern may serve as a predictor for anxiety symptoms in CID patients.

Brain structural correlates of postoperative axial pain in degenerative cervical myelopathy patients following posterior cervical decompression surgery: a voxel-based morphometry study

OBJECTIVE

To investigate the brain structural correlates of postoperative axial pain (PAP) in degenerative cervical myelopathy (DCM) following posterior cervical decompression surgery.

METHODS

Structural images with high-resolution T1 weighting were collected from 62 patients with DCM and analyzed, in addition to 42 age/gender matched subjects who were healthy. Voxel-based morphometry (VBM) was analyzed, grey matter volume (GMV) was computed. One-way ANOVA was performed to reveal the GMV differences among DCM patients with PAP, patients without PAP and healthy controls (HC). Post-hoc analyses were conducted to identify the pair-wise GMV differences among these three groups. Analyses of correlations were conducted to uncover the link between clinical measurements and GMV variations. Last, support vector machine (SVM) was conducted to test the utility of GMV for classifying PAP and nPAP DCM patients.

RESULTS

Three main findings were observed: [1] Compared to healthy controls, DCM patients showed a significantly lower GMV in the precuneus preoperatively. DCM patients with PAP also exhibited a lower GMV within precuneus than those without; [2] In DCM patients with PAP, the precuneus GMV was inversely related to the postoperative pain intensity; [3] Moreover, successful classification between PAP and nPAP were observed via SVM based on precuneus GMV as features.

CONCLUSION

In summary, our results indicate that precuneus GMV may be linked to PAP in DCM, and could be employed to forecast the emergence of PAP in DCM patients.

Alterations of insular dynamic functional connectivity and psychological characteristics in unmedicated bipolar depression patients with a recent suicide attempt

BACKGROUND

Mounting evidence showed that insula contributed to the neurobiological mechanism of suicidal behaviors in bipolar disorder (BD). However, no studies have analyzed the dynamic functional connectivity (dFC) of insular Mubregions and its association with personality traits in BD with suicidal behaviors. Therefore, we investigated the alterations of dFC variability in insular subregions and personality characteristics in BD patients with a recent suicide attempt (SA).

METHODS

Thirty unmedicated BD patients with SA, 38 patients without SA (NSA) and 35 demographically matched healthy controls (HCs) were included. The sliding-window analysis was used to evaluate whole-brain dFC for each insular subregion seed. We assessed between-group differences of psychological characteristics on the Minnesota Multiphasic Personality Inventory-2. Finally, a multivariate regression model was adopted to predict the severity of suicidality.

RESULTS

Compared to NSA and HCs, the SA group exhibited decreased dFC variability values between the left dorsal anterior insula and the left anterior cerebellum. These dFC variability values could also be utilized to predict the severity of suicidality (r = 0.456, p = 0.031), while static functional connectivity values were not appropriate for this prediction. Besides, the SA group scored significantly higher on the schizophrenia clinical scales (p < 0.001) compared with the NSA group.

CONCLUSIONS

Our findings indicated that the dysfunction of insula-cerebellum connectivity may underlie the neural basis of SA in BD patients, and highlighted the dFC variability values could be considered a neuromarker for predictive models of the severity of suicidality. Moreover, the psychiatric features may increase the vulnerability of suicidal behavior.

Stratifying ASD and characterizing the functional connectivity of subtypes in resting-state fMRI

BACKGROUND

Although stratifying autism spectrum disorder (ASD) into different subtypes is a common effort in the research field, few papers have characterized the functional connectivity alterations of ASD subgroups classified by their clinical presentations.

METHODS

This is a case-control rs-fMRI study, based on large samples of open database (Autism Brain Imaging Data Exchange, ABIDE). The rs-MRI data from n=415 ASD patients (males n=357), and n=574 typical development (TD) controls (males n=410) were included. Clinical features of ASD were extracted and classified using data from each patient's Autism Diagnostic Interview-Revised (ADI-R) evaluation. Each subtype of ASD was characterized by local functional connectivity using regional homogeneity (ReHo) for assessment, remote functional connectivity using voxel-mirrored homotopic connectivity (VMHC) for assessment, the whole-brain functional connectivity, and graph theoretical features. These identified imaging properties from each subtype were integrated to create a machine learning model for classifying ASD patients into the subtypes based on their rs-fMRI data, and an independent dataset was used to validate the model.

RESULTS

All ASD participants were classified into Cluster-1 (patients with more severe impairment) and Cluster-2 (patients with moderate impairment) according to the dimensional scores of ADI-R. When compared to the TD group, Cluster-1 demonstrated increased local connection and decreased remote connectivity, and widespread hyper- and hypo-connectivity variations in the whole-brain functional connectivity. Cluster-2 was quite similar to the TD group in both local and remote connectivity. But at the level of whole-brain functional connectivity, the MCC-related connections were specifically impaired in Cluster-2. These properties of functional connectivity were fused to build a machine learning model, which achieved ~75% for identifying ASD subtypes (Cluster-1 accuracy = 81.75%; Cluster-2 accuracy = 76.48%).

CONCLUSIONS

The stratification of ASD by clinical presentations can help to minimize disease heterogeneity and highlight the distinguished properties of brain connectivity in ASD subtypes.

Changes in dynamic and static brain fluctuation distinguish minimal hepatic encephalopathy and cirrhosis patients and predict the severity of liver damage

Purpose

Minimal hepatic encephalopathy (MHE) is characterized by mild neuropsychological and neurophysiological alterations that are not detectable by routine clinical examination. Abnormal brain activity (in terms of the amplitude of low-frequency fluctuation (ALFF) has been observed in MHE patients. However, little is known concerning temporal dynamics of intrinsic brain activity. The present study aimed to investigate the abnormal dynamics of brain activity (dynamic ALFF; dALFF) and static measures [static ALFF; (sALFF)] in MHE patients and to strive for a reliable imaging neuromarkers for distinguishing MHE patients from cirrhosis patients. In addition, the present study also investigated whether intrinsic brain activity predicted the severity of liver damage.

Methods

Thirty-four cirrhosis patients with MHE, 28 cirrhosis patients without MHE, and 33 age-, sex-, and education-matched healthy controls (HCs) underwent resting-state magnetic resonance imaging (rs-fMRI). dALFF was estimated by combining the ALFF method with the sliding-window method, in which temporal variability was quantized over the whole-scan timepoints and then compared among the three groups. Additionally, dALFF, sALFF and both two features were utilized as classification features in a support vector machine (SVM) to distinguish MHE patients from cirrhosis patients. The severity of liver damage was reflected by the Child–Pugh score. dALFF, sALFF and both two features were used to predict Child–Pugh scores in MHE patients using a general linear model.

Results

Compared with HCs, MHE patients showed significantly increased dALFF in the left inferior occipital gyrus, right middle occipital gyrus, and right insula; increased dALFF was also observed in the right posterior lobe of the cerebellum (CPL) and right thalamus. Compared with HCs, noMHE patients exhibited decreased dALFF in the right precuneus. In contrast, compared with noMHE patients, MHE patients showed increased dALFF in the right precuneus, right superior frontal gyrus, and right superior occipital gyrus. Furthermore, the increased dALFF values in the left precuneus were positively associated with poor digit-symbol test (DST) scores (r = 0.356, p = 0.038); however, dALFF in the right inferior temporal gyrus (ITG) was negatively associated with the number connection test–A (NCT-A) scores (r = -0.784, p = 0.000). A significant positive correlation was found between dALFF in the left inferior occipital gyrus (IOG) and high blood ammonia levels (r = 0.424, p = 0.012). Notably, dALFF values yielded a higher classification accuracy than sALFF values in distinguishing MHE patients from cirrhosis patients. Importantly, the dALFF values predicted the Child–Pugh score (r = 0.140, p = 0.030), whereas sALFF values did not in the current dataset. Combining two features had high accuracy in classification in distinguishing MHE patients from cirrhotic patients and yielded prediction in the severity of liver damage.

Conclusion

These findings suggest that combining dALFF and sALFF features is a useful neuromarkers for distinguishing MHE patients from cirrhosis patients and highlights the important role of dALFF feature in predicting the severity of liver damage in MHE.

Both fine-grained and coarse-grained spatial patterns of neural activity measured by functional MRI show preferential encoding of pain in the human brain

How pain emerges from human brain remains an unresolved question in pain neuroscience. Neuroimaging studies have suggested that all brain areas activated by painful stimuli were also activated by tactile stimuli, and vice versa. Nonetheless, pain-preferential spatial patterns of voxel-level activation in the brain have been observed when distinguishing painful and tactile brain activations using multivariate pattern analysis (MVPA). According to two hypotheses, the neural activity pattern preferentially encoding pain could exist at a global, coarse-grained, regional level, corresponding to the "pain connectome" hypothesis proposing that pain-preferential information may be encoded by the synchronized activity across multiple distant brain regions, and/or exist at a local, fine-grained, voxel level, corresponding to the "intermingled specialized/preferential neurons" hypothesis proposing that neurons responding specially or preferentially to pain could be present and intermingled with non-pain neurons within a voxel. Here, we systematically investigated the spatial scales of pain-distinguishing information in the human brain measured by fMRI using machine learning techniques, and found that pain-distinguishing information could be detected at both coarse-grained spatial scales across widely distributed brain regions and fine-grained spatial scales within many local areas. Importantly, the spatial distribution of pain-distinguishing information in the brain varies across individuals and such inter-individual variations may be related to a person's trait about pain perception, particularly the pain vigilance and awareness. These results provide new insights into the long-standing question of how pain is represented in the human brain and help the identification of characteristic neuroimaging measurements of pain.

Characterization of white matter microstructural abnormalities associated with cognitive dysfunction in cerebral small vessel disease with cerebral microbleeds

BACKGROUND

Diffusion tensor imaging (DTI) is recommended as a sensitive method to explore white matter (WM) microstructural alterations. Cerebral small vessel disease (CSVD) may be accompanied by extensive WM microstructural deterioration, while cerebral microbleeds (CMBs) are an important factor affecting CSVD.

METHODS

Fractional anisotropy (FA), mean diffusivity (MD), axial diffusivity (AD) and radial diffusivity (RD) images from 49 CSVD patients with CMBs (CSVD-c), 114 CSVD patients without CMBs (CSVD-n), and 83 controls were analyzed using DTI-derived tract-based spatial statistics to detect WM diffusion changes among groups.

RESULTS

Compared with the CSVD-n and control groups, the CSVD-c group showed a significant FA decrease and AD, RD and MD increases mainly in the cognitive and sensorimotor-related WM tracts. There was no significant difference in any diffusion metric between the CSVD-n and control groups. Furthermore, the widespread regional diffusion alterations among groups were significantly correlated with cognitive parameters in both the CSVD-c and CSVD-n groups. Notably, we applied the multiple kernel learning technique in multivariate pattern analysis to combine multiregion and multiparameter diffusion features, yielding an average accuracy >77 % for three binary classifications, which showed a considerable improvement over the single modality approach.

LIMITATIONS

We only grouped the study according to the presence or absence of CMBs.

CONCLUSIONS

CSVD patients with CMBs have extensive WM microstructural deterioration. Combining DTI-derived diffusivity and anisotropy metrics can provide complementary information for assessing WM alterations associated with cognitive dysfunction and serve as a potential discriminative pattern to detect CSVD at the individual level.

Aberrant dynamic functional connectivity in corticostriatal circuitry in depressed bipolar II disorder with recent suicide attempt

BACKGROUND

The underlying neurobiological mechanisms on suicidal behavior in bipolar disorder remain unclear. We aim to explore the mechanisms of suicide by detecting dynamic functional connectivity (dFC) of corticostriatal circuitry and cognition in depressed bipolar II disorder (BD II) with recent suicide attempt (SA).

METHODS

We analyzed resting-state functional magnetic resonance imaging (fMRI) data from 68 depressed patients with BD-II (30 with SA and 38 without SA) and 35 healthy controls (HCs). The whole-brain dFC variability of corticostriatal circuitry was calculated using a sliding-window analysis. Their correlations with cognitive dysfunction were further detected. Support vector machine (SVM) classification tested the potential of dFC to differentiate BD-II with SA from HCs.

RESULTS

Increased dFC variability between the right vCa and the right insula was found in SA compared to non-SA and HCs, and negatively correlated with speed of processing. Decreased dFC variability between the left dlPu and the right postcentral gyrus was found in non-SA compared to SA and HCs, and positively correlated with reasoning problem-solving. Both SA and non-SA exhibited decreased dFC variability between the right dCa and the left MTG, and between the right dlPu and the right calcarine when compared to HCs. SVM classification achieved an accuracy of 75.24 % and AUC of 0.835 to differentiate SA from non-SA, while combining the abnormal dFC features between SA and non-SA.

CONCLUSIONS

Aberrant dFC variability of corticostriatal circuitry may serve as potential neuromarker for SA in BD-II, which might help to discriminate suicidal BD-II patients from non-suicidal patients and HCs.

SF-MVPA: A from raw data to statistical results and surface space-based MVPA toolbox

Compared with traditional volume space-based multivariate pattern analysis (MVPA), surface space-based MVPA has many advantages and has received increasing attention. However, surface space-based MVPA requires considerable programming and is therefore difficult for people without a programming foundation. To address this, we developed a MATLAB toolbox based on a graphical interactive interface (GUI) called surface space-based multivariate pattern analysis (SF-MVPA) in this manuscript. Unlike the traditional MVPA toolboxes, which often only include MVPA calculation processes after data preprocessing, SF-MVPA covers the complete pipeline of surface space-based MVPA, including raw data format conversion, surface reconstruction, functional magnetic resonance (fMRI) data preprocessing, comparative analysis, surface space-based MVPA, leave one-run out cross validation, and family-wise error correction. With SF-MVPA, users can complete the complete pipeline of surface space-based MVPA without programming. In addition, SF-MVPA is designed for parallel computing and hence has high computational efficiency. After introducing SF-MVPA, we analyzed a sample dataset of tonal working memory load. By comparison with another surface space-based MVPA toolbox named CoSMoMVPA, we found that the two toolboxes obtained consistent results. We hope that through this toolbox, users can more easily implement surface space-based MVPA.

The role of ascending arousal network in patients with chronic insomnia disorder

The ascending arousal system plays a crucial role in individuals' consciousness. Recently, advanced functional magnetic resonance imaging (fMRI) has made it possible to investigate the ascending arousal network (AAN) in vivo. However, the role of AAN in the neuropathology of human insomnia remains unclear. Our study aimed to explore alterations in AAN and its connections with cortical networks in chronic insomnia disorder (CID). Resting-state fMRI data were acquired from 60 patients with CID and 60 good sleeper controls (GSCs). Changes in the brain's functional connectivity (FC) between the AAN and eight cortical networks were detected in patients with CID and GSCs. Multivariate pattern analysis (MVPA) was employed to differentiate CID patients from GSCs and predict clinical symptoms in patients with CID. Finally, these MVPA findings were further verified using an external data set (32 patients with CID and 33 GSCs). Compared to GSCs, patients with CID exhibited increased FC within the AAN, as well as increased FC between the AAN and default mode, cerebellar, sensorimotor, and dorsal attention networks. These AAN-related FC patterns and the MVPA classification model could be used to differentiate CID patients from GSCs with 88% accuracy in the first cohort and 77% accuracy in the validation cohort. Moreover, the MVPA prediction models could separately predict insomnia (data set 1, R²  = .34; data set 2, R²  = .15) and anxiety symptoms (data set 1, R²  = .35; data set 2, R²  = .34) in the two independent cohorts of patients. Our findings indicated that AAN contributed to the neurobiological mechanism of insomnia and highlighted that fMRI-based markers and machine learning techniques might facilitate the evaluation of insomnia and its comorbid mental symptoms.

Widespread aberrant functional connectivity throughout the whole brain in obstructive sleep apnea

Objective

Obstructive sleep apnea (OSA) is a sleep-related breathing disorder with high prevalence and is associated with cognitive impairment. Previous neuroimaging studies have reported abnormal brain functional connectivity (FC) in patients with OSA that might contribute to their neurocognitive impairments. However, it is unclear whether patients with OSA have a characteristic pattern of FC changes that can serve as a neuroimaging biomarker for identifying OSA.

Methods

A total of 21 patients with OSA and 21 healthy controls (HCs) were included in this study and scanned using resting-state functional magnetic resonance imaging (fMRI). The automated anatomical labeling (AAL) atlas was used to divide the cerebrum into 90 regions, and FC between each pair of regions was calculated. Univariate analyses were then performed to detect abnormal FCs in patients with OSA compared with controls, and multivariate pattern analyses (MVPAs) were applied to classify between patients with OSA and controls.

Results

The univariate comparisons did not detect any significantly altered FC. However, the MVPA showed a successful classification between patients with OSA and controls with an accuracy of 83.33% (p = 0.0001). Furthermore, the selected FCs were associated with nearly all brain regions and widely distributed in the whole brain, both within and between, many resting-state functional networks. Among these selected FCs, 3 were significantly correlated with the apnea-hypopnea index (AHI) and 2 were significantly correlated with the percentage of time with the saturation of oxygen (SaO₂) below 90% of the total sleep time (%TST < 90%).

Conclusion

There existed widespread abnormal FCs in the whole brain in patients with OSA. This aberrant FC pattern has the potential to serve as a neurological biomarker of OSA, highlighting its importance for understanding the complex neural mechanism underlying OSA and its cognitive impairment.

Cerebellum regulating cerebral functional cortex through multiple pathways in complete thoracolumbar spinal cord injury

The previous studies have found significant brain structural and functional changes in cerebral regions after spinal cord injury (SCI), but few studies have explored the cerebellar–cerebral circuit changes in SCI. This study aims to study the brain structural changes of cerebellar subregions and its functional connectivity (FC) changes with cerebrum in complete thoracolumbar SCI (CTSCI), and screen out the regions that play relatively important roles in affecting sensorimotor function. Eighteen CTSCI patients and 18 age- and gender-matched healthy controls (HCs) were recruited. Voxel-based morphometry (VBM) was used to characterize the brain structural changes of cerebellar subregions [from the Anatomical Automatic Labeling (AAL116)], seed-based FC was used to evaluate the cerebellar–cerebral FC changes and support vector machine (SVM) analysis was used to search for sensitive imaging indicators. CTSCI patients showed slightly structural atrophy in vermis_3 (p = 0.046) and significantly decreased FC between cerebellum and cerebral sensorimotor-, visual-, cognitive-, and auditory-related regions (cluster-level FWE correction with p < 0.05). Additionally, SVM weight analysis showed that FC values between vermis_10 and right fusiform gyrus had the greatest weight in functional changes of CTSCI. In conclusion, different degrees of structural and functional changes occurred in each subregion of cerebellum following CTSCI, and FC change between vermis_10 and right fusiform gyrus plays the most important role in dysfunction and may become an important neural network index of rehabilitation therapy.

Differential grey matter structure in women with premenstrual dysphoric disorder: evidence from brain morphometry and data-driven classification

Premenstrual dysphoric disorder (PMDD) is a female-specific condition classified in the Diagnostic and Statical Manual-5th edition under depressive disorders. Alterations in grey matter volume, cortical thickness and folding metrics have been associated with a number of mood disorders, though little is known regarding brain morphological alterations in PMDD. Here, women with PMDD and healthy controls underwent magnetic resonance imaging (MRI) during the luteal phase of the menstrual cycle. Differences in grey matter structure between the groups were investigated by use of voxel- and surface-based morphometry. Machine learning and multivariate pattern analysis were performed to test whether MRI data could distinguish women with PMDD from healthy controls. Compared to controls, women with PMDD had smaller grey matter volume in ventral posterior cortices and the cerebellum (Cohen's d = 0.45-0.76). Region-of-interest analyses further indicated smaller volume in the right amygdala and putamen of women with PMDD (Cohen's d = 0.34-0.55). Likewise, thinner cortex was observed in women with PMDD compared to controls, particularly in the left hemisphere (Cohen's d = 0.20-0.74). Classification analyses showed that women with PMDD can be distinguished from controls based on grey matter morphology, with an accuracy up to 74%. In line with the hypothesis of an impaired top-down inhibitory circuit involving limbic structures in PMDD, the present findings point to PMDD-specific grey matter anatomy in regions of corticolimbic networks. Furthermore, the results include widespread cortical and cerebellar regions, suggesting the involvement of distinct networks in PMDD pathophysiology.

Dynamic and Static Amplitude of Low-Frequency Fluctuation Is a Potential Biomarker for Predicting Prognosis of Degenerative Cervical Myelopathy Patients: A Preliminary Resting-State fMRI Study

Objective

The aim of this study was to explore the clinical value of the static amplitude of low-frequency fluctuation (sALFF) and dynamic amplitude of low-frequency fluctuation (dALFF) in the identification of brain functional alterations in degenerative cervical myelopathy (DCM) patients.

Methods

Voxel-wise sALFF and dALFF of 47 DCM patients and 44 healthy controls were calculated using resting-state fMRI data, and an intergroup comparison was performed. The mean of sALFF or dALFF data were extracted within the resultant clusters and the correlation analysis of these data with the clinical measures was performed. Furthermore, whole-brain-wise and region-wise multivariate pattern analyses (MVPAs) were performed to classify DCM patients and healthy controls. sALFF and dALFF were used to predict the prognosis of DCM patients.

Results

The findings showed that (1) DCM patients exhibited higher sALFF within the left thalamus and putamen compared with that of the healthy controls. DCM patients also exhibited lower dALFF within bilateral postcentral gyrus compared with the healthy controls; (2) No significant correlations were observed between brain alterations and clinical measures through univariate correlation analysis; (3) sALFF (91%) and dALFF (95%) exhibited high accuracy in classifying the DCM patients and healthy controls; (4) Region-wise MVPA further revealed brain regions in which functional patterns were associated with prognosis in DCM patients. These regions were mainly located at the frontal lobe and temporal lobe.

Conclusion

In summary, sALFF and dALFF can be used to accurately reveal brain functional alterations in DCM patients. Furthermore, the multivariate approach is a more sensitive method in exploring neuropathology and establishing a prognostic biomarker for DCM compared with the conventional univariate method.

Machine learning approaches for parsing comorbidity/heterogeneity in antisociality and substance use disorders: A primer

By some accounts, as many as 93% of individuals diagnosed with antisocial personality disorder (ASPD) or psychopathy also meet criteria for some form of substance use disorder (SUD). This high level of comorbidity, combined with an overlapping biopsychosocial profile, and potentially interacting features, has made it difficult to delineate the shared/unique characteristics of each disorder. Moreover, while rarely acknowledged, both SUD and antisociality exist as highly heterogeneous disorders in need of more targeted parcellation. While emerging data-driven nosology for psychiatric disorders (e.g., Research Domain Criteria (RDoC), Hierarchical Taxonomy of Psychopathology (HiTOP)) offers the opportunity for a more systematic delineation of the externalizing spectrum, the interrogation of large, complex neuroimaging-based datasets may require data-driven approaches that are not yet widely employed in psychiatric neuroscience. With this in mind, the proposed article sets out to provide an introduction into machine learning methods for neuroimaging that can help parse comorbid, heterogeneous externalizing samples. The modest machine learning work conducted to date within the externalizing domain demonstrates the potential utility of the approach but remains highly nascent. Within the paper, we make suggestions for how future work can make use of machine learning methods, in combination with emerging psychiatric nosology systems, to further diagnostic and etiological understandings of the externalizing spectrum. Finally, we briefly consider some challenges that will need to be overcome to encourage further progress in the field.

A combination of support vector machine and voxel-based morphometry in adult male alcohol use disorder patients with cognitive deficits

Cognitive performance deteriorates with drinking. However, the neural basis of cognitive deficits in alcohol use disorder (AUD) is still incompletely understood. Here we examined the relationship between overall drinking, brain structural alterations and cognitive deficits in AUD. A total of 40 middle-aged AUD males and 40 healthy controls (HC) underwent high-resolution anatomical imaging scans, and the data were analyzed using voxel-based morphometry, support vector machine (SVM) classification and mediation analysis. The AUD patients demonstrated reduced gray matter (GM) volumes that included left amygdala, thalamus, hippocampus, precentral gyrus, cerebellum, calcarine, right supplementary motor area and bilateral superior temporal gyri (voxel-wise p < 0.05, FWE corrected). The SVM results could distinguish AUD from HC with satisfactory classification results (0.8275). GM volumes in the bilateral cerebellum and thalamus, left anterior medial temporal lobe, left nucleus ambiguus + parahippocampus gyrus, left fusiform gyrus, left lingual gyrus, left hippocampus, and right nucleus accumbens had positive correlations with the Montreal Cognitive Assessment (MoCA) scores. Further mediation analysis showed that left cerebellum crus 1 partially mediated the relationship between overall drinking and MoCA scores (standardized beta coefficient = -0.0973, SE = 0.0002, 95% CI = (-0.0006, 0.0000)). Our findings showed widespread GM atrophies and many of these atrophies also mirrored cognitive deficits and were robustly distinguishable. Critically, the left cerebellum crus 1 partially mediated the relationship betweem overall drinking and MoCA scores, suggesting a pathway by which alcohol abuse impairs cognition and accelerates brain ageing in middle-aged AUD males.

Identification and Therapeutic Outcome Prediction of Cervical Spondylotic Myelopathy Based on the Functional Connectivity From Resting-State Functional MRI Data: A Preliminary Machine Learning Study

Currently, strategies to diagnose patients and predict neurological recovery in cervical spondylotic myelopathy (CSM) using MR images of the cervical spine are urgently required. In light of this, this study aimed at exploring potential preoperative brain biomarkers that can be used to diagnose and predict neurological recovery in CSM patients using functional connectivity (FC) analysis of a resting-state functional MRI (rs-fMRI) data. Two independent datasets, including total of 53 patients with CSM and 47 age- and sex-matched healthy controls (HCs), underwent the preoperative rs-fMRI procedure. The FC was calculated from the automated anatomical labeling (AAL) template and used as features for machine learning analysis. After that, three analyses were used, namely, the classification of CSM patients from healthy adults using the support vector machine (SVM) within and across datasets, the prediction of preoperative neurological function in CSM patients via support vector regression (SVR) within and across datasets, and the prediction of neurological recovery in CSM patients via SVR within and across datasets. The results showed that CSM patients could be successfully identified from HCs with high classification accuracies (84.2% for dataset 1, 95.2% for dataset 2, and 73.0% for cross-site validation). Furthermore, the rs-FC combined with SVR could successfully predict the neurological recovery in CSM patients. Additionally, our results from cross-site validation analyses exhibited good reproducibility and generalization across the two datasets. Therefore, our findings provide preliminary evidence toward the development of novel strategies to predict neurological recovery in CSM patients using rs-fMRI and machine learning technique.

The Surface Crack Extraction Method Based on Machine Learning of Image and Quantitative Feature Information Acquisition Method

In order to effectively control the damage caused by surface cracks to a geological environment, we need to find a convenient, efficient, and accurate method to obtain crack information. The existing crack extraction methods based on unmanned air vehicle (UAV) images inevitably have some erroneous pixels because of the complexity of background information. At the same time, there are few researches on crack feature information. In view of this, this article proposes a surface crack extraction method based on machine learning of UAV images, the data preprocessing steps, and the content and calculation methods for crack feature information: length, width, direction, location, fractal dimension, number, crack rate, and dispersion rate. The results show that the method in this article can effectively avoid the interference by vegetation and soil crust. By introducing the concept of dispersion rate, the method combining crack rate and dispersion rate can describe the distribution characteristics of regional cracks more clearly. Compared to field survey data, the calculation result of the crack feature information in this article is close to the true value, which proves that this is a reliable method for obtaining quantitative crack feature information.

A Systematic Characterization of Structural Brain Changes in Schizophrenia

A systematic characterization of the similarities and differences among different methods for detecting structural brain abnormalities in schizophrenia, such as voxel-based morphometry (VBM), tensor-based morphometry (TBM), and projection-based thickness (PBT), is important for understanding the brain pathology in schizophrenia and for developing effective biomarkers for a diagnosis of schizophrenia. However, such studies are still lacking. Here, we performed VBM, TBM, and PBT analyses on T1-weighted brain MR images acquired from 116 patients with schizophrenia and 116 healthy controls. We found that, although all methods detected wide-spread structural changes, different methods captured different information - only 10.35% of the grey matter changes in cortex were detected by all three methods, and VBM only detected 11.36% of the white matter changes detected by TBM. Further, pattern classification between patients and controls revealed that combining different measures improved the classification accuracy (81.9%), indicating that fusion of different structural measures serves as a better neuroimaging marker for the objective diagnosis of schizophrenia.