Multivariate Pattern Recognition for Diagnosis and Prognosis in Clinical Neuroimaging: State of the Art, Current Challenges and Future Trends

Cognitive enrichment through art: a randomized controlled trial on the effect of music or visual arts group practice on cognitive and brain development of young children

BACKGROUND

The optimal stimulation for brain development in the early academic years remains unclear. Current research suggests that musical training has a more profound impact on children's executive functions (EF) compared to other art forms. What is crucially lacking is a large-scale, long-term genuine randomized controlled trial (RCT) in cognitive neuroscience, comparing musical instrumental training (MIP) to another art form, and a control group (CG). This study aims to fill this gap by using machine learning to develop a multivariate model that tracks the interconnected brain and EF development during the academic years, with or without music or other art training.

METHODS

The study plans to enroll 150 children aged 6-8 years and randomly assign them to three groups: Orchestra in Class (OC), Visual Arts (VA), and a control group (CG). Anticipating a 30% attrition rate, each group aims to retain at least 35 participants. The research consists of three analytical stages: 1) baseline analysis correlating EF, brain data, age, gender, and socioeconomic status, 2) comparison between groups and over time of EF brain and behavioral development and their interactions, including hypothesis testing, and 3) exploratory analysis combining behavioral and brain data. The intervention includes intensive art classes once a week, and incremental home training over two years, with the CG receiving six annual cultural outings.

DISCUSSION

This study examines the potential benefits of intensive group arts education, especially contrasting music with visual arts, on EF development in children. It will investigate how artistic enrichment potentially influences the presumed typical transition from a more unified to a more multifaceted EF structure around age eight, comparing these findings against a minimally enriched active control group. This research could significantly influence the incorporation of intensive art interventions in standard curricula.

TRIAL REGISTRATION

The project was accepted after peer-review by the Swiss National Science Foundation (SNSF no. 100014_214977) on March 29, 2023. The study protocol received approval from the Cantonal Commission for Ethics in Human Research of Geneva (CCER, BASEC-ID 2023-01016), which is part of Swiss ethics, on October 25, 2023. The study is registered at clinicaltrials.gov (NCT05912270).

Artificial intelligence: opportunities and challenges in the clinical applications of triple-negative breast cancer

Triple-negative breast cancer (TNBC) accounts for 15-20% of all invasive breast cancer subtypes. Owing to its clinical characteristics, such as the lack of effective therapeutic targets, high invasiveness, and high recurrence rate, TNBC is difficult to treat and has a poor prognosis. Currently, with the accumulation of large amounts of medical data and the development of computing technology, artificial intelligence (AI), particularly machine learning, has been applied to various aspects of TNBC research, including early screening, diagnosis, identification of molecular subtypes, personalised treatment, and prediction of prognosis and treatment response. In this review, we discussed the general principles of artificial intelligence, summarised its main applications in the diagnosis and treatment of TNBC, and provided new ideas and theoretical basis for the clinical diagnosis and treatment of TNBC.

Musical and psychomotor interventions for cognitive, sensorimotor, and cerebral decline in patients with Mild Cognitive Impairment (COPE): a study protocol for a multicentric randomized controlled study

BACKGROUND

Regular cognitive training can boost or maintain cognitive and brain functions known to decline with age. Most studies administered such cognitive training on a computer and in a lab setting. However, everyday life activities, like musical practice or physical exercise that are complex and variable, might be more successful at inducing transfer effects to different cognitive domains and maintaining motivation. "Body-mind exercises", like Tai Chi or psychomotor exercise, may also positively affect cognitive functioning in the elderly. We will compare the influence of active music practice and psychomotor training over 6 months in Mild Cognitive Impairment patients from university hospital memory clinics on cognitive and sensorimotor performance and brain plasticity. The acronym of the study is COPE (Countervail cOgnitive imPairmEnt), illustrating the aim of the study: learning to better "cope" with cognitive decline.

METHODS

We aim to conduct a randomized controlled multicenter intervention study on 32 Mild Cognitive Impairment (MCI) patients (60-80 years), divided over 2 experimental groups: 1) Music practice; 2) Psychomotor treatment. Controls will consist of a passive test-retest group of 16 age, gender and education level matched healthy volunteers. The training regimens take place twice a week for 45 min over 6 months in small groups, provided by professionals, and patients should exercise daily at home. Data collection takes place at baseline (before the interventions), 3, and 6 months after training onset, on cognitive and sensorimotor capacities, subjective well-being, daily living activities, and via functional and structural neuroimaging. Considering the current constraints of the COVID-19 pandemic, recruitment and data collection takes place in 3 waves.

DISCUSSION

We will investigate whether musical practice contrasted to psychomotor exercise in small groups can improve cognitive, sensorimotor and brain functioning in MCI patients, and therefore provoke specific benefits for their daily life functioning and well-being.

TRIAL REGISTRATION

The full protocol was approved by the Commission cantonale d'éthique de la recherche sur l'être humain de Genève (CCER, no. 2020-00510) on 04.05.2020, and an amendment by the CCER and the Commission cantonale d'éthique de la recherche sur l'être humain de Vaud (CER-VD) on 03.08.2021. The protocol was registered at clinicaltrials.gov (20.09.2020, no. NCT04546451).

Tinnitus and distress: an electroencephalography classification study

There exist no objective markers for tinnitus or tinnitus disorders, which complicates diagnosis and treatments. The combination of EEG with sophisticated classification procedures may reveal biomarkers that can identify tinnitus and accurately differentiate different levels of distress experienced by patients. EEG recordings were obtained from 129 tinnitus patients and 142 healthy controls. Linear support vector machines were used to develop two classifiers: the first differentiated tinnitus patients from controls, while the second differentiated tinnitus patients with low and high distress levels. The classifier for healthy controls and tinnitus patients performed with an average accuracy of 96 and 94% for the training and test sets, respectively. For the distress classifier, these average accuracies were 89 and 84%. Minimal overlap was observed between the features of the two classifiers. EEG-derived features made it possible to accurately differentiate healthy controls and tinnitus patients as well as low and high distress tinnitus patients. The minimal overlap between the features of the two classifiers indicates that the source of distress in tinnitus, which could also be involved in distress related to other conditions, stems from different neuronal mechanisms compared to those causing the tinnitus pathology itself.

Post-stroke reorganization of transient brain activity characterizes deficits and recovery of cognitive functions

Functional magnetic resonance imaging (fMRI) has been widely employed to study stroke pathophysiology. In particular, analyses of fMRI signals at rest were directed at quantifying the impact of stroke on spatial features of brain networks. However, brain networks have intrinsic time features that were, so far, disregarded in these analyses. In consequence, standard fMRI analysis failed to capture temporal imbalance resulting from stroke lesions, hence restricting their ability to reveal the interdependent pathological changes in structural and temporal network features following stroke. Here, we longitudinally analyzed hemodynamic-informed transient activity in a large cohort of stroke patients (n = 103) to assess spatial and temporal changes of brain networks after stroke. Metrics extracted from the hemodynamic-informed transient activity were replicable within- and between-individuals in healthy participants, hence supporting their robustness and their clinical applicability. While large-scale spatial patterns of brain networks were preserved after stroke, their durations were altered, with stroke subjects exhibiting a varied pattern of longer and shorter network activations compared to healthy individuals. Specifically, patients showed a longer duration in the lateral precentral gyrus and anterior cingulum, and a shorter duration in the occipital lobe and in the cerebellum. These temporal alterations were associated with white matter damage in projection and association pathways. Furthermore, they were tied to deficits in specific behavioral domains as restoration of healthy brain dynamics paralleled recovery of cognitive functions (attention, language and spatial memory), but was not significantly correlated to motor recovery. These findings underscore the critical importance of network temporal properties in dissecting the pathophysiology of brain changes after stroke, thus shedding new light on the clinical potential of time-resolved methods for fMRI analysis.

Deep learning model of fMRI connectivity predicts PTSD symptom trajectories in recent trauma survivors

Early intervention following exposure to a traumatic life event could change the clinical path from the development of post traumatic stress disorder (PTSD) to recovery, hence the interest in early detection and underlying biological mechanisms involved in the development of post traumatic sequelae. We introduce a novel end-to-end neural network that employs resting-state and task-based functional MRI (fMRI) datasets, obtained one month after trauma exposure, to predict PTSD symptoms at one-, six- and fourteen-months after the exposure. FMRI data, as well as PTSD status and symptoms, were collected from adults at risk for PTSD development, after admission to emergency room following a traumatic event. Our computational method utilized a per-region encoder to extract brain regions embedding, which were subsequently updated by applying the algorithmic technique of pairwise attention. The affinities obtained between each pair of regions were combined to create a pairwise co-activation map used to perform multi-label classification. The results demonstrate that the novel method's performance in predicting PTSD symptoms, in a prospective manner, outperforms previous analytical techniques reported in the fMRI literature, all trained on the same dataset. We further show a high predictive ability for predicting PTSD symptom clusters and PTSD persistence. To the best of our knowledge, this is the first deep learning method applied on fMRI data with respect to prospective clinical outcomes, to predict PTSD status, severity and symptom clusters. Future work could further delineate the mechanisms that underlie such a prediction, and potentially improve single patient characterization.

Train the brain with music (TBM): brain plasticity and cognitive benefits induced by musical training in elderly people in Germany and Switzerland, a study protocol for an RCT comparing musical instrumental practice to sensitization to music

BACKGROUND

Recent data suggest that musical practice prevents age-related cognitive decline. But experimental evidence remains sparse and no concise information on the neurophysiological bases exists, although cognitive decline represents a major impediment to healthy aging. A challenge in the field of aging is developing training regimens that stimulate neuroplasticity and delay or reverse symptoms of cognitive and cerebral decline. To be successful, these regimens should be easily integrated in daily life and intrinsically motivating. This study combines for the first-time protocolled music practice in elderly with cutting-edge neuroimaging and behavioral approaches, comparing two types of musical education.

METHODS

We conduct a two-site Hannover-Geneva randomized intervention study in altogether 155 retired healthy elderly (64-78) years, (63 in Geneva, 92 in Hannover), offering either piano instruction (experimental group) or musical listening awareness (control group). Over 12 months all participants receive weekly training for 1 hour, and exercise at home for ~ 30 min daily. Both groups study different music styles. Participants are tested at 4 time points (0, 6, and 12 months & post-training (18 months)) on cognitive and perceptual-motor aptitudes as well as via wide-ranging functional and structural neuroimaging and blood sampling.

DISCUSSION

We aim to demonstrate positive transfer effects for faculties traditionally described to decline with age, particularly in the piano group: executive functions, working memory, processing speed, abstract thinking and fine motor skills. Benefits in both groups may show for verbal memory, hearing in noise and subjective well-being. In association with these behavioral benefits we anticipate functional and structural brain plasticity in temporal (medial and lateral), prefrontal and parietal areas and the basal ganglia. We intend exhibiting for the first time that musical activities can provoke important societal impacts by diminishing cognitive and perceptual-motor decline supported by functional and structural brain plasticity.

TRIAL REGISTRATION

The Ethikkomission of the Leibniz Universität Hannover approved the protocol on 14.08.17 (no. 3604-2017), the neuroimaging part and blood sampling was approved by the Hannover Medical School on 07.03.18. The full protocol was approved by the Commission cantonale d'éthique de la recherche de Genève (no. 2016-02224) on 27.02.18 and registered at clinicaltrials.gov on 17.09.18 ( NCT03674931 , no. 81185).

Different patterns of cerebral perfusion in SLE patients with and without neuropsychiatric manifestations

To investigate brain perfusion patterns in systemic lupus erythematosus (SLE) patients with and without neuropsychiatric systemic lupus erythematosus (NPSLE and non-NPSLE, respectively) and to identify biomarkers for the diagnosis of NPSLE using noninvasive three-dimensional (3D) arterial spin labeling (ASL). Thirty-one NPSLE and 24 non-NPSLE patients and 32 age- and sex-matched normal controls (NCs) were recruited. Three-dimensional ASL-MRI was applied to quantify cerebral perfusion. Whole brain, gray (GM) and white matter (WM), and voxel-based analysis (VBA) were performed to explore perfusion characteristics. Correlation analysis was performed to find the relationship between the perfusion measures, lesion volumes, and clinical variables. Receiver operating characteristic (ROC) analysis and support vector machine (SVM) classification were applied to differentiate NPSLE patients from non-NPSLE patients and healthy controls. Compared to NCs, NPSLE patients showed increased cerebral blood flow (CBF) within WM but decreased CBF within GM, while non-NPSLE patients showed increased CBF within both GM and WM. Compared to non-NPSLE patients, NPSLE patients showed significantly reduced CBF in the frontal gyrus, cerebellum, and corpus callosum. CBF within several brain regions such as cingulate and corpus callosum showed significant correlations with the Systemic Lupus Erythematosus Disease Activity Index (SLEDAI) and the Systemic Lupus International Collaborating Clinics (SLICC) damage index scores. ROC analysis showed moderate performance in distinguishing NPSLE from non-NPSLE patients with AUCs > 0.7, while SVM analysis demonstrated that CBF within the corpus callosum achieved an accuracy of 83.6% in distinguishing NPSLE from non-NPSLE patients. Different brain perfusion patterns were observed between NPSLE and non-NPSLE patients. CBF measured by noninvasive 3D ASL could be a useful biomarker for the diagnosis and disease monitoring of NPSLE and non-NPSLE patients.

A totally data-driven whole-brain multimodal pipeline for the discrimination of Parkinson's disease, multiple system atrophy and healthy control

Parkinson's Disease (PD) and Multiple System Atrophy (MSA) are two parkinsonian syndromes that share many symptoms, albeit having very different prognosis. Although previous studies have proposed multimodal MRI protocols combined with multivariate analysis to discriminate between these two populations and healthy controls, studies combining all MRI indexes relevant for these disorders (i.e. grey matter volume, fractional anisotropy, mean diffusivity, iron deposition, brain activity at rest and brain connectivity) with a completely data-driven voxelwise analysis for discrimination are still lacking. In this study, we used such a complete MRI protocol and adapted a fully-data driven analysis pipeline to discriminate between these populations and a healthy controls (HC) group. The pipeline combined several feature selection and reduction steps to obtain interpretable models with a low number of discriminant features that can shed light onto the brain pathology of PD and MSA. Using this pipeline, we could discriminate between PD and HC (best accuracy = 0.78), MSA and HC (best accuracy = 0.94) and PD and MSA (best accuracy = 0.88). Moreover, we showed that indexes derived from resting-state fMRI alone could discriminate between PD and HC, while mean diffusivity in the cerebellum and the putamen alone could discriminate between MSA and HC. On the other hand, a more diverse set of indexes derived by multiple modalities was needed to discriminate between the two disorders. We showed that our pipeline was able to discriminate between distinct pathological populations while delivering sparse model that could be used to better understand the neural underpinning of the pathologies.

•

Structuro-functional MRI can discriminate between parkinsonian syndromes

•

Discriminant MRI modalities vary as a function of the discrimination task

•

fMRI is crucial in discriminating between Parkinson's disease patients and controls

•

Structural MRI discriminate between Multiple System Atrophy patients and controls

Cortical thickness in major depressive disorder: A systematic review and meta-analysis

Neuroimaging studies assessing neurobiological differences between patients with major depressive disorder (MDD) and healthy controls (HC) are often hindered by small sample sizes and heterogeneity of the patient sample. We performed a comprehensive literature search for studies assessing cortical thickness between patient and control groups, including studies investigating treatment effects on cortical thickness. We identified 34 studies meeting criteria for the systematic review and used Seed-based d Mapping to meta-analyze 24 of those that met additional criteria. Analysis of the full sample of subjects (MDD = 1073; HC = 936) revealed significant thinning in the MDD group in the bilateral orbitofrontal gyrus (BA 11), left pars opercularis (BA 45) and left calcarine fissure/lingual gyrus (BA 17), as well as an area of significant thickening in the left supramarginal gyrus (BA 40). These results support other imaging modalities that report disruptions in various frontal and temporal areas in MDD and identify additional areas in all major cerebral lobes likely to be significant when parsing for biomarkers of treatment or relapse.

Multimodal Neuroimaging: Basic Concepts and Classification of Neuropsychiatric Diseases

Neuroimaging techniques are widely used in neuroscience to visualize neural activity, to improve our understanding of brain mechanisms, and to identify biomarkers-especially for psychiatric diseases; however, each neuroimaging technique has several limitations. These limitations led to the development of multimodal neuroimaging (MN), which combines data obtained from multiple neuroimaging techniques, such as electroencephalography, functional magnetic resonance imaging, and yields more detailed information about brain dynamics. There are several types of MN, including visual inspection, data integration, and data fusion. This literature review aimed to provide a brief summary and basic information about MN techniques (data fusion approaches in particular) and classification approaches. Data fusion approaches are generally categorized as asymmetric and symmetric. The present review focused exclusively on studies based on symmetric data fusion methods (data-driven methods), such as independent component analysis and principal component analysis. Machine learning techniques have recently been introduced for use in identifying diseases and biomarkers of disease. The machine learning technique most widely used by neuroscientists is classification-especially support vector machine classification. Several studies differentiated patients with psychiatric diseases and healthy controls with using combined datasets. The common conclusion among these studies is that the prediction of diseases increases when combining data via MN techniques; however, there remain a few challenges associated with MN, such as sample size. Perhaps in the future N-way fusion can be used to combine multiple neuroimaging techniques or nonimaging predictors (eg, cognitive ability) to overcome the limitations of MN.

Recognition of Schizophrenia with Regularized Support Vector Machine and Sequential Region of Interest Selection using Structural Magnetic Resonance Imaging

Structural brain abnormalities in schizophrenia have been well characterized with the application of univariate methods to magnetic resonance imaging (MRI) data. However, these traditional techniques lack sensitivity and predictive value at the individual level. Machine-learning approaches have emerged as potential diagnostic and prognostic tools. We used an anatomically and spatially regularized support vector machine (SVM) framework to categorize schizophrenia and healthy individuals based on whole-brain gray matter densities estimated using voxel-based morphometry from structural MRI scans. The regularized SVM model yielded recognition accuracy of 86.6% in the training set of 127 individuals and validation accuracy of 83.5% in an independent set of 85 individuals. A sequential region-of-interest (ROI) selection step was adopted for feature selection, improving recognition accuracy to 92.0% in the training set and 89.4% in the validation set. The combined model achieved 96.6% sensitivity and 74.1% specificity. Seven ROIs were identified as the optimal discriminatory subset: the occipital fusiform gyrus, middle frontal gyrus, pars opercularis of the inferior frontal gyrus, anterior superior temporal gyrus, superior frontal gyrus, left thalamus and left lateral ventricle. These findings demonstrate the utility of spatial and anatomical priors in SVM for neuroimaging analyses in conjunction with sequential ROI selection in the recognition of schizophrenia.

Machine learning classification of first-episode schizophrenia spectrum disorders and controls using whole brain white matter fractional anisotropy

BACKGROUND

Early diagnosis of schizophrenia could improve the outcome of the illness. Unlike classical between-group comparisons, machine learning can identify subtle disease patterns on a single subject level, which could help realize the potential of MRI in establishing a psychiatric diagnosis. Machine learning has previously been predominantly tested on gray-matter structural or functional MRI data. In this paper we used a machine learning classifier to differentiate patients with a first episode of schizophrenia-spectrum disorder (FES) from healthy controls using diffusion tensor imaging.

METHODS

We applied linear support-vector machine (SVM) and traditional tract based spatial statistics between group analyses to brain fractional anisotropy (FA) data from 77 FES and 77 age and sex matched healthy controls. We also evaluated the effects of medication and symptoms on the SVM classification.

RESULTS

The SVM distinguished between patients and controls with significant accuracy of 62.34% (p = 0.005). Participants with FES showed widespread FA reductions relative to controls in a large cluster (N  = 56,647 voxels, corrected p = 0.002). The white matter regions, which contributed to the correct identification of participants with FES, overlapped with the regions, which showed lower FA in patients relative to controls. There was no association between the classification performance and medication or symptoms.

CONCLUSIONS

Our results provide a proof of concept that SVM might help differentiate FES patients early in the course of illness from healthy controls using white-matter fractional anisotropy. As there was no effect of medications or symptoms, the SVM classification seemed to be based on trait rather than state markers and appeared to capture the lower FA in FES participants relative to controls.

Combination of Structural MRI and FDG-PET of the Brain Improves Diagnostic Accuracy in Newly Manifested Cognitive Impairment in Geriatric Inpatients

BACKGROUND

The cause of cognitive impairment in acutely hospitalized geriatric patients is often unclear. The diagnostic process is challenging but important in order to treat potentially life-threatening etiologies or identify underlying neurodegenerative disease.

OBJECTIVE

To evaluate the add-on diagnostic value of structural and metabolic neuroimaging in newly manifested cognitive impairment in elderly geriatric inpatients.

METHODS

Eighty-one inpatients (55 females, 81.6±5.5 y) without history of cognitive complaints prior to hospitalization were recruited in 10 acute geriatrics clinics. Primary inclusion criterion was a clinical hypothesis of Alzheimer's disease (AD), cerebrovascular disease (CVD), or mixed AD+CVD etiology (MD), which remained uncertain after standard diagnostic workup. Additional procedures performed after enrollment included detailed neuropsychological testing and structural MRI and FDG-PET of the brain. An interdisciplinary expert team established the most probable etiologic diagnosis (non-neurodegenerative, AD, CVD, or MD) integrating all available data. Automatic multimodal classification based on Random Undersampling Boosting was used for rater-independent assessment of the complementary contribution of the additional diagnostic procedures to the etiologic diagnosis.

RESULTS

Automatic 4-class classification based on all diagnostic routine standard procedures combined reproduced the etiologic expert diagnosis in 31% of the patients (p = 0.100, chance level 25%). Highest accuracy by a single modality was achieved by MRI or FDG-PET (both 45%, p≤0.001). Integration of all modalities resulted in 76% accuracy (p≤0.001).

CONCLUSION

These results indicate substantial improvement of diagnostic accuracy in uncertain de novo cognitive impairment in acutely hospitalized geriatric patients with the integration of structural MRI and brain FDG-PET into the diagnostic process.

Sample heterogeneity in unipolar depression as assessed by functional connectivity analyses is dominated by general disease effects

OBJECTIVES

Combinations of resting-state fMRI and machine-learning techniques are increasingly employed to develop diagnostic models for mental disorders. However, little is known about the neurobiological heterogeneity of depression and diagnostic machine learning has mainly been tested in homogeneous samples. Our main objective was to explore the inherent structure of a diverse unipolar depression sample. The secondary objective was to assess, if such information can improve diagnostic classification.

MATERIALS AND METHODS

We analyzed data from 360 patients with unipolar depression and 360 non-depressed population controls, who were subdivided into two independent subsets. Cluster analyses (unsupervised learning) of functional connectivity were used to generate hypotheses about potential patient subgroups from the first subset. The relationship of clusters with demographical and clinical measures was assessed. Subsequently, diagnostic classifiers (supervised learning), which incorporated information about these putative depression subgroups, were trained.

RESULTS

Exploratory cluster analyses revealed two weakly separable subgroups of depressed patients. These subgroups differed in the average duration of depression and in the proportion of patients with concurrently severe depression and anxiety symptoms. The diagnostic classification models performed at chance level.

LIMITATIONS

It remains unresolved, if subgroups represent distinct biological subtypes, variability of continuous clinical variables or in part an overfitting of sparsely structured data.

CONCLUSIONS

Functional connectivity in unipolar depression is associated with general disease effects. Cluster analyses provide hypotheses about potential depression subtypes. Diagnostic models did not benefit from this additional information regarding heterogeneity.

On the integrity of functional brain networks in schizophrenia, Parkinson's disease, and advanced age: Evidence from connectivity-based single-subject classification

Previous whole-brain functional connectivity studies achieved successful classifications of patients and healthy controls but only offered limited specificity as to affected brain systems. Here, we examined whether the connectivity patterns of functional systems affected in schizophrenia (SCZ), Parkinson's disease (PD), or normal aging equally translate into high classification accuracies for these conditions. We compared classification performance between pre-defined networks for each group and, for any given network, between groups. Separate support vector machine classifications of 86 SCZ patients, 80 PD patients, and 95 older adults relative to their matched healthy/young controls, respectively, were performed on functional connectivity in 12 task-based, meta-analytically defined networks using 25 replications of a nested 10-fold cross-validation scheme. Classification performance of the various networks clearly differed between conditions, as those networks that best classified one disease were usually non-informative for the other. For SCZ, but not PD, emotion-processing, empathy, and cognitive action control networks distinguished patients most accurately from controls. For PD, but not SCZ, networks subserving autobiographical or semantic memory, motor execution, and theory-of-mind cognition yielded the best classifications. In contrast, young-old classification was excellent based on all networks and outperformed both clinical classifications. Our pattern-classification approach captured associations between clinical and developmental conditions and functional network integrity with a higher level of specificity than did previous whole-brain analyses. Taken together, our results support resting-state connectivity as a marker of functional dysregulation in specific networks known to be affected by SCZ and PD, while suggesting that aging affects network integrity in a more global way. Hum Brain Mapp 38:5845-5858, 2017. © 2017 Wiley Periodicals, Inc.

Support Vector Machine Analysis of Functional Magnetic Resonance Imaging of Interoception Does Not Reliably Predict Individual Outcomes of Cognitive Behavioral Therapy in Panic Disorder with Agoraphobia

BACKGROUND

The approach to apply multivariate pattern analyses based on neuro imaging data for outcome prediction holds out the prospect to improve therapeutic decisions in mental disorders. Patients suffering from panic disorder with agoraphobia (PD/AG) often exhibit an increased perception of bodily sensations. The purpose of this investigation was to assess whether multivariate classification applied to a functional magnetic resonance imaging (fMRI) interoception paradigm can predict individual responses to cognitive behavioral therapy (CBT) in PD/AG.

METHODS

This analysis is based on pretreatment fMRI data during an interoceptive challenge from a multicenter trial of the German PANIC-NET. Patients with DSM-IV PD/AG were dichotomized as responders (n = 30) or non-responders (n = 29) based on the primary outcome (Hamilton Anxiety Scale Reduction ≥50%) after 6 weeks of CBT (2 h/week). fMRI parametric maps were used as features for response classification with linear support vector machines (SVM) with or without automated feature selection. Predictive accuracies were assessed using cross validation and permutation testing. The influence of methodological parameters and the predictive ability for specific interoception-related symptom reduction were further evaluated.

RESULTS

SVM did not reach sufficient overall predictive accuracies (38.0-54.2%) for anxiety reduction in the primary outcome. In the exploratory analyses, better accuracies (66.7%) were achieved for predicting interoception-specific symptom relief as an alternative outcome domain. Subtle information regarding this alternative response criterion but not the primary outcome was revealed by post hoc univariate comparisons.

CONCLUSION

In contrast to reports on other neurofunctional probes, SVM based on an interoception paradigm was not able to reliably predict individual response to CBT. Results speak against the clinical applicability of this technique.

Diagnostic classification of unipolar depression based on resting-state functional connectivity MRI: effects of generalization to a diverse sample

In small, selected samples, an approach combining resting-state functional connectivity MRI and multivariate pattern analysis has been able to successfully classify patients diagnosed with unipolar depression. Purposes of this investigation were to assess the generalizability of this approach to a large clinically more realistic sample and secondarily to assess the replicability of previously reported methodological feasibility in a more homogeneous subgroup with pronounced depressive symptoms. Two independent subsets were drawn from the depression and control cohorts of the BiDirect study, each with 180 patients with and 180 controls without depression. Functional connectivity either among regions covering the gray matter or selected regions with known alterations in depression was assessed by resting-state fMRI. Support vector machines with and without automated feature selection were used to train classifiers differentiating between individual patients and controls in the entire first subset as well as in the subgroup. Model parameters were explored systematically. The second independent subset was used for validation of successful models. Classification accuracies in the large, heterogeneous sample ranged from 45.0 to 56.1% (chance level 50.0%). In the subgroup with higher depression severity, three out of 90 models performed significantly above chance (60.8-61.7% at independent validation). In conclusion, common classification methods previously successful in small homogenous depression samples do not immediately translate to a more realistic population. Future research to develop diagnostic classification approaches in depression should focus on more specific clinical questions and consider heterogeneity, including symptom severity as an important factor.

Connectivity of the anterior insula differentiates participants with first-episode schizophrenia spectrum disorders from controls: a machine-learning study

BACKGROUND

Early diagnosis of schizophrenia could improve the outcomes and limit the negative effects of untreated illness. Although participants with schizophrenia show aberrant functional connectivity in brain networks, these between-group differences have a limited diagnostic utility. Novel methods of magnetic resonance imaging (MRI) analyses, such as machine learning (ML), may help bring neuroimaging from the bench to the bedside. Here, we used ML to differentiate participants with a first episode of schizophrenia-spectrum disorder (FES) from healthy controls based on resting-state functional connectivity (rsFC).

METHOD

We acquired resting-state functional MRI data from 63 patients with FES who were individually matched by age and sex to 63 healthy controls. We applied linear kernel support vector machines (SVM) to rsFC within the default mode network, the salience network and the central executive network.

RESULTS

The SVM applied to the rsFC within the salience network distinguished the FES from the control participants with an accuracy of 73.0% (p = 0.001), specificity of 71.4% and sensitivity of 74.6%. The classification accuracy was not significantly affected by medication dose, or by the presence of psychotic symptoms. The functional connectivity within the default mode or the central executive networks did not yield classification accuracies above chance level.

CONCLUSIONS

Seed-based functional connectivity maps can be utilized for diagnostic classification, even early in the course of schizophrenia. The classification was probably based on trait rather than state markers, as symptoms or medications were not significantly associated with classification accuracy. Our results support the role of the anterior insula/salience network in the pathophysiology of FES.

Discriminative Analysis of Migraine without Aura: Using Functional and Structural MRI with a Multi-Feature Classification Approach

Magnetic resonance imaging (MRI) is by nature a multi-modality technique that provides complementary information about different aspects of diseases. So far no attempts have been reported to assess the potential of multi-modal MRI in discriminating individuals with and without migraine, so in this study, we proposed a classification approach to examine whether or not the integration of multiple MRI features could improve the classification performance between migraine patients without aura (MWoA) and healthy controls. Twenty-one MWoA patients and 28 healthy controls participated in this study. Resting-state functional MRI data was acquired to derive three functional measures: the amplitude of low-frequency fluctuations, regional homogeneity and regional functional correlation strength; and structural MRI data was obtained to measure the regional gray matter volume. For each measure, the values of 116 pre-defined regions of interest were extracted as classification features. Features were first selected and combined by a multi-kernel strategy; then a support vector machine classifier was trained to distinguish the subjects at individual level. The performance of the classifier was evaluated using a leave-one-out cross-validation method, and the final classification accuracy obtained was 83.67% (with a sensitivity of 92.86% and a specificity of 71.43%). The anterior cingulate cortex, prefrontal cortex, orbitofrontal cortex and the insula contributed the most discriminative features. In general, our proposed framework shows a promising classification capability for MWoA by integrating information from multiple MRI features.

Independent Component Analysis-Support Vector Machine-Based Computer-Aided Diagnosis System for Alzheimer's with Visual Support

Computer-aided diagnosis (CAD) systems constitute a powerful tool for early diagnosis of Alzheimer's disease (AD), but limitations on interpretability and performance exist. In this work, a fully automatic CAD system based on supervised learning methods is proposed to be applied on segmented brain magnetic resonance imaging (MRI) from Alzheimer's disease neuroimaging initiative (ADNI) participants for automatic classification. The proposed CAD system possesses two relevant characteristics: optimal performance and visual support for decision making. The CAD is built in two stages: a first feature extraction based on independent component analysis (ICA) on class mean images and, secondly, a support vector machine (SVM) training and classification. The obtained features for classification offer a full graphical representation of the images, giving an understandable logic in the CAD output, that can increase confidence in the CAD support. The proposed method yields classification results up to 89% of accuracy (with 92% of sensitivity and 86% of specificity) for normal controls (NC) and AD patients, 79% of accuracy (with 82% of sensitivity and 76% of specificity) for NC and mild cognitive impairment (MCI), and 85% of accuracy (with 85% of sensitivity and 86% of specificity) for MCI and AD patients.

Multimodal MRI-Based Classification of Trauma Survivors with and without Post-Traumatic Stress Disorder

Post-traumatic stress disorder (PTSD) is a debilitating psychiatric disorder. It can be difficult to discern the symptoms of PTSD and obtain an accurate diagnosis. Different magnetic resonance imaging (MRI) modalities focus on different aspects, which may provide complementary information for PTSD discrimination. However, none of the published studies assessed the diagnostic potential of multimodal MRI in identifying individuals with and without PTSD. In the current study, we investigated whether the complementary information conveyed by multimodal MRI scans could be combined to improve PTSD classification performance. Structural and resting-state functional MRI (rs-fMRI) scans were conducted on 17 PTSD patients, 20 trauma-exposed controls without PTSD (TEC) and 20 non-traumatized healthy controls (HC). Gray matter volume (GMV), amplitude of low-frequency fluctuations (ALFF), and regional homogeneity were extracted as classification features, and in order to integrate the information of structural and functional MRI data, the extracted features were combined by a multi-kernel combination strategy. Then a support vector machine (SVM) classifier was trained to distinguish the subjects at individual level. The performance of the classifier was evaluated using the leave-one-out cross-validation (LOOCV) method. In the pairwise comparison of PTSD, TEC, and HC groups, classification accuracies obtained by the proposed approach were 2.70, 2.50, and 2.71% higher than the best single feature way, with the accuracies of 89.19, 90.00, and 67.57% for PTSD vs. HC, TEC vs. HC, and PTSD vs. TEC respectively. The proposed approach could improve PTSD identification at individual level. Additionally, it provides preliminary support to develop the multimodal MRI method as a clinical diagnostic aid.

Multivariate pattern analysis strategies in detection of remitted major depressive disorder using resting state functional connectivity

Understanding abnormal resting-state functional connectivity of distributed brain networks may aid in probing and targeting mechanisms involved in major depressive disorder (MDD). To date, few studies have used resting state functional magnetic resonance imaging (rs-fMRI) to attempt to discriminate individuals with MDD from individuals without MDD, and to our knowledge no investigations have examined a remitted (r) population. In this study, we examined the efficiency of support vector machine (SVM) classifier to successfully discriminate rMDD individuals from healthy controls (HCs) in a narrow early-adult age range. We empirically evaluated four feature selection methods including multivariate Least Absolute Shrinkage and Selection Operator (LASSO) and Elastic Net feature selection algorithms. Our results showed that SVM classification with Elastic Net feature selection achieved the highest classification accuracy of 76.1% (sensitivity of 81.5% and specificity of 68.9%) by leave-one-out cross-validation across subjects from a dataset consisting of 38 rMDD individuals and 29 healthy controls. The highest discriminating functional connections were between the left amygdala, left posterior cingulate cortex, bilateral dorso-lateral prefrontal cortex, and right ventral striatum. These appear to be key nodes in the etiopathophysiology of MDD, within and between default mode, salience and cognitive control networks. This technique demonstrates early promise for using rs-fMRI connectivity as a putative neurobiological marker capable of distinguishing between individuals with and without rMDD. These methods may be extended to periods of risk prior to illness onset, thereby allowing for earlier diagnosis, prevention, and intervention.

Toward literature-based feature selection for diagnostic classification: a meta-analysis of resting-state fMRI in depression

Information derived from functional magnetic resonance imaging (fMRI) during wakeful rest has been introduced as a candidate diagnostic biomarker in unipolar major depressive disorder (MDD). Multiple reports of resting state fMRI in MDD describe group effects. Such prior knowledge can be adopted to pre-select potentially discriminating features for diagnostic classification models with the aim to improve diagnostic accuracy. Purpose of this analysis was to consolidate spatial information about alterations of spontaneous brain activity in MDD, primarily to serve as feature selection for multivariate pattern analysis techniques (MVPA). Thirty two studies were included in final analyses. Coordinates extracted from the original reports were assigned to two categories based on directionality of findings. Meta-analyses were calculated using the non-additive activation likelihood estimation approach with coordinates organized by subject group to account for non-independent samples. Converging evidence revealed a distributed pattern of brain regions with increased or decreased spontaneous activity in MDD. The most distinct finding was hyperactivity/hyperconnectivity presumably reflecting the interaction of cortical midline structures (posterior default mode network components including the precuneus and neighboring posterior cingulate cortices associated with self-referential processing and the subgenual anterior cingulate and neighboring medial frontal cortices) with lateral prefrontal areas related to externally-directed cognition. Other areas of hyperactivity/hyperconnectivity include the left lateral parietal cortex, right hippocampus and right cerebellum whereas hypoactivity/hypoconnectivity was observed mainly in the left temporal cortex, the insula, precuneus, superior frontal gyrus, lentiform nucleus and thalamus. Results are made available in two different data formats to be used as spatial hypotheses in future studies, particularly for diagnostic classification by MVPA.

Characterization of post-traumatic stress disorder using resting-state fMRI with a multi-level parametric classification approach

Functional neuroimaging studies have found intra-regional activity and inter-regional connectivity alterations in patients with post-traumatic stress disorder (PTSD). However, the results of these studies are based on group-level statistics and therefore it is unclear whether PTSD can be discriminated at single-subject level, for instance using the machine learning approach. Here, we proposed a novel framework to identify PTSD using multi-level measures derived from resting-state functional MRI (fMRI). Specifically, three levels of measures were extracted as classification features: (1) regional amplitude of low-frequency fluctuations (univariate feature), which represents local spontaneous synchronous neural activity; (2) temporal functional connectivity (bivariate feature), which represents the extent of similarity of local activity between two regions, and (3) spatial functional connectivity (multivariate feature), which represents the extent of similarity of temporal correlation maps between two regions. Our method was evaluated on 20 PTSD patients and 20 demographically matched healthy controls. The experimental results showed that the features of each level could successfully discriminate PTSD patients from healthy controls. Furthermore, the combination of multi-level features using multi-kernel learning can further improve the classification performance. Specifically, the classification accuracy obtained by the proposed framework was 92.5 %, which was an increase of at least 5 and 17.5 % from the two-level and single-level feature based methods, respectively. Particularly, the limbic structure and prefrontal cortex provided the most discriminant features for classification, consistent with results reported in previous studies. Together, this study demonstrated for the first time that patients with PTSD can be identified at the individual level using resting-state fMRI data. The promising classification results indicated that this method may provide a complementary approach for improving the clinical diagnosis of PTSD.