The challenges and promise of neuroimaging in psychiatry

Multimodal predictive modeling of individual treatment outcome in cocaine dependence with combined neuroimaging and behavioral predictors

BACKGROUND

Developing personalized treatments for cocaine dependence remains a significant clinical challenge. Positron emission tomography (PET) has shown that the [(11)C]raclopride signal in the ventral striatum is associated with treatment success in a positively reinforced contingency management program. The present study investigates whether this signal can be used to predict treatment outcome at an individual level.

METHODS

Predictive models were developed using PET signals from 5 regions of the striatum and follow-up data in 24 patients, and evaluated using cross-validation.

RESULTS

The ventral striatal PET signal alone can predict individual treatment response with a substantial degree of accuracy (cross-validated correct rate=82%). Incorporating information from other regions-of-interest (ROIs) in the striatum does not improve predictive performance, except for a small improvement with adding the posterior caudate. The addition of baseline demographic variables, including baseline severity measures, does not improve predictive performance. On the other hand, early treatment response and motivation, reflected by cumulative clinic attendance, performs as well as the PET signal (83%) by week 3 in the 24-week study. The combined model with both PET signals and cumulative clinic attendance demonstrates a significant improvement of performance, peaking at 96% during week 3 of the trial.

CONCLUSIONS

These results suggest that a multimodal model can predict treatment success in cocaine dependence at an individual level, and pose hypotheses for the underlying neural circuitry mechanisms responsible for individual variations in treatment outcome.

Filling the gap: relationship between the serotonin-transporter-linked polymorphic region and amygdala activation

The alleged association between the serotonin-transporter-linked polymorphic region (5-HTTLPR) and amygdala activation forms a cornerstone of the common view that carrying the short allele of this polymorphism is a potential risk factor for affective disorders. The authors of a recent meta-analysis showed that this association is statistically significant (Hedges's g = 0.35) but warned that estimates might be distorted because of publication bias. Here, we report a replication study of this relationship in 120 participants. We failed to find an association of 5-HTTLPR variation with amygdala activation during a widely used emotional-face-matching paradigm. Moreover, when we conducted a meta-analysis that included unpublished studies and data from the current study, the pooled meta-analytic effect size was no longer significant (g = 0.20, p = .06). These findings cast doubt on previously reported substantial effects, suggesting that the 5-HTTLPR-amygdala association is either much smaller than previously thought, conditional on other factors, or nonexistent.

Hippocampal and frontolimbic function as intermediate phenotype for psychosis: evidence from healthy relatives and a common risk variant in CACNA1C

BACKGROUND

Variation in CACNA1C has consistently been associated with psychiatric disease in genome-wide association studies. We have previously shown that healthy carriers of the CACNA1C rs1006737 risk variant exhibit hippocampal and perigenual anterior cingulate (pgACC) dysfunction during episodic memory recall. To test whether this brain systems-level abnormality is a potential intermediate phenotype for psychiatric disorder, we studied unaffected relatives of patients with bipolar disorder, major depression, and schizophrenia.

METHODS

The study population comprised 188 healthy first-degree relatives of patients with bipolar disorder (n=59), major depression (n=73), and schizophrenia (n=56) and 110 comparison subjects from our discovery study who were genotyped for rs1006737 and underwent functional magnetic resonance imaging while performing an episodic memory task and psychological testing. Group comparisons were analyzed using SPM8 and PASW Statistics 20.

RESULTS

Similar to risk allele carriers in the discovery sample, relatives of index patients exhibited hippocampal and pgACC dysfunction as well as increased scores in depression and anxiety measures, correlating negatively with hippocampal activation. Carrying the rs1006737 risk variant resulted in a stronger decrease of hippocampal and pgACC activation in relatives, indicating an additive effect of CACNA1C variation on familial risk.

CONCLUSIONS

Our findings implicate abnormal perigenual and hippocampal activation as a promising intermediate phenotype for psychiatric disease and suggest a pathophysiologic mechanism conferred by a CACNA1C variant being implicated in risk for symptom dimensions shared among bipolar disorder, major depression, and schizophrenia.

Abnormal functional resting-state networks in ADHD: graph theory and pattern recognition analysis of fMRI data

The framework of graph theory provides useful tools for investigating the neural substrates of neuropsychiatric disorders. Graph description measures may be useful as predictor variables in classification procedures. Here, we consider several centrality measures as predictor features in a classification algorithm to identify nodes of resting-state networks containing predictive information that can discriminate between typical developing children and patients with attention-deficit/hyperactivity disorder (ADHD). The prediction was based on a support vector machines classifier. The analyses were performed in a multisite and publicly available resting-state fMRI dataset of healthy children and ADHD patients: the ADHD-200 database. Network centrality measures contained little predictive information for the discrimination between ADHD patients and healthy subjects. However, the classification between inattentive and combined ADHD subtypes was more promising, achieving accuracies higher than 65% (balance between sensitivity and specificity) in some sites. Finally, brain regions were ranked according to the amount of discriminant information and the most relevant were mapped. As hypothesized, we found that brain regions in motor, frontoparietal, and default mode networks contained the most predictive information. We concluded that the functional connectivity estimations are strongly dependent on the sample characteristics. Thus different acquisition protocols and clinical heterogeneity decrease the predictive values of the graph descriptors.

Multimodal assessments of the hippocampal formation in schizophrenia and bipolar disorder: evidences from neurobehavioral measures and functional and structural MRI

A potential clinical and etiological overlap between schizophrenia (SZ) and bipolar disorder (BD) has long been a subject of discussion. Imaging studies imply functional and structural alterations of the hippocampus in both diseases. Thus, imaging this core memory region could provide insight into the pathophysiology of these disorders and the associated cognitive deficits. To examine possible shared alterations in the hippocampus, we conducted a multi-modal assessment, including functional and structural imaging as well as neurobehavioral measures of memory performance in BD and SZ patients compared with healthy controls. We assessed episodic memory performance, using tests of verbal and visual learning (HVLT, BVMT) in three groups of participants: BD patients (n = 21), SZ patients (n = 21) and matched (age, gender, education) healthy control subjects (n = 21). In addition, we examined hippocampal resting state functional connectivity, hippocampal volume using voxel-based morphometry (VBM) and fibre integrity of hippocampal connections using diffusion tensor imaging (DTI). We found memory deficits, changes in functional connectivity within the hippocampal network as well as volumetric reductions and altered white matter fibre integrity across patient groups in comparison with controls. However, SZ patients when directly compared with BD patients were more severely affected in several of the assessed parameters (verbal learning, left hippocampal volumes, mean diffusivity of bilateral cingulum and right uncinated fasciculus). The results of our study suggest a graded expression of verbal learning deficits accompanied by structural alterations within the hippocampus in BD patients and SZ patients, with SZ patients being more strongly affected. Our findings imply that these two disorders may share some common pathophysiological mechanisms. The results could thus help to further advance and integrate current pathophysiological models of SZ and BD.

Visualization and unsupervised predictive clustering of high-dimensional multimodal neuroimaging data

BACKGROUND

Neuroimaging machine learning studies have largely utilized supervised algorithms - meaning they require both neuroimaging scan data and corresponding target variables (e.g. healthy vs. diseased) to be successfully 'trained' for a prediction task. Noticeably, this approach may not be optimal or possible when the global structure of the data is not well known and the researcher does not have an a priori model to fit the data.

NEW METHOD

We set out to investigate the utility of an unsupervised machine learning technique; t-distributed stochastic neighbour embedding (t-SNE) in identifying 'unseen' sample population patterns that may exist in high-dimensional neuroimaging data. Multimodal neuroimaging scans from 92 healthy subjects were pre-processed using atlas-based methods, integrated and input into the t-SNE algorithm. Patterns and clusters discovered by the algorithm were visualized using a 2D scatter plot and further analyzed using the K-means clustering algorithm.

COMPARISON WITH EXISTING METHODS

t-SNE was evaluated against classical principal component analysis.

CONCLUSION

Remarkably, based on unlabelled multimodal scan data, t-SNE separated study subjects into two very distinct clusters which corresponded to subjects' gender labels (cluster silhouette index value=0.79). The resulting clusters were used to develop an unsupervised minimum distance clustering model which identified 93.5% of subjects' gender. Notably, from a neuropsychiatric perspective this method may allow discovery of data-driven disease phenotypes or sub-types of treatment responders.

The impact of psychopharmacology on contemporary clinical psychiatry

Clinical psychiatric evaluations of patients have changed dramatically in recent decades. Both initial assessments and follow-up visits have become brief and superficial, focused on searching for categorical diagnostic criteria from checklists, with limited inquiry into patient-reported symptomatic status and tolerability of treatments. The virtually exclusive therapeutic task has become selecting a plausible psychotropic, usually based on expert consensus guidelines. These guidelines and practice patterns rest mainly on published monotherapy trials that may or may not be applicable to particular patients but are having a profound impact, not only on modern psychiatric practice but also on psychiatric education, research, and theory.

[Brain imaging of first-episode psychosis]

In the last decades, schizophrenia has intensively been studied using various brain imaging techniques. However, several potential confounding factors limited their interpretation power (e.g. chronicity, the impact of antipsychotic medication). By considering psychosis as a continuum of changes starting from mild cognitive impairments to serious psychotic symptoms, it became possible to provide deeper insight in the neurobiological mechanisms underlying the onset of psychosis by focusing on at-risk individuals and first-episodes. Recent brain imaging meta-analyses of the first episode psychosis (FEP), noteworthy reported conjoint bilateral structural and functional differences at the level of the insula, the superior temporal gyrus and the medial frontal gyrus, encompassing the anterior cingulate cortex. In the present review, we thus provide an update of brain imaging studies of FEP with a particular emphasis on more recent anatomical, functional and molecular explorations. Specifically, we provide 1) a review of the common features observed in individuals with high risk for psychosis and changes characterizing the transition to psychosis, 2) a description of the environmental and drug factors influencing these abnormalities, 3) how these findings in FEP may differ from those observed in chronic individuals with schizophrenia, and 4) a short overview of new classification algorithms able to use MRI findings as valuable biomarkers to guide early detection in the prodromal phase of psychosis.

The When and Where of Working Memory Dysfunction in Early-Onset Schizophrenia-A Functional Magnetic Resonance Imaging Study

Behavioral evidence indicates that working memory (WM) in schizophrenia is already impaired at the encoding stage. However, the neurophysiological basis of this primary deficit remains poorly understood. Using event-related fMRI, we assessed differences in brain activation and functional connectivity during the encoding, maintenance and retrieval stages of a visual WM task with 3 levels of memory load in 17 adolescents with early-onset schizophrenia (EOS) and 17 matched controls. The amount of information patients could store in WM was reduced at all memory load levels. During encoding, activation in left ventrolateral prefrontal cortex (VLPFC) and extrastriate visual cortex, which in controls positively correlated with the amount of stored information, was reduced in patients. Additionally, patients showed disturbed functional connectivity between prefrontal and visual areas. During retrieval, right inferior VLPFC hyperactivation was correlated with hypoactivation of left VLPFC in patients during encoding. Visual WM encoding is disturbed by a failure to adequately engage a visual-prefrontal network critical for the transfer of perceptual information into WM. Prefrontal hyperactivation appears to be a secondary consequence of this primary deficit. Isolating the component processes of WM can lead to more specific neurophysiological markers for translational efforts seeking to improve the treatment of cognitive dysfunction in schizophrenia.

Predictive classification of individual magnetic resonance imaging scans from children and adolescents

Neuroimaging techniques are increasingly being explored as potential tools for clinical prediction in psychiatry. There are a wide range of approaches which can be applied to make individual predictions for various aspects of disorders such as diagnostic status, symptom severity scores, identification of patients at risk of developing disorders and estimation of the likelihood of response to treatment. This selective review highlights a popular group of pattern recognition techniques, support vector machines (SVMs) for use with structural magnetic resonance imaging scans. First, however, we outline various practical issues, limitations and techniques which need to be considered before SVM's can be applied. We begin with a discussion on the practicalities of scanning children and adolescent participants and the importance of acquiring high quality images. Scan processing required for inter-subject comparisons is then discussed. We then briefly discuss feature selection and other considerations when applying pattern recognition techniques. Finally, SVMs are described and various studies highlighted to indicate the potential of these techniques for child and adolescent psychiatric research.

Clinical applications of the functional connectome

Central to the development of clinical applications of functional connectomics for neurology and psychiatry is the discovery and validation of biomarkers. Resting state fMRI (R-fMRI) is emerging as a mainstream approach for imaging-based biomarker identification, detecting variations in the functional connectome that can be attributed to clinical variables (e.g., diagnostic status). Despite growing enthusiasm, many challenges remain. Here, we assess evidence of the readiness of R-fMRI based functional connectomics to lead to clinically meaningful biomarker identification through the lens of the criteria used to evaluate clinical tests (i.e., validity, reliability, sensitivity, specificity, and applicability). We focus on current R-fMRI-based prediction efforts, and survey R-fMRI used for neurosurgical planning. We identify gaps and needs for R-fMRI-based biomarker identification, highlighting the potential of emerging conceptual, analytical and cultural innovations (e.g., the Research Domain Criteria Project (RDoC), open science initiatives, and Big Data) to address them. Additionally, we note the need to expand future efforts beyond identification of biomarkers for disease status alone to include clinical variables related to risk, expected treatment response and prognosis.

Multivariate classification of blood oxygen level-dependent FMRI data with diagnostic intention: a clinical perspective

SUMMARY

There has been a recent upsurge of reports about applications of pattern-recognition techniques from the field of machine learning to functional MR imaging data as a diagnostic tool for systemic brain disease or psychiatric disorders. Entities studied include depression, schizophrenia, attention deficit hyperactivity disorder, and neurodegenerative disorders like Alzheimer dementia. We review these recent studies which-despite the optimism from some articles-predominantly constitute explorative efforts at the proof-of-concept level. There is some evidence that, in particular, support vector machines seem to be promising. However, the field is still far from real clinical application, and much work has to be done regarding data preprocessing, model optimization, and validation. Reporting standards are proposed to facilitate future meta-analyses or systematic reviews.

Verbal episodic memory deficits in remitted bipolar patients: a combined behavioural and fMRI study

BACKGROUND

Episodic memory deficits affect the majority of patients with bipolar disorder (BD).

AIMS

The study investigates episodic memory performance through different approaches, including behavioural measures, physiological parameters, and the underlying functional activation patterns with functional neuroimaging (fMRI).

METHODS

26 Remitted BD patients and a matched group of healthy controls underwent a verbal episodic memory test together with monitored autonomic response, psychopathological ratings and functional magnetic resonance imaging (fMRI) during the verbal episodic memory test.

RESULTS

Compared to healthy controls, BD patients performed significantly worse during the episodic memory task. The results further indicate that verbal episodic memory deficits in BD are associated with abnormal functional activity patterns in frontal, occipital and limbic regions, and an increase in stress parameters.

LIMITATIONS

We aimed to minimise sample heterogeneity by setting clear criteria for remission, based on the scores of a depression (BDI II) and mania scale (BRMAS) and on the DSM IV criteria. However, our patients were not symptom-free and scored higher on BDI II scores than the control group.

CONCLUSIONS

The results are of interest for the treatment of cognitive symptoms in BD patients, as persistent cognitive impairment may hamper full rehabilitation.

Triangulating perspectives on functional neuroimaging for disorders of mental health

BACKGROUND

Functional neuroimaging is being used in clinical psychiatry today despite the vigorous objections of many in the research community over issues of readiness. To date, a systematic examination of the perspectives of key stakeholders involved in this debate has not yet been attempted. To this fill this gap, we interviewed investigators who conduct functional neuroimaging studies involving adults with mood disorders, schizophrenia, obsessive compulsive disorder, and/or attention deficit hyperactivity disorder, providers who offer clinical neuroimaging services in the open marketplace, and consumers of these services, in order to understand perspectives underlying different views and practices.

METHODS

Semi-structured interviews were conducted over the telephone. Verbal consent was obtained and all interviews were audio recorded. Interviews of investigators and service providers followed the same interview guide. A separate set of questions was developed for consumers. All interviews were transcribed and made software ready. We applied the qualitative methodology of constant comparison to analyze the data, whereby two researchers independently analyzed the results into textual themes. Coding discrepancies were discussed until consensus was achieved.

RESULTS

Investigators, service providers, and consumers held many common perspectives about the potential or actual risks and benefits of functional neuroimaging for mental illness. However, we also found striking divergences. Service providers focused on the challenges posed by the persistence of symptoms based diagnostic categories, whereas the limitations of the science in this area was the challenge noted most frequently by investigators. The majority of consumers stated that their expectations were met.

CONCLUSION

Our findings point toward a fundamental tension between academic investigators on the one hand, and commercial service providers and their customers on the other. This scenario poses dangers to the communities directly involved, and to public trust in science and medicine more generally. We conclude with recommendations for work that needs to be done to minimize tensions and maximize the potential of neurotechnology through concerted efforts to respect its limitations while leveraging the strengths, investments, and hopes of each stakeholder group.

Complex biomarker discovery in neuroimaging data: Finding a needle in a haystack

Neuropsychiatric disorders such as schizophrenia, bipolar disorder and Alzheimer's disease are major public health problems. However, despite decades of research, we currently have no validated prognostic or diagnostic tests that can be applied at an individual patient level. Many neuropsychiatric diseases are due to a combination of alterations that occur in a human brain rather than the result of localized lesions. While there is hope that newer imaging technologies such as functional and anatomic connectivity MRI or molecular imaging may offer breakthroughs, the single biomarkers that are discovered using these datasets are limited by their inability to capture the heterogeneity and complexity of most multifactorial brain disorders. Recently, complex biomarkers have been explored to address this limitation using neuroimaging data. In this manuscript we consider the nature of complex biomarkers being investigated in the recent literature and present techniques to find such biomarkers that have been developed in related areas of data mining, statistics, machine learning and bioinformatics.

Real-time fMRI neurofeedback: progress and challenges

In February of 2012, the first international conference on real time functional magnetic resonance imaging (rtfMRI) neurofeedback was held at the Swiss Federal Institute of Technology Zurich (ETHZ), Switzerland. This review summarizes progress in the field, introduces current debates, elucidates open questions, and offers viewpoints derived from the conference. The review offers perspectives on study design, scientific and clinical applications, rtfMRI learning mechanisms and future outlook.

Association between an Alzheimer's Disease-Related Index and APOE ε4 Gene Dose

Background

We introduced a hypometabolic convergence index (HCI) to characterize in a single measurement the extent to which a person’s fluorodeoxyglucose positron emission tomogram (FDG PET) corresponds to that in Alzheimer’s disease (AD). Apolipoprotein E ε4 (APOE ε4) gene dose is associated with three levels of risk for late-onset AD. We explored the association between gene dose and HCI in cognitively normal ε4 homozygotes, heterozygotes, and non-carriers.

Methods

An algorithm was used to characterize and compare AD-related HCIs in cognitively normal individuals, including 36 ε4 homozygotes, 46 heterozygotes, and 78 non-carriers.

Results

These three groups differed significantly in their HCIs (ANOVA, p = 0.004), and there was a significant association between HCIs and gene dose (linear trend, p = 0.001).

Conclusions

The HCI is associated with three levels of genetic risk for late-onset AD. This supports the possibility of using a single FDG PET measurement to help in the preclinical detection and tracking of AD.

Pattern classification of valence in depression

Neuroimaging biomarkers of depression have potential to aid diagnosis, identify individuals at risk and predict treatment response or course of illness. Nevertheless none have been identified so far, potentially because no single brain parameter captures the complexity of the pathophysiology of depression. Multi-voxel pattern analysis (MVPA) may overcome this issue as it can identify patterns of voxels that are spatially distributed across the brain. Here we present the results of an MVPA to investigate the neuronal patterns underlying passive viewing of positive, negative and neutral pictures in depressed patients. A linear support vector machine (SVM) was trained to discriminate different valence conditions based on the functional magnetic resonance imaging (fMRI) data of nine unipolar depressed patients. A similar dataset obtained in nine healthy individuals was included to conduct a group classification analysis via linear discriminant analysis (LDA). Accuracy scores of 86% or higher were obtained for each valence contrast via patterns that included limbic areas such as the amygdala and frontal areas such as the ventrolateral prefrontal cortex. The LDA identified two areas (the dorsomedial prefrontal cortex and caudate nucleus) that allowed group classification with 72.2% accuracy. Our preliminary findings suggest that MVPA can identify stable valence patterns, with more sensitivity than univariate analysis, in depressed participants and that it may be possible to discriminate between healthy and depressed individuals based on differences in the brain's response to emotional cues.

Cardio-metabolic risk factors and cortical thickness in a neurologically healthy male population: Results from the psychological, social and biological determinants of ill health (pSoBid) study

INTRODUCTION

Cardio-metabolic risk factors have been associated with poor physical and mental health. Epidemiological studies have shown peripheral risk markers to be associated with poor cognitive functioning in normal healthy population and in disease. The aim of the study was to explore the relationship between cardio-metabolic risk factors and cortical thickness in a neurologically healthy middle aged population-based sample.

METHODS

T1-weighted MRI was used to create models of the cortex for calculation of regional cortical thickness in 40 adult males (average age = 50.96 years), selected from the pSoBid study. The relationship between cardio-vascular risk markers and cortical thickness across the whole brain, was examined using the general linear model. The relationship with various covariates of interest was explored.

RESULTS

Lipid fractions with greater triglyceride content (TAG, VLDL and LDL) were associated with greater cortical thickness pertaining to a number of regions in the brain. Greater C reactive protein (CRP) and intercellular adhesion molecule (ICAM-1) levels were associated with cortical thinning pertaining to perisylvian regions in the left hemisphere. Smoking status and education status were significant covariates in the model.

CONCLUSIONS

This exploratory study adds to a small body of existing literature increasingly showing a relationship between cardio-metabolic risk markers and regional cortical thickness involving a number of regions in the brain in a neurologically normal middle aged sample. A focused investigation of factors determining the inter-individual variations in regional cortical thickness in the adult brain could provide further clarity in our understanding of the relationship between cardio-metabolic factors and cortical structures.

ZNF804A genotype modulates neural activity during working memory for faces

BACKGROUND

Genetic susceptibility to schizophrenia (SZ) has been suggested to influence the cortical systems supporting working memory (WM) and face processing. Genetic imaging studies link the SZ risk variant rs1344706 on the ZNF804A gene to psychosis via alterations in functional brain connectivity during WM, but no work has looked at the effects of ZNF804A on WM with face-processing components.

METHODS

We therefore investigated healthy controls that were genotyped for rs1344706 with a face WM task during functional magnetic resonance imaging. We suggested that variation at the rs1344706 locus would be associated with similar alterations as patients previously tested using the same WM task for faces.

RESULTS

The rs1344706 risk allele was indeed associated with altered activation in the right dorsolateral prefrontal (rDLPFC) cortex. We established that the rDLPFC was activated in a task-dependent manner, suggesting that the differences in activation between rs1344706 genotype groups reflected alterations in task processing. Furthermore, we demonstrated that the rDLPFC region showed significant volumetric overlap with the rDLPFC which had previously been reported to be altered during task processing for patients with SZ.

CONCLUSIONS

The findings support an association between rs1344706 and alterations in DLPFC activity during WM for faces. We further suggest that WM for faces may be a useful intermediate phenotype in the investigation of genetic susceptibility to psychosis.

Integrative neuroimaging in mood disorders

PURPOSE OF REVIEW

Neuroimaging has become a central technique of biological psychiatry and is uniquely suited to assess functional and structural brain changes in psychiatric patients in vivo. In this review, we highlight several recent developments that may enable the transition of psychiatric neuroimaging from laboratory to clinic.

RECENT FINDINGS

We describe recent trends in refining imaging techniques for brain microstructure (diffusion imaging) and neurochemistry (magnetic resonance spectroscopy of neurotransmitters and metabolites) and their application to patients with mood disorders and individuals at risk, such as first-degree relatives. We also survey recent progress in imaging-guided deep brain stimulation (DBS), imaging-based (neurofeedback) therapies and studies looking at their convergent anatomical targets. These new interventional techniques, which aim to modulate brain circuits of emotion and motivation highlighted by functional imaging studies, have shown promising effects in several small studies.

SUMMARY

The mapping of brain patterns associated with risk to develop mood disorders may pave the way for diagnostic/prognostic applications of neuroimaging. The neuromodulation techniques of DBS and neurofeedback, which target dysfunctional or compensatory circuits identified by functional imaging, may take neuroimaging into a new, therapeutic domain.

Integrating optogenetic and pharmacological approaches to study neural circuit function: current applications and future directions

Optogenetic strategies to control genetically distinct populations of neurons with light have been rapidly evolving and widely adopted by the neuroscience community as one of the most important tool sets to study neural circuit function. Although optogenetics have already reshaped neuroscience by allowing for more precise control of circuit function compared with traditional techniques, current limitations of these approaches should be considered. Here, we discuss several strategies that combine optogenetic and contemporary pharmacological techniques to further increase the specificity of neural circuit manipulation. We also discuss recent advances that allow for the selective modulation of cellular function and gene expression with light. In addition, we outline a novel application of optogenetic circuit analysis for causally addressing the role of pathway-specific neural activity in mediating alterations in postsynaptic transcriptional processing in genetically defined neurons. By determining how optogenetic activation of specific neural circuits causally contributes to alterations in gene expression in a high-throughput fashion, novel biologic targets for future pharmacological intervention may be uncovered. Lastly, extending this experimental pipeline to selectively target pharmacotherapies to genetically defined neuronal populations or circuits will not only provide more selective control of neural circuits, but also may lead to the development of neural circuit specific pharmacological therapeutics.

Neuroethical issues in clinical neuroscience research

In this chapter, we use the special features of neuroimaging to illustrate research ethics issues for the clinical neurologic sciences, and focus on one particularly compelling case: studies involving first-episode schizophrenic treatment-naïve individuals (FESTNIs) (Eijkholt et al., 2012). FESTNIs are scanned prior to the administration of medication in order to control for the confounding effects of treatment. By concentrating on this program of research, we capture the distinctive ethical challenges associated with neuroimaging research overall, and foreground the issues particular to neuroimaging research involving FESTNIs that have yet to receive sufficient attention in the literature. We highlight assessment of risks and burdens, including risks associated with treatment delays and incidental findings; assessment of benefit, including direct benefit, social value, and scientific quality; subject selection; justice questions related to responsiveness and poststudy access; and, finally, issues related to consent and capacity.

Systems biology, bioinformatics, and biomarkers in neuropsychiatry

Although neuropsychiatric (NP) disorders are among the top causes of disability worldwide with enormous financial costs, they can still be viewed as part of the most complex disorders that are of unknown etiology and incomprehensible pathophysiology. The complexity of NP disorders arises from their etiologic heterogeneity and the concurrent influence of environmental and genetic factors. In addition, the absence of rigid boundaries between the normal and diseased state, the remarkable overlap of symptoms among conditions, the high inter-individual and inter-population variations, and the absence of discriminative molecular and/or imaging biomarkers for these diseases makes difficult an accurate diagnosis. Along with the complexity of NP disorders, the practice of psychiatry suffers from a "top-down" method that relied on symptom checklists. Although checklist diagnoses cost less in terms of time and money, they are less accurate than a comprehensive assessment. Thus, reliable and objective diagnostic tools such as biomarkers are needed that can detect and discriminate among NP disorders. The real promise in understanding the pathophysiology of NP disorders lies in bringing back psychiatry to its biological basis in a systemic approach which is needed given the NP disorders' complexity to understand their normal functioning and response to perturbation. This approach is implemented in the systems biology discipline that enables the discovery of disease-specific NP biomarkers for diagnosis and therapeutics. Systems biology involves the use of sophisticated computer software "omics"-based discovery tools and advanced performance computational techniques in order to understand the behavior of biological systems and identify diagnostic and prognostic biomarkers specific for NP disorders together with new targets of therapeutics. In this review, we try to shed light on the need of systems biology, bioinformatics, and biomarkers in neuropsychiatry, and illustrate how the knowledge gained through these methodologies can be translated into clinical use providing clinicians with improved ability to diagnose, manage, and treat NP patients.

Insights into multimodal imaging classification of ADHD

Attention deficit hyperactivity disorder (ADHD) currently is diagnosed in children by clinicians via subjective ADHD-specific behavioral instruments and by reports from the parents and teachers. Considering its high prevalence and large economic and societal costs, a quantitative tool that aids in diagnosis by characterizing underlying neurobiology would be extremely valuable. This provided motivation for the ADHD-200 machine learning (ML) competition, a multisite collaborative effort to investigate imaging classifiers for ADHD. Here we present our ML approach, which used structural and functional magnetic resonance imaging data, combined with demographic information, to predict diagnostic status of individuals with ADHD from typically developing (TD) children across eight different research sites. Structural features included quantitative metrics from 113 cortical and non-cortical regions. Functional features included Pearson correlation functional connectivity matrices, nodal and global graph theoretical measures, nodal power spectra, voxelwise global connectivity, and voxelwise regional homogeneity. We performed feature ranking for each site and modality using the multiple support vector machine recursive feature elimination (SVM-RFE) algorithm, and feature subset selection by optimizing the expected generalization performance of a radial basis function kernel SVM (RBF-SVM) trained across a range of the top features. Site-specific RBF-SVMs using these optimal feature sets from each imaging modality were used to predict the class labels of an independent hold-out test set. A voting approach was used to combine these multiple predictions and assign final class labels. With this methodology we were able to predict diagnosis of ADHD with 55% accuracy (versus a 39% chance level in this sample), 33% sensitivity, and 80% specificity. This approach also allowed us to evaluate predictive structural and functional features giving insight into abnormal brain circuitry in ADHD.