Dissecting psychiatric spectrum disorders by generative embedding

Neuroimage Analysis Methods and Artificial Intelligence Techniques for Reliable Biomarkers and Accurate Diagnosis of Schizophrenia: Achievements Made by Chinese Scholars Around the Past Decade

BACKGROUND AND HYPOTHESIS

Schizophrenia (SZ) is characterized by significant cognitive and behavioral disruptions. Neuroimaging techniques, particularly magnetic resonance imaging (MRI), have been widely utilized to investigate biomarkers of SZ, distinguish SZ from healthy conditions or other mental disorders, and explore biotypes within SZ or across SZ and other mental disorders, which aim to promote the accurate diagnosis of SZ. In China, research on SZ using MRI has grown considerably in recent years.

STUDY DESIGN

The article reviews advanced neuroimaging and artificial intelligence (AI) methods using single-modal or multimodal MRI to reveal the mechanism of SZ and promote accurate diagnosis of SZ, with a particular emphasis on the achievements made by Chinese scholars around the past decade.

STUDY RESULTS

Our article focuses on the methods for capturing subtle brain functional and structural properties from the high-dimensional MRI data, the multimodal fusion and feature selection methods for obtaining important and sparse neuroimaging features, the supervised statistical analysis and classification for distinguishing disorders, and the unsupervised clustering and semi-supervised learning methods for identifying neuroimage-based biotypes. Crucially, our article highlights the characteristics of each method and underscores the interconnections among various approaches regarding biomarker extraction and neuroimage-based diagnosis, which is beneficial not only for comprehending SZ but also for exploring other mental disorders.

CONCLUSIONS

We offer a valuable review of advanced neuroimage analysis and AI methods primarily focused on SZ research by Chinese scholars, aiming to promote the diagnosis, treatment, and prevention of SZ, as well as other mental disorders, both within China and internationally.

Influence of noninvasive brain stimulation on connectivity and local activation: a combined tDCS and fMRI study

The effect of transcranial direct current stimulation (tDCS) on neurobiological mechanisms underlying executive function in the human brain remains elusive. This study aims at examining the effect of anodal and cathodal tDCS over the left dorsolateral prefrontal cortex (DLPFC) in comparison with sham stimulation on resting-state connectivity as well as functional activation and working memory performance. We hypothesized perturbed fronto-parietal resting-state connectivity during stimulation and altered working memory performance combined with modified functional working memory-related activation. We applied tDCS with 1 mA for 21 min over the DLPFC inside an fMRI scanner. During stimulation, resting-state fMRI was acquired and task-dependent fMRI during working memory task performance was acquired directly after stimulation. N = 36 healthy subjects were studied in a within-subject design with three different experimental conditions (anodal, cathodal and sham) in a double-blind design. Seed-based functional connectivity analyses and dynamic causal modeling were conducted for the resting-state fMRI data. We found a significant stimulation by region interaction in the seed-based ROI-to-ROI resting-state connectivity, but no effect on effective connectivity. We also did not find an effect of stimulation on task-dependent signal alterations in working memory activation in our regions of interest and no effect on working memory performance parameters. We found effects on measures of seed-based resting-state connectivity, while measures of effective connectivity and task-based connectivity did not show any stimulation effect. We could not replicate previous findings of tDCS stimulation effects on behavioral outcomes. We critically discuss possible methodological limitations and implications for future studies.

Explainable multimodal prediction of treatment-resistance in patients with depression leveraging brain morphometry and natural language processing

Although 20 % of patients with depression receiving treatment do not achieve remission, predicting treatment-resistant depression (TRD) remains challenging. In this study, we aimed to develop an explainable multimodal prediction model for TRD using structured electronic medical record data, brain morphometry, and natural language processing. In total, 247 patients with a new depressive episode were included. TRD-predictive models were developed based on the combination of following parameters: selected tabular dataset features, independent components-map weightings from brain T1-weighted magnetic resonance imaging (MRI), and topic probabilities from clinical notes. All models applied the extreme gradient boosting (XGBoost) algorithm via five-fold cross-validation. The model using all data sources showed the highest area under the receiver operating characteristic of 0.794, followed by models that used combined brain MRI and structured data, brain MRI and clinical notes, clinical notes and structured data, brain MRI only, structured data only, and clinical notes only (0.770, 0.762, 0.728, 0.703, 0.684, and 0.569, respectively). Classifications of TRD were driven by several predictors, such as previous exposure to antidepressants and antihypertensive medications, sensorimotor network, default mode network, and somatic symptoms. Our findings suggest that a combination of clinical data with neuroimaging and natural language processing variables improves the prediction of TRD.

Neuroimaging Biomarkers for Drug Discovery and Development in Schizophrenia

Schizophrenia is a chronic mental illness that affects up to 1% of the population. While efficacious therapies are available for positive symptoms, effective treatment of cognitive and negative symptoms remains an unmet need after decades of research. New developments in the field of neuroimaging are accelerating our knowledge gain regarding the underlying pathophysiology of symptoms in schizophrenia and psychosis spectrum disorders, inspiring new targets for drug development. However, no validated and qualified biomarkers are currently available to support the development of new therapeutics. This review summarizes the current use of neuroimaging technology in clinical drug development for psychotic disorders. As exemplified by drug development programs that target NMDA receptor hypofunction, neuroimaging results play a critical role in target discovery and establishing target engagement and dose selection. Furthermore, pharmacological neuroimaging may provide response biomarkers that allow for early decision making in proof-of-concept studies that leverage pharmacological challenge models in healthy volunteers. That said, while response and predictive biomarkers are starting to be evaluated in patient populations, they continue to play a limited role. Novel approaches to neuroimaging data acquisition and analysis may aid the establishment of biomarkers that are predictive at the individual level in the future. Nevertheless, various gaps in knowledge need to be addressed and biomarkers need to be validated to establish them as "fit for purpose" in drug development.

Inferring Trajectories of Psychotic Disorders Using Dynamic Causal Modeling

Introduction

Illness course plays a crucial role in delineating psychiatric disorders. However, existing nosologies consider only its most basic features (e.g., symptom sequence, duration). We developed a Dynamic Causal Model (DCM) that characterizes course patterns more fully using dense timeseries data. This foundational study introduces the new modeling approach and evaluates its validity using empirical and simulated data.

Methods

A three-level DCM was constructed to model how latent dynamics produce symptoms of depression, mania, and psychosis. This model was fit to symptom scores of nine patients collected prospectively over four years, following first hospitalization. Simulated subjects based on these empirical data were used to evaluate model parameters at the subject-level. At the group-level, we tested the accuracy with which the DCM can estimate the latent course patterns using Parametric Empirical Bayes (PEB) and leave-one-out cross-validation.

Results

Analyses of empirical data showed that DCM accurately captured symptom trajectories for all nine subjects. Simulation results showed that parameters could be estimated accurately (correlations between generative and estimated parameters >= 0.76). Moreover, the model could distinguish different latent course patterns, with PEB correctly assigning simulated patients for eight of nine course patterns. When testing any pair of two specific course patterns using leave-one-out cross-validation, 30 out of 36 pairs showed a moderate or high out-of-samples correlation between the true group-membership and the estimated group-membership values.

Conclusion

DCM has been widely used in neuroscience to infer latent neuronal processes from neuroimaging data. Our findings highlight the potential of adopting this methodology for modeling symptom trajectories to explicate nosologic entities, temporal patterns that define them, and facilitate personalized treatment.

Blind Subgrouping of Task-based fMRI

Significant heterogeneity in network structures reflecting individuals' dynamic processes can exist within subgroups of people (e.g., diagnostic category, gender). This makes it difficult to make inferences regarding these predefined subgroups. For this reason, researchers sometimes wish to identify subsets of individuals who have similarities in their dynamic processes regardless of any predefined category. This requires unsupervised classification of individuals based on similarities in their dynamic processes, or equivalently, in this case, similarities in their network structures of edges. The present paper tests a recently developed algorithm, S-GIMME, that takes into account heterogeneity across individuals with the aim of providing subgroup membership and precise information about the specific network structures that differentiate subgroups. The algorithm has previously provided robust and accurate classification when evaluated with large-scale simulation studies but has not yet been validated on empirical data. Here, we investigate S-GIMME's ability to differentiate, in a purely data-driven manner, between brain states explicitly induced through different tasks in a new fMRI dataset. The results provide new evidence that the algorithm was able to resolve, in an unsupervised data-driven manner, the differences between different active brain states in empirical fMRI data to segregate individuals and arrive at subgroup-specific network structures of edges. The ability to arrive at subgroups that correspond to empirically designed fMRI task conditions, with no biasing or priors, suggests this data-driven approach can be a powerful addition to existing methods for unsupervised classification of individuals based on their dynamic processes.

Dysconnection and cognition in schizophrenia: A spectral dynamic causal modeling study

Schizophrenia (SZ) is a severe mental disorder characterized by failure of functional integration (aka dysconnection) across the brain. Recent functional connectivity (FC) studies have adopted functional parcellations to define subnetworks of large-scale networks, and to characterize the (dys)connection between them, in normal and clinical populations. While FC examines statistical dependencies between observations, model-based effective connectivity (EC) can disclose the causal influences that underwrite the observed dependencies. In this study, we investigated resting state EC within seven large-scale networks, in 66 SZ and 74 healthy subjects from a public dataset. The results showed that a remarkable 33% of the effective connections (among subnetworks) of the cognitive control network had been pathologically modulated in SZ. Further dysconnection was identified within the visual, default mode and sensorimotor networks of SZ subjects, with 24%, 20%, and 11% aberrant couplings. Overall, the proportion of discriminative connections was remarkably larger in EC (24%) than FC (1%) analysis. Subsequently, to study the neural correlates of impaired cognition in SZ, we conducted a canonical correlation analysis between the EC parameters and the cognitive scores of the patients. As such, the self-inhibitions of supplementary motor area and paracentral lobule (in the sensorimotor network) and the excitatory connection from parahippocampal gyrus to inferior temporal gyrus (in the cognitive control network) were significantly correlated with the social cognition, reasoning/problem solving and working memory capabilities of the patients. Future research can investigate the potential of whole-brain EC as a biomarker for diagnosis of brain disorders and for neuroimaging-based cognitive assessment.

Predicting Future Depressive Episodes from Resting-State fMRI with Generative Embedding

After a first episode of major depressive disorder (MDD), there is substantial risk for a long-term remitting-relapsing course. Prevention and early interventions are thus critically important. Various studies have examined the feasibility of detecting at-risk individuals based on out-of-sample predictions about the future occurrence of depression. However, functional magnetic resonance imaging (MRI) has received very little attention for this purpose so far. Here, we explored the utility of generative models (i.e. different dynamic causal models, DCMs) as well as functional connectivity (FC) for predicting future episodes of depression in never-depressed adults, using a large dataset (N=906) of task-free ("resting state") fMRI data from the UK Biobank. Connectivity analyses were conducted using timeseries from pre-computed spatially independent components of different dimensionalities. Over a three year period, 50% of participants showed indications of at least one depressive episode, while the other 50% did not. Using nested cross-validation for training and a held-out test set (80/20 split), we systematically examined the combination of 8 connectivity feature sets and 17 classifiers. We found that a generative embedding procedure based on combining regression DCM (rDCM) with a support vector machine (SVM) enabled the best predictions, both on the training set (0.63 accuracy, 0.66 area under the curve, AUC) and the test set (0.62 accuracy, 0.64 AUC; p<0.001). However, on the test set, rDCM was only slightly superior to predictions based on FC (0.59 accuracy, 0.61 AUC). Interpreting model predictions based on SHAP (SHapley Additive exPlanations) values suggested that the most predictive connections were widely distributed and not confined to specific networks. Overall, our analyses suggest (i) ways of improving future fMRI-based generative embedding approaches for the early detection of individuals at-risk for depression and that (ii) achieving accuracies of clinical utility may require combination of fMRI with other data modalities.

Understanding mental health through computers: An introduction to computational psychiatry

Computational psychiatry recently established itself as a new tool in the study of mental disorders and problems. Integration of different levels of analysis is creating computational phenotypes with clinical and research values, and constructing a way to arrive at precision psychiatry are part of this new branch. It conceptualizes the brain as a computational organ that receives from the environment parameters to respond to challenges through calculations and algorithms in continuous feedback and feedforward loops with a permanent degree of uncertainty. Through this conception, one can seize an understanding of the cerebral and mental processes in the form of theories or hypotheses based on data. Using these approximations, a better understanding of the disorder and its different determinant factors facilitates the diagnostics and treatment by having an individual, ecologic, and holistic approach. It is a tool that can be used to homologate and integrate multiple sources of information given by several theoretical models. In conclusion, it helps psychiatry achieve precision and reproducibility, which can help the mental health field achieve significant advancement. This article is a narrative review of the basis of the functioning of computational psychiatry with a critical analysis of its concepts.

Computational psychiatry: from synapses to sentience

This review considers computational psychiatry from a particular viewpoint: namely, a commitment to explaining psychopathology in terms of pathophysiology. It rests on the notion of a generative model as underwriting (i) sentient processing in the brain, and (ii) the scientific process in psychiatry. The story starts with a view of the brain-from cognitive and computational neuroscience-as an organ of inference and prediction. This offers a formal description of neuronal message passing, distributed processing and belief propagation in neuronal networks; and how certain kinds of dysconnection lead to aberrant belief updating and false inference. The dysconnections in question can be read as a pernicious synaptopathy that fits comfortably with formal notions of how we-or our brains-encode uncertainty or its complement, precision. It then considers how the ensuing process theories are tested empirically, with an emphasis on the computational modelling of neuronal circuits and synaptic gain control that mediates attentional set, active inference, learning and planning. The opportunities afforded by this sort of modelling are considered in light of in silico experiments; namely, computational neuropsychology, computational phenotyping and the promises of a computational nosology for psychiatry. The resulting survey of computational approaches is not scholarly or exhaustive. Rather, its aim is to review a theoretical narrative that is emerging across subdisciplines within psychiatry and empirical scales of investigation. These range from epilepsy research to neurodegenerative disorders; from post-traumatic stress disorder to the management of chronic pain, from schizophrenia to functional medical symptoms.

Cell type-specific manifestations of cortical thickness heterogeneity in schizophrenia

Brain morphology differs markedly between individuals with schizophrenia, but the cellular and genetic basis of this heterogeneity is poorly understood. Here, we sought to determine whether cortical thickness (CTh) heterogeneity in schizophrenia relates to interregional variation in distinct neural cell types, as inferred from established gene expression data and person-specific genomic variation. This study comprised 1849 participants in total, including a discovery (140 cases and 1267 controls) and a validation cohort (335 cases and 185 controls). To characterize CTh heterogeneity, normative ranges were established for 34 cortical regions and the extent of deviation from these ranges was measured for each individual with schizophrenia. CTh deviations were explained by interregional gene expression levels of five out of seven neural cell types examined: (1) astrocytes; (2) endothelial cells; (3) oligodendrocyte progenitor cells (OPCs); (4) excitatory neurons; and (5) inhibitory neurons. Regional alignment between CTh alterations with cell type transcriptional maps distinguished broad patient subtypes, which were validated against genomic data drawn from the same individuals. In a predominantly neuronal/endothelial subtype (22% of patients), CTh deviations covaried with polygenic risk for schizophrenia (sczPRS) calculated specifically from genes marking neuronal and endothelial cells (r = -0.40, p = 0.010). Whereas, in a predominantly glia/OPC subtype (43% of patients), CTh deviations covaried with sczPRS calculated from glia and OPC-linked genes (r = -0.30, p = 0.028). This multi-scale analysis of genomic, transcriptomic, and brain phenotypic data may indicate that CTh heterogeneity in schizophrenia relates to inter-individual variation in cell-type specific functions. Decomposing heterogeneity in relation to cortical cell types enables prioritization of schizophrenia subsets for future disease modeling efforts.

Computational Neuroscience Approach to Psychiatry: A Review on Theory-driven Approaches

Translating progress in neuroscience into clinical benefits for patients with psychiatric disorders is challenging because it involves the brain as the most complex organ and its interaction with a complex environment and condition. Dealing with such complexity requires powerful techniques. Computational neuroscience approach to psychiatry integrates multiple levels and types of simulation, analysis and computation according to the different types of computational models to enhance comprehending, prediction and treatment of psychiatric disorder. This approach comprises two approaches: theory-driven and data-driven. In this review, we focus on recent advances in theory-driven approaches that mathematically and mechanistically examine the relationships between disorder-related changes and behavior at different level of brain organization. We discuss recent progresses in computational neuroscience models that relate to psychiatry and show how principles of neural computational modeling can be employed to explain psychopathology.

AI ethics in computational psychiatry: From the neuroscience of consciousness to the ethics of consciousness

Methods used in artificial intelligence (AI) overlap with methods used in computational psychiatry (CP). Hence, considerations from AI ethics are also relevant to ethical discussions of CP. Ethical issues include, among others, fairness and data ownership and protection. Apart from this, morally relevant issues also include potential transformative effects of applications of AI—for instance, with respect to how we conceive of autonomy and privacy. Similarly, successful applications of CP may have transformative effects on how we categorise and classify mental disorders and mental health. Since many mental disorders go along with disturbed conscious experiences, it is desirable that successful applications of CP improve our understanding of disorders involving disruptions in conscious experience. Here, we discuss prospects and pitfalls of transformative effects that CP may have on our understanding of mental disorders. In particular, we examine the concern that even successful applications of CP may fail to take all aspects of disordered conscious experiences into account.

•

Considerations from AI ethics are also relevant to the ethics of computational psychiatry.

•

Ethical issues include, among others, fairness and data ownership and protection.

•

They also include potential transformative effects.

•

Computational psychiatry may transform conceptions of mental disorders and health.

•

Disordered conscious experiences may pose a particular challenge.

Effective connectivity extracts clinically relevant prognostic information from resting state activity in stroke

Recent resting-state functional MRI studies in stroke patients have identified two robust biomarkers of acute brain dysfunction: a reduction of inter-hemispheric functional connectivity between homotopic regions of the same network, and an abnormal increase of ipsi-lesional functional connectivity between task-negative and task-positive resting-state networks. Whole-brain computational modelling studies, at the individual subject level, using undirected effective connectivity derived from empirically measured functional connectivity, have shown a reduction of measures of integration and segregation in stroke as compared to healthy brains. Here we employ a novel method, first, to infer whole-brain directional effective connectivity from zero-lagged and lagged covariance matrices, then, to compare it to empirically measured functional connectivity for predicting stroke versus healthy status, and patient performance (zero, one, multiple deficits) across neuropsychological tests. We also investigated the accuracy of functional connectivity versus model effective connectivity in predicting the long-term outcome from acute measures. Both functional and effective connectivity predicted healthy from stroke individuals significantly better than the chance-level; however, accuracy for the effective connectivity was significantly higher than for functional connectivity at 1- to 2-week, 3-month and 1-year post-stroke. Predictive functional connections mainly included those reported in previous studies (within-network inter-hemispheric and between task-positive and -negative networks intra-hemispherically). Predictive effective connections included additional between-network links. Effective connectivity was a better predictor than functional connectivity of the number of behavioural domains in which patients suffered deficits, both at 2-week and 1-year post-onset of stroke. Interestingly, patient deficits at 1-year time-point were better predicted by effective connectivity values at 2 weeks rather than at 1-year time-point. Our results thus demonstrate that the second-order statistics of functional MRI resting-state activity at an early stage of stroke, derived from a whole-brain effective connectivity, estimated in a model fitted to reproduce the propagation of neuronal activity, has pertinent information for clinical prognosis.

Systematic Review of Functional MRI Applications for Psychiatric Disease Subtyping

Background: Psychiatric disorders have been historically classified using symptom information alone. Recently, there has been a dramatic increase in research interest not only in identifying the mechanisms underlying defined pathologies but also in redefining their etiology. This is particularly relevant for the field of personalized medicine, which searches for data-driven approaches to improve diagnosis, prognosis, and treatment selection for individual patients. Methods: This review aims to provide a high-level overview of the rapidly growing field of functional magnetic resonance imaging (fMRI) from the perspective of unsupervised machine learning applications for disease subtyping. Following the PRISMA guidelines for protocol reproducibility, we searched the PubMed database for articles describing functional MRI applications used to obtain, interpret, or validate psychiatric disease subtypes. We also employed the active learning framework ASReview to prioritize publications in a machine learning-guided way. Results: From the 20 studies that met the inclusion criteria, five used functional MRI data to interpret symptom-derived disease clusters, four used it to interpret clusters derived from biomarker data other than fMRI itself, and 11 applied clustering techniques involving fMRI directly. Major depression disorder and schizophrenia were the two most frequently studied pathologies (35% and 30% of the retrieved studies, respectively), followed by ADHD (15%), psychosis as a whole (10%), autism disorder (5%), and the consequences of early exposure to violence (5%). Conclusions: The increased interest in personalized medicine and data-driven disease subtyping also extends to psychiatric disorders. However, to date, this subfield is at an incipient exploratory stage, and all retrieved studies were mostly proofs of principle where further validation and increased sample sizes are craved for. Whereas results for all explored diseases are inconsistent, we believe this reflects the need for concerted, multisite data collection efforts with a strong focus on measuring the generalizability of results. Finally, whereas functional MRI is the best way of measuring brain function available to date, its low signal-to-noise ratio and elevated monetary cost make it a poor clinical alternative. Even with technology progressing and costs decreasing, this might incentivize the search for more accessible, clinically ready functional proxies in the future.

Conductance-based dynamic causal modeling: A mathematical review of its application to cross-power spectral densities

Dynamic Causal Modeling (DCM) is a Bayesian framework for inferring on hidden (latent) neuronal states, based on measurements of brain activity. Since its introduction in 2003 for functional magnetic resonance imaging data, DCM has been extended to electrophysiological data, and several variants have been developed. Their biophysically motivated formulations make these models promising candidates for providing a mechanistic understanding of human brain dynamics, both in health and disease. However, due to their complexity and reliance on concepts from several fields, fully understanding the mathematical and conceptual basis behind certain variants of DCM can be challenging. At the same time, a solid theoretical knowledge of the models is crucial to avoid pitfalls in the application of these models and interpretation of their results. In this paper, we focus on one of the most advanced formulations of DCM, i.e. conductance-based DCM for cross-spectral densities, whose components are described across multiple technical papers. The aim of the present article is to provide an accessible exposition of the mathematical background, together with an illustration of the model's behavior. To this end, we include step-by-step derivations of the model equations, point to important aspects in the software implementation of those models, and use simulations to provide an intuitive understanding of the type of responses that can be generated and the role that specific parameters play in the model. Furthermore, all code utilized for our simulations is made publicly available alongside the manuscript to allow readers an easy hands-on experience with conductance-based DCM.

Identification of transdiagnostic psychiatric disorder subtypes using unsupervised learning

Psychiatric disorders show heterogeneous symptoms and trajectories, with current nosology not accurately reflecting their molecular etiology and the variability and symptomatic overlap within and between diagnostic classes. This heterogeneity impedes timely and targeted treatment. Our study aimed to identify psychiatric patient clusters that share clinical and genetic features and may profit from similar therapies. We used high-dimensional data clustering on deep clinical data to identify transdiagnostic groups in a discovery sample (N = 1250) of healthy controls and patients diagnosed with depression, bipolar disorder, schizophrenia, schizoaffective disorder, and other psychiatric disorders. We observed five diagnostically mixed clusters and ordered them based on severity. The least impaired cluster 0, containing most healthy controls, showed general well-being. Clusters 1-3 differed predominantly regarding levels of maltreatment, depression, daily functioning, and parental bonding. Cluster 4 contained most patients diagnosed with psychotic disorders and exhibited the highest severity in many dimensions, including medication load. Depressed patients were present in all clusters, indicating that we captured different disease stages or subtypes. We replicated all but the smallest cluster 1 in an independent sample (N = 622). Next, we analyzed genetic differences between clusters using polygenic scores (PGS) and the psychiatric family history. These genetic variables differed mainly between clusters 0 and 4 (prediction area under the receiver operating characteristic curve (AUC) = 81%; significant PGS: cross-disorder psychiatric risk, schizophrenia, and educational attainment). Our results confirm that psychiatric disorders consist of heterogeneous subtypes sharing molecular factors and symptoms. The identification of transdiagnostic clusters advances our understanding of the heterogeneity of psychiatric disorders and may support the development of personalized treatments.

Clustering Individuals on Limited Features of a Vector Autoregressive Model

Dynamical interplays in emotions have been investigated using vector autoregressive (VAR) models, whose estimates can be used to cluster participants into unknown groups. The present study evaluated a clustering algorithm, the alternating least square (ALS) algorithm, for accuracy in predicting individual group membership. We systematically manipulated (a) the number of variables in a model, (b) the size of group differences in regression coefficients, and (c) the number of regression coefficients that vary across the groups (i.e., effective features). The ALS algorithm works reliably when there are at least 5 effective features with very large group differences in a 5-variable model; and 9 effective features with very large group differences in a 10-variable model. These findings suggest that the ALS algorithm is sensitive to group differences that are present only in several coefficients of a VAR model, but that the group differences have to be large. We also found that the ALS algorithm outperforms another clustering method, Gaussian mixture modeling. The ALS algorithm was further evaluated with unbalanced sample sizes between groups and with a greater number of groups in data (Study 2). A real data application was provided to illustrate how to interpret the detected group differences (Study 3).

Prediction Along a Developmental Perspective in Psychiatry: How Far Might We Go?

Most mental disorders originate in childhood, and once symptoms present, a variety of psychosocial and cognitive maladjustments may arise. Although early childhood problems are generally associated with later mental health impairments and psychopathology, pluripotent transdiagnostic trajectories may manifest. Possible predictors range from behavioral and neurobiological mechanisms, genetic predispositions, environmental and social factors, and psychopathological comorbidity. They may manifest in altered neurodevelopmental trajectories and need to be validated capitalizing on large-scale multi-modal epidemiological longitudinal cohorts. Moreover, clinical and etiological variability between patients with the same disorders represents a major obstacle to develop effective treatments. Hence, in order to achieve stratification of patient samples opening the avenue of adapting and optimizing treatment for the individual, there is a need to integrate data from multi-dimensionally phenotyped clinical cohorts and cross-validate them with epidemiological cohort data. In the present review, we discuss these aspects in the context of externalizing and internalizing disorders summarizing the current state of knowledge, obstacles, and pitfalls. Although a large number of studies have already increased our understanding on neuropsychobiological mechanisms of mental disorders, it became also clear that this knowledge might only be the tip of the Eisberg and that a large proportion still remains unknown. We discuss prediction strategies and how the integration of different factors and methods may provide useful contributions to research and at the same time may inform prevention and intervention.

Schizophrenia: A Survey of Artificial Intelligence Techniques Applied to Detection and Classification

Artificial Intelligence in healthcare employs machine learning algorithms to emulate human cognition in the analysis of complicated or large sets of data. Specifically, artificial intelligence taps on the ability of computer algorithms and software with allowable thresholds to make deterministic approximate conclusions. In comparison to traditional technologies in healthcare, artificial intelligence enhances the process of data analysis without the need for human input, producing nearly equally reliable, well defined output. Schizophrenia is a chronic mental health condition that affects millions worldwide, with impairment in thinking and behaviour that may be significantly disabling to daily living. Multiple artificial intelligence and machine learning algorithms have been utilized to analyze the different components of schizophrenia, such as in prediction of disease, and assessment of current prevention methods. These are carried out in hope of assisting with diagnosis and provision of viable options for individuals affected. In this paper, we review the progress of the use of artificial intelligence in schizophrenia.

TAPAS: An Open-Source Software Package for Translational Neuromodeling and Computational Psychiatry

Psychiatry faces fundamental challenges with regard to mechanistically guided differential diagnosis, as well as prediction of clinical trajectories and treatment response of individual patients. This has motivated the genesis of two closely intertwined fields: (i) Translational Neuromodeling (TN), which develops "computational assays" for inferring patient-specific disease processes from neuroimaging, electrophysiological, and behavioral data; and (ii) Computational Psychiatry (CP), with the goal of incorporating computational assays into clinical decision making in everyday practice. In order to serve as objective and reliable tools for clinical routine, computational assays require end-to-end pipelines from raw data (input) to clinically useful information (output). While these are yet to be established in clinical practice, individual components of this general end-to-end pipeline are being developed and made openly available for community use. In this paper, we present the Translational Algorithms for Psychiatry-Advancing Science (TAPAS) software package, an open-source collection of building blocks for computational assays in psychiatry. Collectively, the tools in TAPAS presently cover several important aspects of the desired end-to-end pipeline, including: (i) tailored experimental designs and optimization of measurement strategy prior to data acquisition, (ii) quality control during data acquisition, and (iii) artifact correction, statistical inference, and clinical application after data acquisition. Here, we review the different tools within TAPAS and illustrate how these may help provide a deeper understanding of neural and cognitive mechanisms of disease, with the ultimate goal of establishing automatized pipelines for predictions about individual patients. We hope that the openly available tools in TAPAS will contribute to the further development of TN/CP and facilitate the translation of advances in computational neuroscience into clinically relevant computational assays.

Model-based prediction of muscarinic receptor function from auditory mismatch negativity responses

Drugs affecting neuromodulation, for example by dopamine or acetylcholine, take centre stage among therapeutic strategies in psychiatry. These neuromodulators can change both neuronal gain and synaptic plasticity and therefore affect electrophysiological measures. An important goal for clinical diagnostics is to exploit this effect in the reverse direction, i.e., to infer the status of specific neuromodulatory systems from electrophysiological measures. In this study, we provide proof-of-concept that the functional status of cholinergic (specifically muscarinic) receptors can be inferred from electrophysiological data using generative (dynamic causal) models. To this end, we used epidural EEG recordings over two auditory cortical regions during a mismatch negativity (MMN) paradigm in rats. All animals were treated, across sessions, with muscarinic receptor agonists and antagonists at different doses. Together with a placebo condition, this resulted in five levels of muscarinic receptor status. Using a dynamic causal model - embodying a small network of coupled cortical microcircuits - we estimated synaptic parameters and their change across pharmacological conditions. The ensuing parameter estimates associated with (the neuromodulation of) synaptic efficacy showed both graded muscarinic effects and predictive validity between agonistic and antagonistic pharmacological conditions. This finding illustrates the potential utility of generative models of electrophysiological data as computational assays of muscarinic function. In application to EEG data of patients from heterogeneous spectrum diseases, e.g. schizophrenia, such models might help identify subgroups of patients that respond differentially to cholinergic treatments. SIGNIFICANCE STATEMENT: In psychiatry, the vast majority of pharmacological treatments affect actions of neuromodulatory transmitters, e.g. dopamine or acetylcholine. As treatment is largely trial-and-error based, one of the goals for computational psychiatry is to construct mathematical models that can serve as "computational assays" and infer the status of specific neuromodulatory systems in individual patients. This translational neuromodeling strategy has great promise for electrophysiological data in particular but requires careful validation. The present study demonstrates that the functional status of cholinergic (muscarinic) receptors can be inferred from electrophysiological data using dynamic causal models of neural circuits. While accuracy needs to be enhanced and our results must be replicated in larger samples, our current results provide proof-of-concept for computational assays of muscarinic function using EEG.

Markov chain Monte Carlo methods for hierarchical clustering of dynamic causal models

In this article, we address technical difficulties that arise when applying Markov chain Monte Carlo (MCMC) to hierarchical models designed to perform clustering in the space of latent parameters of subject‐wise generative models. Specifically, we focus on the case where the subject‐wise generative model is a dynamic causal model (DCM) for functional magnetic resonance imaging (fMRI) and clusters are defined in terms of effective brain connectivity. While an attractive approach for detecting mechanistically interpretable subgroups in heterogeneous populations, inverting such a hierarchical model represents a particularly challenging case, since DCM is often characterized by high posterior correlations between its parameters. In this context, standard MCMC schemes exhibit poor performance and extremely slow convergence. In this article, we investigate the properties of hierarchical clustering which lead to the observed failure of standard MCMC schemes and propose a solution designed to improve convergence but preserve computational complexity. Specifically, we introduce a class of proposal distributions which aims to capture the interdependencies between the parameters of the clustering and subject‐wise generative models and helps to reduce random walk behaviour of the MCMC scheme. Critically, these proposal distributions only introduce a single hyperparameter that needs to be tuned to achieve good performance. For validation, we apply our proposed solution to synthetic and real‐world datasets and also compare it, in terms of computational complexity and performance, to Hamiltonian Monte Carlo (HMC), a state‐of‐the‐art Monte Carlo technique. Our results indicate that, for the specific application domain considered here, our proposed solution shows good convergence performance and superior runtime compared to HMC.

Attenuated Directed Exploration during Reinforcement Learning in Gambling Disorder

Gambling disorder (GD) is a behavioral addiction associated with impairments in value-based decision-making and behavioral flexibility and might be linked to changes in the dopamine system. Maximizing long-term rewards requires a flexible trade-off between the exploitation of known options and the exploration of novel options for information gain. This exploration-exploitation trade-off is thought to depend on dopamine neurotransmission. We hypothesized that human gamblers would show a reduction in directed (uncertainty-based) exploration, accompanied by changes in brain activity in a fronto-parietal exploration-related network. Twenty-three frequent, non-treatment seeking gamblers and twenty-three healthy matched controls (all male) performed a four-armed bandit task during functional magnetic resonance imaging (fMRI). Computational modeling using hierarchical Bayesian parameter estimation revealed signatures of directed exploration, random exploration, and perseveration in both groups. Gamblers showed a reduction in directed exploration, whereas random exploration and perseveration were similar between groups. Neuroimaging revealed no evidence for group differences in neural representations of basic task variables (expected value, prediction errors). Our hypothesis of reduced frontal pole (FP) recruitment in gamblers was not supported. Exploratory analyses showed that during directed exploration, gamblers showed reduced parietal cortex and substantia-nigra/ventral-tegmental-area activity. Cross-validated classification analyses revealed that connectivity in an exploration-related network was predictive of group status, suggesting that connectivity patterns might be more predictive of problem gambling than univariate effects. Findings reveal specific reductions of strategic exploration in gamblers that might be linked to altered processing in a fronto-parietal network and/or changes in dopamine neurotransmission implicated in GD.SIGNIFICANCE STATEMENT Wiehler et al. (2021) report that gamblers rely less on the strategic exploration of unknown, but potentially better rewards during reward learning. This is reflected in a related network of brain activity. Parameters of this network can be used to predict the presence of problem gambling behavior in participants.

GABAergic cortical network physiology in frontotemporal lobar degeneration

The clinical syndromes caused by frontotemporal lobar degeneration are heterogeneous, including the behavioural variant frontotemporal dementia (bvFTD) and progressive supranuclear palsy. Although pathologically distinct, they share many behavioural, cognitive and physiological features, which may in part arise from common deficits of major neurotransmitters such as γ-aminobutyric acid (GABA). Here, we quantify the GABAergic impairment and its restoration with dynamic causal modelling of a double-blind placebo-controlled crossover pharmaco-magnetoencephalography study. We analysed 17 patients with bvFTD, 15 patients with progressive supranuclear palsy, and 20 healthy age- and gender-matched controls. In addition to neuropsychological assessment and structural MRI, participants undertook two magnetoencephalography sessions using a roving auditory oddball paradigm: once on placebo and once on 10 mg of the oral GABA reuptake inhibitor tiagabine. A subgroup underwent ultrahigh-field magnetic resonance spectroscopy measurement of GABA concentration, which was reduced among patients. We identified deficits in frontotemporal processing using conductance-based biophysical models of local and global neuronal networks. The clinical relevance of this physiological deficit is indicated by the correlation between top-down connectivity from frontal to temporal cortex and clinical measures of cognitive and behavioural change. A critical validation of the biophysical modelling approach was evidence from parametric empirical Bayes analysis that GABA levels in patients, measured by spectroscopy, were related to posterior estimates of patients’ GABAergic synaptic connectivity. Further evidence for the role of GABA in frontotemporal lobar degeneration came from confirmation that the effects of tiagabine on local circuits depended not only on participant group, but also on individual baseline GABA levels. Specifically, the phasic inhibition of deep cortico-cortical pyramidal neurons following tiagabine, but not placebo, was a function of GABA concentration. The study provides proof-of-concept for the potential of dynamic causal modelling to elucidate mechanisms of human neurodegenerative disease, and explains the variation in response to candidate therapies among patients. The laminar- and neurotransmitter-specific features of the modelling framework, can be used to study other treatment approaches and disorders. In the context of frontotemporal lobar degeneration, we suggest that neurophysiological restoration in selected patients, by targeting neurotransmitter deficits, could be used to bridge between clinical and preclinical models of disease, and inform the personalized selection of drugs and stratification of patients for future clinical trials.

Advances in the computational understanding of mental illness

Computational psychiatry is a rapidly growing field attempting to translate advances in computational neuroscience and machine learning into improved outcomes for patients suffering from mental illness. It encompasses both data-driven and theory-driven efforts. Here, recent advances in theory-driven work are reviewed. We argue that the brain is a computational organ. As such, an understanding of the illnesses arising from it will require a computational framework. The review divides work up into three theoretical approaches that have deep mathematical connections: dynamical systems, Bayesian inference and reinforcement learning. We discuss both general and specific challenges for the field, and suggest ways forward.

Connectivity-Related Roles of Contralesional Brain Regions for Motor Performance Early after Stroke

Hemiparesis after stroke is associated with increased neural activity not only in the lesioned but also in the contralesional hemisphere. While most studies have focused on the role of contralesional primary motor cortex (M1) activity for motor performance, data on other areas within the unaffected hemisphere are scarce, especially early after stroke. We here combined functional magnetic resonance imaging (fMRI) and transcranial magnetic stimulation (TMS) to elucidate the contribution of contralesional M1, dorsal premotor cortex (dPMC), and anterior intraparietal sulcus (aIPS) for the stroke-affected hand within the first 10 days after stroke. We used “online” TMS to interfere with neural activity at subject-specific fMRI coordinates while recording 3D movement kinematics. Interfering with aIPS activity improved tapping performance in patients, but not healthy controls, suggesting a maladaptive role of this region early poststroke. Analyzing effective connectivity parameters using a Lasso prediction model revealed that behavioral TMS effects were predicted by the coupling of the stimulated aIPS with dPMC and ipsilesional M1. In conclusion, we found a strong link between patterns of frontoparietal connectivity and TMS effects, indicating a detrimental influence of the contralesional aIPS on motor performance early after stroke.

Prediction of the Age at Onset of Spinocerebellar Ataxia Type 3 with Machine Learning

BACKGROUND

In polyglutamine (polyQ) disease, the investigation of the prediction of a patient's age at onset (AAO) facilitates the development of disease-modifying intervention and underpins the delay of disease onset and progression. Few polyQ disease studies have evaluated AAO predicted by machine-learning algorithms and linear regression methods.

OBJECTIVE

The objective of this study was to develop a machine-learning model for AAO prediction in the largest spinocerebellar ataxia type 3/Machado-Joseph disease (SCA3/MJD) population from mainland China.

METHODS

In this observational study, we introduced an innovative approach by systematically comparing the performance of 7 machine-learning algorithms with linear regression to explore AAO prediction in SCA3/MJD using CAG expansions of 10 polyQ-related genes, sex, and parental origin.

RESULTS

Similar prediction performance of testing set and training set in each models were identified and few overfitting of training data was observed. Overall, the machine-learning-based XGBoost model exhibited the most favorable performance in AAO prediction over the traditional linear regression method and other 6 machine-learning algorithms for the training set and testing set. The optimal XGBoost model achieved mean absolute error, root mean square error, and median absolute error of 5.56, 7.13, 4.15 years, respectively, in testing set 1, with mean absolute error (4.78 years), root mean square error (6.31 years), and median absolute error (3.59 years) in testing set 2.

CONCLUSION

Machine-learning algorithms can be used to predict AAO in patients with SCA3/MJD. The optimal XGBoost algorithm can provide a good reference for the establishment and optimization of prediction models for SCA3/MJD or other polyQ diseases. © 2020 International Parkinson and Movement Disorder Society.

The definition and measurement of heterogeneity

Heterogeneity is an important concept in psychiatric research and science more broadly. It negatively impacts effect size estimates under case-control paradigms, and it exposes important flaws in our existing categorical nosology. Yet, our field has no precise definition of heterogeneity proper. We tend to quantify heterogeneity by measuring associated correlates such as entropy or variance: practices which are akin to accepting the radius of a sphere as a measure of its volume. Under a definition of heterogeneity as the degree to which a system deviates from perfect conformity, this paper argues that its proper measure roughly corresponds to the size of a system's event/sample space, and has units known as numbers equivalent. We arrive at this conclusion through focused review of more than 100 years of (re)discoveries of indices by ecologists, economists, statistical physicists, and others. In parallel, we review psychiatric approaches for quantifying heterogeneity, including but not limited to studies of symptom heterogeneity, microbiome biodiversity, cluster-counting, and time-series analyses. We argue that using numbers equivalent heterogeneity measures could improve the interpretability and synthesis of psychiatric research on heterogeneity. However, significant limitations must be overcome for these measures-largely developed for economic and ecological research-to be useful in modern translational psychiatric science.

Towards a brain-based predictome of mental illness

Neuroimaging-based approaches have been extensively applied to study mental illness in recent years and have deepened our understanding of both cognitively healthy and disordered brain structure and function. Recent advancements in machine learning techniques have shown promising outcomes for individualized prediction and characterization of patients with psychiatric disorders. Studies have utilized features from a variety of neuroimaging modalities, including structural, functional, and diffusion magnetic resonance imaging data, as well as jointly estimated features from multiple modalities, to assess patients with heterogeneous mental disorders, such as schizophrenia and autism. We use the term "predictome" to describe the use of multivariate brain network features from one or more neuroimaging modalities to predict mental illness. In the predictome, multiple brain network-based features (either from the same modality or multiple modalities) are incorporated into a predictive model to jointly estimate features that are unique to a disorder and predict subjects accordingly. To date, more than 650 studies have been published on subject-level prediction focusing on psychiatric disorders. We have surveyed about 250 studies including schizophrenia, major depression, bipolar disorder, autism spectrum disorder, attention-deficit hyperactivity disorder, obsessive-compulsive disorder, social anxiety disorder, posttraumatic stress disorder, and substance dependence. In this review, we present a comprehensive review of recent neuroimaging-based predictomic approaches, current trends, and common shortcomings and share our vision for future directions.

The role of norepinephrine in the pathophysiology of schizophrenia

Several lines of evidence have suggested for decades a role for norepinephrine (NE) in the pathophysiology and treatment of schizophrenia. Recent experimental findings reveal anatomical and physiological properties of the locus coeruleus-norepinephrine (LC-NE) system and its involvement in brain function and cognition. Here, we integrate these two lines of evidence. First, we review the functional and structural properties of the LC-NE system and its impact on functional brain networks, cognition, and stress, with special emphasis on recent experimental and theoretical advances. Subsequently, we present an update about the role of LC-associated functions for the pathophysiology of schizophrenia, focusing on the cognitive and motivational deficits. We propose that schizophrenia phenomenology, in particular cognitive symptoms, may be explained by an abnormal interaction between genetic susceptibility and stress-initiated LC-NE dysfunction. This in turn, leads to imbalance between LC activity modes, dysfunctional regulation of brain network integration and neural gain, and deficits in cognitive functions. Finally, we suggest how recent development of experimental approaches can be used to characterize LC function in schizophrenia.

Supervised mixture of experts models for population health

We propose a machine learning driven approach to derive insights from observational healthcare data to improve public health outcomes. Our goal is to simultaneously identify patient subpopulations with differing health risks and to find those risk factors within each subpopulation. We develop two supervised mixture of experts models: a Supervised Gaussian Mixture model (SGMM) for general features and a Supervised Bernoulli Mixture model (SBMM) tailored to binary features. We demonstrate the two approaches on an analysis of high cost drivers of Medicaid expenditures for inpatient stays. We focus on the three diagnostic categories that accounted for the highest percentage of inpatient expenditures in New York State (NYS) in 2016. When compared with state-of-the-art learning methods (random forests, boosting, neural networks), our approaches provide comparable prediction performance while also extracting insightful subpopulation structure and risk factors. For problems with binary features the proposed SBMM provides as good or better performance than alternative methods while offering insightful explanations. Our results indicate the promise of such approaches for extracting population health insights from electronic health care records.

Approaches to Defining Common and Dissociable Neurobiological Deficits Associated With Psychopathology in Youth

Psychiatric disorders show high rates of comorbidity and nonspecificity of presenting clinical symptoms, while demonstrating substantial heterogeneity within diagnostic categories. Notably, many of these psychiatric disorders first manifest in youth. We review progress and next steps in efforts to parse heterogeneity in psychiatric symptoms in youths by identifying abnormalities within neural circuits. To address this fundamental challenge in psychiatry, a number of methods have been proposed. We provide an overview of these methods, broadly organized into dimensional versus categorical approaches and single-view versus multiview approaches. Dimensional approaches including factor analysis and canonical correlation analysis aim to capture dimensional associations between psychopathology and brain measures across a continuous spectrum from health to disease. In contrast, categorical approaches, such as clustering and community detection, aim to identify subtypes of individuals within a class of symptoms or brain features. We highlight several studies that apply these methods to samples of youths and discuss issues to consider when using these approaches. Finally, we end by highlighting avenues for future research.

Data-driven segmentation of audiometric phenotypes across a large clinical cohort

Pure tone audiograms are used to assess the degree and underlying source of hearing loss. Audiograms are typically categorized into a few canonical types, each thought to reflect distinct pathologies of the ear. Here, we analyzed 116,400 patient records from our clinic collected over a 24-year period and found that standard categorization left 46% of patient records unclassified. To better account for the full spectrum of hearing loss profiles, we used a Gaussian Mixture Model (GMM) to segment audiograms without any assumptions about frequency relationships, interaural symmetry or etiology. The GMM converged on ten types, featuring varying degrees of high-frequency hearing loss, flat loss, mixed loss, and notched profiles, with predictable relationships to patient age and sex. A separate GMM clustering of 15,380 audiograms from the National Health and Nutrition Examination Survey (NHANES) identified six similar types, that only lacked the more extreme hearing loss configurations observed in our patient cohort. Whereas traditional approaches distill hearing loss configurations down to a few canonical types by disregarding much of the underlying variability, an objective probabilistic model that accounted for all of the data identified an organized, but more heterogenous set of audiogram types that was consistent across two large clinical databases.

Making Sense of Computational Psychiatry

In psychiatry we often speak of constructing "models." Here we try to make sense of what such a claim might mean, starting with the most fundamental question: "What is (and isn't) a model?" We then discuss, in a concrete measurable sense, what it means for a model to be useful. In so doing, we first identify the added value that a computational model can provide in the context of accuracy and power. We then present limitations of standard statistical methods and provide suggestions for how we can expand the explanatory power of our analyses by reconceptualizing statistical models as dynamical systems. Finally, we address the problem of model building-suggesting ways in which computational psychiatry can escape the potential for cognitive biases imposed by classical hypothesis-driven research, exploiting deep systems-level information contained within neuroimaging data to advance our understanding of psychiatric neuroscience.

Neurostructural Heterogeneity in Youths With Internalizing Symptoms

BACKGROUND

Internalizing disorders such as anxiety and depression are common psychiatric disorders that frequently begin in youth and exhibit marked heterogeneity in treatment response and clinical course. Given that symptom-based classification approaches do not align with underlying neurobiology, an alternative approach is to identify neurobiologically informed subtypes based on brain imaging data.

METHODS

We used a recently developed semisupervised machine learning method (HYDRA [heterogeneity through discriminative analysis]) to delineate patterns of neurobiological heterogeneity within youths with internalizing symptoms using structural data collected at 3T from a sample of 1141 youths.

RESULTS

Using volume and cortical thickness, cross-validation methods indicated 2 highly stable subtypes of internalizing youths (adjusted Rand index = 0.66; permutation-based false discovery rate p < .001). Subtype 1, defined by smaller brain volumes and reduced cortical thickness, was marked by impaired cognitive performance and higher levels of psychopathology than both subtype 2 and typically developing youths. Using resting-state functional magnetic resonance imaging and diffusion images not considered during clustering, we found that subtype 1 also showed reduced amplitudes of low-frequency fluctuations in frontolimbic regions at rest and reduced fractional anisotropy in several white matter tracts. In contrast, subtype 2 showed intact cognitive performance and greater volume, cortical thickness, and amplitudes during rest compared with subtype 1 and typically developing youths, despite still showing clinically significant levels of psychopathology.

CONCLUSIONS

We identified 2 subtypes of internalizing youths differentiated by abnormalities in brain structure, function, and white matter integrity, with one of the subtypes showing poorer functioning across multiple domains. Identification of biologically grounded internalizing subtypes may assist in targeting early interventions and assessing longitudinal prognosis.

Predicting individual clinical trajectories of depression with generative embedding

•

Patients with major depressive disorder (MDD) show variable clinical trajectories.

•

Generative embedding (GE) is used to predict clinical trajectories in MDD patients.

•

GE classifies patients with chronic depression vs. fast remission with 79% accuracy.

•

GE provides mechanistic interpretability and outperforms conventional measures.

•

Proof-of-concept that illustrates the potential of GE for clinical prediction.

Patients with major depressive disorder (MDD) show heterogeneous treatment response and highly variable clinical trajectories: while some patients experience swift recovery, others show relapsing-remitting or chronic courses. Predicting individual clinical trajectories at an early stage is a key challenge for psychiatry and might facilitate individually tailored interventions. So far, however, reliable predictors at the single-patient level are absent. Here, we evaluated the utility of a machine learning strategy – generative embedding (GE) – which combines interpretable generative models with discriminative classifiers. Specifically, we used functional magnetic resonance imaging (fMRI) data of emotional face perception in 85 MDD patients from the NEtherlands Study of Depression and Anxiety (NESDA) who had been followed up over two years and classified into three subgroups with distinct clinical trajectories. Combining a generative model of effective (directed) connectivity with support vector machines (SVMs), we could predict whether a given patient would experience chronic depression vs. fast remission with a balanced accuracy of 79%. Gradual improvement vs. fast remission could still be predicted above-chance, but less convincingly, with a balanced accuracy of 61%. Generative embedding outperformed classification based on conventional (descriptive) features, such as functional connectivity or local activation estimates, which were obtained from the same data and did not allow for above-chance classification accuracy. Furthermore, predictive performance of GE could be assigned to a specific network property: the trial-by-trial modulation of connections by emotional content. Given the limited sample size of our study, the present results are preliminary but may serve as proof-of-concept, illustrating the potential of GE for obtaining clinical predictions that are interpretable in terms of network mechanisms. Our findings suggest that abnormal dynamic changes of connections involved in emotional face processing might be associated with higher risk of developing a less favorable clinical course.

Identifying schizophrenia subgroups using clustering and supervised learning

Schizophrenia has a 1% incidence rate world-wide and those diagnosed present with positive (e.g. hallucinations, delusions), negative (e.g. apathy, asociality), and cognitive symptoms. However, both symptom burden and associated brain alterations are highly heterogeneous and intimately linked to prognosis. In this study, we present a method to predict individual symptom profiles by first deriving clinical subgroups and then using machine learning methods to perform subject-level classification based on magnetic resonance imaging (MRI) derived neuroanatomical measures. Symptomatic and MRI data of 167 subjects were used. Subgroups were defined using hierarchical clustering of clinical data resulting in 3 stable clusters: 1) high symptom burden, 2) predominantly positive symptom burden, and 3) mild symptom burden. Cortical thickness estimates were obtained in 78 regions of interest and were input, along with demographic data, into three machine learning models (logistic regression, support vector machine, and random forest) to predict subgroups. Random forest performance metrics for predicting the group membership of the high and mild symptom burden groups exceeded those of the baseline comparison of the entire schizophrenia population versus normal controls (AUC: 0.81 and 0.78 vs. 0.75). Additionally, an analysis of the most important features in the random forest classification demonstrated consistencies with previous findings of regional impairments and symptoms of schizophrenia.

[Computational psychiatry : Data-driven vs. mechanistic approaches]

The emerging research field of so-called computational psychiatry attempts to contribute to an understanding of complex psychiatric phenomena by applying computational methods and to promote the translation of neuroscientific research results into clinical practice. This article presents this field of research using selected examples based on the distinction between data-driven and theory-driven approaches. Exemplary for a data-driven approach are studies to predict clinical outcome, for example, in persons with a high-risk state for psychosis or on the response to pharmacological treatment for depression. Theory-driven approaches attempt to describe the mechanisms of altered information processing as the cause of psychiatric symptoms at the behavioral and neuronal level. In computational models possible mechanisms can be described that may have produced the measured behavioral or neuronal data. For example, in schizophrenia patients the clinical phenomenon of aberrant salience has been described as learning irrelevant information or cognitive deficits have been linked to connectivity changes in frontoparietal networks. Computational psychiatry can make important contributions to the prediction of individual clinical courses as well as to a mechanistic understanding of psychiatric symptoms. For this a further development of reliable and valid methods across different disciplines is indispensable.

Clinical-learning versus machine-learning for transdiagnostic prediction of psychosis onset in individuals at-risk

Predicting the onset of psychosis in individuals at-risk is based on robust prognostic model building methods including a priori clinical knowledge (also termed clinical-learning) to preselect predictors or machine-learning methods to select predictors automatically. To date, there is no empirical research comparing the prognostic accuracy of these two methods for the prediction of psychosis onset. In a first experiment, no improved performance was observed when machine-learning methods (LASSO and RIDGE) were applied-using the same predictors-to an individualised, transdiagnostic, clinically based, risk calculator previously developed on the basis of clinical-learning (predictors: age, gender, age by gender, ethnicity, ICD-10 diagnostic spectrum), and externally validated twice. In a second experiment, two refined versions of the published model which expanded the granularity of the ICD-10 diagnosis were introduced: ICD-10 diagnostic categories and ICD-10 diagnostic subdivisions. Although these refined versions showed an increase in apparent performance, their external performance was similar to the original model. In a third experiment, the three refined models were analysed under machine-learning and clinical-learning with a variable event per variable ratio (EPV). The best performing model under low EPVs was obtained through machine-learning approaches. The development of prognostic models on the basis of a priori clinical knowledge, large samples and adequate events per variable is a robust clinical prediction method to forecast psychosis onset in patients at-risk, and is comparable to machine-learning methods, which are more difficult to interpret and implement. Machine-learning methods should be preferred for high dimensional data when no a priori knowledge is available.

Subjective estimates of uncertainty during gambling and impulsivity after subthalamic deep brain stimulation for Parkinson's disease

Subthalamic deep brain stimulation (DBS) for Parkinson’s disease (PD) may modulate chronometric and instrumental aspects of choice behaviour, including motor inhibition, decisional slowing, and value sensitivity. However, it is not well known whether subthalamic DBS affects more complex aspects of decision-making, such as the influence of subjective estimates of uncertainty on choices. In this study, 38 participants with PD played a virtual casino prior to subthalamic DBS (whilst ‘on’ medication) and again, 3-months postoperatively (whilst ‘on’ stimulation). At the group level, there was a small but statistically significant decrease in impulsivity postoperatively, as quantified by the Barratt Impulsiveness Scale (BIS). The gambling behaviour of participants (bet increases, slot machine switches and double or nothing gambles) was associated with this self-reported measure of impulsivity. However, there was a large variance in outcome amongst participants, and we were interested in whether individual differences in subjective estimates of uncertainty (specifically, volatility) were related to differences in pre- and postoperative impulsivity. To examine these individual differences, we fit a computational model (the Hierarchical Gaussian Filter, HGF), to choices made during slot machine game play as well as a simpler reinforcement learning model based on the Rescorla-Wagner formalism. The HGF was superior in accounting for the behaviour of our participants, suggesting that participants incorporated beliefs about environmental uncertainty when updating their beliefs about gambling outcome and translating these beliefs into action. A specific aspect of subjective uncertainty, the participant’s estimate of the tendency of the slot machine’s winning probability to change (volatility), increased subsequent to DBS. Additionally, the decision temperature of the response model decreased post-operatively, implying greater stochasticity in the belief-to-choice mapping of participants. Model parameter estimates were significantly associated with impulsivity; specifically, increased uncertainty was related to increased postoperative impulsivity. Moreover, changes in these parameter estimates were significantly associated with the maximum post-operative change in impulsivity over a six month follow up period. Our findings suggest that impulsivity in PD patients may be influenced by subjective estimates of uncertainty (environmental volatility) and implicate a role for the subthalamic nucleus in the modulation of outcome certainty. Furthermore, our work outlines a possible approach to characterising those persons who become more impulsive after subthalamic DBS, an intervention in which non-motor outcomes can be highly variable.

Towards a Unifying Cognitive, Neurophysiological, and Computational Neuroscience Account of Schizophrenia

Psychotic experiences may be understood as altered information processing due to aberrant neural computations. A prominent example of such neural computations is the computation of prediction errors (PEs), which signal the difference between expected and experienced events. Among other areas showing PE coding, hippocampal-prefrontal-striatal neurocircuits play a prominent role in information processing. Dysregulation of dopaminergic signaling, often secondary to psychosocial stress, is thought to interfere with the processing of biologically important events (such as reward prediction errors) and result in the aberrant attribution of salience to irrelevant sensory stimuli and internal representations. Bayesian hierarchical predictive coding offers a promising framework for the identification of dysfunctional neurocomputational processes and the development of a mechanistic understanding of psychotic experience. According to this framework, mismatches between prior beliefs encoded at higher levels of the cortical hierarchy and lower-level (sensory) information can also be thought of as PEs, with important consequences for belief updating. Low levels of precision in the representation of prior beliefs relative to sensory data, as well as dysfunctional interactions between prior beliefs and sensory data in an ever-changing environment, have been suggested as a general mechanism underlying psychotic experiences. Translating the promise of the Bayesian hierarchical predictive coding into patient benefit will come from integrating this framework with existing knowledge of the etiology and pathophysiology of psychosis, especially regarding hippocampal-prefrontal-striatal network function and neural mechanisms of information processing and belief updating.

Progress in psychoradiology, the clinical application of psychiatric neuroimaging

Psychoradiology is an emerging field that applies radiological imaging technologies to psychiatric conditions. In the past three decades, brain imaging techniques have rapidly advanced understanding of illness and treatment effects in psychiatry. Based on these advances, radiologists have become increasingly interested in applying these advances for differential diagnosis and individualized patient care selection for common psychiatric illnesses. This shift from research to clinical practice represents the beginning evolution of psychoradiology. In this review, we provide a summary of recent progress relevant to this field based on their clinical functions, namely the (1) classification and subtyping; (2) prediction and monitoring of treatment outcomes; and (3) treatment selection. In addition, we provide guidelines for the practice of psychoradiology in clinical settings and suggestions for future research to validate broader clinical applications. Given the high prevalence of psychiatric disorders and the importance of increased participation of radiologists in this field, a guide regarding advances in this field and a description of relevant clinical work flow patterns help radiologists contribute to this fast-evolving field.

Towards artificial intelligence in mental health by improving schizophrenia prediction with multiple brain parcellation ensemble-learning

In the literature, there are substantial machine learning attempts to classify schizophrenia based on alterations in resting-state (RS) brain patterns using functional magnetic resonance imaging (fMRI). Most earlier studies modelled patients undergoing treatment, entailing confounding with drug effects on brain activity, and making them less applicable to real-world diagnosis at the point of first medical contact. Further, most studies with classification accuracies >80% are based on small sample datasets, which may be insufficient to capture the heterogeneity of schizophrenia, limiting generalization to unseen cases. In this study, we used RS fMRI data collected from a cohort of antipsychotic drug treatment-naive patients meeting DSM IV criteria for schizophrenia (N = 81) as well as age- and sex-matched healthy controls (N = 93). We present an ensemble model -- EMPaSchiz (read as ‘Emphasis’; standing for ‘Ensemble algorithm with Multiple Parcellations for Schizophrenia prediction’) that stacks predictions from several ‘single-source’ models, each based on features of regional activity and functional connectivity, over a range of different a priori parcellation schemes. EMPaSchiz yielded a classification accuracy of 87% (vs. chance accuracy of 53%), which out-performs earlier machine learning models built for diagnosing schizophrenia using RS fMRI measures modelled on large samples (N > 100). To our knowledge, EMPaSchiz is first to be reported that has been trained and validated exclusively on data from drug-naive patients diagnosed with schizophrenia. The method relies on a single modality of MRI acquisition and can be readily scaled-up without needing to rebuild parcellation maps from incoming training images.

Brain space and time in mental disorders: Paradigm shift in biological psychiatry

Contemporary psychiatry faces serious challenges because it has failed to incorporate accumulated knowledge from basic neuroscience, neurophilosophy, and brain-mind relation studies. As a consequence, it has limited explanatory power, and effective treatment options are hard to come by. A new conceptual framework for understanding mental health based on underlying neurobiological spatial-temporal mechanisms of mental disorders (already gained by the experimental studies) is beginning to emerge.

Development of Neuroimaging-Based Biomarkers in Psychiatry

This chapter presents an overview of accumulating neuroimaging data with emphasis on translational potential. The subject will be described in the context of three disease states, i.e., schizophrenia, bipolar disorder, and major depressive disorder, and for three clinical goals, i.e., disease risk assessment, subtyping, and treatment decision.