Prediction of outcome in the psychosis prodrome using neuroanatomical pattern classification

Illness Phase as a Key Assessment and Intervention Window for Psychosis

The phenotype of schizophrenia, regardless of etiology, represents the most studied psychotic disorder with respect to neurobiology and distinct phases of illness. The early phase of illness represents a unique opportunity to provide effective and individualized interventions that can alter illness trajectories. Developmental age and illness stage, including temporal variation in neurobiology, can be targeted to develop phase-specific clinical assessment, biomarkers, and interventions. We review an earlier model whereby an initial glutamate signaling deficit progresses through different phases of allostatic adaptation, moving from potentially reversible functional abnormalities associated with early psychosis and working memory dysfunction, and ending with difficult-to-reverse structural changes after chronic illness. We integrate this model with evidence of dopaminergic abnormalities, including cortical D1 dysfunction, which develop during adolescence. We discuss how this model and a focus on a potential critical window of intervention in the early stages of schizophrenia impact the approach to research design and clinical care. This impact includes stage-specific considerations for symptom assessment as well as genetic, cognitive, and neurophysiological biomarkers. We examine how phase-specific biomarkers of illness phase and brain development can be incorporated into current strategies for large-scale research and clinical programs implementing coordinated specialty care. We highlight working memory and D1 dysfunction as early treatment targets that can substantially affect functional outcome.

A Review of Potential Neuroimaging Biomarkers of Schizophrenia-Risk

The risk for developing schizophrenia is increased among first-degree relatives of those with psychotic disorders, but the risk is even higher in those meeting established criteria for clinical high risk (CHR), a clinical construct most often comprising of attenuated psychotic experiences. Conversion to psychosis among CHR youth has been reported to be about 15-35% over three years. Accurately identifying individuals whose psychotic symptoms will worsen would facilitate earlier intervention, but this has been difficult to do using behavior measures alone. Brain-based risk markers have the potential to improve the accuracy of predicting outcomes in CHR youth. This narrative review provides an overview of neuroimaging studies used to investigate psychosis risk, including studies involving structural, functional, and diffusion imaging, functional connectivity, positron emission tomography, arterial spin labeling, magnetic resonance spectroscopy, and multi-modality approaches. We present findings separately in those observed in the CHR state and those associated with psychosis progression or resilience. Finally, we discuss future research directions that could improve clinical care for those at high risk for developing psychotic disorders.

Machine Learning and Non-Affective Psychosis: Identification, Differential Diagnosis, and Treatment

PURPOSE OF REVIEW

This review will cover the most relevant findings on the use of machine learning (ML) techniques in the field of non-affective psychosis, by summarizing the studies published in the last three years focusing on illness detection and treatment.

RECENT FINDINGS

Multiple ML tools that include mostly supervised approaches such as support vector machine, gradient boosting, and random forest showed promising results by applying these algorithms to various sources of data: socio-demographic information, EEG, language, digital content, blood biomarkers, neuroimaging, and electronic health records. However, the overall performance, in the binary classification case, varied from 0.49, which is to be considered very low (i.e., noise), to over 0.90. These results are fully justified by different factors, some of which may be attributable to the preprocessing of the data, the wide variety of the data, and the a-priori setting of hyperparameters. One of the main limitations of the field is the lack of stratification of results based on biological sex, given that psychosis presents differently in men and women; hence, the necessity to tailor identification tools and data analytic strategies. Timely identification and appropriate treatment are key factors in reducing the consequences of psychotic disorders. In recent years, the emergence of new analytical tools based on artificial intelligence such as supervised ML approaches showed promises as a potential breakthrough in this field. However, ML applications in everyday practice are still in its infancy.

Predictive models for social functioning in healthy young adults: A machine learning study integrating neuroanatomical, cognitive, and behavioral data

Poor social functioning is an emerging public health problem associated with physical and mental health consequences. Developing prognostic tools is critical to identify individuals at risk for poor social functioning and guide interventions. We aimed to inform prediction models of social functioning by evaluating models relying on bio-behavioral data using machine learning. With data from the Human Connectome Project Healthy Young Adult sample (age 22-35, N = 1,101), we built Support Vector Regression models to estimate social functioning from variable sets of brain morphology to behavior with increasing complexity: 1) brain-only model, 2) brain-cognition model, 3) cognition-behavioral model, and 4) combined brain-cognition-behavioral model. Predictive accuracy of each model was assessed and the importance of individual variables for model performance was determined. The combined and cognition-behavioral models significantly predicted social functioning, whereas the brain-only and brain-cognition models did not. Negative affect, psychological wellbeing, extraversion, withdrawal, and cortical thickness of the rostral middle-frontal and superior-temporal regions were the most important predictors in the combined model. Results demonstrate that social functioning can be accurately predicted using machine learning methods. Behavioral markers may be more significant predictors of social functioning than brain measures for healthy young adults and may represent important leverage points for preventative intervention.

Shape features of working memory-related deep-brain regions differentiate high and low community functioning in schizophrenia

We have previously shown that schizophrenia (SCZ) participants with high community functioning demonstrate better verbal working memory (vWM) performance relative to those with low community functioning. In the present study, we investigated whether neuroanatomical differences in regions supporting vWM also exist between schizophrenia groups that vary on community functioning. Utilizing magnetic resonance imaging, shape features of deep-brain nuclei known to be involved in vWM were calculated in samples of high functioning (HF-SCZ, n = 23) and low functioning schizophrenia participants (LF-SCZ, n = 18), as well as in a group of healthy control participants (CON, n = 45). Large deformation diffeomorphic metric mapping was employed to characterize surface anatomy of the caudate nucleus, globus pallidus, hippocampus, and thalamus. Statistical analyses involved linear mixed-effects models and vertex-wise contrast mapping to assess between-group differences in structural shape features, and Pearson correlations to evaluate relationships between shape metrics and vWM performance. We found significant between-group main effects in deep-brain surface anatomy across all structures. Post-hoc comparisons revealed HF-SCZ and LF-SCZ groups significantly differed on both caudate and hippocampal shape, however, significant correlations with vWM were only observed in hippocampal shape for both SCZ groups. Specifically, more abnormal hippocampal deformation was associated with lower vWM suggesting hippocampal shape is both a neural substrate for vWM deficits and a potential biomarker to predict or monitor the efficacy of cognitive rehabilitation. These findings add to a growing body of literature related to functional outcomes in schizophrenia by demonstrating unique shape patterns across the spectrum of community functioning in SCZ.

•

Deep-brain abnormalities are present in patients regardless of functional severity.

•

Caudate and hippocampal shape differ between community functioning-based groups.

•

Verbal working memory relates to hippocampal shape in both patient groups.

Using combined environmental-clinical classification models to predict role functioning outcome in clinical high-risk states for psychosis and recent-onset depression

BACKGROUND

Clinical high-risk states for psychosis (CHR) are associated with functional impairments and depressive disorders. A previous PRONIA study predicted social functioning in CHR and recent-onset depression (ROD) based on structural magnetic resonance imaging (sMRI) and clinical data. However, the combination of these domains did not lead to accurate role functioning prediction, calling for the investigation of additional risk dimensions. Role functioning may be more strongly associated with environmental adverse events than social functioning.

AIMS

We aimed to predict role functioning in CHR, ROD and transdiagnostically, by adding environmental adverse events-related variables to clinical and sMRI data domains within the PRONIA sample.

METHOD

Baseline clinical, environmental and sMRI data collected in 92 CHR and 95 ROD samples were trained to predict lower versus higher follow-up role functioning, using support vector classification and mixed k-fold/leave-site-out cross-validation. We built separate predictions for each domain, created multimodal predictions and validated them in independent cohorts (74 CHR, 66 ROD).

RESULTS

Models combining clinical and environmental data predicted role outcome in discovery and replication samples of CHR (balanced accuracies: 65.4% and 67.7%, respectively), ROD (balanced accuracies: 58.9% and 62.5%, respectively), and transdiagnostically (balanced accuracies: 62.4% and 68.2%, respectively). The most reliable environmental features for role outcome prediction were adult environmental adjustment, childhood trauma in CHR and childhood environmental adjustment in ROD.

CONCLUSIONS

Findings support the hypothesis that environmental variables inform role outcome prediction, highlight the existence of both transdiagnostic and syndrome-specific predictive environmental adverse events, and emphasise the importance of implementing real-world models by measuring multiple risk dimensions.

Neural Correlates of Smooth Pursuit Eye Movements in Schizotypy and Recent Onset Psychosis: A Multivariate Pattern Classification Approach

Schizotypy refers to a set of personality traits that bear resemblance, at subclinical level, to psychosis. Despite evidence of similarity at multiple levels of analysis, direct comparisons of schizotypy and clinical psychotic disorders are rare. Therefore, we used functional magnetic resonance imaging (fMRI) to examine the neural correlates and task-based functional connectivity (psychophysiological interactions; PPI) of smooth pursuit eye movements (SPEM) in patients with recent onset psychosis (ROP; n = 34), participants with high levels of negative (HNS; n = 46) or positive (HPS; n = 41) schizotypal traits, and low-schizotypy control participants (LS; n = 61) using machine-learning. Despite strong previous evidence that SPEM is a highly reliable marker of psychosis, patients and controls could not be significantly distinguished based on SPEM performance or blood oxygen level dependent (BOLD) signal during SPEM. Classification was, however, significant for the right frontal eye field (FEF) seed region in the PPI analyses but not for seed regions in other key areas of the SPEM network. Applying the right FEF classifier to the schizotypal samples yielded decision scores between the LS and ROP groups, suggesting similarities and dissimilarities of the HNS and HPS samples with the LS and ROP groups. The very small difference between groups is inconsistent with previous studies that showed significant differences between patients with ROP and controls in both SPEM performance and underlying neural mechanisms with large effect sizes. As the current study had sufficient power to detect such differences, other reasons are discussed.

Prediction of transition to psychosis from an at-risk mental state using structural neuroimaging, genetic, and environmental data

INTRODUCTION

Psychosis is usually preceded by a prodromal phase in which patients are clinically identified as being at in an "At Risk Mental State" (ARMS). A few studies have demonstrated the feasibility of predicting psychosis transition from an ARMS using structural magnetic resonance imaging (sMRI) data and machine learning (ML) methods. However, the reliability of these findings is unclear due to possible sampling bias. Moreover, the value of genetic and environmental data in predicting transition to psychosis from an ARMS is yet to be explored.

METHODS

In this study we aimed to predict transition to psychosis from an ARMS using a combination of ML, sMRI, genome-wide genotypes, and environmental risk factors as predictors, in a sample drawn from a pool of 246 ARMS subjects (60 of whom later transitioned to psychosis). First, the modality-specific values in predicting transition to psychosis were evaluated using several: (a) feature types; (b) feature manipulation strategies; (c) ML algorithms; (d) cross-validation strategies, as well as sample balancing and bootstrapping. Subsequently, the modalities whose at least 60% of the classification models showed an balanced accuracy (BAC) statistically better than chance level were included in a multimodal classification model.

RESULTS AND DISCUSSION

Results showed that none of the modalities alone, i.e., neuroimaging, genetic or environmental data, could predict psychosis from an ARMS statistically better than chance and, as such, no multimodal classification model was trained/tested. These results suggest that the value of structural MRI data and genome-wide genotypes in predicting psychosis from an ARMS, which has been fostered by previous evidence, should be reconsidered.

Reduced TMS-evoked fast oscillations in the motor cortex predict the severity of positive symptoms in first-episode psychosis

Accumulating evidence points to neurophysiological abnormalities of the motor cortex in Schizophrenia (SCZ). However, whether these abnormalities represent a core biological feature of psychosis rather than a superimposed neurodegenerative process is yet to be defined, as it is their putative relationship with clinical symptoms. in this study, we used Transcranial Magnetic Stimulation coupled with electroencephalography (TMS-EEG) to probe the intrinsic oscillatory properties of motor (Brodmann Area 4, BA4) and non-motor (posterior parietal, BA7) cortical areas in twenty-three first-episode psychosis (FEP) patients and thirteen age and gender-matched healthy comparison (HC) subjects. Patients underwent clinical evaluation at baseline and six-months after the TMS-EEG session. We found that FEP patients had reduced EEG activity evoked by TMS of the motor cortex in the beta-2 (25-34 Hz) frequency band in a cluster of electrodes overlying BA4, relative to HC participants. Beta-2 deficits in the TMS-evoked EEG response correlated with worse positive psychotic symptoms at baseline and also predicted positive symptoms severity at six-month follow-up assessments. Altogether, these findings indicate that reduced TMS-evoked fast oscillatory activity in the motor cortex is an early neural abnormality that: 1) is present at illness onset; 2) may represent a state marker of psychosis; and 3) could play a role in the development of new tools of outcome prediction in psychotic patients.

Surface area in the insula was associated with 28-month functional outcome in first-episode psychosis

Many studies have tested the relationship between demographic, clinical, and psychobiological measurements and clinical outcomes in ultra-high risk for psychosis (UHR) and first-episode psychosis (FEP). However, no study has investigated the relationship between multi-modal measurements and long-term outcomes for >2 years. Thirty-eight individuals with UHR and 29 patients with FEP were measured using one or more modalities (cognitive battery, electrophysiological response, structural magnetic resonance imaging, and functional near-infrared spectroscopy). We explored the characteristics associated with 13- and 28-month clinical outcomes. In UHR, the cortical surface area in the left orbital part of the inferior frontal gyrus was negatively associated with 13-month disorganized symptoms. In FEP, the cortical surface area in the left insula was positively associated with 28-month global social function. The left inferior frontal gyrus and insula are well-known structural brain characteristics in schizophrenia, and future studies on the pathological mechanism of structural alteration would provide a clearer understanding of the disease.

Multivariate pattern analysis of brain structure predicts functional outcome after auditory-based cognitive training interventions

Cognitive gains following cognitive training interventions are associated with improved functioning in people with schizophrenia (SCZ). However, considerable inter-individual variability is observed. Here, we evaluate the sensitivity of brain structural features to predict functional response to auditory-based cognitive training (ABCT) at a single-subject level. We employed whole-brain multivariate pattern analysis with support vector machine (SVM) modeling to identify gray matter (GM) patterns that predicted higher vs. lower functioning after 40 h of ABCT at the single-subject level in SCZ patients. The generalization capacity of the SVM model was evaluated by applying the original model through an out-of-sample cross-validation analysis to unseen SCZ patients from an independent validation sample who underwent 50 h of ABCT. The whole-brain GM volume-based pattern classification predicted higher vs. lower functioning at follow-up with a balanced accuracy (BAC) of 69.4% (sensitivity 72.2%, specificity 66.7%) as determined by nested cross-validation. The neuroanatomical model was generalizable to an independent cohort with a BAC of 62.1% (sensitivity 90.9%, specificity 33.3%). In particular, greater baseline GM volumes in regions within superior temporal gyrus, thalamus, anterior cingulate, and cerebellum predicted improved functioning at the single-subject level following ABCT in SCZ participants. The present findings provide a structural MRI fingerprint associated with preserved GM volumes at a single baseline timepoint, which predicted improved functioning following an ABCT intervention, and serve as a model for how to facilitate precision clinical therapies for SCZ based on imaging data, operating at the single-subject level.

Multimodal prognosis of negative symptom severity in individuals at increased risk of developing psychosis

Negative symptoms occur frequently in individuals at clinical high risk (CHR) for psychosis and contribute to functional impairments. The aim of this study was to predict negative symptom severity in CHR after 9 months. Predictive models either included baseline negative symptoms measured with the Structured Interview for Psychosis-Risk Syndromes (SIPS-N), whole-brain gyrification, or both to forecast negative symptoms of at least moderate severity in 94 CHR. We also conducted sequential risk stratification to stratify CHR into different risk groups based on the SIPS-N and gyrification model. Additionally, we assessed the models' ability to predict functional outcomes in CHR and their transdiagnostic generalizability to predict negative symptoms in 96 patients with recent-onset psychosis (ROP) and 97 patients with recent-onset depression (ROD). Baseline SIPS-N and gyrification predicted moderate/severe negative symptoms with significant balanced accuracies of 68 and 62%, while the combined model achieved 73% accuracy. Sequential risk stratification stratified CHR into a high (83%), medium (40-64%), and low (19%) risk group regarding their risk of having moderate/severe negative symptoms at 9 months follow-up. The baseline SIPS-N model was also able to predict social (61%), but not role functioning (59%) at above-chance accuracies, whereas the gyrification model achieved significant accuracies in predicting both social (76%) and role (74%) functioning in CHR. Finally, only the baseline SIPS-N model showed transdiagnostic generalization to ROP (63%). This study delivers a multimodal prognostic model to identify those CHR with a clinically relevant negative symptom severity and functional impairments, potentially requiring further therapeutic consideration.

The relationship between cognitive deficits and impaired short-term functional outcome in clinical high-risk for psychosis participants: A machine learning and modelling approach

Poor functional outcomes are common in individuals at clinical high-risk for psychosis (CHR-P), but the contribution of cognitive deficits remains unclear. We examined the potential utility of cognitive variables in predictive models of functioning at baseline and follow-up with machine learning methods. Additional models fitted on baseline functioning variables were used as a benchmark to evaluate model performance. Data were available for 1) 146 CHR-P individuals of whom 118 completed a 6- and/or 12-month follow-up, 2) 47 participants not fulfilling CHR criteria (CHR-Ns) but displaying affective and substance use disorders and 3) 55 healthy controls (HCs). Predictors of baseline global assessment of functioning (GAF) scores were selected by L1-regularised least angle regression and then used to train classifiers to predict functional outcome in CHR-P individuals. In CHR-P participants, cognitive deficits together with clinical and functioning variables explained 41% of the variance in baseline GAF scores while cognitive variables alone explained 12%. These variables allowed classification of functional outcome with an average balanced accuracy (BAC) of 63% in both mixed- and cross-site models. However, higher accuracies (68%-70%) were achieved using classifiers fitted only on baseline functioning variables. Our findings suggest that cognitive deficits, alongside clinical and functioning variables, displayed robust relationships with impaired functioning in CHR-P participants at baseline and follow-up. Moreover, these variables allow for prediction of functional outcome. However, models based on baseline functioning variables showed a similar performance, highlighting the need to develop more accurate algorithms for predicting functional outcome in CHR-P participants.

Social reward processing: A biomarker for predicting psychosis risk?

The desire to obtain social rewards (e.g. positive feedback) features prominently in our lives and relationships, and is relevant to understanding psychopathology - where behavior is often impaired. Investigating social rewards within the psychosis-spectrum offers an especially useful opportunity, given the high rates of impaired social functioning and social isolation. The goal of this study was to investigate hedonic experience associated with social reward processing as a potential biomarker for psychosis risk. This study used a task-based functional magnetic resonance imaging (fMRI) paradigm in adolescents at clinical high-risk for the development of psychosis (CHR, n = 19) and healthy unaffected peers (healthy controls - HC, n = 20). Regional activation and connectivity of the ventromedial prefrontal cortex and ventral striatum were examined in response to receiving positive social feedback relative to an ambiguous feedback condition. Expectations of impaired hedonic processes in CHR youth were generally not supported, as there were no group differences in neural response or task-based connectivity. Although interesting relationships were found linking neural reward response and connectivity with social, anticipatory, and consummatory anhedonia in the CHR group, results are difficult to interpret in light of task limitations. We discuss potential implications for future study designs that seek to investigate social reward processing as a biomarker for psychosis risk.

Structural and functional imaging markers for susceptibility to psychosis

The introduction of clinical criteria for the operationalization of psychosis high risk provided a basis for early detection and treatment of vulnerable individuals. However, about two-thirds of people meeting clinical high-risk (CHR) criteria will never develop a psychotic disorder. In the effort to increase prognostic precision, structural and functional neuroimaging have received growing attention as a potentially useful resource in the prediction of psychotic transition in CHR patients. The present review summarizes current research on neuroimaging biomarkers in the CHR state, with a particular focus on their prognostic utility and limitations. Large, multimodal/multicenter studies are warranted to address issues important for clinical applicability such as generalizability and replicability, standardization of clinical definitions and neuroimaging methods, and consideration of contextual factors (e.g., age, comorbidity).

Individualized Diagnostic and Prognostic Models for Patients With Psychosis Risk Syndromes: A Meta-analytic View on the State of the Art

BACKGROUND

The clinical high risk (CHR) paradigm has facilitated research into the underpinnings of help-seeking individuals at risk for developing psychosis, aiming at predicting and possibly preventing transition to the overt disorder. Statistical methods such as machine learning and Cox regression have provided the methodological basis for this research by enabling the construction of diagnostic models (i.e., distinguishing CHR individuals from healthy individuals) and prognostic models (i.e., predicting a future outcome) based on different data modalities, including clinical, neurocognitive, and neurobiological data. However, their translation to clinical practice is still hindered by the high heterogeneity of both CHR populations and methodologies applied.

METHODS

We systematically reviewed the literature on diagnostic and prognostic models built on Cox regression and machine learning. Furthermore, we conducted a meta-analysis on prediction performances investigating heterogeneity of methodological approaches and data modality.

RESULTS

A total of 44 articles were included, covering 3707 individuals for prognostic studies and 1052 individuals for diagnostic studies (572 CHR patients and 480 healthy control subjects). CHR patients could be classified against healthy control subjects with 78% sensitivity and 77% specificity. Across prognostic models, sensitivity reached 67% and specificity reached 78%. Machine learning models outperformed those applying Cox regression by 10% sensitivity. There was a publication bias for prognostic studies yet no other moderator effects.

CONCLUSIONS

Our results may be driven by substantial clinical and methodological heterogeneity currently affecting several aspects of the CHR field and limiting the clinical implementability of the proposed models. We discuss conceptual and methodological harmonization strategies to facilitate more reliable and generalizable models for future clinical practice.

Predicting individual improvement in schizophrenia symptom severity at 1-year follow-up: Comparison of connectomic, structural, and clinical predictors

In a machine learning setting, this study aims to compare the prognostic utility of connectomic, brain structural, and clinical/demographic predictors of individual change in symptom severity in individuals with schizophrenia. Symptom severity at baseline and 1-year follow-up was assessed in 30 individuals with a schizophrenia-spectrum disorder using the Brief Psychiatric Rating Scale. Structural and functional neuroimaging was acquired in all individuals at baseline. Machine learning classifiers were trained to predict whether individuals improved or worsened with respect to positive, negative, and overall symptom severity. Classifiers were trained using various combinations of predictors, including regional cortical thickness and gray matter volume, static and dynamic resting-state connectivity, and/or baseline clinical and demographic variables. Relative change in overall symptom severity between baseline and 1-year follow-up varied markedly among individuals (interquartile range: 55%). Dynamic resting-state connectivity measured within the default-mode network was the most accurate single predictor of change in positive (accuracy: 87%), negative (83%), and overall symptom severity (77%) at follow-up. Incorporating predictors based on regional cortical thickness, gray matter volume, and baseline clinical variables did not markedly improve prediction accuracy and the prognostic utility of these predictors in isolation was moderate (<70%). Worsening negative symptoms at 1-year follow-up were predicted by hyper-connectivity and hypo-dynamism within the default-mode network at baseline assessment, while hypo-connectivity and hyper-dynamism predicted worsening positive symptoms. Given the modest sample size investigated, we recommend giving precedence to the relative ranking of the predictors investigated in this study, rather than the prediction accuracy estimates.

Neuroimaging Heterogeneity in Psychosis: Neurobiological Underpinnings and Opportunities for Prognostic and Therapeutic Innovation

Heterogeneity in symptom presentation, outcomes, and treatment response has long been problematic for researchers aiming to identify biological markers of schizophrenia or psychosis. However, there is increasing recognition that there may likely be no such general illness markers, which is consistent with the notion of a group of schizophrenia(s) that may have both shared and unique neurobiological pathways. Instead, strategies aiming to capitalize on or leverage such heterogeneity may help uncover neurobiological pathways that may then be used to stratify groups of patients for prognostic purposes or for therapeutic trials. A shift toward larger sample sizes with adequate statistical power to overcome small effect sizes and disentangle the shared variance among different brain-imaging or behavioral variables has become a priority for the field. In addition, recognition that two individuals with the same clinical diagnosis may be more different from each other (at brain, genetic, and behavioral levels) than from another individual in a different disorder or nonclinical control group-coupled with computational advances-has catapulted data-driven efforts forward. Emerging challenges for this new approach include longitudinal stability of new subgroups, demonstration of validity, and replicability. The "litmus test" will be whether computational approaches that are successfully identifying groups of patients who share features in common, more than current DSM diagnostic constructs, also provide better prognostic accuracy over time and in addition lead to enhancements in treatment response and outcomes. These are the factors that matter most to patients, families, providers, and payers.

Selective Review of Neuroimaging Findings in Youth at Clinical High Risk for Psychosis: On the Path to Biomarkers for Conversion

First episode psychosis (FEP), and subsequent diagnosis of schizophrenia or schizoaffective disorder, predominantly occurs during late adolescence, is accompanied by a significant decline in function and represents a traumatic experience for patients and families alike. Prior to first episode psychosis, most patients experience a prodromal period of 1-2 years, during which symptoms first appear and then progress. During that time period, subjects are referred to as being at Clinical High Risk (CHR), as a prodromal period can only be designated in hindsight in those who convert. The clinical high-risk period represents a critical window during which interventions may be targeted to slow or prevent conversion to psychosis. However, only one third of subjects at clinical high risk will convert to psychosis and receive a formal diagnosis of a primary psychotic disorder. Therefore, in order for targeted interventions to be developed and applied, predicting who among this population will convert is of critical importance. To date, a variety of neuroimaging modalities have identified numerous differences between CHR subjects and healthy controls. However, complicating attempts at predicting conversion are increasingly recognized co-morbidities, such as major depressive disorder, in a significant number of CHR subjects. The result of this is that phenotypes discovered between CHR subjects and healthy controls are likely non-specific to psychosis and generalized for major mental illness. In this paper, we selectively review evidence for neuroimaging phenotypes in CHR subjects who later converted to psychosis. We then evaluate the recent landscape of machine learning as it relates to neuroimaging phenotypes in predicting conversion to psychosis.

Brain functional connectivity data enhance prediction of clinical outcome in youth at risk for psychosis

The first episode of psychosis is typically preceded by a prodromal phase with subthreshold symptoms and functional decline. Improved outcome prediction in this stage is needed to allow targeted early intervention. This study assesses a combined clinical and resting-state fMRI prediction model in 137 adolescents and young adults at Clinical High Risk (CHR) for psychosis from the Shanghai At Risk for Psychosis (SHARP) program. Based on outcome at one-year follow-up, participants were separated into three outcome categories including good outcome (symptom remission, N = 71), intermediate outcome (ongoing CHR symptoms, N = 30), and poor outcome (conversion to psychosis or treatment-refractory, N = 36). Validated clinical predictors from the psychosis-risk calculator were combined with measures of resting-state functional connectivity. Using multinomial logistic regression analysis and leave-one-out cross-validation, a clinical-only prediction model did not achieve a significant level of outcome prediction (F₁ = 0.32, p = .154). An imaging-only model yielded a significant prediction model (F₁ = 0.41, p = .016), but a combined model including both clinical and connectivity measures showed the best performance (F₁ = 0.46, p < .001). Influential predictors in this model included functional decline, verbal learning performance, a family history of psychosis, default-mode and frontoparietal within-network connectivity, and between-network connectivity among language, salience, dorsal attention, sensorimotor, and cerebellar networks. These findings suggest that brain changes reflected by alterations in functional connectivity may be useful for outcome prediction in the prodromal stage.

Neuroimaging in Schizophrenia

Schizophrenia is a chronic psychotic disorder with a lifetime prevalence of about 1%. Onset is typically in adolescence or early adulthood; characteristic symptoms include positive symptoms, negative symptoms, and impairments in cognition. Neuroimaging studies have shown substantive evidence of brain structural, functional, and neurochemical alterations that are more pronounced in the association cortex and subcortical regions. These abnormalities are not sufficiently specific to be of diagnostic value, but there may be a role for imaging techniques to provide predictions of outcome. Incorporating multimodal imaging datasets using machine learning approaches may offer better diagnostic and predictive value in schizophrenia.

Discovery and Validation of Prediction Algorithms for Psychosis in Youths at Clinical High Risk

In the past 2 to 3 decades, clinicians have used the clinical high risk for psychosis (CHR-P) paradigm to better understand factors that contribute to the onset of psychotic disorders. While this paradigm is useful to identify individuals at risk, the CHR-P criteria are not sufficient to predict outcomes from the CHR-P population. Because approximately 25% of the CHR-P population will ultimately convert to psychosis, more precise methods of prediction are needed to account for heterogeneity in both risk factors and outcomes in the CHR-P population. To this end, several groups in recent years have used data-driven approaches to refine predictive algorithms to predict both conversion to psychosis and functional outcomes. These models have generally used either clinical and behavioral data, including demographics and measures of symptom severity, neurocognitive functioning, and social functioning, or neuroimaging data, including structural and functional measures, to predict conversion to psychosis in CHR-P samples. This review focuses on the empirical models that have been derived within each of these lines of research and evaluates the performance and methodology of these models. This review also serves to inform best practices for data-driven approaches and directions moving forward to improve our prediction of psychotic disorders and associated outcomes. Because sample size is still the most critical consideration in the current models, we urge that algorithms to predict conversion be conducted using multisite data in order to obtain the power necessary to conclusively determine predictive accuracy without overfitting.

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.

Prediction Models of Functional Outcomes for Individuals in the Clinical High-Risk State for Psychosis or With Recent-Onset Depression: A Multimodal, Multisite Machine Learning Analysis

IMPORTANCE

Social and occupational impairments contribute to the burden of psychosis and depression. There is a need for risk stratification tools to inform personalized functional-disability preventive strategies for individuals in at-risk and early phases of these illnesses.

OBJECTIVE

To determine whether predictors associated with social and role functioning can be identified in patients in clinical high-risk (CHR) states for psychosis or with recent-onset depression (ROD) using clinical, imaging-based, and combined machine learning; assess the geographic, transdiagnostic, and prognostic generalizability of machine learning and compare it with human prognostication; and explore sequential prognosis encompassing clinical and combined machine learning.

DESIGN, SETTING, AND PARTICIPANTS

This multisite naturalistic study followed up patients in CHR states, with ROD, and with recent-onset psychosis, and healthy control participants for 18 months in 7 academic early-recognition services in 5 European countries. Participants were recruited between February 2014 and May 2016, and data were analyzed from April 2017 to January 2018.

AIN OUTCOMES AND MEASURES

Performance and generalizability of prognostic models.

RESULTS

A total of 116 individuals in CHR states (mean [SD] age, 24.0 [5.1] years; 58 [50.0%] female) and 120 patients with ROD (mean [SD] age, 26.1 [6.1] years; 65 [54.2%] female) were followed up for a mean (SD) of 329 (142) days. Machine learning predicted the 1-year social-functioning outcomes with a balanced accuracy of 76.9% of patients in CHR states and 66.2% of patients with ROD using clinical baseline data. Balanced accuracy in models using structural neuroimaging was 76.2% in patients in CHR states and 65.0% in patients with ROD, and in combined models, it was 82.7% for CHR states and 70.3% for ROD. Lower functioning before study entry was a transdiagnostic predictor. Medial prefrontal and temporo-parieto-occipital gray matter volume (GMV) reductions and cerebellar and dorsolateral prefrontal GMV increments had predictive value in the CHR group; reduced mediotemporal and increased prefrontal-perisylvian GMV had predictive value in patients with ROD. Poor prognoses were associated with increased risk of psychotic, depressive, and anxiety disorders at follow-up in patients in the CHR state but not ones with ROD. Machine learning outperformed expert prognostication. Adding neuroimaging machine learning to clinical machine learning provided a 1.9-fold increase of prognostic certainty in uncertain cases of patients in CHR states, and a 10.5-fold increase of prognostic certainty for patients with ROD.

CONCLUSIONS AND RELEVANCE

Precision medicine tools could augment effective therapeutic strategies aiming at the prevention of social functioning impairments in patients with CHR states or with ROD.

Brain morphologic changes in early stages of psychosis: Implications for clinical application and early intervention

To date, a large number of magnetic resonance imaging (MRI) studies have been conducted in schizophrenia, which generally demonstrate gray matter reduction, predominantly in the frontal and temporo-limbic regions, as well as gross brain abnormalities (e.g., a deviated sulcogyral pattern). Although the causes as well as timing and course of these findings remain elusive, these morphologic changes (especially gross brain abnormalities and medial temporal lobe atrophy) are likely present at illness onset, possibly reflecting early neurodevelopmental abnormalities. In addition, longitudinal MRI studies suggest that patients with schizophrenia and related psychoses also have progressive gray matter reduction during the transition period from prodrome to overt psychosis, as well as initial periods after psychosis onset, while such changes may become almost stable in the chronic stage. These active brain changes during the early phases seem to be relevant to the development of clinical symptoms in a region-specific manner (e.g., superior temporal gyrus atrophy and positive psychotic symptoms), but may be at least partly ameliorated by antipsychotic medication. Recently, increasing evidence from MRI findings in individuals at risk for developing psychosis has suggested that those who subsequently develop psychosis have baseline brain changes, which could be at least partly predictive of later transition into psychosis. In this article, we selectively review previous MRI findings during the course of psychosis and also refer to the possible clinical applicability of these neuroimaging research findings, especially in the diagnosis of schizophrenia and early intervention for psychosis.

Detecting Neuroimaging Biomarkers for Depression: A Meta-analysis of Multivariate Pattern Recognition Studies

BACKGROUND

Multiple studies have examined functional and structural brain alteration in patients diagnosed with major depressive disorder (MDD). The introduction of multivariate statistical methods allows investigators to utilize data concerning these brain alterations to generate diagnostic models that accurately differentiate patients with MDD from healthy control subjects (HCs). However, there is substantial heterogeneity in the reported results, the methodological approaches, and the clinical characteristics of participants in these studies.

METHODS

We conducted a meta-analysis of all studies using neuroimaging (volumetric measures derived from T1-weighted images, task-based functional magnetic resonance imaging [MRI], resting-state MRI, or diffusion tensor imaging) in combination with multivariate statistical methods to differentiate patients diagnosed with MDD from HCs.

RESULTS

Thirty-three (k = 33) samples including 912 patients with MDD and 894 HCs were included in the meta-analysis. Across all studies, patients with MDD were separated from HCs with 77% sensitivity and 78% specificity. Classification based on resting-state MRI (85% sensitivity, 83% specificity) and on diffusion tensor imaging data (88% sensitivity, 92% specificity) outperformed classifications based on structural MRI (70% sensitivity, 71% specificity) and task-based functional MRI (74% sensitivity, 77% specificity).

CONCLUSIONS

Our results demonstrate the high representational capacity of multivariate statistical methods to identify neuroimaging-based biomarkers of depression. Future studies are needed to elucidate whether multivariate neuroimaging analysis has the potential to generate clinically useful tools for the differential diagnosis of affective disorders and the prediction of both treatment response and functional outcome.

Using clinical information to make individualized prognostic predictions in people at ultra high risk for psychosis

Recent studies have reported an association between psychopathology and subsequent clinical and functional outcomes in people at ultra-high risk (UHR) for psychosis. This has led to the suggestion that psychopathological information could be used to make prognostic predictions in this population. However, because the current literature is based on inferences at group level, the translational value of the findings for everyday clinical practice is unclear. Here we examined whether psychopathological information could be used to make individualized predictions about clinical and functional outcomes in people at UHR. Participants included 416 people at UHR followed prospectively at the Personal Assessment and Crisis Evaluation (PACE) Clinic in Melbourne, Australia. The data were analysed using Support Vector Machine (SVM), a supervised machine learning technique that allows inferences at the individual level. SVM predicted transition to psychosis with a specificity of 60.6%, a sensitivity of 68.6% and an accuracy of 64.6% (p<0.001). In addition, SVM predicted functioning with a specificity of 62.5%, a sensitivity of 62.5% and an accuracy of 62.5% (p=0.008). Prediction of transition was driven by disorder of thought content, attenuated positive symptoms and functioning, whereas functioning was best predicted by attention disturbances, anhedonia-asociality and disorder of thought content. These results indicate that psychopathological information allows individualized prognostic predictions with statistically significant accuracy. However, this level of accuracy may not be sufficient for clinical translation in real-world clinical practice. Accuracy might be improved by combining psychopathological information with other types of data using a multivariate machine learning framework.

Cortical abnormalities in adults and adolescents with major depression based on brain scans from 20 cohorts worldwide in the ENIGMA Major Depressive Disorder Working Group

The neuro-anatomical substrates of major depressive disorder (MDD) are still not well understood, despite many neuroimaging studies over the past few decades. Here we present the largest ever worldwide study by the ENIGMA (Enhancing Neuro Imaging Genetics through Meta-Analysis) Major Depressive Disorder Working Group on cortical structural alterations in MDD. Structural T1-weighted brain magnetic resonance imaging (MRI) scans from 2148 MDD patients and 7957 healthy controls were analysed with harmonized protocols at 20 sites around the world. To detect consistent effects of MDD and its modulators on cortical thickness and surface area estimates derived from MRI, statistical effects from sites were meta-analysed separately for adults and adolescents. Adults with MDD had thinner cortical gray matter than controls in the orbitofrontal cortex (OFC), anterior and posterior cingulate, insula and temporal lobes (Cohen's d effect sizes: -0.10 to -0.14). These effects were most pronounced in first episode and adult-onset patients (>21 years). Compared to matched controls, adolescents with MDD had lower total surface area (but no differences in cortical thickness) and regional reductions in frontal regions (medial OFC and superior frontal gyrus) and primary and higher-order visual, somatosensory and motor areas (d: -0.26 to -0.57). The strongest effects were found in recurrent adolescent patients. This highly powered global effort to identify consistent brain abnormalities showed widespread cortical alterations in MDD patients as compared to controls and suggests that MDD may impact brain structure in a highly dynamic way, with different patterns of alterations at different stages of life.

Pattern of activation during delayed matching to sample task predicts functional outcome in people at ultra high risk for psychosis

BACKGROUND

Clinical outcomes in people identified as at ultra-high risk (UHR) for psychosis are remarkably heterogeneous, and are difficult to predict on the basis of the presenting clinical features. Individuals at UHR are at risk of poor functional outcome regardless of development of psychotic disorder. The aim of the present study was to assess whether there is a relationship between functional neuroimaging measures at presentation and functional outcome as measured by the GAF three years after scanning.

METHODS

Functional magnetic resonance imaging (fMRI) data were collected during an object working memory task in 34 ultra-high risk (UHR) subjects and 20 healthy controls. On the basis of their GAF scores at follow up, the UHR participants were divided into subgroups with good and poor functional outcomes, respectively.

RESULTS

At baseline, the UHR group differed from controls in showing altered frontal and cuneus/posterior cingulate activation. Significant group x task interactions were found in the left cuneus and posterior cingulate gyrus, reflecting differential responses to the task conditions. Within the UHR sample, the subgroup with a poor functional outcome exhibited altered activation in frontal, temporal and striatal regions, and reduced deactivation within default-mode network regions, relative to those with a good outcome. Within the whole UHR sample, in these regions the local task response was correlated with the GAF score at follow up.

CONCLUSIONS

The findings suggest a potential role of functional neuroimaging in the prediction of outcomes in people at high clinical risk of psychosis.

Neuroanatomical Predictors of Functional Outcome in Individuals at Ultra-High Risk for Psychosis

Most individuals at ultra-high risk (UHR) for psychosis do not transition to frank illness. Nevertheless, many have poor clinical outcomes and impaired psychosocial functioning. This study used voxel-based morphometry to investigate if baseline grey and white matter brain densities at identification as UHR were associated with functional outcome at medium- to long-term follow-up. Participants were help-seeking UHR individuals (n = 109, 54M:55F) who underwent magnetic resonance imaging at baseline; functional outcome was assessed an average of 9.2 years later. Primary analysis showed that lower baseline grey matter density, but not white matter density, in bilateral frontal and limbic areas, and left cerebellar declive were associated with poorer functional outcome (Social and Occupational Functioning Assessment Scale [SOFAS]). These findings were independent of transition to psychosis or persistence of the at-risk mental state. Similar regions were significantly associated with lower self-reported levels of social functioning and increased negative symptoms at follow-up. Exploratory analyses showed that lower baseline grey matter densities in middle and inferior frontal gyri were significantly associated with decline in Global Assessment of Functioning (GAF) score over follow-up. There was no association between baseline grey matter density and IQ or positive symptoms at follow-up. The current findings provide novel evidence that those with the poorest functional outcomes have the lowest grey matter densities at identification as UHR, regardless of transition status or persistence of the at-risk mental state. Replication and validation of these findings may allow for early identification of poor functional outcome and targeted interventions.

Individual prediction of long-term outcome in adolescents at ultra-high risk for psychosis: Applying machine learning techniques to brain imaging data

An important focus of studies of individuals at ultra-high risk (UHR) for psychosis has been to identify biomarkers to predict which individuals will transition to psychosis. However, the majority of individuals will prove to be resilient and go on to experience remission of their symptoms and function well. The aim of this study was to investigate the possibility of using structural MRI measures collected in UHR adolescents at baseline to quantitatively predict their long-term clinical outcome and level of functioning. We included 64 UHR individuals and 62 typically developing adolescents (12-18 years old at recruitment). At six-year follow-up, we determined resilience for 43 UHR individuals. Support Vector Regression analyses were performed to predict long-term functional and clinical outcome from baseline MRI measures on a continuous scale, instead of the more typical binary classification. This led to predictive correlations of baseline MR measures with level of functioning, and negative and disorganization symptoms. The highest correlation (r = 0.42) was found between baseline subcortical volumes and long-term level of functioning. In conclusion, our results show that structural MRI data can be used to quantitatively predict long-term functional and clinical outcome in UHR individuals with medium effect size, suggesting that there may be scope for predicting outcome at the individual level. Moreover, we recommend classifying individual outcome on a continuous scale, enabling the assessment of different functional and clinical scales separately without the need to set a threshold. Hum Brain Mapp 38:704-714, 2017. © 2016 Wiley Periodicals, Inc.

The Fallacy of Univariate Solutions to Complex Systems Problems

Complex biological systems, by definition, are composed of multiple components that interact non-linearly. The human brain constitutes, arguably, the most complex biological system known. Yet most investigation of the brain and its function is carried out using assumptions appropriate for simple systems—univariate design and linear statistical approaches. This heuristic must change before we can hope to discover and test interventions to improve the lives of individuals with complex disorders of brain development and function. Indeed, a movement away from simplistic models of biological systems will benefit essentially all domains of biology and medicine. The present brief essay lays the foundation for this argument.

Hyperactivity of caudate, parahippocampal, and prefrontal regions during working memory in never-medicated persons at clinical high-risk for psychosis

BACKGROUND

Deficits in working memory (WM) are a core feature of schizophrenia (SZ) and other psychotic disorders. We examined brain activity during WM in persons at clinical high risk (CHR) for psychosis.

METHODS

Thirty-seven CHR and 34 healthy control participants underwent functional MRI (fMRI) on a 3.0T scanner while performing an N-back WM task. The sample included a sub-sample of CHR participants who had no lifetime history of treatment with psychotropic medications (n=11). Data were analyzed using SPM8 (2-back>0-back contrast). Pearson correlations between brain activity, symptoms, and WM performance were examined.

RESULTS

The total CHR group and medication-naive CHR sub-sample were comparable to controls in most demographic features and in N-back WM performance, but had significantly lower IQ. Relative to controls, medication-naïve CHR showed hyperactivity in the left parahippocampus (PHP) and the left caudate during performance of the N-back WM task. Relative to medication-exposed CHR, medication naïve CHR exhibited hyperactivity in the left caudate and the right dorsolateral prefrontal cortex (DLPFC). DLPFC activity was significantly negatively correlated with WM performance. PHP, caudate and DLPFC activity correlated strongly with symptoms, but results did not withstand FDR-correction for multiple comparisons. When all CHR participants were combined (regardless of medication status), only trend-level PHP hyperactivity was observed in CHR relative to controls.

CONCLUSIONS

Medication-naïve CHR exhibit hyperactivity in regions that subserve WM. These regions are implicated in studies of schizophrenia and risk for psychosis. Results emphasize the importance of medication status in the interpretation of task - induced brain activity.

Detecting Neuroimaging Biomarkers for Psychiatric Disorders: Sample Size Matters

In a recent review, it was suggested that much larger cohorts are needed to prove the diagnostic value of neuroimaging biomarkers in psychiatry. While within a sample, an increase of diagnostic accuracy of schizophrenia (SZ) with number of subjects (N) has been shown, the relationship between N and accuracy is completely different between studies. Using data from a recent meta-analysis of machine learning (ML) in imaging SZ, we found that while low-N studies can reach 90% and higher accuracy, above N/2 = 50 the maximum accuracy achieved steadily drops to below 70% for N/2 > 150. We investigate the role N plays in the wide variability in accuracy results in SZ studies (63-97%). We hypothesize that the underlying cause of the decrease in accuracy with increasing N is sample heterogeneity. While smaller studies more easily include a homogeneous group of subjects (strict inclusion criteria are easily met; subjects live close to study site), larger studies inevitably need to relax the criteria/recruit from large geographic areas. A SZ prediction model based on a heterogeneous group of patients with presumably a heterogeneous pattern of structural or functional brain changes will not be able to capture the whole variety of changes, thus being limited to patterns shared by most patients. In addition to heterogeneity (sample size), we investigate other factors influencing accuracy and introduce a ML effect size. We derive a simple model of how the different factors, such as sample heterogeneity and study setup determine this ML effect size, and explain the variation in prediction accuracies found from the literature, both in cross-validation and independent sample testing. From this, we argue that smaller-N studies may reach high prediction accuracy at the cost of lower generalizability to other samples. Higher-N studies, on the other hand, will have more generalization power, but at the cost of lower accuracy. In conclusion, when comparing results from different ML studies, the sample sizes should be taken into account. To assess the generalizability of the models, validation (by direct application) of the prediction models should be tested in independent samples. The prediction of more complex measures such as outcome, which are expected to have an underlying pattern of more subtle brain abnormalities (lower effect size), will require large samples.

Classifying individuals at high-risk for psychosis based on functional brain activity during working memory processing

The psychosis high-risk state is accompanied by alterations in functional brain activity during working memory processing. We used binary automatic pattern-classification to discriminate between the at-risk mental state (ARMS), first episode psychosis (FEP) and healthy controls (HCs) based on n-back WM-induced brain activity. Linear support vector machines and leave-one-out-cross-validation were applied to fMRI data of matched ARMS, FEP and HC (19 subjects/group).

The HC and ARMS were correctly classified, with an accuracy of 76.2% (sensitivity 89.5%, specificity 63.2%, p = 0.01) using a verbal working memory network mask. Only 50% and 47.4% of individuals were classified correctly for HC vs. FEP (p = 0.46) or ARMS vs. FEP (p = 0.62), respectively. Without mask, accuracy was 65.8% for HC vs. ARMS (p = 0.03) and 65.8% for HC vs. FEP (p = 0.0047), and 57.9% for ARMS vs. FEP (p = 0.18). Regions in the medial frontal, paracingulate, cingulate, inferior frontal and superior frontal gyri, inferior and superior parietal lobules, and precuneus were particularly important for group separation.

These results suggest that FEP and HC or FEP and ARMS cannot be accurately separated in small samples under these conditions. However, ARMS can be identified with very high sensitivity in comparison to HC. This might aid classification and help to predict transition in the ARMS.

•

The ARMS was accurately identified based on an individual patient's response within a WM network.

•

Regional cortical activations were particularly important for group separation.

•

Based on WM alterations, FEP and HC or FEP and ARMS could not be accurately separated in small samples.