Search

Importance: Diverse models have been developed to predict psychosis in patients with clinical high-risk (CHR) states. Whether prediction can be improved by efficiently combining clinical and biological models and by broadening the risk spectrum to young patients with depressive syndromes remains unclear., Objectives: To evaluate whether psychosis transition can be predicted in patients with CHR or recent-onset depression (ROD) using multimodal machine learning that optimally integrates clinical and neurocognitive data, structural magnetic resonance imaging (sMRI), and polygenic risk scores (PRS) for schizophrenia; to assess models' geographic generalizability; to test and integrate clinicians' predictions; and to maximize clinical utility by building a sequential prognostic system., Design, Setting, and Participants: This multisite, longitudinal prognostic study performed in 7 academic early recognition services in 5 European countries followed up patients with CHR syndromes or ROD and healthy volunteers. The referred sample of 167 patients with CHR syndromes and 167 with ROD was recruited from February 1, 2014, to May 31, 2017, of whom 26 (23 with CHR syndromes and 3 with ROD) developed psychosis. Patients with 18-month follow-up (n = 246) were used for model training and leave-one-site-out cross-validation. The remaining 88 patients with nontransition served as the validation of model specificity. Three hundred thirty-four healthy volunteers provided a normative sample for prognostic signature evaluation. Three independent Swiss projects contributed a further 45 cases with psychosis transition and 600 with nontransition for the external validation of clinical-neurocognitive, sMRI-based, and combined models. Data were analyzed from January 1, 2019, to March 31, 2020., Main Outcomes and Measures: Accuracy and generalizability of prognostic systems., Results: A total of 668 individuals (334 patients and 334 controls) were included in the analysis (mean [SD] age, 25.1 [5.8] years; 354 [53.0%] female and 314 [47.0%] male). Clinicians attained a balanced accuracy of 73.2% by effectively ruling out (specificity, 84.9%) but ineffectively ruling in (sensitivity, 61.5%) psychosis transition. In contrast, algorithms showed high sensitivity (76.0%-88.0%) but low specificity (53.5%-66.8%). A cybernetic risk calculator combining all algorithmic and human components predicted psychosis with a balanced accuracy of 85.5% (sensitivity, 84.6%; specificity, 86.4%). In comparison, an optimal prognostic workflow produced a balanced accuracy of 85.9% (sensitivity, 84.6%; specificity, 87.3%) at a much lower diagnostic burden by sequentially integrating clinical-neurocognitive, expert-based, PRS-based, and sMRI-based risk estimates as needed for the given patient. Findings were supported by good external validation results., Conclusions and Relevance: These findings suggest that psychosis transition can be predicted in a broader risk spectrum by sequentially integrating algorithms' and clinicians' risk estimates. For clinical translation, the proposed workflow should undergo large-scale international validation.

At present, risk assessment for alcohol withdrawal syndrome relies on clinical judgment. Our aim was to develop accurate machine learning tools to predict alcohol withdrawal outcomes at the individual subject level using information easily attainable at patients' admission. An observational machine learning analysis using nested cross-validation and out-of-sample validation was applied to alcohol-dependent patients at two major detoxification wards (LMU, n = 389; TU, n = 805). 121 retrospectively derived clinical, blood-derived, and sociodemographic measures were used to predict 1) moderate to severe withdrawal defined by the alcohol withdrawal scale, 2) delirium tremens, and 3) withdrawal seizures. Mild and more severe withdrawal cases could be separated with significant, although highly variable accuracy in both samples (LMU, balanced accuracy [BAC] = 69.4%; TU, BAC = 55.9%). Poor outcome predictions were associated with higher cumulative clomethiazole doses during the withdrawal course. Delirium tremens was predicted in the TU cohort with BAC of 75%. No significant model predicting withdrawal seizures could be found. Our models were unique to each treatment site and thus did not generalize. For both treatment sites and withdrawal outcome different variable sets informed our models' decisions. Besides previously described variables (most notably, thrombocytopenia), we identified new predictors (history of blood pressure abnormalities, urine screening for benzodiazepines and educational attainment). In conclusion, machine learning approaches may facilitate generalizable, individualized predictions for alcohol withdrawal severity. Since predictive patterns highly vary for different outcomes of withdrawal severity and across treatment sites, prediction tools should not be recommended for clinical practice unless adequately validated in specific cohorts., Competing Interests: Conflict of interest All authors declare no conflict of interest., (Copyright © 2020 Elsevier Ltd. All rights reserved.)

Neurobiological heterogeneity in schizophrenia is poorly understood and confounds current analyses. We investigated neuroanatomical subtypes in a multi-institutional multi-ethnic cohort, using novel semi-supervised machine learning methods designed to discover patterns associated with disease rather than normal anatomical variation. Structural MRI and clinical measures in established schizophrenia (n = 307) and healthy controls (n = 364) were analysed across three sites of PHENOM (Psychosis Heterogeneity Evaluated via Dimensional Neuroimaging) consortium. Regional volumetric measures of grey matter, white matter, and CSF were used to identify distinct and reproducible neuroanatomical subtypes of schizophrenia. Two distinct neuroanatomical subtypes were found. Subtype 1 showed widespread lower grey matter volumes, most prominent in thalamus, nucleus accumbens, medial temporal, medial prefrontal/frontal and insular cortices. Subtype 2 showed increased volume in the basal ganglia and internal capsule, and otherwise normal brain volumes. Grey matter volume correlated negatively with illness duration in Subtype 1 (r = -0.201, P = 0.016) but not in Subtype 2 (r = -0.045, P = 0.652), potentially indicating different underlying neuropathological processes. The subtypes did not differ in age (t = -1.603, df = 305, P = 0.109), sex (chi-square = 0.013, df = 1, P = 0.910), illness duration (t = -0.167, df = 277, P = 0.868), antipsychotic dose (t = -0.439, df = 210, P = 0.521), age of illness onset (t = -1.355, df = 277, P = 0.177), positive symptoms (t = 0.249, df = 289, P = 0.803), negative symptoms (t = 0.151, df = 289, P = 0.879), or antipsychotic type (chi-square = 6.670, df = 3, P = 0.083). Subtype 1 had lower educational attainment than Subtype 2 (chi-square = 6.389, df = 2, P = 0.041). In conclusion, we discovered two distinct and highly reproducible neuroanatomical subtypes. Subtype 1 displayed widespread volume reduction correlating with illness duration, and worse premorbid functioning. Subtype 2 had normal and stable anatomy, except for larger basal ganglia and internal capsule, not explained by antipsychotic dose. These subtypes challenge the notion that brain volume loss is a general feature of schizophrenia and suggest differential aetiologies. They can facilitate strategies for clinical trial enrichment and stratification, and precision diagnostics., (© The Author(s) (2020). Published by Oxford University Press on behalf of the Guarantors of Brain. All rights reserved. For permissions, please email: journals.permissions@oup.com.)

Imaging studies have characterized functional and structural brain abnormalities in adults after premature birth, but these investigations have mostly used univariate methods that do not account for hypothesized interdependencies between brain regions or quantify accuracy in identifying individuals. To overcome these limitations, we used multivariate machine learning to identify gray matter volume (GMV) and amplitude of low frequency fluctuations (ALFF) brain patterns that best classify young adults born very preterm/very low birth weight (VP/VLBW; n = 94) from those born full-term (FT; n = 92). We then compared the spatial maps of the structural and functional brain signatures and validated them by assessing associations with clinical birth history and basic cognitive variables. Premature birth could be predicted with a balanced accuracy of 80.7% using GMV and 77.4% using ALFF. GMV predictions were mediated by a pattern of subcortical and middle temporal reductions and volumetric increases of the lateral prefrontal, medial prefrontal, and superior temporal gyrus regions. ALFF predictions were characterized by a pattern including increases in the thalamus, pre- and post-central gyri, and parietal lobes, in addition to decreases in the superior temporal gyri bilaterally. Decision scores from each classification, assessing the degree to which an individual was classified as a VP/VLBW case, were predicted by the number of days in neonatal hospitalization and birth weight. ALFF decision scores also contributed to the prediction of general IQ, which highlighted their potential clinical significance. Combined, the results clarified previous research and suggested that primary subcortical and temporal damage may be accompanied by disrupted neurodevelopment of the cortex., (© 2019 Wiley Periodicals, Inc.)

Despite its initial promise, neuroimaging has not been widely translated into clinical psychiatry to assist in the prediction of diagnoses, prognoses, and optimal therapeutic strategies. Machine learning approaches may enhance the translational potential of neuroimaging because they specifically focus on overcoming biases by optimizing the generalizability of pipelines that measure complex brain patterns to predict targets at a single-subject level. This article introduces some fundamentals of a translational machine learning approach before selectively reviewing literature to-date. Promising initial results are then balanced by the description of limitations that should be considered in order to interpret existing research and maximize the possibility of future translation. Future directions are then presented in order to inspire further research and progress the field towards clinical translation., (Copyright © 2018. Published by Elsevier Inc.)

Identifying distinctive subtypes of schizophrenia could ultimately enhance diagnostic and prognostic accuracy. We aimed to uncover neuroanatomical subtypes of chronic schizophrenia patients to test whether stratification can enhance computer-aided discrimination of patients from control subjects. Unsupervised, data-driven clustering of structural MRI (sMRI) data was used to identify 2 subtypes of schizophrenia patients drawn from a US-based open science repository (n = 71) and we quantified classification improvements compared to controls (n = 74) using supervised machine learning. We externally validated the unsupervised and supervised learning models in a heterogeneous German validation sample (n = 316), and characterized symptom, cognition, and longitudinal symptom change signatures. Stratification improved classification accuracies from 68.5% to 73% (subgroup 1) and 78.8% (subgroup 2), respectively. Increased accuracy was also found when models were externally validated, and an average gain of 9% was found in supplementary analyses. The first subgroup was associated with cortical and subcortical volume reductions coupled with substantially longer illness duration, whereas the second subgroup was mainly characterized by cortical reductions, reduced illness duration, and comparatively less negative symptoms. Individuals within each subgroup could be identified using just 10 clinical questions at an accuracy of 81.2%, and differential cognitive and symptom course signatures were suggested in multivariate analyses. Our findings suggest that sMRI-based subtyping enhances the neuroanatomical discrimination of schizophrenia by identifying generalizable brain patterns that align with a clinical staging model of the disorder. These findings could be used to improve illness stratification for biomarker-based computer-aided diagnoses.

Machine learning approaches for clinical psychology and psychiatry explicitly focus on learning statistical functions from multidimensional data sets to make generalizable predictions about individuals. The goal of this review is to provide an accessible understanding of why this approach is important for future practice given its potential to augment decisions associated with the diagnosis, prognosis, and treatment of people suffering from mental illness using clinical and biological data. To this end, the limitations of current statistical paradigms in mental health research are critiqued, and an introduction is provided to critical machine learning methods used in clinical studies. A selective literature review is then presented aiming to reinforce the usefulness of machine learning methods and provide evidence of their potential. In the context of promising initial results, the current limitations of machine learning approaches are addressed, and considerations for future clinical translation are outlined.

BackgroundTransition to psychosis is among the most adverse outcomes of clinical high-risk (CHR) syndromes encompassing ultra-high risk (UHR) and basic symptom states. Clinical risk calculators may facilitate an early and individualized interception of psychosis, but their real-world implementation requires thorough validation across diverse risk populations, including young patients with depressive syndromes.MethodsWe validated the previously described NAPLS-2 (North American Prodrome Longitudinal Study 2) calculator in 334 patients (26 with transition to psychosis) with CHR or recent-onset depression (ROD) drawn from the multisite European PRONIA (Personalised Prognostic Tools for Early Psychosis Management) study. Patients were categorized into three risk enrichment levels, ranging from UHR, over CHR, to a broad-risk population comprising patients with CHR or ROD (CHR|ROD). We assessed how risk enrichment and different predictive algorithms influenced prognostic performance using reciprocal external validation.ResultsAfter calibration, the NAPLS-2 model predicted psychosis with a balanced accuracy (BAC) (sensitivity, specificity) of 68% (73%, 63%) in the PRONIA-UHR cohort, 67% (74%, 60%) in the CHR cohort, and 70% (73%, 66%) in patients with CHR|ROD. Multiple model derivation in PRONIA-CHR|ROD and validation in NAPLS-2-UHR patients confirmed that broader risk definitions produced more accurate risk calculators (CHR|ROD-based vs. UHR-based performance: 67% [68%, 66%] vs. 58% [61%, 56%]). Support vector machines were superior in CHR|ROD (BAC = 71%), while ridge logistic regression and support vector machines performed similarly in CHR (BAC = 67%) and UHR cohorts (BAC = 65%). Attenuated psychotic symptoms predicted psychosis across risk levels, while younger age and reduced processing speed became increasingly relevant for broader risk cohorts.ConclusionsClinical-neurocognitive machine learning models operating in young patients with affective and CHR syndromes facilitate a more precise and generalizable prediction of psychosis. Future studies should investigate their therapeutic utility in large-scale clinical trials.

Objective: The prevalence and significance of schizophrenia-related phenotypes at the population level is debated in the literature. Here, the authors assessed whether two recently reported neuroanatomical signatures of schizophrenia-signature 1, with widespread reduction of gray matter volume, and signature 2, with increased striatal volume-could be replicated in an independent schizophrenia sample, and investigated whether expression of these signatures can be detected at the population level and how they relate to cognition, psychosis spectrum symptoms, and schizophrenia genetic risk.Methods: This cross-sectional study used an independent schizophrenia-control sample (N=347; ages 16-57 years) for replication of imaging signatures, and then examined two independent population-level data sets: typically developing youths and youths with psychosis spectrum symptoms in the Philadelphia Neurodevelopmental Cohort (N=359; ages 16-23 years) and adults in the UK Biobank study (N=836; ages 44-50 years). The authors quantified signature expression using support-vector machine learning and compared cognition, psychopathology, and polygenic risk between signatures.Results: Two neuroanatomical signatures of schizophrenia were replicated. Signature 1 but not signature 2 was significantly more common in youths with psychosis spectrum symptoms than in typically developing youths, whereas signature 2 frequency was similar in the two groups. In both youths and adults, signature 1 was associated with worse cognitive performance than signature 2. Compared with adults with neither signature, adults expressing signature 1 had elevated schizophrenia polygenic risk scores, but this was not seen for signature 2.Conclusions: The authors successfully replicated two neuroanatomical signatures of schizophrenia and describe their prevalence in population-based samples of youths and adults. They further demonstrated distinct relationships of these signatures with psychosis symptoms, cognition, and genetic risk, potentially reflecting underlying neurobiological vulnerability. [ABSTRACT FROM AUTHOR]

Depression frequently occurs in first-episode psychosis (FEP) and predicts longer-term negative outcomes. It is possible that this depression is seen primarily in a distinct subgroup, which if identified could allow targeted treatments. We hypothesize that patients with recent-onset psychosis (ROP) and comorbid depression would be identifiable by symptoms and neuroanatomical features similar to those seen in recent-onset depression (ROD). Data were extracted from the multisite PRONIA study: 154 ROP patients (FEP within 3 months of treatment onset), of whom 83 were depressed (ROP+D) and 71 who were not depressed (ROP−D), 146 ROD patients, and 265 healthy controls (HC). Analyses included a (1) principal component analysis that established the similar symptom structure of depression in ROD and ROP+D, (2) supervised machine learning (ML) classification with repeated nested cross-validation based on depressive symptoms separating ROD vs ROP+D, which achieved a balanced accuracy (BAC) of 51%, and (3) neuroanatomical ML-based classification, using regions of interest generated from ROD subjects, which identified BAC of 50% (no better than chance) for separation of ROP+D vs ROP−D. We conclude that depression at a symptom level is broadly similar with or without psychosis status in recent-onset disorders; however, this is not driven by a separable depressed subgroup in FEP. Depression may be intrinsic to early stages of psychotic disorder, and thus treating depression could produce widespread benefit. [ABSTRACT FROM AUTHOR]

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. 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. 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. 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. [ABSTRACT FROM AUTHOR]

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. [ABSTRACT FROM AUTHOR]

Background: Childhood maltreatment (CM) is a major psychiatric risk factor and leads to long-lasting physical and mental health implications throughout the affected individual’s lifespan. Nonetheless, the neuroanatomical correlates of CM and their specific clinical impact remain elusive. This might be attributed to the complex, multidimensional nature of CM as well as to the restrictions of traditional analysis pipelines using nosological grouping, univariate analysis and region-of-interest approaches. To overcome these issues, we present a novel transdiagnostic and naturalistic machine learning approach towards a better and more comprehensive understanding of the clinical and neuroanatomical complexity of CM. Methods: We acquired our dataset from the multi-center European PRONIA cohort (www.pronia.eu). Specifically, we selected 649 male and female individuals, comprising young, minimally medicated patients with clinical high-risk states for psychosis as well as recent-onset of depression or psychosis and healthy volunteers. As part of our analysis approach, we created a new Matlab Toolbox, which performs multivariate Sparse Partial Least Squares Analysis in a robust machine learning framework. We employed this algorithm to detect multi-layered associations between combinations of items from the Childhood Trauma Questionnaire (CTQ) and grey matter volume (GMV) and assessed their generalizability via nested cross-validation. The clinical relevance of these CM signatures was assessed by correlating them to a wide range of clinical measurements, including current functioning (GAF, GF), depressivity (BDI), quality of life (WHOQOL-BREF) and personality traits (NEO-FFI). Results: Overall, we detected three distinct signatures of sexual, physical and emotional maltreatment. The first signature consisted of an agedependent sexual abuse pattern and a corresponding GMV pattern along the prefronto-thalamo-cerebellar axis. The second signature yielded a sexdependent physical and sexual abuse pattern with a corresponding GMV pattern in parietal, occipital and subcortical regions. The third signature was a global emotional trauma signature, independent of age or sex, and projected to a brain structural pattern in sensory and limbic brain regions. Regarding the clinical impact of these signatures, the emotional trauma signature was most strongly associated with massively impaired state- and trait-level characteristics. Both on a phenomenological and on a brain structural level, the emotional trauma pattern was significantly correlated with lower levels of functioning, higher depression scores, decreased quality of life and maladaptive personality traits. Discussion: Our findings deliver multimodal, data-driven evidence for a differential impact of sexual, physical and emotional trauma on brain structure and clinical state- and trait-level phenotypes. They also highlight the multidimensional nature of CM, which consists of multiple layers of highly complex trauma-brain patterns. In broader terms, our study emphasizes the potential of machine learning approaches in generating novel insights into long-standing psychiatric topics. [ABSTRACT FROM AUTHOR]