Distinct multivariate brain morphological patterns and their added predictive value with cognitive and polygenic risk scores in mental disorders

Multiscale heterogeneity of white matter morphometry in psychiatric disorders

BACKGROUND

Inter-individual variability in the neurobiological and clinical characteristics of mental illnesses are often overlooked by classical group-mean case-control studies. Studies using normative modelling to infer person-specific deviations of grey matter volume have indicated that group means are not representative of most individuals. The extent to which this variability is present in white matter morphometry, which is integral to brain function, remains unclear.

METHODS

We applied Warped Bayesian Linear Regression normative models to T1-weighted magnetic resonance imaging data and mapped inter-individual variability in person-specific white matter volume deviations in 1,294 cases (58% male) diagnosed with one of six disorders (attention-deficit/hyperactivity, autism, bipolar, major depressive, obsessive-compulsive and schizophrenia) and 1,465 matched controls (54% male) recruited across 25 scan sites. We developed a framework to characterize deviation heterogeneity at multiple spatial scales, from individual voxels, through inter-regional connections, specific brain regions, and spatially extended brain networks.

RESULTS

The specific locations of white matter volume deviations were highly heterogeneous across participants, affecting the same voxel in fewer than 8% of individuals with the same diagnosis. For autism and schizophrenia, negative deviations (i.e., areas where volume is lower than normative expectations) aggregated into common tracts, regions, and large-scale networks in up to 69% of individuals.

CONCLUSIONS

The prevalence of white matter volume deviations was lower than previously observed in grey matter, and the specific location of these deviations was highly heterogeneous when considering voxel-wise spatial resolution. Evidence of aggregation within common pathways and networks was apparent in schizophrenia and autism, but not other disorders.

Beyond traditional orthopaedic data analysis: AI, multimodal models and continuous monitoring

Multimodal artificial intelligence (AI) has the potential to revolutionise healthcare by enabling the simultaneous processing and integration of various data types, including medical imaging, electronic health records, genomic information and real-time data. This review explores the current applications and future potential of multimodal AI across healthcare, with a particular focus on orthopaedic surgery. In presurgical planning, multimodal AI has demonstrated significant improvements in diagnostic accuracy and risk prediction, with studies reporting an Area under the receiving operator curve presenting good to excellent performance across various orthopaedic conditions. Intraoperative applications leverage advanced imaging and tracking technologies to enhance surgical precision, while postoperative care has been advanced through continuous patient monitoring and early detection of complications. Despite these advances, significant challenges remain in data integration, standardisation, and privacy protection. Technical solutions such as federated learning (allowing decentralisation of models) and edge computing (allowing data analysis to happen on site or closer to site instead of multipurpose datacenters) are being developed to address these concerns while maintaining compliance with regulatory frameworks. As this field continues to evolve, the integration of multimodal AI promises to advance personalised medicine, improve patient outcomes, and transform healthcare delivery through more comprehensive and nuanced analysis of patient data. Level of Evidence: Level V.

Machine learning with multiple modalities of brain magnetic resonance imaging data to identify the presence of bipolar disorder

BACKGROUND

Bipolar disorder (BD) is a chronic psychiatric mood disorder that is solely diagnosed based on clinical symptoms. These symptoms often overlap with other psychiatric disorders. Efforts to use machine learning (ML) to create predictive models for BD based on data from brain imaging are expanding but have often been limited using only a single modality and the exclusion of the cerebellum, which may be relevant in BD.

METHODS

In this study, we sought to improve ML classification of BD by combining information from structural, functional, and diffusion-weighted imaging. Participants (108 BD I, 78 control) with BD type I and matched controls were recruited into an imaging study. This dataset was randomly divided into training and testing sets. For each of the three modalities, a separate ML model was selected, trained, and then used to generate a prediction of the class of each test subject. Majority voting was used to combine results from the three models to make a final prediction of whether a subject had BD. An independent replication sample was used to evaluate the ability of the ML classification to generalize to data collected at other sites.

RESULTS

Combining the three machine learning models through majority voting resulted in an accuracy of 89.5 % for classification of the test subjects as being in the BD or control group. Bootstrapping resulted in a 95 % confidence interval of 78.9 %-97.4 % for test accuracy. Performance was reduced when only using 2 of the 3 modalities. Analysis of feature importance revealed that the cerebellum and nodes of the emotional control network were among the most important regions for classification. The machine learning model performed at chance on the independent replication sample.

CONCLUSION

BD I could be identified with high accuracy in our relatively small sample by combining structural, functional, and diffusion-weighted imaging data within a single site but not generalize well to an independent replication sample. Future studies using harmonized imaging protocols may facilitate generalization of ML models.

Biomarker discovery using machine learning in the psychosis spectrum

The past decade witnessed substantial discoveries related to the psychosis spectrum. Many of these discoveries resulted from pursuits of objective and quantifiable biomarkers in tandem with the application of analytical tools such as machine learning. These approaches provided exciting new insights that significantly helped improve precision in diagnosis, prognosis, and treatment. This article provides an overview of how machine learning has been employed in recent biomarker discovery research in the psychosis spectrum, which includes schizophrenia, schizoaffective disorders, bipolar disorder with psychosis, first episode psychosis, and clinical high risk for psychosis. It highlights both human and animal model studies and explores a varying range of the most impactful biomarkers including cognition, neuroimaging, electrophysiology, and digital markers. We specifically highlight new applications and opportunities for machine learning to impact noninvasive symptom monitoring, prediction of future diagnosis and treatment outcomes, integration of new methods with traditional clinical research and practice, and personalized medicine approaches.

Psychiatric Polygenic Risk Scores Across Youth With Bipolar Disorder, Youth at High Risk for Bipolar Disorder, and Controls

OBJECTIVE

There is a pronounced gap in knowledge regarding polygenic underpinnings of youth bipolar disorder (BD). This study aimed to compare polygenic risk scores (PRSs) in youth with BD, youth at high clinical and/or familial risk for BD (HR), and controls.

METHOD

Participants were 344 youths of European ancestry (13-20 years old), including 136 youths with BD, 121 HR youths, and 87 controls. PRSs for BD, schizophrenia, major depressive disorder, and attention-deficit/hyperactivity disorder were constructed using independent genome-wide summary statistics from adult cohorts. Multinomial logistic regression was used to examine the association between each PRS and diagnostic status (BD vs HR vs controls). All genetic analyses controlled for age, sex, and 2 genetic principal components.

RESULTS

The BD group showed significantly higher BD-PRS than the control group (odds ratio = 1.54, 95% CI = 1.13-2.10, p = .006), with the HR group numerically intermediate. BD-PRS explained 7.9% of phenotypic variance. PRSs for schizophrenia, major depressive disorder, and attention-deficit/hyperactivity disorder were not significantly different among groups. In the BD group, BD-PRS did not significantly differ in relation to BD subtype, age of onset, psychosis, or family history of BD.

CONCLUSION

BD-PRS derived from adult genome-wide summary statistics is elevated in youth with BD. Absence of significant between-group differences in PRSs for other psychiatric disorders supports the specificity of BD-PRS in youth. These findings add to the biological validation of BD in youth and could have implications for early identification and diagnosis. To enhance clinical utility, future genome-wide association studies that focus specifically on early-onset BD are warranted, as are studies integrating additional genetic and environmental factors.

PLAIN LANGUAGE SUMMARY

Polygenic risk scores estimate an individual's genetic susceptibility to develop a disorder, such as bipolar disorder (BD). In this study, the authors constructed polygenic risk scores from previous adult studies. Youth with BD had elevated polygenic risk scores for BD compared to youth without bipolar disorder. Youth at high risk for BD had intermediate polygenic risk scores. To evaluate the specificity of polygenic risk scores for BD, the authors estimated risk scores for other mental health disorders including schizophrenia, major depressive disorder, and attention-deficit/hyperactivity disorder. These other polygenic risk scores did not differ between youth with and without BD. These findings support the biological validation of BD in youth, with potential implications for early identification and diagnosis.

DIVERSITY & INCLUSION STATEMENT

We worked to ensure sex and gender balance in the recruitment of human participants. One or more of the authors of this paper self-identifies as a member of one or more historically underrepresented sexual and/or gender groups in science. One or more of the authors of this paper self-identifies as a member of one or more historically underrepresented racial and/or ethnic groups in science. We actively worked to promote sex and gender balance in our author group. We actively worked to promote inclusion of historically underrepresented racial and/or ethnic groups in science in our author group. The author list of this paper includes contributors from the location and/or community where the research was conducted who participated in the data collection, design, analysis, and/or interpretation of the work.

Multiscale heterogeneity of white matter morphometry in psychiatric disorders

Background

Inter-individual variability in neurobiological and clinical characteristics in mental illness is often overlooked by classical group-mean case-control studies. Studies using normative modelling to infer person-specific deviations of grey matter volume have indicated that group means are not representative of most individuals. The extent to which this variability is present in white matter morphometry, which is integral to brain function, remains unclear.

Methods

We applied Warped Bayesian Linear Regression normative models to T1-weighted magnetic resonance imaging data and mapped inter-individual variability in person-specific white matter volume deviations in 1,294 cases (58% male) diagnosed with one of six disorders (attention-deficit/hyperactivity, autism, bipolar, major depressive, obsessive-compulsive and schizophrenia) and 1,465 matched controls (54% male) recruited across 25 scan sites. We developed a framework to characterize deviation heterogeneity at multiple spatial scales, from individual voxels, through inter-regional connections, specific brain regions, and spatially extended brain networks.

Results

The specific locations of white matter volume deviations were highly heterogeneous across participants, affecting the same voxel in fewer than 8% of individuals with the same diagnosis. For autism and schizophrenia, negative deviations (i.e., areas where volume is lower than normative expectations) aggregated into common tracts, regions and large-scale networks in up to 35% of individuals.

Conclusions

The prevalence of white matter volume deviations was lower than previously observed in grey matter, and the specific location of these deviations was highly heterogeneous when considering voxel-wise spatial resolution. Evidence of aggregation within common pathways and networks was apparent in schizophrenia and autism but not other disorders.

Cortical similarities in psychiatric and mood disorders identified in federated VBM analysis via COINSTAC

Structural neuroimaging studies have identified a combination of shared and disorder-specific patterns of gray matter (GM) deficits across psychiatric disorders. Pooling large data allows for examination of a possible common neuroanatomical basis that may identify a certain vulnerability for mental illness. Large-scale collaborative research is already facilitated by data repositories, institutionally supported databases, and data archives. However, these data-sharing methodologies can suffer from significant barriers. Federated approaches augment these approaches by enabling access or more sophisticated, shareable and scaled-up analyses of large-scale data. We examined GM alterations using Collaborative Informatics and Neuroimaging Suite Toolkit for Anonymous Computation, an open-source, decentralized analysis application. Through federated analysis of eight sites, we identified significant overlap in the GM patterns (n = 4,102) of individuals with schizophrenia, major depressive disorder, and autism spectrum disorder. These results show cortical and subcortical regions that may indicate a shared vulnerability to psychiatric disorders.

Combining Clinical With Cognitive or Magnetic Resonance Imaging Data for Predicting Transition to Psychosis in Ultra High-Risk Patients: Data From the PACE 400 Cohort

BACKGROUND

Multimodal modeling that combines biological and clinical data shows promise in predicting transition to psychosis in individuals who are at ultra-high risk. Individuals who transition to psychosis are known to have deficits at baseline in cognitive function and reductions in gray matter volume in multiple brain regions identified by magnetic resonance imaging.

METHODS

In this study, we used Cox proportional hazards regression models to assess the additive predictive value of each modality-cognition, cortical structure information, and the neuroanatomical measure of brain age gap-to a previously developed clinical model using functioning and duration of symptoms prior to service entry as predictors in the Personal Assessment and Crisis Evaluation (PACE) 400 cohort. The PACE 400 study is a well-characterized cohort of Australian youths who were identified as ultra-high risk of transitioning to psychosis using the Comprehensive Assessment of At Risk Mental States (CAARMS) and followed for up to 18 years; it contains clinical data (from N = 416 participants), cognitive data (n = 213), and magnetic resonance imaging cortical parameters extracted using FreeSurfer (n = 231).

RESULTS

The results showed that neuroimaging, brain age gap, and cognition added marginal predictive information to the previously developed clinical model (fraction of new information: neuroimaging 0%-12%, brain age gap 7%, cognition 0%-16%).

CONCLUSIONS

In summary, adding a second modality to a clinical risk model predicting the onset of a psychotic disorder in the PACE 400 cohort showed little improvement in the fit of the model for long-term prediction of transition to psychosis.

Distinguishing schizophrenia and bipolar disorder through a Multiclass Classification model based on multimodal neuroimaging data

This study aimed to identify neural biomarkers for schizophrenia (SZ) and bipolar disorder (BP) by analyzing multimodal neuroimaging. Utilizing data from structural magnetic resonance imaging (sMRI), diffusion tensor imaging (DTI), and resting-state functional magnetic resonance imaging (rs-fMRI), multiclass classification models were created for SZ, BP, and healthy controls (HC). A total of 113 participants (BP: 31, SZ: 39, and HC: 43) were recruited under strict enrollment control, from which 272, 200, and 1875 features were extracted from sMRI, DTI, and rs-fMRI data, respectively. A support vector machine (SVM) with recursive feature elimination (RFE) was employed to build the models using a one-against-one approach and leave-one-out cross-validation, achieving a classification accuracy of 70.8%. The most discriminative features were primarily from rs-fMRI, along with significant findings in sMRI and DTI. Key biomarkers identified included the increased thickness of the left cuneus cortex and decreased regional functional connectivity strength (rFCS) in the left supramarginal gyrus as shared indicators for BP and SZ. Additionally, decreased fractional anisotropy in the left superior fronto-occipital fasciculus was suggested as specific to BP, while decreased rFCS in the left inferior parietal area might serve as a specific biomarker for SZ. These findings underscore the potential of multimodal neuroimaging in distinguishing between BP and SZ and contribute to the understanding of their neural underpinnings.

A primer on the use of machine learning to distil knowledge from data in biological psychiatry

Applications of machine learning in the biomedical sciences are growing rapidly. This growth has been spurred by diverse cross-institutional and interdisciplinary collaborations, public availability of large datasets, an increase in the accessibility of analytic routines, and the availability of powerful computing resources. With this increased access and exposure to machine learning comes a responsibility for education and a deeper understanding of its bases and bounds, borne equally by data scientists seeking to ply their analytic wares in medical research and by biomedical scientists seeking to harness such methods to glean knowledge from data. This article provides an accessible and critical review of machine learning for a biomedically informed audience, as well as its applications in psychiatry. The review covers definitions and expositions of commonly used machine learning methods, and historical trends of their use in psychiatry. We also provide a set of standards, namely Guidelines for REporting Machine Learning Investigations in Neuropsychiatry (GREMLIN), for designing and reporting studies that use machine learning as a primary data-analysis approach. Lastly, we propose the establishment of the Machine Learning in Psychiatry (MLPsych) Consortium, enumerate its objectives, and identify areas of opportunity for future applications of machine learning in biological psychiatry. This review serves as a cautiously optimistic primer on machine learning for those on the precipice as they prepare to dive into the field, either as methodological practitioners or well-informed consumers.

Association of polygenic risk for bipolar disorder with resting-state network functional connectivity in youth with and without bipolar disorder

Little is known regarding the polygenic underpinnings of anomalous resting-state functional connectivity (rsFC) in youth bipolar disorder (BD). The current study examined the association of polygenic risk for BD (BD-PRS) with whole-brain rsFC at the large-scale network level in youth with and without BD. 99 youth of European ancestry (56 BD, 43 healthy controls [HC]), ages 13-20 years, completed resting-state fMRI scans. BD-PRS was calculated using summary statistics from the latest adult BD genome-wide association study. Data-driven independent component analyses of the resting-state fMRI data were implemented to examine the association of BD-PRS with rsFC in the overall sample, and separately in BD and HC. In the overall sample, higher BD-PRS was associated with lower rsFC of the salience network and higher rsFC of the frontoparietal network with frontal and parietal regions. Within the BD group, higher BD-PRS was associated with higher rsFC of the default mode network with orbitofrontal cortex, and altered rsFC of the visual network with frontal and occipital regions. Within the HC group, higher BD-PRS was associated with altered rsFC of the frontoparietal network with frontal, temporal and occipital regions. In conclusion, the current study found that BD-PRS generated based on adult genetic data was associated with altered rsFC patterns of brain networks in youth. Our findings support the usefulness of BD-PRS to investigate genetically influenced neuroimaging markers of vulnerability to BD, which can be observed in youth with BD early in their course of illness as well as in healthy youth.

The overlapping genetic architecture of psychiatric disorders and cortical brain structure

Both psychiatric vulnerability and cortical structure are shaped by the cumulative effect of common genetic variants across the genome. However, the shared genetic underpinnings between psychiatric disorders and brain structural phenotypes, such as thickness and surface area of the cerebral cortex, remains elusive. In this study, we employed pleiotropy-informed conjunctional false discovery rate analysis to investigate shared loci across genome-wide association scans of regional cortical thickness, surface area, and seven psychiatric disorders in approximately 700,000 individuals of European ancestry. Aggregating regional measures, we identified 50 genetic loci shared between psychiatric disorders and surface area, as well as 26 genetic loci shared with cortical thickness. Risk alleles exhibited bidirectional effects on both cortical thickness and surface area, such that some risk alleles for each disorder increased regional brain size while other risk alleles decreased regional brain size. Due to bidirectional effects, in many cases we observed extensive pleiotropy between an imaging phenotype and a psychiatric disorder even in the absence of a significant genetic correlation between them. The impact of genetic risk for psychiatric disorders on regional brain structure did exhibit a consistent pattern across highly comorbid psychiatric disorders, with 80% of the genetic loci shared across multiple disorders displaying consistent directions of effect. Cortical patterning of genetic overlap revealed a hierarchical genetic architecture, with the association cortex and sensorimotor cortex representing two extremes of shared genetic influence on psychiatric disorders and brain structural variation. Integrating multi-scale functional annotations and transcriptomic profiles, we observed that shared genetic loci were enriched in active genomic regions, converged on neurobiological and metabolic pathways, and showed differential expression in postmortem brain tissue from individuals with psychiatric disorders. Cumulatively, these findings provide a significant advance in our understanding of the overlapping polygenic architecture between psychopathology and cortical brain structure.

Regional, circuit and network heterogeneity of brain abnormalities in psychiatric disorders

The substantial individual heterogeneity that characterizes people with mental illness is often ignored by classical case-control research, which relies on group mean comparisons. Here we present a comprehensive, multiscale characterization of the heterogeneity of gray matter volume (GMV) differences in 1,294 cases diagnosed with one of six conditions (attention-deficit/hyperactivity disorder, autism spectrum disorder, bipolar disorder, depression, obsessive-compulsive disorder and schizophrenia) and 1,465 matched controls. Normative models indicated that person-specific deviations from population expectations for regional GMV were highly heterogeneous, affecting the same area in <7% of people with the same diagnosis. However, these deviations were embedded within common functional circuits and networks in up to 56% of cases. The salience-ventral attention system was implicated transdiagnostically, with other systems selectively involved in depression, bipolar disorder, schizophrenia and attention-deficit/hyperactivity disorder. Phenotypic differences between cases assigned the same diagnosis may thus arise from the heterogeneous localization of specific regional deviations, whereas phenotypic similarities may be attributable to the dysfunction of common functional circuits and networks.

Supervised machine learning classification of psychosis biotypes based on brain structure: findings from the Bipolar-Schizophrenia network for intermediate phenotypes (B-SNIP)

Traditional diagnostic formulations of psychotic disorders have low correspondence with underlying disease neurobiology. This has led to a growing interest in using brain-based biomarkers to capture biologically-informed psychosis constructs. Building upon our prior work on the B-SNIP Psychosis Biotypes, we aimed to examine whether structural MRI (an independent biomarker not used in the Biotype development) can effectively classify the Biotypes. Whole brain voxel-wise grey matter density (GMD) maps from T1-weighted images were used to train and test (using repeated randomized train/test splits) binary L2-penalized logistic regression models to discriminate psychosis cases (n = 557) from healthy controls (CON, n = 251). A total of six models were evaluated across two psychosis categorization schemes: (i) three Biotypes (B1, B2, B3) and (ii) three DSM diagnoses (schizophrenia (SZ), schizoaffective (SAD) and bipolar (BD) disorders). Above-chance classification accuracies were observed in all Biotype (B1 = 0.70, B2 = 0.65, and B3 = 0.56) and diagnosis (SZ = 0.64, SAD = 0.64, and BD = 0.59) models. However, the only model that showed evidence of specificity was B1, i.e., the model was able to discriminate B1 vs. CON and did not misclassify other psychosis cases (B2 or B3) as B1 at rates above nominal chance. The GMD-based classifier evidence for B1 showed a negative association with an estimate of premorbid general intellectual ability, regardless of group membership, i.e. psychosis or CON. Our findings indicate that, complimentary to clinical diagnoses, the B-SNIP Psychosis Biotypes may offer a promising approach to capture specific aspects of psychosis neurobiology.

Sampling inequalities affect generalization of neuroimaging-based diagnostic classifiers in psychiatry

BACKGROUND

The development of machine learning models for aiding in the diagnosis of mental disorder is recognized as a significant breakthrough in the field of psychiatry. However, clinical practice of such models remains a challenge, with poor generalizability being a major limitation.

METHODS

Here, we conducted a pre-registered meta-research assessment on neuroimaging-based models in the psychiatric literature, quantitatively examining global and regional sampling issues over recent decades, from a view that has been relatively underexplored. A total of 476 studies (n = 118,137) were included in the current assessment. Based on these findings, we built a comprehensive 5-star rating system to quantitatively evaluate the quality of existing machine learning models for psychiatric diagnoses.

RESULTS

A global sampling inequality in these models was revealed quantitatively (sampling Gini coefficient (G) = 0.81, p < .01), varying across different countries (regions) (e.g., China, G = 0.47; the USA, G = 0.58; Germany, G = 0.78; the UK, G = 0.87). Furthermore, the severity of this sampling inequality was significantly predicted by national economic levels (β =  - 2.75, p < .001, R²_adj = 0.40; r =  - .84, 95% CI: - .41 to - .97), and was plausibly predictable for model performance, with higher sampling inequality for reporting higher classification accuracy. Further analyses showed that lack of independent testing (84.24% of models, 95% CI: 81.0-87.5%), improper cross-validation (51.68% of models, 95% CI: 47.2-56.2%), and poor technical transparency (87.8% of models, 95% CI: 84.9-90.8%)/availability (80.88% of models, 95% CI: 77.3-84.4%) are prevailing in current diagnostic classifiers despite improvements over time. Relating to these observations, model performances were found decreased in studies with independent cross-country sampling validations (all p < .001, BF₁₀ > 15). In light of this, we proposed a purpose-built quantitative assessment checklist, which demonstrated that the overall ratings of these models increased by publication year but were negatively associated with model performance.

CONCLUSIONS

Together, improving sampling economic equality and hence the quality of machine learning models may be a crucial facet to plausibly translating neuroimaging-based diagnostic classifiers into clinical practice.

Developmental manifestations of polygenic risk for bipolar disorder from infancy to middle childhood

Knowledge on how genetic risk for bipolar disorder manifests in developmental, emotional or behavioral traits during childhood is lacking. This issue is important to address to inform early detection and intervention efforts. We investigated whether polygenic risk for bipolar disorder is associated with developmental outcomes during early to middle childhood in the general population, and if associations differ between boys and girls. Our sample consisted of 28 001 children from the Norwegian Mother, Father and Child Cohort study, a prospective pregnancy cohort with available genotype and developmental data. Mothers reported on a range of developmental outcomes in their children at 6 and 18 months, 3, 5 and 8 years. Polygenic risk scores reflecting common variant liability to bipolar disorder were calculated. Linear regression models were used in a multi-group framework to investigate associations between polygenic risk score and developmental outcomes, using sex as a grouping variable. We found robust evidence for an association between polygenic risk scores for bipolar disorder and conduct difficulties (β = 0.041, CI = 0.020-0.062) and oppositional defiant difficulties (β = 0.032, CI = 0.014-0.051) at 8 years. Associations with most other outcomes were estimated within the region of practical equivalence to zero (equivalence range D = -0.1 to 0.1), with the exceptions of negative association for activity levels (β = -0.028, CI = -0.047- -0.010) at age 5 and benevolence (β = -0.025, CI = -0.043 to -0.008) at age 8, and positive association for motor difficulties (β = 0.025, CI = 0.008-0.043) at age 3, inattention (β = 0.021, CI = 0.003-0.041) and hyperactivity (β = 0.025, CI = 0.006-0.044) at age 8. Our results suggest that genetic risk for bipolar disorder manifests as disruptive behaviors like oppositional defiant and conduct difficulties in childhood in the general population.

An Overview of Bipolar Disorder Diagnosis Using Machine Learning Approaches: Clinical Opportunities and Challenges

Objective: Automatic diagnosis of psychiatric disorders such as bipolar disorder (BD) through machine learning techniques has attracted substantial attention from psychiatric and artificial intelligence communities. These approaches mostly rely on various biomarkers extracted from electroencephalogram (EEG) or magnetic resonance imaging (MRI)/functional MRI (fMRI) data. In this paper, we provide an updated overview of existing machine learning-based methods for bipolar disorder (BD) diagnosis using MRI and EEG data. Method : This study is a short non-systematic review with the aim of describing the current situation in automatic diagnosis of BD using machine learning methods. Therefore, an appropriate literature search was conducted via relevant keywords for original EEG/MRI studies on distinguishing BD from other conditions, particularly from healthy peers, in PubMed, Web of Science, and Google Scholar databases. Results: We reviewed 26 studies, including 10 EEG studies and 16 MRI studies (including structural and functional MRI), that used traditional machine learning methods and deep learning algorithms to automatically detect BD. The reported accuracies for EEG studies is about 90%, while the reported accuracies for MRI studies remains below the minimum level for clinical relevance, i.e. about 80% of the classification outcome for traditional machine learning methods. However, deep learning techniques have generally achieved accuracies higher than 95%. Conclusion: Research utilizing machine learning applied to EEG signals and brain images has provided proof of concept for how this innovative technique can help psychiatrists distinguish BD patients from healthy people. However, the results have been somewhat contradictory and we must keep away from excessive optimistic interpretations of the findings. Much progress is still needed to reach the level of clinical practice in this field.

Evaluation of Risk of Bias in Neuroimaging-Based Artificial Intelligence Models for Psychiatric Diagnosis: A Systematic Review

IMPORTANCE

Neuroimaging-based artificial intelligence (AI) diagnostic models have proliferated in psychiatry. However, their clinical applicability and reporting quality (ie, feasibility) for clinical practice have not been systematically evaluated.

OBJECTIVE

To systematically assess the risk of bias (ROB) and reporting quality of neuroimaging-based AI models for psychiatric diagnosis.

EVIDENCE REVIEW

PubMed was searched for peer-reviewed, full-length articles published between January 1, 1990, and March 16, 2022. Studies aimed at developing or validating neuroimaging-based AI models for clinical diagnosis of psychiatric disorders were included. Reference lists were further searched for suitable original studies. Data extraction followed the CHARMS (Checklist for Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modeling Studies) and PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses) guidelines. A closed-loop cross-sequential design was used for quality control. The PROBAST (Prediction Model Risk of Bias Assessment Tool) and modified CLEAR (Checklist for Evaluation of Image-Based Artificial Intelligence Reports) benchmarks were used to systematically evaluate ROB and reporting quality.

FINDINGS

A total of 517 studies presenting 555 AI models were included and evaluated. Of these models, 461 (83.1%; 95% CI, 80.0%-86.2%) were rated as having a high overall ROB based on the PROBAST. The ROB was particular high in the analysis domain, including inadequate sample size (398 of 555 models [71.7%; 95% CI, 68.0%-75.6%]), poor model performance examination (with 100% of models lacking calibration examination), and lack of handling data complexity (550 of 555 models [99.1%; 95% CI, 98.3%-99.9%]). None of the AI models was perceived to be applicable to clinical practices. Overall reporting completeness (ie, number of reported items/number of total items) for the AI models was 61.2% (95% CI, 60.6%-61.8%), and the completeness was poorest for the technical assessment domain with 39.9% (95% CI, 38.8%-41.1%).

CONCLUSIONS AND RELEVANCE

This systematic review found that the clinical applicability and feasibility of neuroimaging-based AI models for psychiatric diagnosis were challenged by a high ROB and poor reporting quality. Particularly in the analysis domain, ROB in AI diagnostic models should be addressed before clinical application.

Multimodal fusion analysis of functional, cerebrovascular and structural neuroimaging in healthy aging subjects

Brain aging is a complex process that requires a multimodal approach. Neuroimaging can provide insights into brain morphology, functional organization, and vascular dynamics. However, most neuroimaging studies of aging have focused on each imaging modality separately, limiting the understanding of interrelations between processes identified by different modalities and their relevance to cognitive decline in aging. Here, we used a data-driven multimodal approach, linked independent component analysis (ICA), to jointly analyze magnetic resonance imaging (MRI) of grey matter volume, cerebrovascular, and functional network topographies in relation to measures of fluid intelligence. Neuroimaging and cognitive data from the Cambridge Centre for Ageing and Neuroscience study were used, with healthy participants aged 18-88 years (main dataset n = 215 and secondary dataset n = 433). Using linked ICA, functional network activities were characterized in independent components but not captured in the same component as structural and cerebrovascular patterns. Split-sample (n = 108/107) and out-of-sample (n = 433) validation analyses using linked ICA were also performed. Global grey matter volume with regional cerebrovascular changes and the right frontoparietal network activity were correlated with age-related and individual differences in fluid intelligence. This study presents the insights from linked ICA to bring together measurements from multiple imaging modalities, with independent and additive information. We propose that integrating multiple neuroimaging modalities allows better characterization of brain pattern variability and changes associated with healthy aging.

A deep learning based model using RNN-LSTM for the Detection of Schizophrenia from EEG data

Normal life can be ensured for schizophrenic patients if diagnosed early. Electroencephalogram (EEG) carries information about the brain network connectivity which can be used to detect brain anomalies that are indicative of schizophrenia. Since deep learning is capable of automatically extracting the significant features and make classifications, the authors proposed a deep learning based model using RNN-LSTM to analyze the EEG signal data to diagnose schizophrenia. The proposed model used three dense layers on top of a 100 dimensional LSTM. EEG signal data of 45 schizophrenic patients and 39 healthy subjects were used in the study. Dimensionality reduction algorithm was used to obtain an optimal feature set and the classifier was run with both sets of data. An accuracy of 98% and 93.67% were obtained with the complete feature set and the reduced feature set respectively. The robustness of the model was evaluated using model performance measure and combined performance measure. Outcomes were compared with the outcome obtained with traditional machine learning classifiers such as Random Forest, SVM, FURIA, and AdaBoost, and the proposed model was found to perform better with the complete dataset. When compared with the result of the researchers who worked with the same set of data using either CNN or RNN, the proposed model's accuracy was either better or comparable to theirs.

Deviations from normative brain white and gray matter structure are associated with psychopathology in youth

Combining imaging modalities and metrics that are sensitive to various aspects of brain structure and maturation may help identify individuals that show deviations in relation to same-aged peers, and thus benefit early-risk-assessment for mental disorders. We used one timepoint multimodal brain imaging, cognitive, and questionnaire data from 1280 eight- to twenty-one-year-olds from the Philadelphia Neurodevelopmental Cohort. We estimated age-related gray and white matter properties and estimated individual deviation scores using normative modeling. Next, we tested for associations between the estimated deviation scores, and with psychopathology domain scores and cognition. More negative deviations in DTI-based fractional anisotropy (FA) and the first principal eigenvalue of the diffusion tensor (L1) were associated with higher scores on psychosis positive and prodromal symptoms and general psychopathology. A more negative deviation in cortical thickness (CT) was associated with a higher general psychopathology score. Negative deviations in global FA, surface area, L1 and CT were also associated with poorer cognitive performance. No robust associations were found between the deviation scores based on CT and DTI. The low correlations between the different multimodal magnetic resonance imaging-based deviation scores suggest that psychopathological burden in adolescence can be mapped onto partly distinct neurobiological features.

Clinical and cortical similarities identified between bipolar disorder I and schizophrenia: A multivariate approach

BACKGROUND

Structural neuroimaging studies have identified similarities in the brains of individuals diagnosed with schizophrenia (SZ) and bipolar I disorder (BP), with overlap in regions of gray matter (GM) deficits between the two disorders. Recent studies have also shown that the symptom phenotypes associated with SZ and BP may allow for a more precise categorization than the current diagnostic criteria. In this study, we sought to identify GM alterations that were unique to each disorder and whether those alterations were also related to unique symptom profiles.

MATERIALS AND METHODS

We analyzed the GM patterns and clinical symptom presentations using independent component analysis (ICA), hierarchical clustering, and n-way biclustering in a large (N ∼ 3,000), merged dataset of neuroimaging data from healthy volunteers (HV), and individuals with either SZ or BP.

RESULTS

Component A showed a SZ and BP < HV GM pattern in the bilateral insula and cingulate gyrus. Component B showed a SZ and BP < HV GM pattern in the cerebellum and vermis. There were no significant differences between diagnostic groups in these components. Component C showed a SZ < HV and BP GM pattern bilaterally in the temporal poles. Hierarchical clustering of the PANSS scores and the ICA components did not yield new subgroups. N-way biclustering identified three unique subgroups of individuals within the sample that mapped onto different combinations of ICA components and symptom profiles categorized by the PANSS but no distinct diagnostic group differences.

CONCLUSION

These multivariate results show that diagnostic boundaries are not clearly related to structural differences or distinct symptom profiles. Our findings add support that (1) BP tend to have less severe symptom profiles when compared to SZ on the PANSS without a clear distinction, and (2) all the gray matter alterations follow the pattern of SZ < BP < HV without a clear distinction between SZ and BP.

Multimodal machine learning in precision health: A scoping review

Machine learning is frequently being leveraged to tackle problems in the health sector including utilization for clinical decision-support. Its use has historically been focused on single modal data. Attempts to improve prediction and mimic the multimodal nature of clinical expert decision-making has been met in the biomedical field of machine learning by fusing disparate data. This review was conducted to summarize the current studies in this field and identify topics ripe for future research. We conducted this review in accordance with the PRISMA extension for Scoping Reviews to characterize multi-modal data fusion in health. Search strings were established and used in databases: PubMed, Google Scholar, and IEEEXplore from 2011 to 2021. A final set of 128 articles were included in the analysis. The most common health areas utilizing multi-modal methods were neurology and oncology. Early fusion was the most common data merging strategy. Notably, there was an improvement in predictive performance when using data fusion. Lacking from the papers were clear clinical deployment strategies, FDA-approval, and analysis of how using multimodal approaches from diverse sub-populations may improve biases and healthcare disparities. These findings provide a summary on multimodal data fusion as applied to health diagnosis/prognosis problems. Few papers compared the outputs of a multimodal approach with a unimodal prediction. However, those that did achieved an average increase of 6.4% in predictive accuracy. Multi-modal machine learning, while more robust in its estimations over unimodal methods, has drawbacks in its scalability and the time-consuming nature of information concatenation.

Machine learning algorithm performance evaluation in structural magnetic resonance imaging-based classification of pediatric bipolar disorders type I patients

The diagnosis based on clinical assessment of pediatric bipolar disorder (PBD) may sometimes lead to misdiagnosis in clinical practice. For the past several years, machine learning (ML) methods were introduced for the classification of bipolar disorder (BD), which were helpful in the diagnosis of BD. In this study, brain cortical thickness and subcortical volume of 33 PBD-I patients and 19 age-sex matched healthy controls (HCs) were extracted from the magnetic resonance imaging (MRI) data and set as features for classification. The dimensionality reduced feature subset, which was filtered by Lasso or f_classif, was sent to the six classifiers (logistic regression (LR), support vector machine (SVM), random forest classifier, naïve Bayes, k-nearest neighbor, and AdaBoost algorithm), and the classifiers were trained and tested. Among all the classifiers, the top two classifiers with the highest accuracy were LR (84.19%) and SVM (82.80%). Feature selection was performed in the six algorithms to obtain the most important variables including the right middle temporal gyrus and bilateral pallidum, which is consistent with structural and functional anomalous changes in these brain regions in PBD patients. These findings take the computer-aided diagnosis of BD a step forward.

The effect of polygenic risk scores for major depressive disorder, bipolar disorder and schizophrenia on morphological brain measures: A systematic review of the evidence

BACKGROUND

Major depressive disorder (MDD), bipolar disorder (BD) and schizophrenia (SCZ) share clinical features and genetic bases. Magnetic Resonance Imaging (MRI) studies assessing the effect of polygenic risk score (PRS) for psychiatric disorders on brain structure show heterogeneous results. Therefore, we provided an overview of the existing evidence on the association between PRS for MDD, BD and SCZ and MRI abnormalities in clinical and healthy populations.

METHODS

A search on PubMed, Web of Science and Scopus was performed to identify the studies exploring the effect of PRS for MDD, BD and SCZ on MRI measures. A total of 25 studies were included (N = 13 on healthy individuals and N = 12 on clinical populations).

RESULTS

Both in affected and unaffected individuals, PRS for BD and SCZ showed either positive or negative correlations with cortical thickness (CT), mostly involving fronto-temporal areas, whereas PRS for MDD was associated with cortical alterations in prefrontal regions in healthy subjects.

LIMITATIONS

The heterogeneity in the methods limits the conclusions of this review.

CONCLUSIONS

Overall the evidence on the effect of PRS for MDD, BD and SCZ on brain is considerably heterogeneous and far to be conclusive. However, from the results emerged that PRS for MDD, BD and SCZ were associated with widespread cortical abnormalities in all the populations explored, suggesting that genetic risk for MDD, BD and SCZ might affect neurodevelopmental processes, resulting in cortical alterations that transcend diagnostic boundaries and seem to be independent from the clinical status.

Machine learning for prediction of schizophrenia using genetic and demographic factors in the UK biobank

Machine learning (ML) holds promise for precision psychiatry, but its predictive performance is unclear. We assessed whether ML provided added value over logistic regression for prediction of schizophrenia, and compared models built using polygenic risk scores (PRS) or clinical/demographic factors. LASSO and ridge-penalised logistic regression, support vector machines (SVM), random forests, boosting, neural networks and stacked models were trained to predict schizophrenia, using PRS for schizophrenia (PRS_SZ), sex, parental depression, educational attainment, winter birth, handedness and number of siblings as predictors. Models were evaluated for discrimination using area under the receiver operator characteristic curve (AUROC) and relative importance of predictors using permutation feature importance (PFI). In a secondary analysis, fitted models were tested for association with schizophrenia-related traits which had not been used in model development. Following learning curve analysis, 738 cases and 3690 randomly sampled controls were selected from the UK Biobank. ML models combining all predictors showed the highest discrimination (linear SVM, AUROC = 0.71), but did not significantly outperform logistic regression. AUROC was robust over 100 random resamples of controls. PFI identified PRS_SZ as the most important predictor. Highest variance in fitted models was explained by schizophrenia-related traits including fluid intelligence (most associated: linear SVM), digit symbol substitution (RBF SVM), BMI (XGBoost), smoking status (XGBoost) and deprivation (linear SVM). In conclusion, ML approaches did not provide substantial added value for prediction of schizophrenia over logistic regression, as indexed by AUROC; however, risk scores derived with different ML approaches differ with respect to association with schizophrenia-related traits.

Machine learning approaches for prediction of bipolar disorder based on biological, clinical and neuropsychological markers: a systematic review and meta-analysis

Applying machine learning (ML) to objective markers may overcome prognosis uncertainty due to the subjective nature of the diagnosis of bipolar disorder (BD). This PRISMA-compliant meta-analysis provides new systematic evidence of the BD classification accuracy reached by different markers and ML algorithms. We focused on neuroimaging, electrophysiological techniques, peripheral biomarkers, genetic data, neuropsychological or clinical measures, and multimodal approaches. PubMed, Embase and Scopus were searched through 3^rd December 2020. Meta-analyses were performed using random-effect models. Overall, 81 studies were included in this systematic review and 65 in the meta-analysis (11,336 participants, 3,903 BD). The overall pooled classification accuracy was 0.77 (95%CI[0.75;0.80]). Despite subgroup analyses for diagnostic comparison group, psychiatric disorders, marker, ML algorithm, and validation procedure were not significant, linear discriminant analysis significantly outperformed support vector machine for peripheral biomarkers (p=0.03). Sample size was inversely related to accuracy. Evidence of publication bias was detected. Ultimately, although ML reached a high accuracy in differentiating BD from other psychiatric disorders, best practices in methodology are needed for the advancement of future studies.

Decreased BOLD signals elicited by 40-Hz auditory stimulation of the right primary auditory cortex in bipolar disorder: An fMRI study

BACKGROUND

A number studies have been conducted on abnormalities in the cortical circuitry of gamma oscillations, including deficit in auditory steady-state response (ASSR) to gamma-frequency (≧ 30-Hz) stimulation, in patients with bipolar disorder (BD). In the current study, we investigated neural responses during click stimulation by blood oxygen level-dependent (BOLD) signals. We focused on Broadman 41 and 42, the main sources of ASSR.

MATERIALS AND METHODS

We acquired BOLD responses elicited by click trains of 80-, 40-, 30- and 20-Hz frequencies from 25 patients with BD to 27 healthy controls (HC) with normal hearing between 22 and 59 years of age assessed via a standard general linear-model-based analysis. We extracted contrast values by identifying the primary auditory cortex and Brodmann areas 41 and 42 as regions of interest (ROI)s.

RESULTS

BD group showed significantly decreased ASSR-BOLD signals in response to 40-Hz stimuli compared to the HC group in the right Brodmann areas 41 and 42. We found significant negative correlations between the BOLD change in the right Brodmann areas 41 and 42 and Structured Interview Guide for the Hamilton Depression Rating Scale (SIGH-D) scores, also the BOLD change in the right Brodmann areas 41 and 42 and the Positive and Negative Syndrome Scale (PANSS)-Negative scores.

CONCLUSION

The observed decrease in BOLD signal patterns in the right primary auditory cortex during 40-Hz ASSR may be a potential biomarker option for bipolar disorder.

Genetic risk for bipolar disorder and schizophrenia predicts structure and function of the ventromedial prefrontal cortex

BACKGROUND

Bipolar disorder is highly heritable and polygenic. The polygenic risk for bipolar disorder overlaps with that of schizophrenia, and polygenic scores are normally distributed in the population. Bipolar disorder has been associated with structural brain abnormalities, but it is unknown how these are linked to genetic risk factors for psychotic disorders.

METHODS

We tested whether polygenic risk scores for bipolar disorder and schizophrenia predict structural brain alterations in 98 patients with bipolar disorder and 81 healthy controls. We derived brain cortical thickness, surface area and volume from structural MRI scans. In post-hoc analyses, we correlated polygenic risk with functional hub strength, derived from resting-state functional MRI and brain connectomics.

RESULTS

Higher polygenic risk scores for both bipolar disorder and schizophrenia were associated with a thinner ventromedial prefrontal cortex (vmPFC). We found these associations in the combined group, and separately in patients and drug-naive controls. Polygenic risk for bipolar disorder was correlated with the functional hub strength of the vmPFC within the default mode network.

LIMITATIONS

Polygenic risk is a cumulative measure of genomic burden. Detailed genetic mechanisms underlying brain alterations and their cognitive consequences still need to be determined.

CONCLUSION

Our multimodal neuroimaging study linked genomic burden and brain endophenotype by demonstrating an association between polygenic risk scores for bipolar disorder and schizophrenia and the structure and function of the vmPFC. Our findings suggest that genetic factors might confer risk for psychotic disorders by influencing the integrity of the vmPFC, a brain region involved in self-referential processes and emotional regulation. Our study may also provide an imaging-genetics vulnerability marker that can be used to help identify individuals at risk for developing bipolar disorder.

Classification of first-episode psychosis using cortical thickness: A large multicenter MRI study

Machine learning classifications of first-episode psychosis (FEP) using neuroimaging have predominantly analyzed brain volumes. Some studies examined cortical thickness, but most of them have used parcellation approaches with data from single sites, which limits claims of generalizability. To address these limitations, we conducted a large-scale, multi-site analysis of cortical thickness comparing parcellations and vertex-wise approaches. By leveraging the multi-site nature of the study, we further investigated how different demographical and site-dependent variables affected predictions. Finally, we assessed relationships between predictions and clinical variables. 428 subjects (147 females, mean age 27.14) with FEP and 448 (230 females, mean age 27.06) healthy controls were enrolled in 8 centers by the ClassiFEP group. All subjects underwent a structural MRI and were clinically assessed. Cortical thickness parcellation (68 areas) and full cortical maps (20,484 vertices) were extracted. Linear Support Vector Machine was used for classification within a repeated nested cross-validation framework. Vertex-wise thickness maps outperformed parcellation-based methods with a balanced accuracy of 66.2% and an Area Under the Curve of 72%. By stratifying our sample for MRI scanner, we increased generalizability across sites. Temporal brain areas resulted as the most influential in the classification. The predictive decision scores significantly correlated with age at onset, duration of treatment, and positive symptoms. In conclusion, although far from the threshold of clinical relevance, temporal cortical thickness proved to classify between FEP subjects and healthy individuals. The assessment of site-dependent variables permitted an increase in the across-site generalizability, thus attempting to address an important machine learning limitation.

Replicating extensive brain structural heterogeneity in individuals with schizophrenia and bipolar disorder

Identifying brain processes involved in the risk and development of mental disorders is a major aim. We recently reported substantial interindividual heterogeneity in brain structural aberrations among patients with schizophrenia and bipolar disorder. Estimating the normative range of voxel-based morphometry (VBM) data among healthy individuals using a Gaussian process regression (GPR) enables us to map individual deviations from the healthy range in unseen datasets. Here, we aim to replicate our previous results in two independent samples of patients with schizophrenia (n1 = 94; n2 = 105), bipolar disorder (n1 = 116; n2 = 61), and healthy individuals (n1 = 400; n2 = 312). In line with previous findings with exception of the cerebellum our results revealed robust group level differences between patients and healthy individuals, yet only a small proportion of patients with schizophrenia or bipolar disorder exhibited extreme negative deviations from normality in the same brain regions. These direct replications support that group level-differences in brain structure disguise considerable individual differences in brain aberrations, with important implications for the interpretation and generalization of group-level brain imaging findings to the individual with a mental disorder.

Genetic neuroimaging of bipolar disorder: a systematic 2017-2020 update

There is evidence of genetic polymorphism influences on brain structure and function, genetic risk in bipolar disorder (BD), and neuroimaging correlates of BD. How genetic influences related to BD could be reflected on brain changes in BD has been efficiently reviewed in a 2017 systematic review. We aimed to confirm and extend these findings through a Preferred Reporting Items for Systematic reviews and Meta-Analyses-based systematic review. Our study allowed us to conclude that there is no replicated finding in the timeframe considered. We were also unable to further confirm prior results of the BDNF gene polymorphisms to affect brain structure and function in BD. The most consistent finding is an influence of the CACNA1C rs1006737 polymorphism in brain connectivity and grey matter structure and function. There was a tendency of undersized studies to obtain positive results and large, genome-wide polygenic risk studies to find negative results in BD. The neuroimaging genetics in BD field is rapidly expanding.

Sparse deep neural networks on imaging genetics for schizophrenia case-control classification

Deep learning methods hold strong promise for identifying biomarkers for clinical application. However, current approaches for psychiatric classification or prediction do not allow direct interpretation of original features. In the present study, we introduce a sparse deep neural network (DNN) approach to identify sparse and interpretable features for schizophrenia (SZ) case–control classification. An L₀‐norm regularization is implemented on the input layer of the network for sparse feature selection, which can later be interpreted based on importance weights. We applied the proposed approach on a large multi‐study cohort with gray matter volume (GMV) and single nucleotide polymorphism (SNP) data for SZ classification. A total of 634 individuals served as training samples, and the classification model was evaluated for generalizability on three independent datasets of different scanning protocols (N = 394, 255, and 160, respectively). We examined the classification power of pure GMV features, as well as combined GMV and SNP features. Empirical experiments demonstrated that sparse DNN slightly outperformed independent component analysis + support vector machine (ICA + SVM) framework, and more effectively fused GMV and SNP features for SZ discrimination, with an average error rate of 28.98% on external data. The importance weights suggested that the DNN model prioritized to select frontal and superior temporal gyrus for SZ classification with high sparsity, with parietal regions further included with lower sparsity, echoing previous literature. The results validate the application of the proposed approach to SZ classification, and promise extended utility on other data modalities and traits which ultimately may result in clinically useful tools.

A machine-learning framework for robust and reliable prediction of short- and long-term treatment response in initially antipsychotic-naïve schizophrenia patients based on multimodal neuropsychiatric data

The reproducibility of machine-learning analyses in computational psychiatry is a growing concern. In a multimodal neuropsychiatric dataset of antipsychotic-naïve, first-episode schizophrenia patients, we discuss a workflow aimed at reducing bias and overfitting by invoking simulated data in the design process and analysis in two independent machine-learning approaches, one based on a single algorithm and the other incorporating an ensemble of algorithms. We aimed to (1) classify patients from controls to establish the framework, (2) predict short- and long-term treatment response, and (3) validate the methodological framework. We included 138 antipsychotic-naïve, first-episode schizophrenia patients with data on psychopathology, cognition, electrophysiology, and structural magnetic resonance imaging (MRI). Perinatal data and long-term outcome measures were obtained from Danish registers. Short-term treatment response was defined as change in Positive And Negative Syndrome Score (PANSS) after the initial antipsychotic treatment period. Baseline diagnostic classification algorithms also included data from 151 matched controls. Both approaches significantly classified patients from healthy controls with a balanced accuracy of 63.8% and 64.2%, respectively. Post-hoc analyses showed that the classification primarily was driven by the cognitive data. Neither approach predicted short- nor long-term treatment response. Validation of the framework showed that choice of algorithm and parameter settings in the real data was successfully guided by results from the simulated data. In conclusion, this novel approach holds promise as an important step to minimize bias and obtain reliable results with modest sample sizes when independent replication samples are not available.

Testing relationships between multimodal modes of brain structural variation and age, sex and polygenic scores for neuroticism in children and adolescents

Human brain development involves spatially and temporally heterogeneous changes, detectable across a wide range of magnetic resonance imaging (MRI) measures. Investigating the interplay between multimodal MRI and polygenic scores (PGS) for personality traits associated with mental disorders in youth may provide new knowledge about typical and atypical neurodevelopment. We derived independent components across cortical thickness, cortical surface area, and grey/white matter contrast (GWC) (n = 2596, 3-23 years), and tested for associations between these components and age, sex and-, in a subsample (n = 878), PGS for neuroticism. Age was negatively associated with a single-modality component reflecting higher global GWC, and additionally with components capturing common variance between global thickness and GWC, and several multimodal regional patterns. Sex differences were found for components primarily capturing global and regional surface area (boys > girls), but also regional cortical thickness. For PGS for neuroticism, we found weak and bidirectional associations with a component reflecting right prefrontal surface area. These results indicate that multimodal fusion is sensitive to age and sex differences in brain structure in youth, but only weakly to polygenic load for neuroticism.

Will machine learning applied to neuroimaging in bipolar disorder help the clinician? A critical review and methodological suggestions

OBJECTIVES

The existence of anatomofunctional brain abnormalities in bipolar disorder (BD) is now well established by magnetic resonance imaging (MRI) studies. To create diagnostic and prognostic tools, as well as identifying biologically valid subtypes of BD, research has recently turned towards the use of machine learning (ML) techniques. We assessed both supervised ML and unsupervised ML studies in BD to evaluate their robustness, reproducibility and the potential need for improvement.

METHOD

We systematically searched for studies using ML algorithms based on MRI data of patients with BD until February 2019.

RESULT

We identified 47 studies, 45 using supervised ML techniques and 2 including unsupervised ML analyses. Among supervised studies, 43 focused on diagnostic classification. The reported accuracies for classification of BD ranged between (a) 57% and 100%, for BD vs healthy controls; (b) 49.5% and 93.1% for BD vs patients with major depressive disorder; and (c) 50% and 96.2% for BD vs patients with schizophrenia. Reported accuracies for discriminating subjects genetically at risk for BD (either from control or from patients with BD) ranged between 64.3% and 88.93%.

CONCLUSIONS

Although there are strong methodological limitations in previous studies and an important need for replication in large multicentric samples, the conclusions of our review bring hope of future computer-aided diagnosis of BD and pave the way for other applications, such as treatment response prediction. To reinforce the reliability of future results we provide methodological suggestions for good practice in conducting and reporting MRI-based ML studies in BD.

Neurocognitive and psychiatric disorders-related axonal degeneration in Parkinson's disease

Neurocognitive and psychiatric disorders have significant consequences for quality of life in patients with Parkinson's disease (PD). In the current study, we evaluated microstructural white matter (WM) alterations associated with neurocognitive and psychiatric disorders in PD using neurite orientation dispersion and density imaging (NODDI) and linked independent component analysis (LICA). The indices of NODDI were compared between 20 and 19 patients with PD with and without neurocognitive and psychiatric disorders, respectively, and 25 healthy controls using tract‐based spatial statistics and tract‐of‐interest analyses. LICA was applied to model inter‐subject variability across measures. A widespread reduction in axonal density (indexed by intracellular volume fraction [ICVF]) was demonstrated in PD patients with and without neurocognitive and psychiatric disorders, as compared with healthy controls. Compared with patients without neurocognitive and psychiatric disorders, patients with neurocognitive and psychiatric disorders exhibited more extensive (posterior predominant) decreases in axonal density. Using LICA, ICVF demonstrated the highest contribution (59% weight) to the main effects of diagnosis that reflected widespread decreases in axonal density. These findings suggest that axonal loss is a major factor underlying WM pathology related to neurocognitive and psychiatric disorders in PD, whereas patients with neurocognitive and psychiatric disorders had broader axonal pathology, as compared with those without. LICA suggested that the ICVF can be used as a useful biomarker of microstructural changes in the WM related to neurocognitive and psychiatric disorders in PD.

Longitudinal Cortical Thickness Changes in Bipolar Disorder and the Relationship to Genetic Risk, Mania, and Lithium Use

BACKGROUND

Bipolar disorder (BD) is a highly heritable psychiatric disorder characterized by episodes of manic and depressed mood states and associated with cortical brain abnormalities. Although the course of BD is often progressive, longitudinal brain imaging studies are scarce. It remains unknown whether brain abnormalities are static traits of BD or result from pathological changes over time. Moreover, the genetic effect on implicated brain regions remains unknown.

METHODS

Patients with BD and healthy control (HC) subjects underwent structural magnetic resonance imaging at baseline (123 patients, 83 HC subjects) and after 6 years (90 patients, 61 HC subjects). Cortical thickness maps were generated using FreeSurfer. Using linear mixed effects models, we compared longitudinal changes in cortical thickness between patients with BD and HC subjects across the whole brain. We related our findings to genetic risk for BD and tested for effects of demographic and clinical variables.

RESULTS

Patients showed abnormal cortical thinning of temporal cortices and thickness increases in visual/somatosensory brain areas. Thickness increases were related to genetic risk and lithium use. Patients who experienced hypomanic or manic episodes between time points showed abnormal thinning in inferior frontal cortices. Cortical changes did not differ between diagnostic BD subtypes I and II.

CONCLUSIONS

In the largest longitudinal BD study to date, we detected abnormal cortical changes with high anatomical resolution. We delineated regional effects of clinical symptoms, genetic factors, and medication that may explain progressive brain changes in BD. Our study yields important insights into disease mechanisms and suggests that neuroprogression plays a role in BD.

Using structural MRI to identify bipolar disorders - 13 site machine learning study in 3020 individuals from the ENIGMA Bipolar Disorders Working Group

Bipolar disorders (BDs) are among the leading causes of morbidity and disability. Objective biological markers, such as those based on brain imaging, could aid in clinical management of BD. Machine learning (ML) brings neuroimaging analyses to individual subject level and may potentially allow for their diagnostic use. However, fair and optimal application of ML requires large, multi-site datasets. We applied ML (support vector machines) to MRI data (regional cortical thickness, surface area, subcortical volumes) from 853 BD and 2167 control participants from 13 cohorts in the ENIGMA consortium. We attempted to differentiate BD from control participants, investigated different data handling strategies and studied the neuroimaging/clinical features most important for classification. Individual site accuracies ranged from 45.23% to 81.07%. Aggregate subject-level analyses yielded the highest accuracy (65.23%, 95% CI = 63.47-67.00, ROC-AUC = 71.49%, 95% CI = 69.39-73.59), followed by leave-one-site-out cross-validation (accuracy = 58.67%, 95% CI = 56.70-60.63). Meta-analysis of individual site accuracies did not provide above chance results. There was substantial agreement between the regions that contributed to identification of BD participants in the best performing site and in the aggregate dataset (Cohen's Kappa = 0.83, 95% CI = 0.829-0.831). Treatment with anticonvulsants and age were associated with greater odds of correct classification. Although short of the 80% clinically relevant accuracy threshold, the results are promising and provide a fair and realistic estimate of classification performance, which can be achieved in a large, ecologically valid, multi-site sample of BD participants based on regional neurostructural measures. Furthermore, the significant classification in different samples was based on plausible and similar neuroanatomical features. Future multi-site studies should move towards sharing of raw/voxelwise neuroimaging data.

Transdiagnostic Multimodal Neuroimaging in Psychosis: Structural, Resting-State, and Task Magnetic Resonance Imaging Correlates of Cognitive Control

BACKGROUND

Disorders with psychotic features, including schizophrenia and some bipolar disorders, are associated with impairments in regulation of goal-directed behavior, termed cognitive control. Cognitive control-related neural alterations have been studied in psychosis. However, studies are typically unimodal, and relationships across modalities of brain function and structure remain unclear. Thus, we performed transdiagnostic multimodal analyses to examine cognitive control-related neural variation in psychosis.

METHODS

Structural, resting, and working memory task imaging for 31 control participants, 27 participants with bipolar disorder, and 23 participants with schizophrenia were collected and processed identically to the Human Connectome Project, enabling identification of relationships with prior multimodal work. Two cognitive control-related independent components (ICs) derived from the Human Connectome Project using multiset canonical correlation analysis with joint IC analysis were used to predict performance in psychosis. De novo multiset canonical correlation analysis with joint IC analysis was performed, and the results were correlated with cognitive control.

RESULTS

A priori working memory and cortical thickness maps significantly predicted cognitive control in psychosis. De novo multiset canonical correlation analysis with joint IC analysis identified an IC correlated with cognitive control that also discriminated groups. Structural contributions included insular and cingulate regions; task contributions included precentral, posterior parietal, cingulate, and visual regions; and resting-state contributions highlighted canonical network organization. Follow-up analyses suggested that correlations with cognitive control were primarily influenced by participants with schizophrenia.

CONCLUSIONS

A priori and de novo imaging replicably identified a set of interrelated patterns across modalities and the healthy-to-psychosis spectrum, suggesting robustness of these features. Relationships between imaging and cognitive control performance suggest that shared symptomatology may be key to identifying transdiagnostic relationships in psychosis.

Distinct structural brain circuits indicate mood and apathy profiles in bipolar disorder

Bipolar disorder (BD) is a severe manic-depressive illness. Patients with BD have been shown to have gray matter (GM) deficits in prefrontal, frontal, parietal, and temporal regions; however, the relationship between structural effects and clinical profiles has proved elusive when considered on a region by region or voxel by voxel basis. In this study, we applied parallel independent component analysis (pICA) to structural neuroimaging measures and the positive and negative syndrome scale (PANSS) in 110 patients (mean age 34.9 ± 11.65) with bipolar disorder, to examine networks of brain regions that relate to symptom profiles. The pICA revealed two distinct symptom profiles and associated GM concentration alteration circuits. The first PANSS pICA profile mainly involved anxiety, depression and guilty feelings, reflecting mood symptoms. Reduced GM concentration in right temporal regions predicted worse mood symptoms in this profile. The second PANSS pICA profile generally covered blunted affect, emotional withdrawal, passive/apathetic social withdrawal, depression and active social avoidance, exhibiting a withdrawal or apathy dominating component. Lower GM concentration in bilateral parietal and frontal regions showed worse symptom severity in this profile. In summary, a pICA decomposition suggested BD patients showed distinct mood and apathy profiles differing from the original PANSS subscales, relating to distinct brain structural networks.

•

Structural relationships with symptoms in bipolar disorder are complex.

•

A parallel ICA analysis of PANSS questions and structural images finds two correlated profiles.

•

The first pair links mood symptoms with right temporal regions.

•

The second pair highlights social withdrawal and apathy symptoms linked to bilateral frontal and parietal regions.

Brain Heterogeneity in Schizophrenia and Its Association With Polygenic Risk

Importance

Between-individual variability in brain structure is determined by gene-environment interactions, possibly reflecting differential sensitivity to environmental and genetic perturbations. Magnetic resonance imaging (MRI) studies have revealed thinner cortices and smaller subcortical volumes in patients with schizophrenia. However, group-level comparisons may mask considerable within-group heterogeneity, which has largely remained unnoticed in the literature.

Objectives

To compare brain structural variability between individuals with schizophrenia and healthy controls and to test whether respective variability reflects the polygenic risk score (PRS) for schizophrenia in an independent sample of healthy controls.

Design, Setting, and Participants

This case-control and polygenic risk analysis compared MRI-derived cortical thickness and subcortical volumes between healthy controls and patients with schizophrenia across 16 cohorts and tested for associations between PRS and MRI features in a control cohort from the UK Biobank. Data were collected from October 27, 2004, through April 12, 2018, and analyzed from December 3, 2017, through August 1, 2018.

Main Outcomes and Measures

Mean and dispersion parameters were estimated using double generalized linear models. Vertex-wise analysis was used to assess cortical thickness, and regions-of-interest analyses were used to assess total cortical volume, total surface area, and white matter, subcortical, and hippocampal subfield volumes. Follow-up analyses included within-sample analysis, test of robustness of the PRS threshold, population covariates, outlier removal, and control for image quality.

Results

A comparison of 1151 patients with schizophrenia (mean [SD] age, 33.8 [10.6] years; 68.6% male [n = 790] and 31.4% female [n = 361]) with 2010 healthy controls (mean [SD] age, 32.6 [10.4] years; 56.0% male [n = 1126] and 44.0% female [n = 884]) revealed higher heterogeneity in schizophrenia for cortical thickness and area (t = 3.34), cortical (t = 3.24) and ventricle (t range, 3.15-5.78) volumes, and hippocampal subfields (t range, 2.32-3.55). In the UK Biobank sample of 12 490 participants (mean [SD] age, 55.9 [7.5] years; 48.2% male [n = 6025] and 51.8% female [n = 6465]), higher PRS was associated with thinner frontal and temporal cortices and smaller left CA2/3 (t = -3.00) but was not significantly associated with dispersion.

Conclusions and Relevance

This study suggests that schizophrenia is associated with substantial brain structural heterogeneity beyond the mean differences. These findings may reflect higher sensitivity to environmental and genetic perturbations in patients, supporting the heterogeneous nature of schizophrenia. A higher PRS was associated with thinner frontotemporal cortices and smaller hippocampal subfield volume, but not heterogeneity. This finding suggests that brain variability in schizophrenia results from interactions between environmental and genetic factors that are not captured by the PRS. Factors contributing to heterogeneity in frontotemporal cortices and hippocampus are key to furthering our understanding of how genetic and environmental factors shape brain biology in schizophrenia.

Transdiagnostic neuroimaging in psychiatry: A review

Transdiagnostic approach has a long history in neuroimaging, predating its recent ascendance as a paradigm for new psychiatric nosology. Various psychiatric disorders have been compared for commonalities and differences in neuroanatomical features and activation patterns, with different aims and rationales. This review covers both structural and functional neuroimaging publications with direct comparison of different psychiatric disorders, including schizophrenia, bipolar disorder, major depressive disorder, autism spectrum disorder, obsessive-compulsive disorder, attention-deficit/hyperactivity disorder, conduct disorder, anorexia nervosa, and bulimia nervosa. Major findings are systematically presented along with specific rationales for each comparison.

Association of antidepressants with brain morphology in early stages of psychosis: an imaging genomics approach

Depressive symptoms in subjects at Clinical High Risk for Psychosis (CHR-P) or at first-episode psychosis (FEP) are often treated with antidepressants. Our cross-sectional study investigated whether brain morphology is altered by antidepressant medication. High-resolution T₁-weighted structural MRI scans of 33 CHR-P and FEP subjects treated with antidepressants, 102 CHR-P and FEP individuals without antidepressant treatment and 55 controls, were automatically segmented using Freesurfer 6.0. Linear mixed-effects modelling was applied to assess the differences in subcortical volume, surface area and cortical thickness in treated, non-treated and healthy subjects, taking into account converted dosages of antidepressants. Increasing antidepressant dose was associated with larger volume of the pallidum and the putamen, and larger surface of the left inferior temporal gyrus. In a pilot subsample of separately studied subjects of known genomic risk loci, we found that in the right postcentral gyrus, the left paracentral lobule and the precentral gyrus antidepressant dose-associated surface increase depended on polygenic schizophrenia-related-risk score. As the reported regions are linked to the symptoms of psychosis, our findings reflect the possible beneficial effects of antidepressant treatment on an emerging psychosis.

Structural Variability in the Human Brain Reflects Fine-Grained Functional Architecture at the Population Level

Human brain structure topography is thought to be related in part to functional specialization. However, the extent of such relationships is unclear. Here, using a data-driven, multimodal approach for studying brain structure across the lifespan (N = 484, n = 260 females), we demonstrate that numerous structural networks, covering the entire brain, follow a functionally meaningful architecture. These gray matter networks (GMNs) emerge from the covariation of gray matter volume and cortical area at the population level. We further reveal fine-grained anatomical signatures of functional connectivity. For example, within the cerebellum, a structural separation emerges between lobules that are functionally connected to distinct, mainly sensorimotor, cognitive and limbic regions of the cerebral cortex and subcortex. Structural modes of variation also replicate the fine-grained functional architecture seen in eight well defined visual areas in both task and resting-state fMRI. Furthermore, our study shows a structural distinction corresponding to the established segregation between anterior and posterior default-mode networks (DMNs). These fine-grained GMNs further cluster together to form functionally meaningful larger-scale organization. In particular, we identify a structural architecture bringing together the functional posterior DMN and its anticorrelated counterpart. In summary, our results demonstrate that the relationship between structural and functional connectivity is fine-grained, widespread across the entire brain, and driven by covariation in cortical area, i.e. likely differences in shape, depth, or number of foldings. These results suggest that neurotrophic events occur during development to dictate that the size and folding pattern of distant, functionally connected brain regions should vary together across subjects.SIGNIFICANCE STATEMENT Questions about the relationship between structure and function in the human brain have engaged neuroscientists for centuries in a debate that continues to this day. Here, by investigating intersubject variation in brain structure across a large number of individuals, we reveal modes of structural variation that map onto fine-grained functional organization across the entire brain, and specifically in the cerebellum, visual areas, and default-mode network. This functionally meaningful structural architecture emerges from the covariation of gray matter volume and cortical folding. These results suggest that the neurotrophic events at play during development, and possibly evolution, which dictate that the size and folding pattern of distant brain regions should vary together across subjects, might also play a role in functional cortical specialization.

Composite Pain Biomarker Signatures for Objective Assessment and Effective Treatment

Pain is a subjective sensory experience that can, mostly, be reported but cannot be directly measured or quantified. Nevertheless, a suite of biomarkers related to mechanisms, neural activity, and susceptibility offer the possibility-especially when used in combination-to produce objective pain-related indicators with the specificity and sensitivity required for diagnosis and for evaluation of risk of developing pain and of analgesic efficacy. Such composite biomarkers will also provide improved understanding of pain pathophysiology.

Reproducible grey matter patterns index a multivariate, global alteration of brain structure in schizophrenia and bipolar disorder

Schizophrenia is a severe mental disorder characterized by numerous subtle changes in brain structure and function. Machine learning allows exploring the utility of combining structural and functional brain magnetic resonance imaging (MRI) measures for diagnostic application, but this approach has been hampered by sample size limitations and lack of differential diagnostic data. Here, we performed a multi-site machine learning analysis to explore brain structural patterns of T1 MRI data in 2668 individuals with schizophrenia, bipolar disorder or attention-deficit/ hyperactivity disorder, and healthy controls. We found reproducible changes of structural parameters in schizophrenia that yielded a classification accuracy of up to 76% and provided discrimination from ADHD, through it lacked specificity against bipolar disorder. The observed changes largely indexed distributed grey matter alterations that could be represented through a combination of several global brain-structural parameters. This multi-site machine learning study identified a brain-structural signature that could reproducibly differentiate schizophrenia patients from controls, but lacked specificity against bipolar disorder. While this currently limits the clinical utility of the identified signature, the present study highlights that the underlying alterations index substantial global grey matter changes in psychotic disorders, reflecting the biological similarity of these conditions, and provide a roadmap for future exploration of brain structural alterations in psychiatric patients.