Multivariate neuroanatomical classification of cognitive subtypes in schizophrenia: a support vector machine learning approach

Magnetic resonance imaging-based machine learning classification of schizophrenia spectrum disorders: a meta-analysis

BACKGROUND

Recent advances in multivariate pattern recognition have fostered the search for reliable neuroimaging-based biomarkers in psychiatric conditions, including schizophrenia. These approaches consider the complex pattern of alterations in brain function and structure, overcoming the limitations of traditional univariate methods. To assess the reliability of neuroimaging-based biomarkers and the contribution of study characteristics in distinguishing individuals with schizophrenia spectrum disorder (SSD) from healthy controls (HCs), we conducted a systematic review of the studies that used multivariate pattern recognition for this objective.

METHODS

We systematically searched PubMed, Scopus, and Web of Science for studies on SSD classification using multivariate pattern analysis on magnetic resonance imaging data. We employed a bivariate random-effects meta-analytic model to explore the classification of sensitivity (SE) and specificity (SP) across studies while also evaluating the moderator effects of clinical and non-clinical variables.

RESULTS

A total of 119 studies (with 12,723 patients with SSD and 13,196 HCs) were identified. The meta-analysis estimated a SE of 79.1% (95% confidence interval [CI], 77.1%-81.0%) and a SP of 80.0% (95% CI, 77.8%-82.0%). In particular, the Positive and Negative Syndrome Scale and the Global Assessment of Functioning scores, age, age of onset, duration of untreated psychosis, deep learning, algorithm type, features selection, and validation methods had significant effects on classification performance.

CONCLUSIONS

Multivariate pattern analysis reliably identifies neuroimaging-based biomarkers of SSD, achieving ∼80% SE and SP. Despite clinical heterogeneity, discernible brain modifications effectively differentiate SSD from HCs. Classification performance depends on patient-related and methodological factors crucial for the development, validation, and application of prospective models in clinical settings.

Cognitive subgroups of affective and non-affective psychosis show differences in medication and cortico-subcortical brain networks

Cognitive deficits are prevalent in individuals with psychosis and are associated with neurobiological changes, potentially serving as an endophenotype for psychosis. Using the HCP-Early-Psychosis-dataset (n = 226), we aimed to investigate cognitive subtypes (deficit/intermediate/spared) through data-driven clustering in affective (AP) and non-affective psychosis patients (NAP) and controls (HC). We explored differences between three clusters in symptoms, cognition, medication, and grey matter volume. Applying principal component analysis, we selected features for clustering. Features that explained most variance were scores for intelligence, verbal recognition and comprehension, auditory attention, working memory, reasoning and executive functioning. Fuzzy K-Means clustering on those features revealed that the subgroups significantly varied in cognitive impairment, clinical symptoms, and, importantly, also in medication and grey matter volume in fronto-parietal and subcortical networks. The spared cluster (86%HC, 37%AP, 17%NAP) exhibited unimpaired cognition, lowest symptoms/medication, and grey matter comparable to controls. The deficit cluster (4%HC, 10%AP, 47%NAP) had impairments across all domains, highest symptoms scores/medication dosage, and pronounced grey matter alterations. The intermediate deficit cluster (11%HC, 54%AP, 36%NAP) showed fewer deficits than the second cluster, but similar symptoms/medication/grey matter to the spared cluster. Controlling for medication, cognitive scores correlated with grey matter changes and negative symptoms across all patients. Our findings generally emphasize the interplay between cognition, brain structure, symptoms, and medication in AP and NAP, and specifically suggest a possible mediating role of cognition, highlighting the potential of screening cognitive changes to aid tailoring treatments and interventions.

Characterizing cognitive subtypes in schizophrenia using cortical curvature

Cognitive deficits are a core symptom of schizophrenia, but research on their neural underpinnings has been challenged by the heterogeneity in deficits' severity among patients. Here, we address this issue by combining logistic regression and random forest to classify two neuropsychological profiles of patients with high (HighCog) and low (LowCog) cognitive performance in two independent samples. We based our analysis on the cortical features grey matter volume (VOL), cortical thickness (CT), and mean curvature (MC) of N = 57 patients (discovery sample) and validated the classification in an independent sample (N = 52). We investigated which cortical feature would yield the best classification results and expected that the 10 most important features would include frontal and temporal brain regions. The model based on MC had the best performance with area under the curve (AUC) values of 76% and 73%, and identified fronto-temporal and occipital brain regions as the most important features for the classification. Moreover, subsequent comparison analyses could reveal significant differences in MC of single brain regions between the two cognitive profiles. The present study suggests MC as a promising neuroanatomical parameter for characterizing schizophrenia cognitive subtypes.

Transdiagnostic subgroups of cognitive impairment in early affective and psychotic illness

Cognitively impaired and spared patient subgroups were identified in psychosis and depression, and in clinical high-risk for psychosis (CHR). Studies suggest differences in underlying brain structural and functional characteristics. It is unclear whether cognitive subgroups are transdiagnostic phenomena in early stages of psychotic and affective disorder which can be validated on the neural level. Patients with recent-onset psychosis (ROP; N = 140; female = 54), recent-onset depression (ROD; N = 130; female = 73), CHR (N = 128; female = 61) and healthy controls (HC; N = 270; female = 165) were recruited through the multi-site study PRONIA. The transdiagnostic sample and individual study groups were clustered into subgroups based on their performance in eight cognitive domains and characterized by gray matter volume (sMRI) and resting-state functional connectivity (rsFC) using support vector machine (SVM) classification. We identified an impaired subgroup (NROP = 79, NROD = 30, NCHR = 37) showing cognitive impairment in executive functioning, working memory, processing speed and verbal learning (all p < 0.001). A spared subgroup (NROP = 61, NROD = 100, NCHR = 91) performed comparable to HC. Single-disease subgroups indicated that cognitive impairment is stronger pronounced in impaired ROP compared to impaired ROD and CHR. Subgroups in ROP and ROD showed specific symptom- and functioning-patterns. rsFC showed superior accuracy compared to sMRI in differentiating transdiagnostic subgroups from HC (BACimpaired = 58.5%; BACspared = 61.7%, both: p < 0.01). Cognitive findings were validated in the PRONIA replication sample (N = 409). Individual cognitive subgroups in ROP, ROD and CHR are more informative than transdiagnostic subgroups as they map onto individual cognitive impairment and specific functioning- and symptom-patterns which show limited overlap in sMRI and rsFC. CLINICAL TRIAL REGISTRY NAME: German Clinical Trials Register (DRKS). Clinical trial registry URL: https://www.drks.de/drks_web/ . Clinical trial registry number: DRKS00005042.

Estimating multimodal brain variability in schizophrenia spectrum disorders: A worldwide ENIGMA study

Objective

Schizophrenia is a multifaceted disorder associated with structural brain heterogeneity. Despite its relevance for identifying illness subtypes and informative biomarkers, structural brain heterogeneity in schizophrenia remains incompletely understood. Therefore, the objective of this study was to provide a comprehensive insight into the structural brain heterogeneity associated with schizophrenia.

Methods

This meta- and mega-analysis investigated the variability of multimodal structural brain measures of white and gray matter in individuals with schizophrenia versus healthy controls. Using the ENIGMA dataset of MRI-based brain measures from 22 international sites with up to 6139 individuals for a given brain measure, we examined variability in cortical thickness, surface area, folding index, subcortical volume and fractional anisotropy.

Results

We found that individuals with schizophrenia are distinguished by higher heterogeneity in the frontotemporal network with regard to multimodal structural measures. Moreover, individuals with schizophrenia showed higher homogeneity of the folding index, especially in the left parahippocampal region.

Conclusions

Higher multimodal heterogeneity in frontotemporal regions potentially implies different subtypes of schizophrenia that converge on impaired frontotemporal interaction as a core feature of the disorder. Conversely, more homogeneous folding patterns in the left parahippocampal region might signify a consistent characteristic of schizophrenia shared across subtypes. These findings underscore the importance of structural brain variability in advancing our neurobiological understanding of schizophrenia, and aid in identifying illness subtypes as well as informative biomarkers.

Prevalence of cognitive impairments and strengths in the early course of psychosis and depression

BACKGROUND

Studies investigating cognitive impairments in psychosis and depression have typically compared the average performance of the clinical group against healthy controls (HC), and do not report on the actual prevalence of cognitive impairments or strengths within these clinical groups. This information is essential so that clinical services can provide adequate resources to supporting cognitive functioning. Thus, we investigated this prevalence in individuals in the early course of psychosis or depression.

METHODS

A comprehensive cognitive test battery comprising 12 tests was completed by 1286 individuals aged 15-41 (mean age 25.07, s.d. 5.88) from the PRONIA study at baseline: HC (N = 454), clinical high risk for psychosis (CHR; N = 270), recent-onset depression (ROD; N = 267), and recent-onset psychosis (ROP; N = 295). Z-scores were calculated to estimate the prevalence of moderate or severe deficits or strengths (>2 s.d. or 1-2 s.d. below or above HC, respectively) for each cognitive test.

RESULTS

Impairment in at least two cognitive tests was as follows: ROP (88.3% moderately, 45.1% severely impaired), CHR (71.2% moderately, 22.4% severely impaired), ROD (61.6% moderately, 16.2% severely impaired). Across clinical groups, impairments were most prevalent in tests of working memory, processing speed, and verbal learning. Above average performance (>1 s.d.) in at least two tests was present for 40.5% ROD, 36.1% CHR, 16.1% ROP, and was >2 SDs in 1.8% ROD, 1.4% CHR, and 0% ROP.

CONCLUSIONS

These findings suggest that interventions should be tailored to the individual, with working memory, processing speed, and verbal learning likely to be important transdiagnostic targets.

Evaluating the clinical utility of speech analysis and machine learning in schizophrenia: A pilot study

BACKGROUND

Schizophrenia is a serious mental disorder that significantly impacts social functioning and quality of life. However, current diagnostic methods lack objective biomarker support. While some studies have indicated differences in audio features between patients with schizophrenia and healthy controls, these findings are influenced by demographic information and variations in experimental paradigms. Therefore, it is crucial to explore stable and reliable audio biomarkers for an auxiliary diagnosis and disease severity prediction of schizophrenia.

METHOD

A total of 130 individuals (65 patients with schizophrenia and 65 healthy controls) read three fixed texts containing positive, neutral, and negative emotions, and recorded them. All audio signals were preprocessed and acoustic features were extracted by a librosa-0.9.2 toolkit. Independent sample t-tests were performed on two sets of acoustic features, and Pearson correlation on the acoustic features and Positive and Negative Syndrome Scale (PANSS) scores of the schizophrenia group. Classification algorithms in scikit-learn were used to diagnose schizophrenia and predict the level of negative symptoms.

RESULTS

Significant differences were observed between the two groups in the mfcc_8, mfcc_11, and mfcc_33 of mel-frequency cepstral coefficient (MFCC). Furthermore, a significant correlation was found between mfcc_7 and the negative PANSS scores. Through acoustic features, we could not only differentiate patients with schizophrenia from healthy controls with an accuracy of 0.815 but also predict the grade of the negative symptoms in schizophrenia with an average accuracy of 0.691.

CONCLUSIONS

The results demonstrated the considerable potential of acoustic characteristics as reliable biomarkers for diagnosing schizophrenia and predicting clinical symptoms.

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.

Brain Morphological Characteristics of Cognitive Subgroups of Schizophrenia-Spectrum Disorders and Bipolar Disorder: A Systematic Review with Narrative Synthesis

Despite a growing body of research, there is yet to be a cohesive synthesis of studies examining differences in brain morphology according to patterns of cognitive function among both schizophrenia-spectrum disorder (SSD) and bipolar disorder (BD) individuals. We aimed to provide a systematic overview of the morphological differences-inclusive of grey and white matter volume, cortical thickness, and cortical surface area-between cognitive subgroups of these disorders and healthy controls, and between cognitive subgroups themselves. An initial search of PubMed and Scopus databases resulted in 1486 articles of which 20 met inclusion criteria and were reviewed in detail. The findings of this review do not provide strong evidence that cognitive subgroups of SSD or BD map to unique patterns of brain morphology. There is preliminary evidence to suggest that reductions in cortical thickness may be more strongly associated with cognitive impairment, whilst volumetric deficits may be largely tied to the presence of disease.

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.

Automatic recognition of schizophrenia from facial videos using 3D convolutional neural network

Schizophrenia affects patients and their families and society because of chronic impairments in cognition, behavior, and emotion. However, its clinical diagnosis mainly depends on the clinicians' knowledge of the patients' symptoms. Other auxiliary diagnostic methods such as MRI and EEG are cumbersome and time-consuming. Recently, the convolutional neural network (CNN) has been applied to the auxiliary diagnosis of psychiatry. Hence, in this study, a method based on deep learning and facial videos is proposed for the rapid detection of schizophrenia. Herein, 125 videos from 125 schizophrenic patients and 75 videos from 75 healthy controls based on emotional stimulation tasks were obtained. The video preprocessing included the experiment clips extraction, face detection, facial region cropping, resizing to 500 × 500 pixel size, and uniform sampling of 100 frames. The preprocessed facial videos were used to train the Resnet18_3D. We utilized ten-fold cross-validation, and held-out testing set to evaluate the model with the accuracy, the precision, the sensitivity, the specificity, the balanced accuracy, and the AUC. The Resnet18_3D trained on Film_order achieved the best performance with accuracy, sensitivity, specificity, balanced accuracy, and AUC of 89.00%, 96.80%, 76.00%, 86.40% and 0.9397. The neural network model indeed recognizes healthy controls and schizophrenic patients through the changes in the area of the face. The results show that facial video under emotional stimulation can be used to classify schizophrenic patients and help clinicians with diagnosis in the clinical environment. Among the different types of stimuli, the video stimuli with fixed emotional order showed the best classification performance.

A morphological study of schizophrenia with magnetic resonance imaging, advanced analytics, and machine learning

We have performed a morphological analysis of patients with schizophrenia and compared them with healthy controls. Our analysis includes the use of publicly available automated extraction tools to assess regional cortical thickness (inclusive of within region cortical thickness variability) from structural magnetic resonance imaging (MRI), to characterize group-wise abnormalities associated with schizophrenia based on a publicly available dataset. We have also performed a correlation analysis between the automatically extracted biomarkers and a variety of patient clinical variables available. Finally, we also present the results of a machine learning analysis. Results demonstrate regional cortical thickness abnormalities in schizophrenia. We observed a correlation (rho = 0.474) between patients’ depression and the average cortical thickness of the right medial orbitofrontal cortex. Our leading machine learning technology evaluated was the support vector machine with stepwise feature selection, yielding a sensitivity of 92% and a specificity of 74%, based on regional brain measurements, including from the insula, superior frontal, caudate, calcarine sulcus, gyrus rectus, and rostral middle frontal regions. These results imply that advanced analytic techniques combining MRI with automated biomarker extraction can be helpful in characterizing patients with schizophrenia.

Structural MRI-Based Schizophrenia Classification Using Autoencoders and 3D Convolutional Neural Networks in Combination with Various Pre-Processing Techniques

Schizophrenia is a severe neuropsychiatric disease whose diagnosis, unfortunately, lacks an objective diagnostic tool supporting a thorough psychiatric examination of the patient. We took advantage of today’s computational abilities, structural magnetic resonance imaging, and modern machine learning methods, such as stacked autoencoders (SAE) and 3D convolutional neural networks (3D CNN), to teach them to classify 52 patients with schizophrenia and 52 healthy controls. The main aim of this study was to explore whether complex feature extraction methods can help improve the accuracy of deep learning-based classifiers compared to minimally preprocessed data. Our experiments employed three commonly used preprocessing steps to extract three different feature types. They included voxel-based morphometry, deformation-based morphometry, and simple spatial normalization of brain tissue. In addition to classifier models, features and their combination, other model parameters such as network depth, number of neurons, number of convolutional filters, and input data size were also investigated. Autoencoders were trained on feature pools of 1000 and 5000 voxels selected by Mann-Whitney tests, and 3D CNNs were trained on whole images. The most successful model architecture (autoencoders) achieved the highest average accuracy of 69.62% (sensitivity 68.85%, specificity 70.38%). The results of all experiments were statistically compared (the Mann-Whitney test). In conclusion, SAE outperformed 3D CNN, while preprocessing using VBM helped SAE improve the results.

Computing schizophrenia: ethical challenges for machine learning in psychiatry

Recent advances in machine learning (ML) promise far-reaching improvements across medical care, not least within psychiatry. While to date no psychiatric application of ML constitutes standard clinical practice, it seems crucial to get ahead of these developments and address their ethical challenges early on. Following a short general introduction concerning ML in psychiatry, we do so by focusing on schizophrenia as a paradigmatic case. Based on recent research employing ML to further the diagnosis, treatment, and prediction of schizophrenia, we discuss three hypothetical case studies of ML applications with view to their ethical dimensions. Throughout this discussion, we follow the principlist framework by Tom Beauchamp and James Childress to analyse potential problems in detail. In particular, we structure our analysis around their principles of beneficence, non-maleficence, respect for autonomy, and justice. We conclude with a call for cautious optimism concerning the implementation of ML in psychiatry if close attention is paid to the particular intricacies of psychiatric disorders and its success evaluated based on tangible clinical benefit for patients.

Identification of transdiagnostic psychiatric disorder subtypes using unsupervised learning

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

Cognitive Deficit in Schizophrenia: From Etiology to Novel Treatments

Schizophrenia is a major mental illness characterized by positive and negative symptoms, and by cognitive deficit. Although cognitive impairment is disabling for patients, it has been largely neglected in the treatment of schizophrenia. There are several reasons for this lack of treatments for cognitive deficit, but the complexity of its etiology-in which neuroanatomic, biochemical and genetic factors concur-has contributed to the lack of effective treatments. In the last few years, there have been several attempts to develop novel drugs for the treatment of cognitive impairment in schizophrenia. Despite these efforts, little progress has been made. The latest findings point to the importance of developing personalized treatments for schizophrenia which enhance neuroplasticity, and of combining pharmacological treatments with non-pharmacological measures.

Hold-out validation for the assessment of stability and reliability of multivariable regression demonstrated with magnetic resonance imaging of patients with schizophrenia

Neuroscience studies are very often tasked with identifying measurable differences between two groups of subjects, typically one group with a pathological condition and one group representing control subjects. It is often expected that the measurements acquired for comparing groups are also affected by a variety of additional patient characteristics such as sex, age, and comorbidities. Multivariable regression (MVR) is a statistical analysis technique commonly employed in neuroscience studies to "control for" or "adjust for" secondary effects (such as sex, age, and comorbidities) in order to ensure that the main study findings are focused on actual differences between the groups of interest associated with the condition under investigation. It is common practice in the neuroscience literature to utilize MVR to control for secondary effects; however, at present, it is not typically possible to assess whether the MVR adjustments correct for more error than they introduce. In common neuroscience practice, MVR models are not validated and no attempt to characterize deficiencies in the MVR model is made. In this article, we demonstrate how standard hold-out validation techniques (commonly used in machine learning analyses) that involve repeatedly randomly dividing datasets into training and testing samples can be adapted to the assessment of stability and reliability of MVR models with a publicly available neurological magnetic resonance imaging (MRI) dataset of patients with schizophrenia. Results demonstrate that MVR can introduce measurement error up to 30.06% and, on average across all considered measurements, introduce 9.84% error on this dataset. When hold-out validated MVR does not agree with the results of the standard use of MVR, the use of MVR in the given application is unstable. Thus, this paper helps evaluate the extent to which the simplistic use of MVR introduces study error in neuroscientific analyses with an analysis of patients with schizophrenia.

Cognitive subtypes in recent onset psychosis: distinct neurobiological fingerprints?

In schizophrenia, neurocognitive subtypes can be distinguished based on cognitive performance and they are associated with neuroanatomical alterations. We investigated the existence of cognitive subtypes in shortly medicated recent onset psychosis patients, their underlying gray matter volume patterns and clinical characteristics. We used a K-means algorithm to cluster 108 psychosis patients from the multi-site EU PRONIA (Prognostic tools for early psychosis management) study based on cognitive performance and validated the solution independently (N = 53). Cognitive subgroups and healthy controls (HC; n = 195) were classified based on gray matter volume (GMV) using Support Vector Machine classification. A cognitively spared (N = 67) and impaired (N = 41) subgroup were revealed and partially independently validated (N_spared = 40, N_impaired = 13). Impaired patients showed significantly increased negative symptomatology (p_fdr = 0.003), reduced cognitive performance (p_fdr < 0.001) and general functioning (p_fdr < 0.035) in comparison to spared patients. Neurocognitive deficits of the impaired subgroup persist in both discovery and validation sample across several domains, including verbal memory and processing speed. A GMV pattern (balanced accuracy = 60.1%, p = 0.01) separating impaired patients from HC revealed increases and decreases across several fronto-temporal-parietal brain areas, including basal ganglia and cerebellum. Cognitive and functional disturbances alongside brain morphological changes in the impaired subgroup are consistent with a neurodevelopmental origin of psychosis. Our findings emphasize the relevance of tailored intervention early in the course of psychosis for patients suffering from the likely stronger neurodevelopmental character of the disease.

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

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

Brain morphology does not clearly map to cognition in individuals on the bipolar-schizophrenia-spectrum: a cross-diagnostic study of cognitive subgroups

BACKGROUND

Characterisation of brain morphological features common to cognitively similar individuals with bipolar disorder (BD) and schizophrenia spectrum disorders (SSD) may be key to understanding their shared neurobiological deficits. In the current study we examined whether three previously characterised cross-diagnostic cognitive subgroups differed among themselves and in comparison to healthy controls across measures of brain morphology.

METHOD

T1-weighted structural magnetic resonance imaging scans were obtained for 143 individuals; 65 healthy controls and 78 patients (SSD, n = 40; BD I, n = 38) classified into three cross-diagnostic cognitive subgroups: Globally Impaired (n = 24), Selectively Impaired (n = 32), and Superior/Near-Normal (n = 22). Cognitive subgroups were compared to each other and healthy controls on three separate analyses investigating (1) global, (2) regional, and (3) vertex-wise comparisons of brain volume, thickness, and surface area.

RESULTS

No significant subgroup differences were evident in global measures of brain morphology. In region of interest analyses, the Selectively Impaired subgroup had greater right accumbens volume than those Superior/Near-Normal subgroup and healthy controls, and the Superior/Near-Normal subgroup had reduced volume of the left entorhinal region compared to all other groups. In vertex-wise comparisons, the Globally Impaired subgroup had greater right precentral volume than the Selectively Impaired subgroup, and thicker cortex in the postcentral region relative to the Superior/Near-Normal subgroup.

LIMITATIONS

Exploration of medication effects was limited in our data.

CONCLUSIONS

Although some differences were evident in this sample, generally cross-diagnostic cognitive subgroups of individuals with SSD and BD did not appear to be clearly distinguished by patterns in brain morphology.

Promises and pitfalls of deep neural networks in neuroimaging-based psychiatric research

By promising more accurate diagnostics and individual treatment recommendations, deep neural networks and in particular convolutional neural networks have advanced to a powerful tool in medical imaging. Here, we first give an introduction into methodological key concepts and resulting methodological promises including representation and transfer learning, as well as modelling domain-specific priors. After reviewing recent applications within neuroimaging-based psychiatric research, such as the diagnosis of psychiatric diseases, delineation of disease subtypes, normative modeling, and the development of neuroimaging biomarkers, we discuss current challenges. This includes for example the difficulty of training models on small, heterogeneous and biased data sets, the lack of validity of clinical labels, algorithmic bias, and the influence of confounding variables.

Identification of Children at Risk of Schizophrenia via Deep Learning and EEG Responses

The prospective identification of children likely to develop schizophrenia is a vital tool to support early interventions that can mitigate the risk of progression to clinical psychosis. Electroencephalographic (EEG) patterns from brain activity and deep learning techniques are valuable resources in achieving this identification. We propose automated techniques that can process raw EEG waveforms to identify children who may have an increased risk of schizophrenia compared to typically developing children. We also analyse abnormal features that remain during developmental follow-up over a period of   ∼ 4 years in children with a vulnerability to schizophrenia initially assessed when aged 9 to 12 years. EEG data from participants were captured during the recording of a passive auditory oddball paradigm. We undertake a holistic study to identify brain abnormalities, first by exploring traditional machine learning algorithms using classification methods applied to hand-engineered features (event-related potential components). Then, we compare the performance of these methods with end-to-end deep learning techniques applied to raw data. We demonstrate via average cross-validation performance measures that recurrent deep convolutional neural networks can outperform traditional machine learning methods for sequence modeling. We illustrate the intuitive salient information of the model with the location of the most relevant attributes of a post-stimulus window. This baseline identification system in the area of mental illness supports the evidence of developmental and disease effects in a pre-prodromal phase of psychosis. These results reinforce the benefits of deep learning to support psychiatric classification and neuroscientific research more broadly.

Machine-learning classification using neuroimaging data in schizophrenia, autism, ultra-high risk and first-episode psychosis

Neuropsychiatric disorders are diagnosed based on behavioral criteria, which makes the diagnosis challenging. Objective biomarkers such as neuroimaging are needed, and when coupled with machine learning, can assist the diagnostic decision and increase its reliability. Sixty-four schizophrenia, 36 autism spectrum disorder (ASD), and 106 typically developing individuals were analyzed. FreeSurfer was used to obtain the data from the participant's brain scans. Six classifiers were utilized to classify the subjects. Subsequently, 26 ultra-high risk for psychosis (UHR) and 17 first-episode psychosis (FEP) subjects were run through the trained classifiers. Lastly, the classifiers' output of the patient groups was correlated with their clinical severity. All six classifiers performed relatively well to distinguish the subject groups, especially support vector machine (SVM) and Logistic regression (LR). Cortical thickness and subcortical volume feature groups were most useful for the classification. LR and SVM were highly consistent with clinical indices of ASD. When UHR and FEP groups were run with the trained classifiers, majority of the cases were classified as schizophrenia, none as ASD. Overall, SVM and LR were the best performing classifiers. Cortical thickness and subcortical volume were most useful for the classification, compared to surface area. LR, SVM, and DT's output were clinically informative. The trained classifiers were able to help predict the diagnostic category of both UHR and FEP Individuals.

Towards a brain-based predictome of mental illness

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

Identificacion of MRI-based psychosis subtypes: Replication and refinement

The identification of the cerebral substrates of psychoses such as schizophrenia and bipolar disorder is likely hampered by its biological heterogeneity, which may contribute to the low replication of results in the field. In this study we aimed to replicate in a completely new sample and supplement the results of a previous study with additional data on this topic. In the aforementioned study we identified a schizophrenia cluster characterized by high mean cortical curvature and low cortical thickness, subcortical hypometabolism and progressive negative symptoms. Here, we have used magnetic resonance images from 61 schizophrenia and 28 bipolar patients, as well as 51 healthy controls and a cluster analysis to search for possible subgroups primarily characterized by cerebral structural data. Diffusion tensor imaging (fractional anisotropy, FA), cognition, clinical data and electroencephalographic (EEG) modulation during a P300 task were used to validate the possible clusters. Two clusters of patients were identified. The first cluster (29 schizophrenia and 18 bipolar patients) showed decreased cortical thickness and area values, as well as lower subcortical volumes and higher cortical curvature in some regions, as compared to the second cluster. This first cluster also showed decreased FA in frontal lobe connections and worse cognitive performance. Although this cluster also showed longer illness duration, there were first episode patients in both clusters and treatment doses and types were not different between clusters. Both clusters of patients showed decreased EEG task-related modulation. In conclusion, our data give additional support to a distinct biologically based cluster encompassing schizophrenia and bipolar disorder patients with cortical and subcortical alterations, hampered cortical connectivity and lower cognitive performance.

Polygenic disruption of retinoid signalling in schizophrenia and a severe cognitive deficit subtype

Retinoid metabolites of vitamin A are intrinsically linked to neural development, connectivity and plasticity, and have been implicated in the pathophysiology of schizophrenia. We hypothesised that a greater burden of common and rare genomic variation in genes involved with retinoid biogenesis and signalling could be associated with schizophrenia and its cognitive symptoms. Common variants associated with schizophrenia in the largest genome-wide association study were aggregated in retinoid genes and used to formulate a polygenic risk score (PRSRet) for each participant in the Australian Schizophrenia Research Bank. In support of our hypothesis, we found PRSRet to be significantly associated with the disorder. Cases with severe cognitive deficits, while not further differentiated by PRSRet, were enriched with rare variation in the retinoic acid receptor beta gene RARB, detected through whole-genome sequencing. RARB rare variant burden was also associated with reduced cerebellar volume in the cases with marked cognitive deficit, and with covariation in grey matter throughout the brain. An excess of rare variation was further observed in schizophrenia in retinoic acid response elements proximal to target genes, which we show are differentially expressed in the disorder in two RNA sequencing datasets. Our results suggest that genomic variation may disrupt retinoid signalling in schizophrenia, with particular significance for cases with severe cognitive impairment.

Morphological Profiling of Schizophrenia: Cluster Analysis of MRI-Based Cortical Thickness Data

The diagnosis of schizophrenia is thought to embrace several distinct subgroups. The manifold entities in a single clinical patient group increase the variance of biological measures, deflate the group-level estimates of causal factors, and mask the presence of treatment effects. However, reliable neurobiological boundaries to differentiate these subgroups remain elusive. Since cortical thinning is a well-established feature in schizophrenia, we investigated if individuals (patients and healthy controls) with similar patterns of regional cortical thickness form naturally occurring morphological subtypes. K-means algorithm clustering was applied to regional cortical thickness values obtained from 256 structural MRI scans (179 patients with schizophrenia and 77 healthy controls [HCs]). GAP statistics revealed three clusters with distinct regional thickness patterns. The specific patterns of cortical thinning, clinical characteristics, and cognitive function of each clustered subgroup were assessed. The three clusters based on thickness patterns comprised of a morphologically impoverished subgroup (25% patients, 1% HCs), an intermediate subgroup (47% patients, 46% HCs), and an intact subgroup (28% patients, 53% HCs). The differences of clinical features among three clusters pertained to age-of-onset, N-back performance, duration exposure to treatment, total burden of positive symptoms, and severity of delusions. Particularly, the morphologically impoverished group had deficits in N-back performance and less severe positive symptom burden. The data-driven neuroimaging approach illustrates the occurrence of morphologically separable subgroups in schizophrenia, with distinct clinical characteristics. We infer that the anatomical heterogeneity of schizophrenia arises from both pathological deviance and physiological variance. We advocate using MRI-guided stratification for clinical trials as well as case-control investigations in schizophrenia.

Heterogeneity of Striatal Dopamine Function in Schizophrenia: Meta-analysis of Variance

BACKGROUND

It has been hypothesized that dopamine function in schizophrenia exhibits heterogeneity in excess of that seen in the general population. However, no previous study has systematically tested this hypothesis.

METHODS

We employed meta-analysis of variance to investigate interindividual variability of striatal dopaminergic function in patients with schizophrenia and in healthy control subjects. We included 65 studies that reported molecular imaging measures of dopamine synthesis or release capacities, dopamine D_2/3 receptor (D2/3R) or dopamine transporter (DAT) availabilities, or synaptic dopamine levels in 983 patients and 968 control subjects. Variability differences were quantified using variability ratio (VR) and coefficient of variation ratio.

RESULTS

Interindividual variability of striatal D2/3R (VR = 1.26, p < .0001) and DAT (VR = 1.31, p = .01) availabilities and synaptic dopamine levels (VR = 1.38, p = .045) but not dopamine synthesis (VR = 1.12, p = .13) or release (VR = 1.08, p = .70) capacities were significantly greater in patients than in control subjects. Findings were robust to variability measure. Mean dopamine synthesis (g = 0.65, p = .004) and release (g = 0.66, p = .03) capacities, as well as synaptic levels (g = 0.78, p = .0006), were greater in patients overall, but mean synthesis capacity did not differ from that of control subjects in treatment-resistant patients (p > .3). Mean D2/3R (g = 0.17, p = .14) and DAT (g = -0.20, p = .28) availabilities did not differ between groups.

CONCLUSIONS

Our findings demonstrate significant heterogeneity of striatal dopamine function in schizophrenia. They suggest that while elevated dopamine synthesis and release capacities may be core features of the disorder, altered D2/3R and DAT availabilities and synaptic dopamine levels may occur only in a subgroup of patients. This heterogeneity may contribute to variation in treatment response and side effects.

Polygenic Risk Scores for Subtyping of Schizophrenia

Schizophrenia is a complex disorder with many comorbid conditions. In this study, we used polygenic risk scores (PRSs) from schizophrenia and comorbid traits to explore consistent cluster structure in schizophrenia patients. With 10 comorbid traits, we found a stable 4-cluster structure in two datasets (MGS and SSCCS). When the same traits and parameters were applied for the patients in a clinical trial of antipsychotics, the CATIE study, a 5-cluster structure was observed. One of the 4 clusters found in the MGS and SSCCS was further split into two clusters in CATIE, while the other 3 clusters remained unchanged. For the 5 CATIE clusters, we evaluated their association with the changes of clinical symptoms, neurocognitive functions, and laboratory tests between the enrollment baseline and the end of Phase I trial. Class I was found responsive to treatment, with significant reduction for the total, positive, and negative symptoms (p = 0.0001, 0.0099, and 0.0028, respectively), and improvement for cognitive functions (VIGILANCE, p = 0.0099; PROCESSING SPEED, p = 0.0006; WORKING MEMORY, p = 0.0023; and REASONING, p = 0.0015). Class II had modest reduction of positive symptoms (p = 0.0492) and better PROCESSING SPEED (p = 0.0071). Class IV had a specific reduction of negative symptoms (p = 0.0111) and modest cognitive improvement for all tested domains. Interestingly, Class IV was also associated with decreased lymphocyte counts and increased neutrophil counts, an indication of ongoing inflammation or immune dysfunction. In contrast, Classes III and V showed no symptom reduction but a higher level of phosphorus. Overall, our results suggest that PRSs from schizophrenia and comorbid traits can be utilized to classify patients into subtypes with distinctive clinical features. This genetic susceptibility based subtyping may be useful to facilitate more effective treatment and outcome prediction.

A Systematic Review of Studies Reporting Data-Driven Cognitive Subtypes across the Psychosis Spectrum

The delineation of cognitive subtypes of schizophrenia and bipolar disorder may offer a means of determining shared genetic markers and neuropathology among individuals with these conditions. We systematically reviewed the evidence from published studies reporting the use of data-driven (i.e., unsupervised) clustering methods to delineate cognitive subtypes among adults diagnosed with schizophrenia, schizoaffective disorder, or bipolar disorder. We reviewed 24 studies in total, contributing data to 13 analyses of schizophrenia spectrum patients, 8 analyses of bipolar disorder, and 5 analyses of mixed samples of schizophrenia and bipolar disorder participants. Studies of bipolar disorder most consistently revealed a 3-cluster solution, comprising a subgroup with 'near-normal' (cognitively spared) cognition and two other subgroups demonstrating graded deficits across cognitive domains. In contrast, there was no clear consensus regarding the number of cognitive subtypes among studies of cognitive subtypes in schizophrenia, while four of the five studies of mixed diagnostic groups reported a 4-cluster solution. Common to all cluster solutions was a severe cognitive deficit subtype with cognitive impairments of moderate to large effect size relative to healthy controls. Our review highlights several key factors (e.g., symptom profile, sample size, statistical procedures, and cognitive domains examined) that may influence the results of data-driven clustering methods, and which were largely inconsistent across the studies reviewed. This synthesis of findings suggests caution should be exercised when interpreting the utility of particular cognitive subtypes for biological investigation, and demonstrates much heterogeneity among studies using unsupervised clustering approaches to cognitive subtyping within and across the psychosis spectrum.

Improving individual predictions: Machine learning approaches for detecting and attacking heterogeneity in schizophrenia (and other psychiatric diseases)

Psychiatric diseases are very heterogeneous both in clinical manifestation and etiology. With the recent rise of using machine learning techniques to attempt to diagnose and prognose these disorders, the issue of heterogeneity becomes increasingly important. With the growing interest in personalized medicine, it becomes even more important to not only classify someone as a patient with a certain disorder, its treatment needs a more precise definition of the underlying neurobiology, since different biological origins of the same disease may require (very) different treatments. We review the possible contributions that machine learning techniques could make to explore the heterogeneous nature of psychiatric disorders with a focus on schizophrenia. First we will review how heterogeneity shows up and how machine learning, or multivariate pattern recognition methods in general, can be used to discover it. Secondly, we will discuss the possible uses of these techniques to attack heterogeneity, leading to improved predictions and understanding of the neurobiological background of the disorder.

Identifying schizophrenia subgroups using clustering and supervised learning

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

Decentralized distribution-sampled classification models with application to brain imaging

BACKGROUND

In this age of big data, certain models require very large data stores in order to be informative and accurate. In many cases however, the data are stored in separate locations requiring data transfer between local sites which can cause various practical hurdles, such as privacy concerns or heavy network load. This is especially true for medical imaging data, which can be constrained due to the health insurance portability and accountability act (HIPAA) which provides security protocols for medical data. Medical imaging datasets can also contain many thousands or millions of features, requiring heavy network load.

NEW METHOD

Our research expands upon current decentralized classification research by implementing a new singleshot method for both neural networks and support vector machines. Our approach is to estimate the statistical distribution of the data at each local site and pass this information to the other local sites where each site resamples from the individual distributions and trains a model on both locally available data and the resampled data. The model for each local site produces its own accuracy value which are then averaged together to produce the global average accuracy.

RESULTS

We show applications of our approach to handwritten digit classification as well as to multi-subject classification of brain imaging data collected from patients with schizophrenia and healthy controls. Overall, the results showed comparable classification accuracy to the centralized model with lower network load than multishot methods.

COMPARISON WITH EXISTING METHODS

Many decentralized classifiers are multishot, requiring heavy network traffic. Our model attempts to alleviate this load while preserving prediction accuracy.

CONCLUSIONS

We show that our proposed approach performs comparably to a centralized approach while minimizing network traffic compared to multishot methods.

Brain Morphometry Methods for Feature Extraction in Random Subspace Ensemble Neural Network Classification of First-Episode Schizophrenia

Machine learning (ML) is a growing field that provides tools for automatic pattern recognition. The neuroimaging community currently tries to take advantage of ML in order to develop an auxiliary diagnostic tool for schizophrenia diagnostics. In this letter, we present a classification framework based on features extracted from magnetic resonance imaging (MRI) data using two automatic whole-brain morphometry methods: voxel-based (VBM) and deformation-based morphometry (DBM). The framework employs a random subspace ensemble-based artificial neural network classifier—in particular, a multilayer perceptron (MLP). The framework was tested on data from first-episode schizophrenia patients and healthy controls. The experiments differed in terms of feature extraction methods, using VBM, DBM, and a combination of both morphometry methods. Thus, features of different types were available for model adaptation. As we expected, the combination of features increased the MLP classification accuracy up to 73.12%—an improvement of 5% versus MLP-based only on VBM or DBM features. To further verify the findings, other comparisons using support vector machines in place of MLPs were made within the framework. However, it cannot be concluded that any classifier was better than another.

Machine Learning in Acute Ischemic Stroke Neuroimaging

Machine Learning (ML) through pattern recognition algorithms is currently becoming an essential aid for the diagnosis, treatment, and prediction of complications and patient outcomes in a number of neurological diseases. The evaluation and treatment of Acute Ischemic Stroke (AIS) have experienced a significant advancement over the past few years, increasingly requiring the use of neuroimaging for decision-making. In this review, we offer an insight into the recent developments and applications of ML in neuroimaging focusing on acute ischemic stroke.

Combined structural and functional patterns discriminating upper limb motor disability in multiple sclerosis using multivariate approaches

A structural or functional pattern of neuroplasticity that could systematically discriminate between people with impaired and preserved motor performance could help us to understand the brain networks contributing to preservation or compensation of behavior in multiple sclerosis (MS). This study aimed to (1) investigate whether a machine learning-based technique could accurately classify MS participants into groups defined by upper extremity function (i.e. motor function preserved (MP) vs. motor function impaired (MI)) based on their regional grey matter measures (GMM, cortical thickness and deep grey matter volume) and inter-regional functional connection (FC), (2) investigate which features (GMM, FC, or GMM + FC) could classify groups more accurately, and (3) identify the multivariate patterns of GMM and FCs that are most discriminative between MP and MI participants, and between each of these groups and the healthy controls (HCs). With 26 MP, 25 MI, and 21 HCs (age and sex matched) underwent T1-weighted and resting-state functional MRI at 3 T, we applied support vector machine (SVM) based classification to learn discriminant functions indicating regions in which GMM or between which FCs were most discriminative between groups. This study demonstrates that there exist structural and FC patterns sufficient for correct classification of upper limb motor ability of people with MS. The classifier with GMM + FC features yielded the highest accuracy of 85.61 % (p < 0.001) to distinguish between the MS groups using leave-one-out cross-validation. It suggests that a machine-learning approach combining structural and functional features is useful for identifying the specific neural substrates that are necessary and sufficient to preserve motor function among people with MS.

Decoding diagnosis and lifetime consumption in alcohol dependence from grey-matter pattern information

OBJECTIVE

We investigated the potential of computer-based models to decode diagnosis and lifetime consumption in alcohol dependence (AD) from grey-matter pattern information. As machine-learning approaches to psychiatric neuroimaging have recently come under scrutiny due to unclear generalization and the opacity of algorithms, our investigation aimed to address a number of methodological criticisms.

METHOD

Participants were adult individuals diagnosed with AD (N = 119) and substance-naïve controls (N = 97) ages 20-65 who underwent structural MRI. Machine-learning models were applied to predict diagnosis and lifetime alcohol consumption.

RESULTS

A classification scheme based on regional grey matter attained 74% diagnostic accuracy and predicted lifetime consumption with high accuracy (r = 0.56, P < 10^-10 ). A key advantage of the classification scheme was its algorithmic transparency, revealing cingulate, insular and inferior frontal cortices as important brain areas underlying classification. Validation of the classification scheme on data of an independent trial was successful with nearly identical accuracy, addressing the concern of generalization. Finally, compared to a blinded radiologist, computer-based classification showed higher accuracy and sensitivity, reduced age and gender biases, but lower specificity.

CONCLUSION

Computer-based models applied to whole-brain grey-matter predicted diagnosis and lifetime consumption in AD with good accuracy. Computer-based classification may be particularly suited as a screening tool with high sensitivity.

Segmentation and texture analysis of structural biomarkers using neighborhood-clustering-based level set in MRI of the schizophrenic brain

OBJECTIVE

Schizophrenia (SZ) is a psychiatric disorder that especially affects individuals during their adolescence. There is a need to study the subanatomical regions of SZ brain on magnetic resonance images (MRI) based on morphometry. In this work, an attempt was made to analyze alterations in structure and texture patterns in images of the SZ brain using the level-set method and Laws texture features.

MATERIALS AND METHODS

T1-weighted MRI of the brain from Center of Biomedical Research Excellence (COBRE) database were considered for analysis. Segmentation was carried out using the level-set method. Geometrical and Laws texture features were extracted from the segmented brain stem, corpus callosum, cerebellum, and ventricle regions to analyze pattern changes in SZ.

RESULTS

The level-set method segmented multiple brain regions, with higher similarity and correlation values compared with an optimized method. The geometric features obtained from regions of the corpus callosum and ventricle showed significant variation (p < 0.00001) between normal and SZ brain. Laws texture feature identified a heterogeneous appearance in the brain stem, corpus callosum and ventricular regions, and features from the brain stem were correlated with Positive and Negative Syndrome Scale (PANSS) score (p < 0.005).

CONCLUSION

A framework of geometric and Laws texture features obtained from brain subregions can be used as a supplement for diagnosis of psychiatric disorders.

Heterogeneity and Homogeneity of Regional Brain Structure in Schizophrenia: A Meta-analysis

IMPORTANCE

Schizophrenia is associated with alterations in mean regional brain volumes. However, it is not known whether the clinical heterogeneity seen in the disorder is reflected at the neurobiological level, for example, in differences in the interindividual variability of these brain volumes relative to control individuals.

OBJECTIVE

To investigate whether patients with first-episode schizophrenia exhibit greater variability of regional brain volumes in addition to mean volume differences.

DATA SOURCES

Studies that reported regional brain volumetric measures in patients and controls by using magnetic resonance imaging in the MEDLINE, EMBASE, and PsycINFO databases from inception to October 1, 2016, were examined.

STUDY SELECTION

Case-control studies that reported regional brain volumes in patients with first-episode schizophrenia and healthy controls by using magnetic resonance imaging were selected.

DATA EXTRACTION AND SYNTHESIS

Means and variances (SDs) were extracted for each measure to calculate effect sizes, which were combined using multivariate meta-analysis.

MAIN OUTCOMES AND MEASURES

Relative variability of regional brain volumetric measurements in patients compared with control groups as indexed by the variability ratio (VR) and coefficient of variation ratio (CVR). Hedges g was used to quantify mean differences.

RESULTS

A total of 108 studies that reported measurements from 3901 patients (1272 [32.6%] female) with first-episode schizophrenia and 4040 controls (1613 [39.9%] female) were included in the analyses. Variability of putamen (VR, 1.13; 95% CI, 1.03-1.24; P = .01), temporal lobe (VR, 1.12; 95% CI, 1.04-1.21; P = .004), thalamus (VR, 1.16; 95% CI, 1.07-1.26; P < .001), and third ventricle (VR, 1.43; 95% CI, 1.20-1.71; P < 1 × 10-5) volume was significantly greater in patients, whereas variability of anterior cingulate cortex volume was lower (VR, 0.89; 95% CI, 0.81-0.98; P = .02). These findings were robust to choice of outcome measure. There was no evidence of altered variability of caudate nucleus or frontal lobe volumes. Mean volumes of the lateral (g = 0.40; 95% CI, 0.29-0.51; P < .001) and third ventricles (g = 0.43; 95% CI, 0.26-0.59; P < .001) were greater, whereas mean volumes of the amygdala (g = -0.46; -0.65 to -0.26; P < .001), anterior cingulate cortex (g = -0.26; 95% CI, -0.43 to -0.10; P = .005), frontal lobe (g = -0.31; 95% CI, -0.44 to -0.19; P = .001), hippocampus (g = -0.66; 95% CI, -0.84 to -0.47; P < .001), temporal lobe (g = -0.22; 95% CI, -0.36 to -0.09; P = .001), and thalamus (g = -0.36; 95% CI, -0.57 to -0.15; P = .001) were lower in patients. There was no evidence of altered mean volume of caudate nucleus or putamen.

CONCLUSIONS AND RELEVANCE

In addition to altered mean volume of many brain structures, schizophrenia is associated with significantly greater variability of temporal cortex, thalamus, putamen, and third ventricle volumes, consistent with biological heterogeneity in these regions, but lower variability of anterior cingulate cortex volume. This finding indicates greater homogeneity of anterior cingulate volume and, considered with the significantly lower mean volume of this region, suggests that this is a core region affected by the disorder.

Auditory prediction errors as individual biomarkers of schizophrenia

Schizophrenia is a complex psychiatric disorder, typically diagnosed through symptomatic evidence collected through patient interview. We aim to develop an objective biologically-based computational tool which aids diagnosis and relies on accessible imaging technologies such as electroencephalography (EEG). To achieve this, we used machine learning techniques and a combination of paradigms designed to elicit prediction errors or Mismatch Negativity (MMN) responses. MMN, an EEG component elicited by unpredictable changes in sequences of auditory stimuli, has previously been shown to be reduced in people with schizophrenia and this is arguably one of the most reproducible neurophysiological markers of schizophrenia.

EEG data were acquired from 21 patients with schizophrenia and 22 healthy controls whilst they listened to three auditory oddball paradigms comprising sequences of tones which deviated in 10% of trials from regularly occurring standard tones. Deviant tones shared the same properties as standard tones, except for one physical aspect: 1) duration - the deviant stimulus was twice the duration of the standard; 2) monaural gap - deviants had a silent interval omitted from the standard, or 3) inter-aural timing difference, which caused the deviant location to be perceived as 90° away from the standards.

We used multivariate pattern analysis, a machine learning technique implemented in the Pattern Recognition for Neuroimaging Toolbox (PRoNTo) to classify images generated through statistical parametric mapping (SPM) of spatiotemporal EEG data, i.e. event-related potentials measured on the two-dimensional surface of the scalp over time. Using support vector machine (SVM) and Gaussian processes classifiers (GPC), we were able classify individual patients and controls with balanced accuracies of up to 80.48% (p-values = 0.0326, FDR corrected) and an ROC analysis yielding an AUC of 0.87. Crucially, a GP regression revealed that MMN predicted global assessment of functioning (GAF) scores (correlation = 0.73, R² = 0.53, p = 0.0006).

•

The diagnostic utility of multiple auditory oddball stimulus paradigms is assessed.

•

Greatest classification accuracy was achieved using a monaural gap stimulus paradigm.

•

The full post-stimulus epoch contains relevant discriminatory components.

Multi-center machine learning in imaging psychiatry: A meta-model approach

One of the biggest problems in automated diagnosis of psychiatric disorders from medical images is the lack of sufficiently large samples for training. Sample size is especially important in the case of highly heterogeneous disorders such as schizophrenia, where machine learning models built on relatively low numbers of subjects may suffer from poor generalizability. Via multicenter studies and consortium initiatives researchers have tried to solve this problem by combining data sets from multiple sites. The necessary sharing of (raw) data is, however, often hindered by legal and ethical issues. Moreover, in the case of very large samples, the computational complexity might become too large. The solution to this problem could be distributed learning. In this paper we investigated the possibility to create a meta-model by combining support vector machines (SVM) classifiers trained on the local datasets, without the need for sharing medical images or any other personal data. Validation was done in a 4-center setup comprising of 480 first-episode schizophrenia patients and healthy controls in total. We built SVM models to separate patients from controls based on three different kinds of imaging features derived from structural MRI scans, and compared models built on the joint multicenter data to the meta-models. The results showed that the combined meta-model had high similarity to the model built on all data pooled together and comparable classification performance on all three imaging features. Both similarity and performance was superior to that of the local models. We conclude that combining models is thus a viable alternative that facilitates data sharing and creating bigger and more informative models.

Single subject prediction of brain disorders in neuroimaging: Promises and pitfalls

Neuroimaging-based single subject prediction of brain disorders has gained increasing attention in recent years. Using a variety of neuroimaging modalities such as structural, functional and diffusion MRI, along with machine learning techniques, hundreds of studies have been carried out for accurate classification of patients with heterogeneous mental and neurodegenerative disorders such as schizophrenia and Alzheimer's disease. More than 500 studies have been published during the past quarter century on single subject prediction focused on a multiple brain disorders. In the first part of this study, we provide a survey of more than 200 reports in this field with a focus on schizophrenia, mild cognitive impairment (MCI), Alzheimer's disease (AD), depressive disorders, autism spectrum disease (ASD) and attention-deficit hyperactivity disorder (ADHD). Detailed information about those studies such as sample size, type and number of extracted features and reported accuracy are summarized and discussed. To our knowledge, this is by far the most comprehensive review of neuroimaging-based single subject prediction of brain disorders. In the second part, we present our opinion on major pitfalls of those studies from a machine learning point of view. Common biases are discussed and suggestions are provided. Moreover, emerging trends such as decentralized data sharing, multimodal brain imaging, differential diagnosis, disease subtype classification and deep learning are also discussed. Based on this survey, there is extensive evidence showing the great potential of neuroimaging data for single subject prediction of various disorders. However, the main bottleneck of this exciting field is still the limited sample size, which could be potentially addressed by modern data sharing models such as the ones discussed in this paper. Emerging big data technologies and advanced data-intensive machine learning methodologies such as deep learning have coincided with an increasing need for accurate, robust and generalizable single subject prediction of brain disorders during an exciting time. In this report, we survey the past and offer some opinions regarding the road ahead.