Using biological indices to classify schizophrenia and other psychotic patients

Multikernel Capsule Network for Schizophrenia Identification

Schizophrenia seriously affects the quality of life. To date, both simple (e.g., linear discriminant analysis) and complex (e.g., deep neural network) machine-learning methods have been utilized to identify schizophrenia based on functional connectivity features. The existing simple methods need two separate steps (i.e., feature extraction and classification) to achieve the identification, which disables simultaneous tuning for the best feature extraction and classifier training. The complex methods integrate two steps and can be simultaneously tuned to achieve optimal performance, but these methods require a much larger amount of data for model training. To overcome the aforementioned drawbacks, we proposed a multikernel capsule network (MKCapsnet), which was developed by considering the brain anatomical structure. Kernels were set to match partition sizes of the brain anatomical structure in order to capture interregional connectivities at the varying scales. With the inspiration of the widely used dropout strategy in deep learning, we developed capsule dropout in the capsule layer to prevent overfitting of the model. The comparison results showed that the proposed method outperformed the state-of-the-art methods. Besides, we compared performances using different parameters and illustrated the routing process to reveal characteristics of the proposed method. MKCapsnet is promising for schizophrenia identification. Our study first utilized the capsule neural network for analyzing functional connectivity of magnetic resonance imaging (MRI) and proposed a novel multikernel capsule structure with the consideration of brain anatomical parcellation, which could be a new way to reveal brain mechanisms. In addition, we provided useful information in the parameter setting, which is informative for further studies using a capsule network for other neurophysiological signal classification.

Psychosis Biotypes: Replication and Validation from the B-SNIP Consortium

Current clinical phenomenological diagnosis in psychiatry neither captures biologically homologous disease entities nor allows for individualized treatment prescriptions based on neurobiology. In this report, we studied two large samples of cases with schizophrenia, schizoaffective, and bipolar I disorder with psychosis, presentations with clinical features of hallucinations, delusions, thought disorder, affective, or negative symptoms. A biomarker approach to subtyping psychosis cases (called psychosis Biotypes) captured neurobiological homology that was missed by conventional clinical diagnoses. Two samples (called "B-SNIP1" with 711 psychosis and 274 healthy persons, and the "replication sample" with 717 psychosis and 198 healthy persons) showed that 44 individual biomarkers, drawn from general cognition (BACS), motor inhibitory (stop signal), saccadic system (pro- and anti-saccades), and auditory EEG/ERP (paired-stimuli and oddball) tasks of psychosis-relevant brain functions were replicable (r's from .96-.99) and temporally stable (r's from .76-.95). Using numerical taxonomy (k-means clustering) with nine groups of integrated biomarker characteristics (called bio-factors) yielded three Biotypes that were virtually identical between the two samples and showed highly similar case assignments to subgroups based on cross-validations (88.5%-89%). Biotypes-1 and -2 shared poor cognition. Biotype-1 was further characterized by low neural response magnitudes, while Biotype-2 was further characterized by overactive neural responses and poor sensory motor inhibition. Biotype-3 was nearly normal on all bio-factors. Construct validation of Biotype EEG/ERP neurophysiology using measures of intrinsic neural activity and auditory steady state stimulation highlighted the robustness of these outcomes. Psychosis Biotypes may yield meaningful neurobiological targets for treatments and etiological investigations.

Bimodal distribution of tone-matching deficits indicates discrete pathophysiological entities within the syndrome of schizophrenia

To date, no measures are available that permit differentiation of discrete, clinically distinct subtypes of schizophrenia (SZ) with potential differential underlying pathophysiologies. Over recent years, there has been increasing recognition that SZ is heterogeneously associated with deficits in early auditory processing (EAP), as demonstrated using clinically applicable tasks such as tone-matching task (TMT). Here, we pooled TMT performances across 310 SZ individuals and 219 healthy controls (HC), along with clinical, cognitive, and resting-state functional-connectivity MRI (rsFC-MRI) measures. In addition, TMT was measured in a group of 24 patients at symptomatic clinical high risk (CHR) for SZ and 24 age-matched HC (age range 7-27 years). We provide the first demonstration that the EAP deficits are bimodally distributed across SZ subjects (P < 0.0001 vs. unimodal distribution), with one group showing entirely unimpaired TMT performance (SZ-EAP+), and a second showing an extremely large TMT impairment (SZ-EAP-), relative to both controls (d = 2.1) and SZ-EAP+ patients (d = 3.4). The SZ-EAP- group predominated among samples drawn from inpatient sites, showed higher levels of cognitive symptoms (PANSS), worse social cognition and a differential deficit in neurocognition (MATRICS battery), and reduced functional capacity. rsFC-MRI analyses showed significant reduction in SZ-EAP- relative to controls between subcortical and cortical auditory regions. As opposed to SZ, CHR patients showed intact EAP function. In HC age-matched to CHR, EAP ability was shown to increase across the age range of vulnerability preceding SZ onset. These results indicate that EAP measure segregates between discrete SZ subgroups. As TMT can be readily implemented within routine clinical settings, its use may be critical to account for the heterogeneity of clinical outcomes currently observed across SZ patients, as well as for pre-clinical detection and efficacious treatment selection.

Motor clusters reveal differences in risk for psychosis, cognitive functioning, and thalamocortical connectivity: evidence for vulnerability subtypes

Abnormal development of parallel cortical-striatal networks may contribute to abnormal motor, cognitive, and affective behavior prior to the onset of psychosis. Partitioning individuals at clinical high-risk (CHR) using motor behavior may provide a novel perspective on different etiological pathways or patient subtypes. A K-means cluster analysis was conducted in CHR (N=69; 42% female, mean age=18.67 years) young adults using theoretically distinct measures of motor behavior. The resulting subtypes were then compared on positive and negative symptoms at baseline, and 2-year risk of psychosis conversion. CHR participants were followed for 2 years to determine conversion to psychosis. CHR subtypes and healthy controls (N=61; 57% female, mean age=18.58 years) were compared on multiple cognitive domains and cortical-striatal connectivity. Results suggest 3 vulnerability subtypes of CHR individuals with different profiles of motor performance, symptoms, risk for conversion to psychosis, cognition, and thalamocortical connectivity. This approach may reflect a novel strategy for promoting tailored risk assessment as well as future research developing individualized medicine.

Single-subject classification of schizophrenia using event-related potentials obtained during auditory and visual oddball paradigms

In the search for the biomarkers of schizophrenia, event-related potential (ERP) deficits obtained by applying the classic oddball paradigm are among the most consistent findings. However, the single-subject classification rate based on these parameters remains to be determined. Here, we present a data-driven approach by applying machine learning classifiers to relevant oddball ERPs. Twenty-four schizophrenic patients and 24 matched healthy controls finished auditory and visual oddball tasks while high-density electrophysiological recordings were applied. The N1 component in response to standards and target as well as the P3 component following targets were submitted to different machine learning algorithms and the resulting ERP features were submitted to further correlation analyses. We obtained a classification accuracy of 72.4 % using only two ERP components. Latencies of parietal N1 components to visual standard stimuli at electrode positions Pz and P1 were sufficient for classification. Further analysis revealed a high correlation of these features in controls and an intermediate correlation in schizophrenia patients. These data exemplarily show how automated inference may be applied to classify a pathological state in single subjects without prior knowledge of their diagnoses and illustrate the potential of machine learning algorithms for the identification of potential biomarkers. Moreover, this approach assesses the discriminative accuracy of one of the most consistent findings in schizophrenia research by means of single-subject classification.

Single-subject classification of schizophrenia by event-related potentials during selective attention

BACKGROUND

Executive dysfunction has repeatedly been proposed as a robust and promising substrate of analytical approaches in the research of neurocognitive markers of schizophrenia. Here, we present a mixed model- and data-driven classification approach by applying a task that targets executive dysfunction in schizophrenia and by investigating relevant event-related potential (ERP) features with machine learning classifiers.

METHODS

Forty schizophrenic patients and forty matched healthy controls completed the Attention Network Test while an electroencephalogram was recorded. Target-locked N1 and P3 ERP components were constructed and submitted to different classification analyses without a priori hypotheses. Standardized source localization was applied to estimate neural sources of N1 and P3 deficits in schizophrenia.

RESULTS

We obtained a classification accuracy of 79% using only very few ERP components. Central P3 components following compatible and incompatible trials and right parietal N1 latencies averaged across targets and were sufficient for classification. P3 deficits were associated with anterior cingulate cortex dysfunction, while right posterior current density deficits were observed in schizophrenia during the N1 time frame.

CONCLUSIONS

The data exemplarily show how automated inference may be applied to classify a pathological state in single subjects without prior knowledge of their diagnoses. While classification accuracy may be optimized by application of other executive paradigms, this approach illustrates the potential of machine learning algorithms for the identification of biomarkers that are independent of clinical assessments. Conversely, data suggest a pathophysiological mechanism that includes early visual and late executive deficits during response inhibition in schizophrenia.

A Hybrid Machine Learning Method for Fusing fMRI and Genetic Data: Combining both Improves Classification of Schizophrenia

We demonstrate a hybrid machine learning method to classify schizophrenia patients and healthy controls, using functional magnetic resonance imaging (fMRI) and single nucleotide polymorphism (SNP) data. The method consists of four stages: (1) SNPs with the most discriminating information between the healthy controls and schizophrenia patients are selected to construct a support vector machine ensemble (SNP-SVME). (2) Voxels in the fMRI map contributing to classification are selected to build another SVME (Voxel-SVME). (3) Components of fMRI activation obtained with independent component analysis (ICA) are used to construct a single SVM classifier (ICA-SVMC). (4) The above three models are combined into a single module using a majority voting approach to make a final decision (Combined SNP-fMRI). The method was evaluated by a fully validated leave-one-out method using 40 subjects (20 patients and 20 controls). The classification accuracy was: 0.74 for SNP-SVME, 0.82 for Voxel-SVME, 0.83 for ICA-SVMC, and 0.87 for Combined SNP-fMRI. Experimental results show that better classification accuracy was achieved by combining genetic and fMRI data than using either alone, indicating that genetic and brain function representing different, but partially complementary aspects, of schizophrenia etiopathology. This study suggests an effective way to reassess biological classification of individuals with schizophrenia, which is also potentially useful for identifying diagnostically important markers for the disorder.

Anhedonia as a phenotype for the Val158Met COMT polymorphism in relatives of patients with schizophrenia

The Val(158)Met polymorphism of the catechol-O-methyltransferase (COMT) gene has been associated with aspects of schizophrenia that are possibly related to the disorder's pathogenesis. The present study investigated the Val(158)Met polymorphism in relation to anhedonia--a construct central to negative schizotypy. Anhedonia and other schizotypal characteristics were assessed in relatives of patients with schizophrenia, relatives of patients with bipolar disorder, and nonpsychiatric controls using the Chapman schizotypy scales and the Schizotypal Personality Questionnaire. Compared with controls, relatives of individuals with schizophrenia had elevated scores on Chapman scales for social anhedonia and physical anhedonia, while relatives of patients with bipolar disorder exhibited only increased scores on the Social Anhedonia Scale. As a group, relatives of patients with schizophrenia who were homozygous for the val allele of the COMT polymorphism showed the highest elevations in self-reported social and physical anhedonia. Associations with the COMT polymorphism were absent in relatives of patients with bipolar disorder and control participants. Findings suggest that anhedonia is manifest in individuals who carry genetic liability for schizophrenia and is associated with the Val(158)Met polymorphism of the COMT gene.

Multivariate voxel-based morphometry successfully differentiates schizophrenia patients from healthy controls

Currently available laboratory procedures might provide additional information to psychiatric diagnostic systems for more valid classifications of mental disorders. To identify the correlative pattern of gray matter distribution that best discriminates schizophrenia patients from healthy subjects, we applied discriminant function analysis techniques using the multivariate linear model and the voxel-based morphometry. The first analysis was conducted to obtain a statistical model that classified 30 male healthy subjects and 30 male schizophrenia patients diagnosed according to current operational criteria. The second analysis was performed to prospectively validate the statistical model by successfully classifying a new cohort that consisted of 16 male healthy subjects and 16 male schizophrenia patients. Inferences about the structural relevance of the gray matter distribution could be made if the individual profile of pattern expression could be linked to the specific diagnosis of each subject. The result was that 90% of the subjects were correctly classified by the eigenimage, and the Jackknife approach revealed well above chance accuracy. The pattern of the eigenimage was characterized by positive loadings indicating gray matter decline in the patients in the lateral and medial prefrontal regions, insula, lateral temporal regions, medial temporal structures, and thalamus as well as the negative loadings reflecting gray matter increase in the patients in the putamen and cerebellum. When the eigenimage derived from the original cohort was applied to classify data from the second cohort, it correctly assigned more than 80% of the healthy subjects and schizophrenia patients. These findings suggest that the characteristic distribution of gray matter changes may be of diagnostic value for schizophrenia.

Subtyping schizophrenia: implications for genetic research

Phenotypic variability and likely extensive genetic heterogeneity have been confounding the search for the causes of schizophrenia since the inception of the diagnostic category. The inconsistent results of genetic linkage and association studies using the diagnostic category as the sole schizophrenia phenotype suggest that the current broad concept of schizophrenia does not demarcate a homogeneous disease entity. Approaches involving subtyping and stratification by covariates to reduce heterogeneity have been successful in the genetic study of other complex disorders, but rarely applied in schizophrenia research. This article reviews past and present attempts at delineating schizophrenia subtypes based on clinical features, statistically derived measures, putative genetic indicators, and intermediate phenotypes, highlighting the potential utility of multidomain neurocognitive endophenotypes.

Sensitivity and specificity of select biological indices in characterizing psychotic patients and their relatives

BACKGROUND

Although studies have detailed biological abnormalities in schizophrenia patients and their first-degree biological relatives, few studies have directly compared the utility of biological indices in these individuals.

METHODS

Measures of global smooth-pursuit ocular motor (OM) function, low frequency and alpha band electroencephalogram (EEG) power, and nonspecific fluctuations (NSF) in electrodermal activity and visibility of the plexus in the nailfold were collected from 136 schizophrenia patients and 67 of their first-degree biological relatives, 71 affective disorder psychotic patients and 68 of their first-degree biological relatives, and 169 nonpsychiatric comparison subjects. We conducted receiver operator characteristic (ROC) analyses to determine how well each index differentiated the patient groups and the groups of first-degree relatives.

RESULTS

Smooth-pursuit ocular motor function, low frequency and alpha band EEG power, and nailfold plexus visibility differentiated schizophrenia patients from nonpsychiatric comparison subjects. Nailfold plexus visibility was the only measure that significantly differentiated schizophrenia patients from both nonpsychiatric controls and affective patients. Smooth-pursuit ocular motor function and the number of electrodermal nonspecific fluctuations differentiated relatives of schizophrenia patients from nonpsychiatric comparison subjects.

CONCLUSION

Increased nailfold plexus visibility may mark a process associated with abnormal brain development leading to schizophrenia. Smooth-pursuit dysfunction may mark genetic vulnerability that is relatively specific to schizophrenia.