Neural Correlates of Schizotypal Personality Disorder: a Systematic Review of Neuroimaging and EEG Studies

Optimizing graph neural network architectures for schizophrenia spectrum disorder prediction using evolutionary algorithms

BACKGROUND AND OBJECTIVE

The accurate diagnosis of schizophrenia spectrum disorder plays an important role in improving patient outcomes, enabling timely interventions, and optimizing treatment plans. Functional connectivity analysis, utilizing functional magnetic resonance imaging data, has been demonstrated to offer invaluable biomarkers conducive to clinical diagnosis. However, previous studies mainly focus on traditional machine learning methods or hand-crafted neural networks, which may not fully capture the spatial topological relationship between brain regions.

METHODS

This paper proposes an evolutionary algorithm (EA) based graph neural architecture search (GNAS) method. EA-GNAS has the ability to search for high-performance graph neural networks for schizophrenia spectrum disorder prediction. Moreover, we adopt GNNExplainer to investigate the explainability of the acquired architectures, ensuring that the model's predictions are both accurate and comprehensible.

RESULTS

The results suggest that the graph neural network model, derived using genetic algorithm search, outperforms under five-fold cross-validation, achieving a fitness of 0.1850. Relative to conventional machine learning and other deep learning approaches, the proposed method yields superior accuracy, F1 score, and AUC values of 0.8246, 0.8438, and 0.8258, respectively.

CONCLUSION

Based on a multi-site dataset from schizophrenia spectrum disorder patients, the findings reveal an enhancement over prior methods, advancing our comprehension of brain function and potentially offering a biomarker for diagnosing schizophrenia spectrum disorder.

Cognitive Processes and Resting-State Functional Neuroimaging Findings in High Schizotypal Individuals and Schizotypal Personality Disorder Patients: A Systematic Review

Ample research findings indicate that there is altered brain functioning in the schizophrenia spectrum. Nevertheless, functional neuroimaging findings remain ambiguous for healthy individuals expressing high schizotypal traits and patients with schizotypal personality disorder (SPD). The purpose of this systematic review was to identify patterns of task-related and resting-state neural abnormalities across these conditions. MEDLINE-PubMed and PsycINFO were systematically searched and forty-eight studies were selected. Forty studies assessed healthy individuals with high schizotypal traits and eight studies examined SPD patients with functional neuroimaging techniques (fNIRS; fMRI; Resting-state fMRI). Functional alterations in striatal, frontal and temporal regions were found in healthy individuals with high schizotypal traits. Schizotypal personality disorder was associated with default mode network abnormalities but further research is required in order to better conceive its neural correlates. There was also evidence for functional compensatory mechanisms associated with both conditions. To conclude, the findings suggest that brain dysfunctions are evident in individuals who lie along the subclinical part of the spectrum, further supporting the continuum model for schizophrenia susceptibility. Additional research is required in order to delineate the counterbalancing processes implicated in the schizophrenia spectrum, as this approach will provide promising insights for both conversion and protection from conversion into schizophrenia.

Tensor factorization approach for ERP-based assessment of schizotypy in a novel auditory oddball task on perceived family stress

Objective. Schizotypy, a potential phenotype for schizophrenia, is a personality trait that depicts psychosis-like signs in the normal range of psychosis continuum. Family communication may affect the social functioning of people with schizotypy. Greater family stress, such as irritability, criticism and less praise, is perceived at a higher level of schizotypy. This study aims to determine the differences between people with high and low levels of schizotypy using electroencephalography (EEG) during criticism, praise and neutral comments. EEGs were recorded from 29 participants in the general community who varied from low schizotypy to high schizotypy (HS) during a novel emotional auditory oddball task.Approach. We consider the difference in event-related potential parameters, namely the amplitude and latency of P300 subcomponents (P3a and P3b), between pairs of target words (standard, positive, negative and neutral). A model based on tensor factorization is then proposed to detect these components from the EEG using the CANDECOMP/PARAFAC decomposition technique. Finally, we employ the mutual information estimation method to select influential features for classification.Main results.The highest classification accuracy, sensitivity, and specificity of 93.1%, 94.73%, and 90% are obtained via leave-one-out cross validation.Significance. This is the first attempt to investigate the identification of individuals with psychometrically-defined HS from brain responses that are specifically associated with perceiving family stress and schizotypy. By measuring these brain responses to social stress, we achieve the goal of improving the accuracy in detection of early episodes of psychosis.

The Influence of Filters on EEG-ERP Testing: Analysis of Motor Cortex in Healthy Subjects

The raw EEG signal is always contaminated with many different artifacts, such as muscle movements (electromyographic artifacts), eye blinking (electrooculographic artifacts) or power line disturbances. All artifacts must be removed for correct data interpretation. However, various noise reduction methods significantly influence the final shape of the EEG signal and thus its characteristic values, latency and amplitude. There are several types of filters to eliminate noise early in the processing of EEG data. However, there is no gold standard for their use. This article aims to verify and compare the influence of four various filters (FIR, IIR, FFT, NOTCH) on the latency and amplitude of the EEG signal. By presenting a comparison of selected filters, the authors intend to raise awareness among researchers as regards the effects of known filters on latency and amplitude in a selected area-the sensorimotor area.