Single-trial analysis of auditory evoked potentials improves separation of normal and schizophrenia subjects

Is auditory processing measured by the N100 an endophenotype for psychosis? A family study and a meta-analysis

BACKGROUND

The N100, an early auditory event-related potential, has been found to be altered in patients with psychosis. However, it is unclear if the N100 is a psychosis endophenotype that is also altered in the relatives of patients.

METHODS

We conducted a family study using the auditory oddball paradigm to compare the N100 amplitude and latency across 243 patients with psychosis, 86 unaffected relatives, and 194 controls. We then conducted a systematic review and a random-effects meta-analysis pooling our results and 14 previously published family studies. We compared data from a total of 999 patients, 1192 relatives, and 1253 controls in order to investigate the evidence and degree of N100 differences.

RESULTS

In our family study, patients showed reduced N100 amplitudes and prolonged N100 latencies compared to controls, but no significant differences were found between unaffected relatives and controls. The meta-analysis revealed a significant reduction of the N100 amplitude and delay of the N100 latency in both patients with psychosis (standardized mean difference [s.m.d.] = -0.48 for N100 amplitude and s.m.d. = 0.43 for N100 latency) and their relatives (s.m.d. = - 0.19 for N100 amplitude and s.m.d. = 0.33 for N100 latency). However, only the N100 latency changes in relatives remained significant when excluding studies with affected relatives.

CONCLUSIONS

N100 changes, especially prolonged N100 latencies, are present in both patients with psychosis and their relatives, making the N100 a promising endophenotype for psychosis. Such changes in the N100 may reflect changes in early auditory processing underlying the etiology of psychosis.

Machine Learning Techniques for the Diagnosis of Schizophrenia Based on Event-Related Potentials

Antecedent

The event-related potential (ERP) components P300 and mismatch negativity (MMN) have been linked to cognitive deficits in patients with schizophrenia. The diagnosis of schizophrenia could be improved by applying machine learning procedures to these objective neurophysiological biomarkers. Several studies have attempted to achieve this goal, but no study has examined Multiple Kernel Learning (MKL) classifiers. This algorithm finds optimally a combination of kernel functions, integrating them in a meaningful manner, and thus could improve diagnosis.

Objective

This study aimed to examine the efficacy of the MKL classifier and the Boruta feature selection method for schizophrenia patients (SZ) and healthy controls (HC) single-subject classification.

Methods

A cohort of 54 SZ and 54 HC participants were studied. Three sets of features related to ERP signals were calculated as follows: peak related features, peak to peak related features, and signal related features. The Boruta algorithm was used to evaluate the impact of feature selection on classification performance. An MKL algorithm was applied to address schizophrenia detection.

Results

A classification accuracy of 83% using the whole dataset, and 86% after applying Boruta feature selection was obtained. The variables that contributed most to the classification were mainly related to the latency and amplitude of the auditory P300 paradigm.

Conclusion

This study showed that MKL can be useful in distinguishing between schizophrenic patients and controls when using ERP measures. Moreover, the use of the Boruta algorithm provides an improvement in classification accuracy and computational cost.

A hybrid decision support system for automatic detection of Schizophrenia using EEG signals

BACKGROUND

Schizophrenia (SCZ) is a serious neurological condition in which people suffer with distorted perception of reality. SCZ may result in a combination of delusions, hallucinations, disordered thinking, and behavior. This causes permanent disability and hampers routine functioning. Trained neurologists use interviewing and visual inspection techniques for the detection and diagnosis of SCZ. These techniques are manual, time-consuming, subjective, and error-prone. Therefore, there is a need to develop an automatic model for SCZ classification. The aim of this study is to develop an automated SCZ classification model using electroencephalogram (EEG) signals. The EEG signals can capture the changes in neural dynamics of human cognition during SCZ.

METHOD

Based on the nature of the SCZ condition, the EEG signals must be examined. For accurate interpretation of EEG signals during SCZ, an automated model integrating a robust variational mode decomposition (RVMD) and an optimized extreme learning machine (OELM) classifier is developed. Traditional VMD suffers from noisy mode generation, mode duplication, under segmentation, and mode discarding. These problems are suppressed in RVMD by automating the selection of quadratic penalty factor (α) and a number of modes (L). The hyperparameters (HPM) of the OELM classifier are automatically selected to ensure maximum accuracy for each mode without overfitting or underfitting. For the selection of α and L in RVMD and HPM in the OELM classifier, a whale optimization algorithm is used. The root mean square error is minimized for RVMD and classification accuracy of each mode is maximized for the OELM classifier. The EEG signals of three conditions performing basic sensory tasks have been analyzed to detect SCZ.

RESULTS

The Kruskal Wallis test is used to select different features extracted from the modes produced by RVMD. An OELM classifier is tested using a ten-fold cross-validation technique. An accuracy, precision, specificity, F-1 measure, sensitivity, and Cohen's Kappa of 92.93%, 93.94%, 91.06% 94.07%, 97.15%, and 85.32% are obtained.

CONCLUSION

The third mode's chaotic features helped to capture the significant changes that occurred during the SCZ state. The findings of the RVMD-OELM-based hybrid decision support system can help neuro-experts for the accurate identification of SCZ in real-time scenarios.

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.

Decomposing P300 into correlates of genetic risk and current symptoms in schizophrenia: An inter-trial variability analysis

BACKGROUND

The P300 event-related potential (ERP) component, which reflects cognitive processing, is a candidate biomarker for schizophrenia. However, the role of P300 in the pathophysiology of schizophrenia remains unclear because averaged P300 amplitudes reflect both genetic predisposition and current clinical status. Thus, we sought to identify which aspects of P300 are associated with genetic risk versus symptomatic status via an inter-trial variability analysis.

METHODS

Auditory P300, clinical symptoms, and neurocognitive function assessments were obtained from forty-five patients with schizophrenia, thirty-two subjects at genetic high risk (GHR), thirty-two subjects at clinical high risk (CHR), and fifty-two healthy control (HC) participants. Both conventional averaging and inter-trial variability analyses were conducted for P300, and results were compared across groups using analysis of variance (ANOVA). Pearson's correlation was utilized to determine associations among inter-trial variability for P300, current symptoms and neurocognitive status.

RESULTS

Average P300 amplitude was reduced in the GHR, CHR, and schizophrenia groups compared with that in the HC group. P300 inter-trial variability was elevated in the CHR and schizophrenia groups but relatively normal in the GHR and HC groups. Furthermore, P300 inter-trial variability was significantly related to negative symptom severity and neurocognitive performance results in schizophrenia patients.

CONCLUSIONS

These results suggest that P300 amplitude is an endophenotype for schizophrenia and that greater inter-trial variability of P300 is associated with more severe negative and cognitive symptoms in schizophrenia patients.

The Dissociation between Polarity, Semantic Orientation, and Emotional Tone as an Early Indicator of Cognitive Impairment

The present study aims to identify early cognitive impairment through the efficient use of therapies that can improve the quality of daily life and prevent disease progress. We propose a methodology based on the hypothesis that the dissociation between oral semantic expression and the physical expressions, facial gestures, or emotions transmitted in a person's tone of voice is a possible indicator of cognitive impairment. Experiments were carried out with phrases, analyzing the semantics of the message, and the tone of the voice of patients through unstructured interviews in healthy people and patients at an early Alzheimer's stage. The results show that the dissociation in cognitive impairment was an effective indicator, arising from patterns of inconsistency between the analyzed elements. Although the results of our study are encouraging, we believe that further studies are necessary to confirm that this dissociation is a probable indicator of cognitive impairment.

Single-subject classification of schizophrenia patients based on a combination of oddball and mismatch evoked potential paradigms

OBJECTIVE

The diagnostic process for schizophrenia is mainly clinical and has to be performed by an experienced psychiatrist, relying primarily on clinical signs and symptoms. Current neurophysiological measurements can distinguish groups of healthy controls and groups of schizophrenia patients. Individual classification based on neurophysiological measurements mostly shows moderate accuracy. We wanted to examine whether it is possible to distinguish controls and patients individually with a good accuracy. To this end we used a combination of features extracted from the auditory and visual P300 paradigms and the mismatch negativity paradigm.

METHODS

We selected 54 patients and 54 controls, matched for age and gender, from the data available at the UPC Kortenberg. The EEG-data were high- and low-pass filtered, epoched and averaged. Features (latencies and amplitudes of component peaks) were extracted from the averaged signals. The resulting dataset was used to train and test classification algorithms. First on separate paradigms and then on all combinations, we applied Naïve Bayes, Support Vector Machine and Decision Tree, with two of its improvements: Adaboost and Random Forest.

RESULTS

For at least two classifiers the performance increased significantly by combining paradigms compared to single paradigms. The classification accuracy increased from at best 79.8% when trained on features from single paradigms, to 84.7% when trained on features from all three paradigms.

CONCLUSION

A combination of features originating from three evoked potential paradigms allowed us to accurately classify individual subjects as either control or patient. Classification accuracy was mostly above 80% for the machine learners evaluated in this study and close to 85% at best.

Cognitive and neuroplasticity mechanisms by which congenital or early blindness may confer a protective effect against schizophrenia

Several authors have noted that there are no reported cases of people with schizophrenia who were born blind or who developed blindness shortly after birth, suggesting that congenital or early (C/E) blindness may serve as a protective factor against schizophrenia. By what mechanisms might this effect operate? Here, we hypothesize that C/E blindness offers protection by strengthening cognitive functions whose impairment characterizes schizophrenia, and by constraining cognitive processes that exhibit excessive flexibility in schizophrenia. After briefly summarizing evidence that schizophrenia is fundamentally a cognitive disorder, we review areas of perceptual and cognitive function that are both impaired in the illness and augmented in C/E blindness, as compared to healthy sighted individuals. We next discuss: (1) the role of neuroplasticity in driving these cognitive changes in C/E blindness; (2) evidence that C/E blindness does not confer protective effects against other mental disorders; and (3) evidence that other forms of C/E sensory loss (e.g., deafness) do not reduce the risk of schizophrenia. We conclude by discussing implications of these data for designing cognitive training interventions to reduce schizophrenia-related cognitive impairment, and perhaps to reduce the likelihood of the development of the disorder itself.

Consistency of the auditory evoked response: the presence of aberrant responses and their effect on N100 localization

The structure and distribution of the sources underlying the generation of evoked potentials (EPs) is often very complex. In an effort to improve localization accuracy of the auditory N100 (negative response occurring around 100ms poststimulus) component, we analyzed 13 datasets of single-trial EPs obtained from normal subjects using an iterative independent component analysis procedure which allowed us to detect a clear N100 component in each single trial and to study gross changes in component morphology across trials. We found that single-trial N100 amplitude was most often negative in polarity, as expected, but occasionally exhibited a marked reversal to become positive. The average N100, however, showed the typical negative polarity, in all subjects. Based on this observation, we separated the processed single trials in two groups of typical and aberrant responses, and from each group, we computed a partial EP that was used to localize the underlying intracranial sources. Additionally, we localized the classical ensemble average EP. Before processing, the N100 sources were identified correctly in the primary auditory cortex in only four datasets, while after processing, all 13 datasets yielded correct localizations, and the confidence volume of the sources improved by about 80%. Further analysis demonstrated that in nine datasets the improvement was mostly due to the typical responses, while the aberrant responses had an antagonistic effect. Our results suggest that aberrant responses should not be included in source localizations, especially when EEG-based brain mapping is intended as a clinical tool.