Identification of Major Psychiatric Disorders From Resting-State Electroencephalography Using a Machine Learning Approach

EEG-based Signatures of Schizophrenia, Depression, and Aberrant Aging: A Supervised Machine Learning Investigation

BACKGROUND

Electroencephalography (EEG) is a noninvasive, cost-effective, and robust tool, which directly measures in vivo neuronal mass activity with high temporal resolution. Combined with state-of-the-art machine learning (ML) techniques, EEG recordings could potentially yield in silico biomarkers of severe mental disorders.

HYPOTHESIS

Pathological and physiological aging processes influence the electrophysiological signatures of schizophrenia (SCZ) and major depressive disorder (MDD).

STUDY DESIGN

From a single-center cohort (N = 735, 51.6% male) comprising healthy control individuals (HC, N = 245) and inpatients suffering from SCZ (N = 250) or MDD (N = 240), we acquired resting-state 19 channel-EEG recordings. Using repeated nested cross-validation, support vector machine models were trained to (1) classify patients with SCZ or MDD and HC individuals and (2) predict age in HC individuals. The age model was applied to patient groups to calculate Electrophysiological Age Gap Estimation (EphysAGE) as the difference between predicted and chronological age. The links between EphysAGE, diagnosis, and medication were then further explored.

STUDY RESULTS

The classification models robustly discriminated SCZ from HC (balanced accuracy, BAC = 72.7%, P < .001), MDD from HC (BAC = 67.0%, P < .001), and SCZ from MDD individuals (BAC = 63.2%, P < .001). Notably, central alpha (8-11 Hz) power decrease was the most consistently predictive feature for SCZ and MDD. Higher EphysAGE was associated with an increased likelihood of being misclassified as SCZ in HC and MDD (ρHC = 0.23, P < .001; ρMDD = 0.17, P = .01).

CONCLUSIONS

ML models can extract electrophysiological signatures of MDD and SCZ for potential clinical use. However, the impact of aging processes on diagnostic separability calls for timely application of such models, possibly in early recognition settings.

Multichannel convolutional transformer for detecting mental disorders using electroancephalogrpahy records

Mental disorders represent a critical global health challenge that affects millions around the world and significantly disrupts daily life. Early and accurate detection is paramount for timely intervention, which can lead to improved treatment outcomes. Electroencephalography (EEG) provides the non-invasive means for observing brain activity, making it a useful tool for detecting potential mental disorders. Recently, deep learning techniques have gained prominence for their ability to analyze complex datasets, such as electroencephalography recordings. In this study, we introduce a novel deep-learning architecture for the classification of mental disorders such as post-traumatic stress disorder, depression, or anxiety, using electroencephalography data. Our proposed model, the multichannel convolutional transformer, integrates the strengths of both convolutional neural networks and transformers. Before feeding the model as low-level features, the input is pre-processed using a common spatial pattern filter, a signal space projection filter, and a wavelet denoising filter. Then the EEG signals are transformed using continuous wavelet transform to obtain a time-frequency representation. The convolutional layers tokenize the input signals transformed by our pre-processing pipeline, while the Transformer encoder effectively captures long-range temporal dependencies across sequences. This architecture is specifically tailored to process EEG data that has been preprocessed using continuous wavelet transform, a technique that provides a time-frequency representation, thereby enhancing the extraction of relevant features for classification. We evaluated the performance of our proposed model on three datasets: the EEG Psychiatric Dataset, the MODMA dataset, and the EEG and Psychological Assessment dataset. Our model achieved classification accuracies of 87.40% on the EEG and Psychological Assessment dataset, 89.84% on the MODMA dataset, and 92.28% on the EEG Psychiatric dataset. Our approach outperforms every concurrent approaches on the datasets we used, without showing any sign of over-fitting. These results underscore the potential of our proposed architecture in delivering accurate and reliable mental disorder detection through EEG analysis, paving the way for advancements in early diagnosis and treatment strategies.

Decentralized EEG-based detection of major depressive disorder via transformer architectures and split learning

Introduction

Major Depressive Disorder (MDD) remains a critical mental health concern, necessitating accurate detection. Traditional approaches to diagnosing MDD often rely on manual Electroencephalography (EEG) analysis to identify potential disorders. However, the inherent complexity of EEG signals along with the human error in interpreting these readings requires the need for more reliable, automated methods of detection.

Methods

This study utilizes EEG signals to classify MDD and healthy individuals through a combination of machine learning, deep learning, and split learning approaches. State of the art machine learning models i.e., Random Forest, Support Vector Machine, and Gradient Boosting are utilized, while deep learning models such as Transformers and Autoencoders are selected for their robust feature-extraction capabilities. Traditional methods for training machine learning and deep learning models raises data privacy concerns and require significant computational resources. To address these issues, the study applies a split learning framework. In this framework, an ensemble learning technique has been utilized that combines the best performing machine and deep learning models.

Results

Results demonstrate a commendable classification performance with certain ensemble methods, and a Transformer-Random Forest combination achieved 99% accuracy. In addition, to address data-sharing constraints, a split learning framework is implemented across three clients, yielding high accuracy (over 95%) while preserving privacy. The best client recorded 96.23% accuracy, underscoring the robustness of combining Transformers with Random Forest under resource-constrained conditions.

Discussion

These findings demonstrate that distributed deep learning pipelines can deliver precise MDD detection from EEG data without compromising data security. Proposed framework keeps data on local nodes and only exchanges intermediate representations. This approach meets institutional privacy requirements while providing robust classification outcomes.

Quantitative EEG Insights Into A Hundred Adult ADHD Patients: A Deep Dive Into Test of Variables of Attention (TOVA) Correlations and Attention Dynamics

OBJECTIVE

This study aimed to enhance the diagnostic accuracy of attention-deficit/hyperactivity disorder (ADHD) by integrating quantitative electroencephalography (qEEG) power bands with the test of variables of attention (TOVA) and self-reported psychiatric symptoms. We examined the relationship between TOVA scores, qEEG findings-particularly the theta-beta ratio-and comorbid psychiatric conditions to assess their role in refining ADHD diagnoses according to the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-V).

METHOD

A total of 100 participants were assessed using TOVA, qEEG, and psychological scales, including the Beck Anxiety Inventory (BAI), Beck Depression Inventory (BDI), Maudsley Obsessive-Compulsive Inventory (MOCI), and the Mood Disorder Questionnaire (MDQ). Participants were categorized into groups based on their Attention Comparison Scores (ACS) above or below the zero threshold. Mann-Whitney U-tests, correlation analyses, and predictive modeling using automatic linear modeling (ALM) were conducted to evaluate group differences, age-related changes, and predictor variables for attention performance.

RESULTS

All participants met the diagnostic criteria for ADHD. Among them, 37% exhibited anxiety, 60% depression, 26% obsessive-compulsive, and 35% mood disorder symptoms. The group with ACS above zero was significantly older (p = 0.034) and performed better on all Tests of Variables of Attention (TOVA) measures (p < 0.05). Age negatively correlated with attention scores (r = -0.371, p < 0.001), response time variability (r = -0.241, p = 0.016), and response time (r = -0.311, p = 0.002). qEEG showed significant age-related changes in theta-to-beta and delta-to-beta ratios (p < 0.005). TOVA and qEEG ratios, particularly beta and delta activity, predicted attention and response time variability, with adjusted R2 values between 71.5% and 87.1%.

CONCLUSION

The findings highlight that attention performance in ADHD is shaped by age, neuropsychological factors, and qEEG-measured brain activity. Higher attention scores correlate with better TOVA results, particularly in response time and error rates. Age-related declines in attention align with reductions in theta-to-beta and delta-to-beta ratios. Predictive modeling underscores the value of combining TOVA and qEEG to identify key predictors like response time variability, omission errors, and specific beta and delta activity. This integration enhances the evaluation of attention deficits and brain dynamics, benefiting both clinical and research applications.

Zipper Pattern: An Investigation into Psychotic Criminal Detection Using EEG Signals

Background: Electroencephalography (EEG) signal-based machine learning models are among the most cost-effective methods for information retrieval. In this context, we aimed to investigate the cortical activities of psychotic criminal subjects by deploying an explainable feature engineering (XFE) model using an EEG psychotic criminal dataset. Methods: In this study, a new EEG psychotic criminal dataset was curated, containing EEG signals from psychotic criminal and control groups. To extract meaningful findings from this dataset, we presented a new channel-based feature extraction function named Zipper Pattern (ZPat). The proposed ZPat extracts features by analyzing the relationships between channels. In the feature selection phase of the proposed XFE model, an iterative neighborhood component analysis (INCA) feature selector was used to choose the most distinctive features. In the classification phase, we employed an ensemble and iterative distance-based classifier to achieve high classification performance. Therefore, a t-algorithm-based k-nearest neighbors (tkNN) classifier was used to obtain classification results. The Directed Lobish (DLob) symbolic language was used to derive interpretable results from the identities of the selected feature vectors in the final phase of the proposed ZPat-based XFE model. Results: To obtain the classification results from the ZPat-based XFE model, leave-one-record-out (LORO) and 10-fold cross-validation (CV) methods were used. The proposed ZPat-based model achieved over 95% classification accuracy on the curated EEG psychotic criminal dataset. Moreover, a cortical connectome diagram related to psychotic criminal detection was created using a DLob-based explainable artificial intelligence (XAI) method. Conclusions: In this regard, the proposed ZPat-based XFE model achieved both high classification performance and interpretability. Thus, the model contributes to feature engineering, psychiatry, neuroscience, and forensic sciences. Moreover, the presented ZPat-based XFE model is one of the pioneering XAI models for investigating psychotic criminal/criminal individuals.

Abnormal nonlinear features of EEG microstate sequence in obsessive-compulsive disorder

BACKGROUND

At present, only a few studies have explored electroencephalography (EEG) microstates of patients with obsessive-compulsive disorder (OCD) and the results are inconsistent. Additionally, the nonlinear features of EEG microstate sequences contain rich information about the brain, yet how the nonlinear features of EEG microstate sequences abnormally change in patients with OCD is still unknown.

METHODS

Resting-state EEG data were collected from 48 OCD patients and macheted 48 healthy controls (HC). Subsequently, EEG microstate analysis was used to extract the microstate temporal parameters (duration, occurrence, coverage) and nonlinear features of EEG microstate sequences (sample entropy, Lempel-Ziv complexity, Hurst index). Finally, the temporal parameters and nonlinear features of EEG microstate sequences were sent to three kinds of machine learning models to classify OCD patients.

RESULTS

Both groups obtained four typical EEG microstate topographies. The duration of microstates A, B, and C in OCD patients decreased significantly, while the occurrence of microstate D increased significantly compared to HC. Sample entropy and Lempel-Ziv complexity of microstate sequences in OCD patients increased significantly, while Hurst index decreased significantly compared to HC. The classification accuracy using the nonlinear features of microstate sequences reached up to 85%, significantly higher than that based on microstate temporal parameter models.

CONCLUSION

This study provides supplementary findings on EEG microstates in OCD patients with a larger sample size. We found that the nonlinear features of EEG microstate sequences in OCD patients can serve as potential electrophysiological biomarkers for distinguishing OCD patients.

Psychiatric disorders from EEG signals through deep learning models

Psychiatric disorders present diagnostic challenges due to individuals concealing their genuine emotions, and traditional methods relying on neurophysiological signals have limitations. Our study proposes an improved EEG-based diagnostic model employing Deep Learning (DL) techniques to address this. By experimenting with DL models on EEG data, we aimed to enhance psychiatric disorder diagnosis, offering promising implications for medical advancements. We utilized a dataset of 945 individuals, including 850 patients and 95 healthy subjects, focusing on six main and nine specific disorders. Quantitative EEG data were analyzed during resting states, featuring power spectral density (PSD) and functional connectivity (FC) across various frequency bands. Employing artificial neural networks (ANN), K nearest neighbors (KNN), Long short-term memory (LSTM), bidirectional Long short-term memory (Bi LSTM), and a hybrid CNN-LSTM model, we performed binary classification. Remarkably, all proposed models outperformed previous approaches, with the ANN achieving 96.83 % accuracy for obsessive-compulsive disorder using entire band features. CNN-LSTM attained the same accuracy for adjustment disorder, while KNN and LSTM achieved 98.94 % accuracy for acute stress disorder using specific feature sets. Notably, KNN and Bi-LSTM models reached 97.88 % accuracy for predicting obsessive-compulsive disorder. These findings underscore the potential of EEG as a cost-effective and accessible diagnostic tool for psychiatric disorders, complementing traditional methods like MRI. Our study's advanced DL models show promise in enhancing psychiatric disorder detection and monitoring, with significant implications for clinical application, inspiring hope for improved patient care and outcomes. The potential of EEG as a diagnostic tool for psychiatric disorders is substantial, as it can lead to improved patient care and outcomes in the field of psychiatry.

Revisiting the role of computational neuroimaging in the era of integrative neuroscience

Computational models have become integral to human neuroimaging research, providing both mechanistic insights and predictive tools for human cognition and behavior. However, concerns persist regarding the ecological validity of lab-based neuroimaging studies and whether their spatiotemporal resolution is not sufficient for capturing neural dynamics. This review aims to re-examine the utility of computational neuroimaging, particularly in light of the growing prominence of alternative neuroscientific methods and the growing emphasis on more naturalistic behaviors and paradigms. Specifically, we will explore how computational modeling can both enhance the analysis of high-dimensional imaging datasets and, conversely, how neuroimaging, in conjunction with other data modalities, can inform computational models through the lens of neurobiological plausibility. Collectively, this evidence suggests that neuroimaging remains critical for human neuroscience research, and when enhanced by computational models, imaging can serve an important role in bridging levels of analysis and understanding. We conclude by proposing key directions for future research, emphasizing the development of standardized paradigms and the integrative use of computational modeling across neuroimaging techniques.

Machine-Learning-Based Depression Detection Model from Electroencephalograph (EEG) Data Obtained by Consumer-Grade EEG Device

Background/Objectives: There have been attempts to detect depression using medical-grade electroencephalograph (EEG) data based on a machine learning approach. EEG has garnered interest as a method for assessing brainwaves by attaching electrodes to the scalp to obtain electrical activity in the brain. Recently, machine learning has been applied to the EEG data to detect depression, with encouraging results. Specifically, studies using medical-grade EEG data have shown that depression can be accurately detected. However, there is a need to expand the range of applications by achieving a score with machine learning using simpler consumer-grade brain wave sensors. At present, a sufficient score has not been achieved.; Methods: To improve the score of depression detection, we quantified various EEG indices to train models such as power spectrum, asymmetry, complexity, and functional connectivity. In addition, feature selection was performed to ensure that the model learns only promising EEG indices for depression detection. The feature selection methods were Light Gradient Boosting Machine (LightGBM) feature importance, mutual information, ReliefF and ElasticNet coefficients. The selected EEG indices were learned by the LightGBM model, which is reported to be as accurate as the latest deep learning models. In cross-validation, the independence of test and training data was ensured to avoid excessively calculated score; Results: The results showed that the Macro F1 score was 91.59%, suggesting that a consumer-grade EEG can detect depression. In addition, analysis of the EEG indices selected by feature selection indicated that the Macro F1 score was about 80% for single EEG indices such as differential entropy in the frequency band β and functional connectivity in the left frontal region in the frequency band 1-128 Hz; Conclusions: Although the data were obtained from a consumer-grade EEG, the results suggest that these EEG indices are promising for detection depression.

Multi-modal EEG NEO-FFI with Trained Attention Layer (MENTAL) for mental disorder prediction

Early detection and accurate diagnosis of mental disorders is difficult due to the complexity of the diagnostic process, resulting in conditions being left undiagnosed or misdiagnosed. Previous studies have demonstrated that features of Electroencephalogram (EEG) data, such as Power Spectral Density (PSD), are useful biomarkers for indicating the onset of various mental disorders. Existing models using EEG data are typically trained to distinguish between healthy and afflicted individuals, and they are unable to distinguish between individuals with different disorders. We propose MENTAL (Multi-modal EEG NEO-FFI with Trained Attention Layer) to predict an individual's mental state using both EEG and Neo-Five Factor Inventory (NEO-FFI) personality data. We include an attention layer that captures the interactions between personality traits and PSD features, and emphasizes the important PSD features. MENTAL features a Recurrent Neural Network (RNN) to model the temporal nature of EEG data. We train our model with the Two Decades Brainclinics Research Archive for Insights in Neuroscience (TDBRAIN) dataset, which consists of 1274 healthy and psychiatric individuals including over 30 different diagnoses. MENTAL is able to achieve 93.3% accuracy when trained to classify between healthy individuals and those with ADHD. When trained to identify individuals with ADHD from among 33 disorder classes, MENTAL improves accuracy from 70.5 to 81.7%. MENTAL also demonstrates over 20% improvement in predictive accuracy when trained for MDD prediction. For the multi-class classification task of three classes, MENTAL improves accuracy by 4%, and for five classes, by nearly 8%. MENTAL is the first multi-modal model that utilizes both EEG and NEO-FFI data for the task of mental disorder prediction. We are one of the first groups to utilize TDBRAIN for automated disorder classification. MENTAL is accessible and cost-effective, as EEG is more affordable than other neuroimaging methods such as MRI, and the NEO-FFI is a self- reported survey. Our model can lead to acceptance and support for individuals living with mental health challenges and improve quality of life in our society.

Disentangling periodic and aperiodic resting EEG correlates of personality

Previous studies of resting electroencephalography (EEG) correlates of personality traits have conflated periodic and aperiodic sources of EEG signals. Because these are associated with different underlying neural dynamics, disentangling them can avoid measurement confounds and clarify findings. In a large sample (n = 300), we investigated how disentangling these activities impacts findings related to two research programs within personality neuroscience. In Study 1 we examined associations between Extraversion and two putative markers of reward sensitivity-Left Frontal Alpha asymmetry (LFA) and Frontal-Posterior Theta (FPT). In Study 2 we used machine learning to predict personality trait scores from resting EEG. In both studies, power within each EEG frequency bin was quantified as both total power and separate contributions of periodic and aperiodic activity. In Study 1, total power LFA and FPT correlated negatively with Extraversion (r ∼ -0.14), but there was no relation when LFA and FPT were derived only from periodic activity. In Study 2, all Big Five traits could be decoded from periodic power (r ∼ 0.20), and Agreeableness could also be decoded from total power and from aperiodic indices. Taken together, these results show how separation of periodic and aperiodic activity in resting EEG may clarify findings in personality neuroscience. Disentangling these signals allows for more reliable findings relating to periodic EEG markers of personality, and highlights novel aperiodic markers to be explored in future research.

Integrating EEG and Ensemble Learning for Accurate Grading and Quantification of Generalized Anxiety Disorder: A Novel Diagnostic Approach

Current assessments for generalized anxiety disorder (GAD) are often subjective and do not rely on a standardized measure to evaluate the GAD across its severity levels. The lack of objective and multi-level quantitative diagnostic criteria poses as a significant challenge for individualized treatment strategies. To address this need, this study aims to establish a GAD grading and quantification diagnostic model by integrating an electroencephalogram (EEG) and ensemble learning. In this context, a total of 39 normal subjects and 80 GAD patients were recruited and divided into four groups: normal control, mild GAD, moderate GAD, and severe GAD. Ten minutes resting state EEG data were collected for every subject. Functional connectivity features were extracted from each EEG segment with different time windows. Then, ensemble learning was employed for GAD classification studies and brain mechanism analysis. Hence, the results showed that the Catboost model with a 10 s time window achieved an impressive 98.1% accuracy for four-level classification. Particularly, it was found that those functional connections situated between the frontal and temporal lobes were significantly more abundant than in other regions, with the beta rhythm being the most prominent. The analysis framework and findings of this study provide substantial evidence for the applications of artificial intelligence in the clinical diagnosis of GAD.

Enhancing generalized anxiety disorder diagnosis precision: MSTCNN model utilizing high-frequency EEG signals

Generalized Anxiety Disorder (GAD) is a prevalent mental disorder on the rise in modern society. It is crucial to achieve precise diagnosis of GAD for improving the treatments and averting exacerbation. Although a growing number of researchers beginning to explore the deep learning algorithms for detecting mental disorders, there is a dearth of reports concerning precise GAD diagnosis. This study proposes a multi-scale spatial-temporal local sequential and global parallel convolutional model, named MSTCNN, which designed to achieve highly accurate GAD diagnosis using high-frequency electroencephalogram (EEG) signals. To this end, 10-min resting EEG data were collected from 45 GAD patients and 36 healthy controls (HC). Various frequency bands were extracted from the EEG data as the inputs of the MSTCNN. The results demonstrate that the proposed MSTCNN, combined with the attention mechanism of Squeeze-and-Excitation Networks, achieves outstanding classification performance for GAD detection, with an accuracy of 99.48% within the 4-30 Hz EEG data, which is competitively related to state-of-art methods in terms of GAD classification. Furthermore, our research unveils an intriguing revelation regarding the pivotal role of high-frequency band in GAD diagnosis. As the frequency band increases, diagnostic accuracy improves. Notably, high-frequency EEG data ranging from 10-30 Hz exhibited an accuracy rate of 99.47%, paralleling the performance of the broader 4-30 Hz band. In summary, these findings move a step forward towards the practical application of automatic diagnosis of GAD and provide basic theory and technical support for the development of future clinical diagnosis system.

A comprehensive review for machine learning on neuroimaging in obsessive-compulsive disorder

Obsessive-compulsive disorder (OCD) is a common mental disease, which can exist as a separate disease or become one of the symptoms of other mental diseases. With the development of society, statistically, the incidence rate of obsessive-compulsive disorder has been increasing year by year. At present, in the diagnosis and treatment of OCD, The clinical performance of patients measured by scales is no longer the only quantitative indicator. Clinical workers and researchers are committed to using neuroimaging to explore the relationship between changes in patient neurological function and obsessive-compulsive disorder. Through machine learning and artificial learning, medical information in neuroimaging can be better displayed. In this article, we discuss recent advancements in artificial intelligence related to neuroimaging in the context of Obsessive-Compulsive Disorder.

Systematic review of machine learning in PTSD studies for automated diagnosis evaluation

Post-traumatic stress disorder (PTSD) is frequently underdiagnosed due to its clinical and biological heterogeneity. Worldwide, many people face barriers to accessing accurate and timely diagnoses. Machine learning (ML) techniques have been utilized for early assessments and outcome prediction to address these challenges. This paper aims to conduct a systematic review to investigate if ML is a promising approach for PTSD diagnosis. In this review, statistical methods were employed to synthesize the outcomes of the included research and provide guidance on critical considerations for ML task implementation. These included (a) selection of the most appropriate ML model for the available dataset, (b) identification of optimal ML features based on the chosen diagnostic method, (c) determination of appropriate sample size based on the distribution of the data, and (d) implementation of suitable validation tools to assess the performance of the selected ML models. We screened 3186 studies and included 41 articles based on eligibility criteria in the final synthesis. Here we report that the analysis of the included studies highlights the potential of artificial intelligence (AI) in PTSD diagnosis. However, implementing AI-based diagnostic systems in real clinical settings requires addressing several limitations, including appropriate regulation, ethical considerations, and protection of patient privacy.

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.

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.

Feature Signature Discovery for Autism Detection: An Automated Machine Learning Based Feature Ranking Framework

Autism spectrum disorder is the most used umbrella term for a myriad of neuro-degenerative/developmental conditions typified by inappropriate social behavior, lack of communication/comprehension skills, and restricted mental and emotional maturity. The intriguing factor of this disorder is attributed to the fact that it can be detected only by close monitoring of developmental milestones after childbirth. Moreover, the exact causes for the occurrence of this neurodevelopmental condition are still unknown. Besides, autism is prevalent across individuals irrespective of ethnicity, genetic/familial history, and economic/educational background. Although research suggests that autism is genetic in nature and early detection of this disorder can greatly enhance the independent lifestyle and societal adaptability of affected individuals, there is still a great dearth of information to support the statement of proven facts and figures. This research work places emphasis on the application of automated machine learning incorporated with feature ranking techniques to generate significant feature signatures for the early detection of autism. Publicly available datasets based on the Q-chat scores of individuals across diverse age groups-toddlers, children, adolescents, and adults have been employed in this study. A machine learning framework based on automated hyperparameter optimization is proposed in this work to rank the potential nonclinical markers for autism. Moreover, this study aimed at ranking the AutoML models based on Mathew's correlation coefficient and balanced accuracy via which nonclinical markers were identified from these datasets. Besides, the feature signatures and their significance in distinguishing between classes are being reported for the first time in autism detection. The proposed framework yielded ∼90% MCC and ∼95% balanced accuracy across all four age groups of autism datasets. Deep learning approaches have yielded a maximum of 92.7% accuracy on the same datasets but are limited in their ability to extract significant markers, have not reported on MCC for unbalanced data, and cannot adapt automatically to new data entries. However, AutoML approaches are more flexible, easier to implement, and provide automated optimization, thereby yielding the highest accuracy with minimal user intervention.

Developmental Manipulation-Induced Changes in Cognitive Functioning

Schizophrenia is a complex neurodevelopmental disorder with as-yet no identified cause. The use of animals has been critical to teasing apart the potential individual and intersecting roles of genetic and environmental risk factors in the development of schizophrenia. One way to recreate in animals the cognitive impairments seen in people with schizophrenia is to disrupt the prenatal or neonatal environment of laboratory rodent offspring. This approach can result in congruent perturbations in brain physiology, learning, memory, attention, and sensorimotor domains. Experimental designs utilizing such animal models have led to a greatly improved understanding of the biological mechanisms that could underlie the etiology and symptomology of schizophrenia, although there is still more to be discovered. The implementation of the Research and Domain Criterion (RDoC) has been critical in taking a more comprehensive approach to determining neural mechanisms underlying abnormal behavior in people with schizophrenia through its transdiagnostic approach toward targeting mechanisms rather than focusing on symptoms. Here, we describe several neurodevelopmental animal models of schizophrenia using an RDoC perspective approach. The implementation of animal models, combined with an RDoC framework, will bolster schizophrenia research leading to more targeted and likely effective therapeutic interventions resulting in better patient outcomes.

Use of machine learning to diagnose somatic symptom disorder: Are the biomarkers beneficial for the diagnosis?

OBJECTIVES

We used machine learning to incorporate three types of biomarkers (respiratory sinus arrhythmia, RSA; skin conductance, SC; finger temperature, FT) for examining the performance of diagnosing somatic symptom disorder (SSD).

METHODS

We recruited 97 SSD subjects and 96 controls without psychiatric history or somatic distress. The values of RSA, SC and FT were recorded in three situations (resting state, under a cognitive task and under paced breathing) and compared for the two populations. We used machine learning to combine the biological signals and then applied receiver operating characteristic curve analysis to examine the performance of diagnosing SSD regarding the distinct indicators and situations. Subgroup analysis for subjects without depression/anxiety was also conducted.

RESULTS

FT was significantly different between SSD patients and controls, especially in the resting state and under paced breathing. However, the biomarkers (0.75-0.76) did not reveal an area under the curve (AUC) comparable with the psychological questionnaires (0.86). Combining the biological and psychological indicators gave a high AUC (0.86-0.92). When excluding individuals with depression/anxiety, combining three biomarkers (0.79-0.83) and adopting psychological questionnaires (0.78) revealed a similar AUC.

CONCLUSIONS

The performance of RSA/SC/FT was unsatisfactory for diagnosing SSD but became comparable when excluding comorbid depression/anxiety.

Difference analysis of multidimensional electroencephalogram characteristics between young and old patients with generalized anxiety disorder

Growing evidences indicate that age plays an important role in the development of mental disorders, but few studies focus on the neuro mechanisms of generalized anxiety disorder (GAD) in different age groups. Therefore, this study attempts to reveal the neurodynamics of Young_GAD (patients with GAD under the age of 50) and Old_GAD (patients with GAD over 50 years old) through statistical analysis of multidimensional electroencephalogram (EEG) features and machine learning models. In this study, 10-min resting-state EEG data were collected from 45 Old_GAD and 33 Young_GAD. And multidimensional EEG features were extracted, including absolute power (AP), fuzzy entropy (FE), and phase-lag-index (PLI), on which comparison and analyses were performed later. The results showed that Old_GAD exhibited higher power spectral density (PSD) value and FE value in beta rhythm compared to theta, alpha1, and alpha2 rhythms, and functional connectivity (FC) also demonstrated significant reorganization of brain function in beta rhythm. In addition, the accuracy of machine learning classification between Old_GAD and Young_GAD was 99.67%, further proving the feasibility of classifying GAD patients by age. The above findings provide an objective basis in the field of EEG for the age-specific diagnosis and treatment of GAD.

Aberrated Multidimensional EEG Characteristics in Patients with Generalized Anxiety Disorder: A Machine-Learning Based Analysis Framework

Although increasing evidences support the notion that psychiatric disorders are associated with abnormal communication between brain regions, scattered studies have investigated brain electrophysiological disconnectivity of patients with generalized anxiety disorder (GAD). To this end, this study intends to develop an analysis framework for automatic GAD detection through incorporating multidimensional EEG feature extraction and machine learning techniques. Specifically, resting-state EEG signals with a duration of 10 min were obtained from 45 patients with GAD and 36 healthy controls (HC). Then, an analysis framework of multidimensional EEG characteristics (including univariate power spectral density (PSD) and fuzzy entropy (FE), and multivariate functional connectivity (FC), which can decode the EEG information from three different dimensions) were introduced for extracting aberrated multidimensional EEG features via statistical inter-group comparisons. These aberrated features were subsequently fused and fed into three previously validated machine learning methods to evaluate classification performance for automatic patient detection. We showed that patients exhibited a significant increase in beta rhythm and decrease in alpha1 rhythm of PSD, together with the reduced long-range FC between frontal and other brain areas in all frequency bands. Moreover, these aberrated features contributed to a very good classification performance with 97.83 ± 0.40% of accuracy, 97.55 ± 0.31% of sensitivity, 97.78 ± 0.36% of specificity, and 97.95 ± 0.17% of F1. These findings corroborate previous hypothesis of disconnectivity in psychiatric disorders and further shed light on distribution patterns of aberrant spatio-spectral EEG characteristics, which may lead to potential application of automatic diagnosis of GAD.

Diagnosis of Schizophrenia Based on the Data of Various Modalities: Biomarkers and Machine Learning Techniques (Review)

Schizophrenia is a socially significant mental disorder resulting frequently in severe forms of disability. Diagnosis, choice of treatment tactics, and rehabilitation in clinical psychiatry are mainly based on the assessment of behavioral patterns, socio-demographic data, and other investigations such as clinical observations and neuropsychological testing including examination of patients by the psychiatrist, self-reports, and questionnaires. In many respects, these data are subjective and therefore a large number of works have appeared in recent years devoted to the search for objective characteristics (indices, biomarkers) of the processes going on in the human body and reflected in the behavioral and psychoneurological patterns of patients. Such biomarkers are based on the results of instrumental and laboratory studies (neuroimaging, electro-physiological, biochemical, immunological, genetic, and others) and are successfully being used in neurosciences for understanding the mechanisms of the emergence and development of nervous system pathologies. Presently, with the advent of new effective neuroimaging, laboratory, and other methods of investigation and also with the development of modern methods of data analysis, machine learning, and artificial intelligence, a great number of scientific and clinical studies is being conducted devoted to the search for the markers which have diagnostic and prognostic value and may be used in clinical practice to objectivize the processes of establishing and clarifying the diagnosis, choosing and optimizing treatment and rehabilitation tactics, predicting the course and outcome of the disease. This review presents the analysis of the works which describe the correlates between the diagnosis of schizophrenia, established by health professionals, various manifestations of the psychiatric disorder (its subtype, variant of the course, severity degree, observed symptoms, etc.), and objectively measured characteristics/quantitative indicators (anatomical, functional, immunological, genetic, and others) obtained during instrumental and laboratory examinations of patients. A considerable part of these works has been devoted to correlates/biomarkers of schizophrenia based on the data of structural and functional (at rest and under cognitive load) MRI, EEG, tractography, and immunological data. The found correlates/biomarkers reflect anatomic disorders in the specific brain regions, impairment of functional activity of brain regions and their interconnections, specific microstructure of the brain white matter and the levels of connectivity between the tracts of various structures, alterations of electrical activity in various parts of the brain in different EEG spectral ranges, as well as changes in the innate and adaptive links of immunity. Current methods of data analysis and machine learning to search for schizophrenia biomarkers using the data of diverse modalities and their application during building and interpretation of predictive diagnostic models of schizophrenia have been considered in the present review.

Predictions of tDCS treatment response in PTSD patients using EEG based classification

Transcranial direct current stimulation (tDCS) is an emerging therapeutic tool for treating posttraumatic stress disorder (PTSD). Prior studies have shown that tDCS responses are highly individualized, thus necessitating the individualized optimization of treatment configurations. To date, an effective tool for predicting tDCS treatment outcomes in patients with PTSD has not yet been proposed. Therefore, we aimed to build and validate a tool for predicting tDCS treatment outcomes in patients with PTSD. Forty-eight patients with PTSD received 20 min of 2 mA tDCS stimulation in position of the anode over the F3 and cathode over the F4 region. Non-responders were defined as those with less than 50% improvement after reviewing clinical symptoms based on the Clinician-Administered DSM-5 PTSD Scale (before and after stimulation). Resting-state electroencephalograms were recorded for 3 min before and after stimulation. We extracted power spectral densities (PSDs) for five frequency bands. A support vector machine (SVM) model was used to predict responders and non-responders using PSDs obtained before stimulation. We investigated statistical differences in PSDs before and after stimulation and found statistically significant differences in the F8 channel in the theta band (p = 0.01). The SVM model had an area under the ROC curve (AUC) of 0.93 for predicting responders and non-responders using PSDs. To our knowledge, this study provides the first empirical evidence that PSDs can be useful biomarkers for predicting the tDCS treatment response, and that a machine learning model can provide robust prediction performance. Machine learning models based on PSDs can be useful for informing treatment decisions in tDCS treatment for patients with PTSD.