ADHD identification and its interpretation of functional connectivity using deep self-attention factorization

Graphical Insight: Revolutionizing Seizure Detection with EEG Representation

Epilepsy is characterized by recurring seizures that result from abnormal electrical activity in the brain. These seizures manifest as various symptoms including muscle contractions and loss of consciousness. The challenging task of detecting epileptic seizures involves classifying electroencephalography (EEG) signals into ictal (seizure) and interictal (non-seizure) classes. This classification is crucial because it distinguishes between the states of seizure and seizure-free periods in patients with epilepsy. Our study presents an innovative approach for detecting seizures and neurological diseases using EEG signals by leveraging graph neural networks. This method effectively addresses EEG data processing challenges. We construct a graph representation of EEG signals by extracting features such as frequency-based, statistical-based, and Daubechies wavelet transform features. This graph representation allows for potential differentiation between seizure and non-seizure signals through visual inspection of the extracted features. To enhance seizure detection accuracy, we employ two models: one combining a graph convolutional network (GCN) with long short-term memory (LSTM) and the other combining a GCN with balanced random forest (BRF). Our experimental results reveal that both models significantly improve seizure detection accuracy, surpassing previous methods. Despite simplifying our approach by reducing channels, our research reveals a consistent performance, showing a significant advancement in neurodegenerative disease detection. Our models accurately identify seizures in EEG signals, underscoring the potential of graph neural networks. The streamlined method not only maintains effectiveness with fewer channels but also offers a visually distinguishable approach for discerning seizure classes. This research opens avenues for EEG analysis, emphasizing the impact of graph representations in advancing our understanding of neurodegenerative diseases.

TanhReLU -based convolutional neural networks for MDD classification

Major Depression Disorder (MDD), a complex mental health disorder, poses significant challenges in accurate diagnosis. In addressing the issue of gradient vanishing in the classification of MDD using current data-driven electroencephalogram (EEG) data, this study introduces a TanhReLU-based Convolutional Neural Network (CNN). By integrating the TanhReLU activation function, which combines the characteristics of the hyperbolic tangent (Tanh) and rectified linear unit (ReLU) activations, the model aims to improve performance in identifying patterns associated with MDD while alleviating the issue of model overfitting and gradient vanishing. Experimental results demonstrate promising outcomes in the task of MDD classification upon the publicly available EEG data, suggesting potential clinical applications.

Prediction of cognitive impairment using higher order item response theory and machine learning models

Timely detection of cognitive impairment (CI) is critical for the wellbeing of elderly individuals. The MyCog assessment employs two validated iPad-based measures from the NIH Toolbox® for Assessment of Neurological and Behavioral Function (NIH Toolbox). These measures assess pivotal cognitive domains: Picture Sequence Memory (PSM) for episodic memory and Dimensional Change Card Sort Test (DCCS) for cognitive flexibility. The study involved 86 patients and explored diverse machine learning models to enhance CI prediction. This encompassed traditional classifiers and neural-network-based methods. After 100 bootstrap replications, the Random Forest model stood out, delivering compelling results: precision at 0.803, recall at 0.758, accuracy at 0.902, F1 at 0.742, and specificity at 0.951. Notably, the model incorporated a composite score derived from a 2-parameter higher order item response theory (HOIRT) model that integrated DCCS and PSM assessments. The study's pivotal finding underscores the inadequacy of relying solely on a fixed composite score cutoff point. Instead, it advocates for machine learning models that incorporate HOIRT-derived scores and encompass relevant features such as age. Such an approach promises more effective predictive models for CI, thus advancing early detection and intervention among the elderly.

Finding neural correlates of depersonalisation/derealisation disorder via explainable CNN-based analysis guided by clinical assessment scores

Mental health disorders are typically diagnosed based on subjective reports (e.g., through questionnaires) followed by clinical interviews to evaluate the self-reported symptoms. Therefore, considering the interconnected nature of psychiatric disorders, their accurate diagnosis is a real challenge without indicators of underlying physiological dysfunction. Depersonalisation/derealisation disorder (DPD) is an example of dissociative disorder affecting 1-2 % of the population. DPD is characterised mainly by persistent disembodiment, detachment from surroundings, and feelings of emotional numbness, which can significantly impact patients' quality of life. The underlying neural correlates of DPD have been investigated for years to understand and help with a more accurate and in-time diagnosis of the disorder. However, in terms of EEG studies, which hold great importance due to their convenient and inexpensive nature, the literature has often been based on hypotheses proposed by experts in the field, which require prior knowledge of the disorder. In addition, participants' labelling in research experiments is often derived from the outcome of the Cambridge Depersonalisation Scale (CDS), a subjective assessment to quantify the level of depersonalisation/derealisation, the threshold and reliability of which might be challenged. As a result, we aimed to propose a novel end-to-end EEG processing pipeline based on deep neural networks for DPD biomarker discovery, which requires no prior handcrafted labelled data. Alternatively, it can assimilate knowledge from clinical outcomes like CDS as well as data-driven patterns that differentiate individual brain responses. In addition, the structure of the proposed model targets the uncertainty in CDS scores by using them as prior information only to guide the unsupervised learning task in a multi-task learning scenario. A comprehensive evaluation has been done to confirm the significance of the proposed deep structure, including new ways of network visualisation to investigate spectral, spatial, and temporal information derived in the learning process. We argued that the proposed EEG analytics could also be applied to investigate other psychological and mental disorders currently indicated on the basis of clinical assessment scores. The code to reproduce the results presented in this paper is openly accessible at https://github.com/AbbasSalami/DPD_Analysis.

Transdiagnostic clustering of self-schema from self-referential judgements identifies subtypes of healthy personality and depression

Introduction

The heterogeneity of depressive and anxiety disorders complicates clinical management as it may account for differences in trajectory and treatment response. Self-schemas, which can be determined by Self-Referential Judgements (SRJs), are heterogeneous yet stable. SRJs have been used to characterize personality in the general population and shown to be prognostic in depressive and anxiety disorders.

Methods

In this study, we used SRJs from a Self-Referential Encoding Task (SRET) to identify clusters from a clinical sample of 119 patients recruited from the Institute of Mental Health presenting with depressive or anxiety symptoms and a non-clinical sample of 115 healthy adults. The generated clusters were examined in terms of most endorsed words, cross-sample correspondence, association with depressive symptoms and the Depressive Experiences Questionnaire and diagnostic category.

Results

We identify a 5-cluster solution in each sample and a 7-cluster solution in the combined sample. When perturbed, metrics such as optimum cluster number, criterion value, likelihood, DBI and CHI remained stable and cluster centers appeared stable when using BIC or ICL as criteria. Top endorsed words in clusters were meaningful across theoretical frameworks from personality, psychodynamic concepts of relatedness and self-definition, and valence in self-referential processing. The clinical clusters were labeled "Neurotic" (C1), "Extraverted" (C2), "Anxious to please" (C3), "Self-critical" (C4), "Conscientious" (C5). The non-clinical clusters were labeled "Self-confident" (N1), "Low endorsement" (N2), "Non-neurotic" (N3), "Neurotic" (N4), "High endorsement" (N5). The combined clusters were labeled "Self-confident" (NC1), "Externalising" (NC2), "Neurotic" (NC3), "Secure" (NC4), "Low endorsement" (NC5), "High endorsement" (NC6), "Self-critical" (NC7). Cluster differences were observed in endorsement of positive and negative words, latency biases, recall biases, depressive symptoms, frequency of depressive disorders and self-criticism.

Discussion

Overall, clusters endorsing more negative words tended to endorse fewer positive words, showed more negative biases in reaction time and negative recall bias, reported more severe depressive symptoms and a higher frequency of depressive disorders and more self-criticism in the clinical population. SRJ-based clustering represents a novel transdiagnostic framework for subgrouping patients with depressive and anxiety symptoms that may support the future translation of the science of self-referential processing, personality and psychodynamic concepts of self-definition to clinical applications.

Application of graph frequency attention convolutional neural networks in depression treatment response

Depression, a prevalent global mental health disorder, necessitates precise treatment response prediction for the improvement of personalized care and patient prognosis. The Graph Convolutional Neural Networks (GCNs) have emerged as a promising technique for handling intricate signals and classification tasks owing to their end-to-end neural architecture and nonlinear processing capabilities. In this context, this article proposes a model named the Graph Frequency Attention Convolutional Neural Network (GFACNN). Primarily, the model transforms the EEG signals into graphs to depict the connections between electrodes and brain regions, while integrating a frequency attention module to accentuate brain rhythm information. The proposed approach delves into the application of graph neural networks in the classification of EEG data, aiming to evaluate the response to antidepressant treatment and discern between treatment-resistant and treatment-responsive cases. Experimental results obtained from an EEG dataset at Peking University People's Hospital demonstrate the notable performance of GFACNN in distinguishing treatment responses among depression patients, surpassing deep learning methodologies including CapsuleNet and GoogLeNet. This highlights the efficacy of graph neural networks in leveraging the connections within EEG signal data. Overall, GFACNN exhibits potential for the classification of depression EEG signals, thereby potentially aiding clinical diagnosis and treatment.

Seed correlation analysis based on brain region activation for ADHD diagnosis in a large-scale resting state data set

Background

Attention-deficit/hyperactivity disorder (ADHD) is a neurodevelopmental disorder of multifactorial pathogenesis, which is often accompanied by dysfunction in several brain functional connectivity. Resting-state functional MRI have been used in ADHD, and they have been proposed as a possible biomarker of diagnosis information. This study's primary aim was to offer an effective seed-correlation analysis procedure to investigate the possible biomarker within resting state brain networks as diagnosis information.

Method

Resting-state functional magnetic resonance imaging (rs-fMRI) data of 149 childhood ADHD were analyzed. In this study, we proposed a two-step hierarchical analysis method to extract functional connectivity features and evaluation by linear classifiers and random sampling validation.

Result

The data-driven method-ReHo provides four brain regions (mPFC, temporal pole, motor area, and putamen) with regional homogeneity differences as second-level seeds for analyzing functional connectivity differences between distant brain regions. The procedure reduces the difficulty of seed selection (location, shape, and size) in estimations of brain interconnections, improving the search for an effective seed; The features proposed in our study achieved a success rate of 83.24% in identifying ADHD patients through random sampling (saving 25% as the test set, while the remaining data was the training set) validation (using a simple linear classifier), surpassing the use of traditional seeds.

Conclusion

This preliminary study examines the feasibility of diagnosing ADHD by analyzing the resting-state fMRI data from the ADHD-200 NYU dataset. The data-driven model provides a precise way to find reliable seeds. Data-driven models offer precise methods for finding reliable seeds and are feasible across different datasets. Moreover, this phenomenon may reveal that using a data-driven approach to build a model specific to a single data set may be better than combining several data and creating a general model.

Insight into ADHD diagnosis with deep learning on Actimetry: Quantitative interpretation of occlusion maps in age and gender subgroups

Attention Deficit/Hyperactivity Disorder (ADHD) is a prevalent neurodevelopmental disorder in childhood that often persists into adulthood. Objectively diagnosing ADHD can be challenging due to the reliance on subjective questionnaires in clinical assessment. Fortunately, recent advancements in artificial intelligence (AI) have shown promise in providing objective diagnoses through the analysis of medical images or activity recordings. These AI-based techniques have demonstrated accurate ADHD diagnosis; however, the growing complexity of deep learning models has introduced a lack of interpretability. These models often function as black boxes, unable to offer meaningful insights into the data patterns that characterize ADHD.

OBJECTIVE

This paper proposes a methodology to interpret the output of an AI-based diagnosis system for combined ADHD in age and gender-stratified populations.

METHODS

Our system is based on the analysis of 24 hour-long activity records using Convolutional Neural Networks (CNNs) to classify spectrograms of activity windows. These windows are interpreted using occlusion maps to highlight the time-frequency patterns explaining ADHD activity.

RESULTS

Significant differences in the frequency patterns between ADHD and controls both in diurnal and nocturnal activity were found for all the populations. Temporal dispersion also presented differences in the male population.

CONCLUSION

The proposed interpretation techniques for CNNs highlighted gender- and age-related differences between ADHD patients and controls. Leveraging these differences could potentially lead to improved diagnostic accuracy, especially if a larger and more balanced dataset is utilized.

SIGNIFICANCE

Our findings pave the way for the development of an AI-based diagnosis system for ADHD that offers interpretability, thereby providing valuable insights into the underlying etiology of the disease.

Interpretation of deep non-linear factorization for autism

Autism, a neurodevelopmental disorder, presents significant challenges for diagnosis and classification. Despite the widespread use of neural networks in autism classification, the interpretability of their models remains a crucial issue. This study aims to address this concern by investigating the interpretability of neural networks in autism classification using the deep symbolic regression and brain network interpretative methods. Specifically, we analyze publicly available autism fMRI data using our previously developed Deep Factor Learning model on a Hibert Basis tensor (HB-DFL) method and extend the interpretative Deep Symbolic Regression method to identify dynamic features from factor matrices, construct brain networks from generated reference tensors, and facilitate the accurate diagnosis of abnormal brain network activity in autism patients by clinicians. Our experimental results show that our interpretative method effectively enhances the interpretability of neural networks and identifies crucial features for autism classification.

Association between mental health symptoms and behavioral performance in younger vs. older online workers

Background

The COVID-19 pandemic has been associated with increased rates of mental health problems, particularly in younger people.

Objective

We quantified mental health of online workers before and during the COVID-19 pandemic, and cognition during the early stages of the pandemic in 2020. A pre-registered data analysis plan was completed, testing the following three hypotheses: reward-related behaviors will remain intact as age increases; cognitive performance will decline with age; mood symptoms will worsen during the pandemic compared to before. We also conducted exploratory analyses including Bayesian computational modeling of latent cognitive parameters.

Methods

Self-report depression (Patient Health Questionnaire 8) and anxiety (General Anxiety Disorder 7) prevalence were compared from two samples of Amazon Mechanical Turk (MTurk) workers ages 18-76: pre-COVID 2018 (N = 799) and peri-COVID 2020 (N = 233). The peri-COVID sample also completed a browser-based neurocognitive test battery.

Results

We found support for two out of three pre-registered hypotheses. Notably our hypothesis that mental health symptoms would increase in the peri-COVID sample compared to pre-COVID sample was not supported: both groups reported high mental health burden, especially younger online workers. Higher mental health symptoms were associated with negative impacts on cognitive performance (speed/accuracy tradeoffs) in the peri-COVID sample. We found support for two hypotheses: reaction time slows down with age in two of three attention tasks tested, whereas reward function and accuracy appear to be preserved with age.

Conclusion

This study identified high mental health burden, particularly in younger online workers, and associated negative impacts on cognitive function.

EEG-responses to mood induction interact with seasonality and age

The EEG is suggested as a potential diagnostic and prognostic biomarker for seasonal affective disorder (SAD). As a pre-clinical form of SAD, seasonality is operationalized as seasonal variation in mood, appetite, weight, sleep, energy, and socializing. Importantly, both EEG biomarkers and seasonality interact with age. Inducing sad mood to assess cognitive vulnerability was suggested to improve the predictive value of summer assessments for winter depression. However, no EEG studies have been conducted on induced sad mood in relation to seasonality, and no studies so far have controlled for age. We recorded EEG and calculated bandpower in 114 participants during rest and during induced sad mood in summer. Participants were grouped by age and based on a seasonality score as obtained with the seasonal pattern assessment questionnaire (SPAQ). Participants with high seasonality scores showed significantly larger changes in EEG power from rest to sad mood induction, specifically in the alpha frequency range (p = 0.027), compared to participants with low seasonality scores. Furthermore, seasonality interacted significantly with age (p < 0.001), with lower activity in individuals with high seasonality scores that were older than 50 years but the opposite pattern in individuals up to 50 years. Effects of sad mood induction on brain activity are related to seasonality and can therefore be consider as potential predicting biomarkers for SAD. Future studies should control for age as a confounding factor, and more studies are needed to elaborate on the characteristics of EEG biomarkers in participants above 50 years.