Examining Effects of Schizophrenia on EEG with Explainable Deep Learning Models

Ellis, Charles A.; Sattiraju, Abhinav; Miller, Robyn L.; Calhoun, Vince D.

doi:10.1109/bibe55377.2022.00068

Cited by 30 publications

(44 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We used a resting-state scalp EEG dataset with 47 individuals with SZ (SZs) [14]. The dataset has been used for multiple classification studies [1]. All study participants gave written informed consent, and data collection was approved by the Hartford Hospital Internal Review Board.…”

Section: A Description Of Dataset and Preprocessingmentioning

confidence: 99%

See 1 more Smart Citation

A Convolutional Autoencoder-based Explainable Clustering Approach for Resting-State EEG Analysis

Ellis

Miller

Calhoun

2023

Preprint

Self Cite

View full text Add to dashboard Cite

Machine learning methods have frequently been applied to electroencephalography (EEG) data. However, while supervised EEG classification is well-developed, relatively few studies have clustered EEG, which is problematic given the potential for clustering EEG to identify novel subtypes or patterns of dynamics that could improve our understanding of neuropsychiatric disorders. There are established methods for clustering EEG using manually extracted features that reduce the richness of the feature space for clustering, but only a couple studies have sought to use deep learning-based approaches with automated feature learning to cluster EEG. Those studies involve separately training an autoencoder and then performing clustering on the extracted features, and the separation of those steps can lead to poor quality clustering. In this study, we propose an explainable convolutional autoencoder-based approach that combines model training with clustering to yield high quality clusters. We apply the approach within the context of schizophrenia (SZ), identifying 8 EEG states characterized by varying levels of δ activity. We also find that individuals who spend more time outside of the dominant state tend to have increased negative symptom severity. Our approach represents a significant step forward for clustering resting-state EEG data and has the potential to lead to novel findings across a variety of neurological and neuropsychological disorders in future years.

show abstract

Section: A Description Of Dataset and Preprocessingmentioning

confidence: 99%

“…The standard 10-20 format of 64 electrodes was used during collection. Like previous studies [1], we only used 19 electrodes. Data was recorded at 1000 Hertz (Hz) for 5 minutes.…”

Section: A Description Of Dataset and Preprocessingmentioning

confidence: 99%

A Convolutional Autoencoder-based Explainable Clustering Approach for Resting-State EEG Analysis

Ellis

Miller

Calhoun

2023

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…While that is an important distinction, it also makes our subtyping highly dependent upon our classifier. Our results could be affected by (1) model performance or (2) even the random initialization of model weights during training. Our model test performance was quite high but not perfect.…”

Section: Limitations and Next Stepsmentioning

confidence: 99%

“…The identification of novel neuropsychiatric disorder subtypes based on brain recordings has many potential benefits including enhancing the scientific understanding of disorders, enabling improved data-driven diagnosis that goes beyond symptoms [1], or enabling personalized treatment. In this study, we identify several subtypes of schizophrenia (SZ) using a novel approach that combines explainable deep learning classifiers and traditional clustering methods to analyze resting-state functional magnetic resonance imaging (rs-fMRI) dynamic functional network connectivity (FNC).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Neuropsychiatric Disorder Subtyping Via Clustered Deep Learning Classifier Explanations

Ellis

Miller

Calhoun

2022

Preprint

Self Cite

View full text Add to dashboard Cite

Identifying subtypes of neuropsychiatric disorders based on characteristics of their brain activity has tremendous potential to contribute to a better understanding of those disorders and to the development of new diagnostic and personalized treatment approaches. Many studies focused on neuropsychiatric disorders examine the interaction of brain networks over time using dynamic functional network connectivity (dFNC) extracted from resting-state functional magnetic resonance imaging data. Some of these studies involve the use of either deep learning classifiers or traditional clustering approaches, but usually not both. In this study, we present a novel approach for subtyping individuals with neuropsychiatric disorders within the context of schizophrenia (SZ). We train an explainable deep learning classifier to differentiate between dFNC data from individuals with SZ and controls, obtaining a test accuracy of 79%. We next make use of cross-validation to obtain robust average explanations for SZ training participants across folds, identifying 5 SZ subtypes that each differ from controls in a distinct manner and that have different degrees of symptom severity. These subtypes specifically differ from one another in their interaction between the visual network and the subcortical, sensorimotor, and auditory networks and between the cerebellar network and the cognitive control and subcortical networks. Additionally, there are statistically significant differences in negative symptom scores between the subtypes. It is our hope that the proposed novel subtyping approach will contribute to the improved understanding and characterization of SZ and other neuropsychiatric disorders.

show abstract

Identifying EEG Biomarkers of Depression with Novel Explainable Deep Learning Architectures

Ellis,

Sancho,

Miller

et al. 2024

Communications in Computer and Information Science

View full text Add to dashboard Cite

Examining Effects of Schizophrenia on EEG with Explainable Deep Learning Models

Cited by 30 publications

References 10 publications

A Convolutional Autoencoder-based Explainable Clustering Approach for Resting-State EEG Analysis

A Convolutional Autoencoder-based Explainable Clustering Approach for Resting-State EEG Analysis

Neuropsychiatric Disorder Subtyping Via Clustered Deep Learning Classifier Explanations

Identifying EEG Biomarkers of Depression with Novel Explainable Deep Learning Architectures

Contact Info

Product

Resources

About