Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Insights into Schizophrenia: Leveraging Machine Learning for Early Identification via EEG, ERP, and Demographic Attributes

Sara Alkhalifa

TL;DR

This work tackles early, accurate schizophrenia identification by combining EEG/ERP features with demographic attributes and applying multiple ML classifiers. Using a public Ford2014 EEG dataset of 81 participants (32 healthy controls, 49 SZ), the study evaluates SVM, k-NN, and a Decision Tree on ERP-inclusive and ERP-exclusive data, achieving up to 99.930% accuracy with all features. It further investigates entropy-based feature selection and incremental feature removal to identify the most informative ERP/EEG/demographic markers, finding ERP components and specific electrode locations to be highly discriminative. The results support the value of integrating ERP data with EEG and demographic information for AI-assisted SZ screening, while highlighting the balance between model performance and interpretability and the importance of robust feature selection.

Abstract

The research presents a machine learning (ML) classifier designed to differentiate between schizophrenia patients and healthy controls by utilising features extracted from electroencephalogram (EEG) data, specifically focusing on event-related potentials (ERPs) and certain demographic variables. The dataset comprises data from 81 participants, encompassing 32 healthy controls and 49 schizophrenia patients, all sourced from an online dataset. After preprocessing the dataset, our ML model achieved an accuracy of 99.930%. This performance outperforms earlier research, including those that used deep learning methods. Additionally, an analysis was conducted to assess individual features' contribution to improving classification accuracy. This involved systematically excluding specific features from the original dataset one at a time, and another technique involved an iterative process of removing features based on their entropy scores incrementally. The impact of these removals on model performance was evaluated to identify the most informative features.

Insights into Schizophrenia: Leveraging Machine Learning for Early Identification via EEG, ERP, and Demographic Attributes

TL;DR

This work tackles early, accurate schizophrenia identification by combining EEG/ERP features with demographic attributes and applying multiple ML classifiers. Using a public Ford2014 EEG dataset of 81 participants (32 healthy controls, 49 SZ), the study evaluates SVM, k-NN, and a Decision Tree on ERP-inclusive and ERP-exclusive data, achieving up to 99.930% accuracy with all features. It further investigates entropy-based feature selection and incremental feature removal to identify the most informative ERP/EEG/demographic markers, finding ERP components and specific electrode locations to be highly discriminative. The results support the value of integrating ERP data with EEG and demographic information for AI-assisted SZ screening, while highlighting the balance between model performance and interpretability and the importance of robust feature selection.

Abstract

The research presents a machine learning (ML) classifier designed to differentiate between schizophrenia patients and healthy controls by utilising features extracted from electroencephalogram (EEG) data, specifically focusing on event-related potentials (ERPs) and certain demographic variables. The dataset comprises data from 81 participants, encompassing 32 healthy controls and 49 schizophrenia patients, all sourced from an online dataset. After preprocessing the dataset, our ML model achieved an accuracy of 99.930%. This performance outperforms earlier research, including those that used deep learning methods. Additionally, an analysis was conducted to assess individual features' contribution to improving classification accuracy. This involved systematically excluding specific features from the original dataset one at a time, and another technique involved an iterative process of removing features based on their entropy scores incrementally. The impact of these removals on model performance was evaluated to identify the most informative features.

Paper Structure

This paper contains 16 sections, 2 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Research aims
  3. Research questions
  4. Contributions
  5. Paper structure
  6. Related work
  7. Overview of research in EEG
  8. Machine Learning Model in Psychiatry
  9. Methodology
  10. Data Collection
  11. CLASSIFICATION USING MACHINE LEARNING MODELS
  12. Attribute Selection Measures
  13. Results
  14. Discussion
  15. Conclusions
  16. ...and 1 more sections

Figures (2)

  • Figure 1: An overview of EEG.
  • Figure 2: An overview of SVM.