Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Systematic Review of EEG-based Machine Intelligence Algorithms for Depression Diagnosis, and Monitoring

Amir Nassibi, Christos Papavassiliou, Ildar Rakhmatulin, Danilo Mandic, S. Farokh Atashzar

TL;DR

This systematic review addresses EEG-based depression diagnosis using machine intelligence by aggregating 139 studies from an initial 938 records (1985–Dec 2022) through PRISMA. It organizes findings around data acquisition, preprocessing, feature extraction, and ML methods (notably SVM, KNN, LR, CNNs) and highlights how various pipelines yield varying accuracies and clinical usefulness. The review identifies pervasive heterogeneity in datasets, channels, and labels, emphasizing the lack of standardized benchmarks and publicly accessible datasets, which hampers cross-study comparisons. It advocates for a universal anonymized EEG depression dataset, multi-label clinical phenotyping, and scalable wearable/cloud-based platforms to enable in-home monitoring and broader clinical adoption of EEG-based depression diagnostics.

Abstract

Depression disorder is a serious health condition that has affected the lives of millions of people around the world. Diagnosis of depression is a challenging practice that relies heavily on subjective studies and, in most cases, suffers from late findings. Electroencephalography (EEG) biomarkers have been suggested and investigated in recent years as a potential transformative objective practice. In this article, for the first time, a detailed systematic review of EEG-based depression diagnosis approaches is conducted using advanced machine learning techniques and statistical analyses. For this, 938 potentially relevant articles (since 1985) were initially detected and filtered into 139 relevant articles based on the review scheme 'preferred reporting items for systematic reviews and meta-analyses (PRISMA).' This article compares and discusses the selected articles and categorizes them according to the type of machine learning techniques and statistical analyses. Algorithms, preprocessing techniques, extracted features, and data acquisition systems are discussed and summarized. This review paper explains the existing challenges of the current algorithms and sheds light on the future direction of the field. This systematic review outlines the issues and challenges in machine intelligence for the diagnosis of EEG depression that can be addressed in future studies and possibly in future wearable technologies.

A Systematic Review of EEG-based Machine Intelligence Algorithms for Depression Diagnosis, and Monitoring

TL;DR

This systematic review addresses EEG-based depression diagnosis using machine intelligence by aggregating 139 studies from an initial 938 records (1985–Dec 2022) through PRISMA. It organizes findings around data acquisition, preprocessing, feature extraction, and ML methods (notably SVM, KNN, LR, CNNs) and highlights how various pipelines yield varying accuracies and clinical usefulness. The review identifies pervasive heterogeneity in datasets, channels, and labels, emphasizing the lack of standardized benchmarks and publicly accessible datasets, which hampers cross-study comparisons. It advocates for a universal anonymized EEG depression dataset, multi-label clinical phenotyping, and scalable wearable/cloud-based platforms to enable in-home monitoring and broader clinical adoption of EEG-based depression diagnostics.

Abstract

Depression disorder is a serious health condition that has affected the lives of millions of people around the world. Diagnosis of depression is a challenging practice that relies heavily on subjective studies and, in most cases, suffers from late findings. Electroencephalography (EEG) biomarkers have been suggested and investigated in recent years as a potential transformative objective practice. In this article, for the first time, a detailed systematic review of EEG-based depression diagnosis approaches is conducted using advanced machine learning techniques and statistical analyses. For this, 938 potentially relevant articles (since 1985) were initially detected and filtered into 139 relevant articles based on the review scheme 'preferred reporting items for systematic reviews and meta-analyses (PRISMA).' This article compares and discusses the selected articles and categorizes them according to the type of machine learning techniques and statistical analyses. Algorithms, preprocessing techniques, extracted features, and data acquisition systems are discussed and summarized. This review paper explains the existing challenges of the current algorithms and sheds light on the future direction of the field. This systematic review outlines the issues and challenges in machine intelligence for the diagnosis of EEG depression that can be addressed in future studies and possibly in future wearable technologies.

Paper Structure

This paper contains 18 sections, 30 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Methodology
  3. Machine Learning Methodologies
  4. Machine Learning Algorithms
  5. Review of algorithms to diagnose depression
  6. EEG data acquisition
  7. Preprocessing and feature extraction
  8. Data processing and machine learning
  9. Logistic Regression
  10. Support vector machine (SVM)
  11. K-Nearest Neighbor (KNN)
  12. Neural Networks
  13. Statistical Analysis
  14. Information Fusion beyond EEG
  15. Discussion and conclusion
  16. ...and 3 more sections

Figures (30)

  • Figure 1: PRISMA systematic review diagram Dressel2017
  • Figure 2: Summary of the most recent literature based on data processing pipeline, performance metrics, and main results
  • Figure 3: Continued - summary of the most recent literature based on data processing pipeline, performance metrics, and main results
  • Figure 4: Continued - summary of the most recent literature based on data processing pipeline, performance metrics, and main results
  • Figure 5: Continued - summary of the most recent literature based on data processing pipeline, performance metrics, and main results
  • ...and 25 more figures