Abstract
BACKGROUND: Innovations in the analysis of resting-state EEG focused on connectivity and network organization, combined with machine learning, offer new opportunities for treatment response predictions in psychiatry.
AIM: Introduction of analysis methods in this emerging field, description of some promising results, and critical consideration of possibilities and challenges for implementation in clinical practice.
METHOD: Narrative review of the literature.
RESULTS: EEG connectivity and network properties may contain predictive information for treatment response to pharmacological interventions, neurostimulation, and psychotherapeutic treatments. However, the results are currently based on studies with small sample sizes and limited validation in independent datasets. Factors such as placebo effects, natural course and treatment adherence during therapy necessitate a cautious interpretation of promising results.
CONCLUSION: Independent replication studies and research on implementation are needed to determine whether developed algorithms that predict treatment outcomes based on EEG recordings are of value in clinical practice.
| Translated title of the contribution | Resting state-EEG connectivity and machine learning: towards improved treatment response predictions in psychiatry |
|---|---|
| Original language | Dutch |
| Pages (from-to) | 637-640 |
| Number of pages | 4 |
| Journal | Tijdschrift voor Psychiatrie |
| Volume | 65 |
| Issue number | 10 |
| Publication status | Published - 2023 |
Keywords
- Algorithms
- Brain/physiology
- Electroencephalography/methods
- Humans
- Machine Learning
- Psychiatry
- Treatment Outcome