A Deep Learning-Derived Transdiagnostic Signature Indexing Hypoarousal and Impulse Control: Implications for
Hannah Meijs1, Jurjen J Luykx1, Nikita van der Vinne2
1Research Institute Brainclinics, Brainclinics Foundation, Nijmegen, the Netherlands; Department of Cognitive Neuroscience, Faculty of Psychology and Neuroscience, Maastricht University, Maastricht, the Netherlands.
Biological Psychiatry. Cognitive Neuroscience and Neuroimaging
|August 14, 2024
Summary
Research Domain Criteria (RDoC) identified a brainwave pattern linked to impulse control and sleep issues. This biomarker may aid in diagnosing and predicting treatment response in psychiatric disorders.
Area of Science:
- Neuroscience
- Psychiatry
- Computational Biology
Background:
- Traditional psychiatric diagnostic categories have limitations.
- The Research Domain Criteria (RDoC) offers a dimensional approach to psychiatric disorders.
- RDoC aims to understand mental illness via neurobiological and behavioral systems.
Purpose of the Study:
- To investigate electroencephalographic frontal beta activity as a transdiagnostic biomarker.
- To assess its potential in diagnosing and predicting impulse control and sleep problems.
Main Methods:
- Utilized a deep learning algorithm to classify spindling excessive beta in electroencephalography (EEG) data.
- Analyzed EEG data from three independent datasets (total N=4623).
- Examined Brainmarker-III (beta power and spindling excessive beta probability) in relation to diagnoses and treatment outcomes.
Main Results:
- Spindling excessive beta probability correlated with poor sleep maintenance and low impulse control.
- Brainmarker-III associated with attention-deficit/hyperactivity disorder (ADHD) diagnosis.
- Brainmarker-III predicted treatment response to methylphenidate in children with ADHD and antidepressants in adults with major depressive disorder.
Conclusions:
- The Research Domain Criteria (RDoC) approach facilitates biomarker discovery in psychiatry.
- EEG-based biomarkers show potential for diagnosis and treatment prediction.
- Findings support the utility of RDoC in advancing psychiatric research and clinical applications.
