QEEG-Based Ensemble Machine Learning for Anxiety Disorder Classification and Scoring
Poster No:
1103
Submission Type:
Abstract Submission
Authors:
Hyunna Kim1, Seungwan Kang1
Institutions:
1iMediSync Inc., Seoul, Korea, Republic of
First Author:
Co-Author:
Introduction:
Anxiety is a normal response to facing danger. However, when this response becomes overwhelming or persistent, it is considered an anxiety disorder. Anxiety disorders are the most common mental health conditions worldwide and have shown a growing prevalence in recent years. Anxiety symptoms are often underestimated or neglected, which can lead to the development of panic disorders and secondary complications. Therefore, screening and assessing one's level of anxiety is crucial for early intervention. This study aims to develop a model that uses quantitative electroencephalography (qEEG) data to easily screen and quantify anxiety levels.
Methods:
For model training, we utilized eyes-closed EEG data measured at 19 channels based on the 10-20 system during the resting state from 962 subjects. The dataset consisted of 481 subjects with anxiety and 481 healthy controls. Additionally, data from 76 PTSD patients (BCL-5 ≥ 30) from Scott and Baylor's Hospital were used for model validation. The qEEG features included spectral power, power ratios, and alpha peak from the occipital lobe. To eliminate variability arising from age and gender differences, a calculated Z-score was used by comparing and analyzing the standard EEG database matched for age and gender, enabling a common and statistically robust analysis. Among the 962 data points, 70% were used as the training set, and the remaining 30% as the test set. We employed four tree-based models to handle missing values automatically. To reduce overfitting and improve accuracy, we applied an ensemble method. The ensemble utilized the soft voting method to obtain class probabilities, which were subsequently scaled to a 0–100 range using the Min-Max Scaler for scoring.
Results:
The final ensemble model achieved a test accuracy of 88% and an F1-score of 87% for anxiety classification. A threshold score of 40 was used to distinguish between anxiety and normal groups. When PTSD data was included, 67 out of 76 individuals were identified as having anxiety.
Conclusions:
The qEEG-based machine learning classifier developed in this study successfully distinguished between anxiety and control groups. According to the important features identified by the ensemble model, beta waves, particularly in the central and parietal regions, play a significant role. This finding is consistent with previous studies indicating that beta activity(22~30Hz) is associated with anxiety. Therefore, our model could serve as an efficient tool for screening anxiety levels.
Disorders of the Nervous System:
Psychiatric (eg. Depression, Anxiety, Schizophrenia) 2
Modeling and Analysis Methods:
Classification and Predictive Modeling 1
EEG/MEG Modeling and Analysis
Task-Independent and Resting-State Analysis
Keywords:
Anxiety
Data analysis
Electroencephaolography (EEG)
Machine Learning
Modeling
1|2Indicates the priority used for review
Abstract Information
By submitting your proposal, you grant permission for the Organization for Human Brain Mapping (OHBM) to distribute your work in any format, including video, audio print and electronic text through OHBM OnDemand, social media channels, the OHBM website, or other electronic publications and media.
The Open Science Special Interest Group (OSSIG) is introducing a reproducibility challenge for OHBM 2025. This new initiative aims to enhance the reproducibility of scientific results and foster collaborations between labs. Teams will consist of a “source” party and a “reproducing” party, and will be evaluated on the success of their replication, the openness of the source work, and additional deliverables. Click here for more information. Propose your OHBM abstract(s) as source work for future OHBM meetings by selecting one of the following options:
Please indicate below if your study was a "resting state" or "task-activation” study.
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Was this research conducted in the United States?
Were any human subjects research approved by the relevant Institutional Review Board or ethics panel? NOTE: Any human subjects studies without IRB approval will be automatically rejected.
Were any animal research approved by the relevant IACUC or other animal research panel? NOTE: Any animal studies without IACUC approval will be automatically rejected.
Please indicate which methods were used in your research:
Provide references using APA citation style.
Díaz, H. (2019). EEG Beta band frequency domain evaluation for assessing stress and anxiety in resting, eyes closed, basal conditions. Procedia Computer Science, 162, 974-981.