Motif Discovery Framework for Psychiatric EEG Data Classification

TL;DR

The paper tackles the problem of predicting antidepressant treatment response earlier by analyzing day-7 EEG using a motif-based framework. It introduces k-Motiflets to extract discriminative motifs from alpha, beta, and theta bands and builds a distance-based feature matrix for classification, achieving a best mean $F1$ of $0.722$ in the alpha band with a DT classifier. The approach is extended to other psychiatric EEG datasets (schizophrenia, seizures, Alzheimer's, dementia), showing robust performance across conditions. Key contributions include an elbow-driven method to determine $k$ and an area-based criterion for selecting $l$, a motif-based feature engineering pipeline, and evidence of generalizability to diverse EEG datasets. This work has potential to support earlier, data-driven clinical decisions by identifying treatment-responsive patterns in EEG before traditional MADRS-based assessments.

Abstract

In current medical practice, patients undergoing depression treatment must wait four to six weeks before a clinician can assess medication response due to the delayed noticeable effects of antidepressants. Identification of a treatment response at any earlier stage is of great importance, since it can reduce the emotional and economic burden connected with the treatment. We approach the prediction of a patient response to a treatment as a classification problem, by utilizing the dynamic properties of EEG recordings on the 7th day of the treatment. We present a novel framework that applies motif discovery to extract meaningful features from EEG data distinguishing between depression treatment responders and non-responders. We applied our framework also to classification tasks in other psychiatric EEG datasets, namely to patients with symptoms of schizophrenia, pediatric patients with intractable seizures, and Alzheimer disease and dementia. We achieved high classification precision in all data sets. The results demonstrate that the dynamic properties of the EEGs may support clinicians in decision making both in diagnosis and in the prediction depression treatment response as early as on the 7th day of the treatment. To our best knowledge, our work is the first one using motifs in the depression diagnostics in general.

Figures (11)

• Figure 1: Overview on our treatment prediction framework
• Figure 2: Example of motif pairs for each label (label 0, in red - non-responders for MDD and healthy subjects for other sets, and label 1 in blue - responders for MDD and pathology for other sets) from each dataset.
• Figure 3: Example plot of the elbow method of choosing the top motifs from the electrode's alpha band with a motif length of 2s, using the plot_elbowmotifletsrepo.
• Figure 4: An example output of match returning the distances to the best matches (orange) to a motif (red).
• Figure 5: Example of a motif from the beta band and with a higher difference score, together with three examples matches across the two classes.