Stringology-Based Motif Discovery from EEG Signals: an ADHD Case Study

2026-03-05 · Source: cs.NE updates on arXiv.org · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Data Science & Analytics, Health & Medical Research · Depth: Advanced, extended

Summary

A novel computational framework, based on stringology methods, has been developed to analyze electroencephalography (EEG) time series for identifying and characterizing recurrent temporal patterns in neural signals. This framework adapts order preserving matching (OPM) and Cartesian tree matching (CTM) to discover temporal motifs that preserve relative ordering and hierarchical relationships, invariant to amplitude scaling. Applied to a publicly available dataset of multichannel EEG recordings from 61 children with ADHD and 60 matched controls (aged 7-12 years, sampled at 128Hz, downsampled to 32Hz), the framework identified highly recurrent, group-unique motifs (support ≥ 0.9). ADHD participants showed significantly higher motif frequencies, indicating increased repetitiveness. OPM analysis revealed shorter motif lengths and greater gradient instability for the ADHD group, with larger mean and maximal inter-sample amplitude jumps. CTM analysis corroborated these findings, further revealing reduced hierarchical complexity in ADHD, characterized by shallower tree structures and fewer hierarchical levels, despite comparable motif lengths.

Key takeaway

For AI scientists developing diagnostic tools for neurodevelopmental disorders, this stringology-based EEG motif analysis offers a powerful, complementary framework. You should consider integrating OPM and CTM methods to capture fine-scale temporal instability and hierarchical complexity, which traditional spectral analyses often miss. This approach could lead to more objective biomarkers for ADHD diagnosis and monitoring, providing a nuanced view of neural dysregulation beyond amplitude or spectral power changes.

Key insights

Stringology-based motif discovery offers a novel, complementary approach to analyze EEG dynamics in neurodevelopmental disorders like ADHD.

Principles

• Relative ordering and hierarchical structure are key for EEG pattern matching.
• Motif recurrence and structural complexity differentiate neural dynamics.
• Amplitude scaling invariance improves EEG analysis robustness.

Method

The method involves preprocessing EEG signals, treating them as strings, and applying OPM and CTM with a support threshold of ≥0.9 to identify group-unique maximal motifs. Features like length, frequency, gradient dynamics, and tree structure are then extracted and statistically compared.

In practice

• Use OPM for fine-scale ordinal pattern detection in time series.
• Employ CTM for broader waveform shape analysis, tolerant to minor variations.
• Apply FDR correction for multiple comparisons in EEG feature analysis.

Topics

• EEG Signal Analysis
• ADHD
• Motif Discovery
• Order Preserving Matching
• Cartesian Tree Matching

Best for: AI Scientist, AI Researcher, Data Scientist, Research Scientist

Related on AIssential

• Fusion Learning from Dynamic Functional Connectivity: Combining the Amplitude and Phase of fMRI Signals to Identify Brain Disorders — Combining fMRI signal amplitude and phase via fusion learning improves brain disorder detection.
• Neural Response to Familiar Names Predicts Outcome of Comatose ICU Patients: A Prospective Observational Cohort Study — EEG responses to familiar names predict outcomes in comatose ICU patients.
• RECTOR: Masked Region-Channel-Temporal Modeling for Affective and Cognitive Representation Learning — RECTOR unifies region-channel-temporal EEG/sEEG representation learning via self-supervision, adapting functional regions for robust affective and cognitive disorder diagnosis.
• Feature selection leads to divergent neurobiological interpretations of brain-based machine learning biomarkers — Discarded brain features can predict phenotypes with accuracy comparable to top-ranked features, yielding divergent neurobiological interpretations.
• Modeling of ASD/TD Children's Behaviors in Interaction with a Virtual Social Robot During a Music Education Program Using Deep Neural Networks — Deep neural networks can effectively model and distinguish ASD and neurotypical child behaviors during robot interaction.
• DeepTokenEEG Enhancing Mild Cognitive Impairment and Alzheimers Classification via Tokenized EEG Features — DeepTokenEEG offers a lightweight, high-accuracy EEG-based method for Alzheimer's disease detection using spatial and temporal tokenization.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by cs.NE updates on arXiv.org.