Interpretable Concept-Guided Polynomial Tabular Kolmogorov-Arnold Network for EEG-Based Mild Cognitive Impairment Detection

2026-06-24

AI Summary

This study addresses the limitations of conventional approaches in early detection of mild cognitive impairment (MCI) from EEG signals—namely, the loss of neurophysiological structure in feature engineering and the poor interpretability of deep learning models—by proposing the CPTabKAN framework. The method first maps heterogeneous EEG features, such as multiscale entropy and power spectral density, onto interpretable physiological concepts. It then models intra- and inter-concept interactions using quadratic polynomials and employs a Fourier-parameterized TabKAN classifier to learn nonlinear decision boundaries. Evaluated on sleep EEG data from 372 participants, CPTabKAN achieves a weighted F1-score of 0.9038, significantly outperforming Gradient Boosting by 5.65 percentage points. The results further highlight the dominant role of specific physiological concepts and their higher-order interactions in MCI discrimination, marking the first integration of physiological concept encoding, high-order interaction modeling, and interpretable tabular learning in this domain.

Abstract

Early and scalable detection of mild cognitive impairment (MCI) remains an unresolved clinical challenge. Existing EEG-based screening approaches are constrained by handcrafted feature pipelines that discard neurophysiologically meaningful domain structure and deep learning classifiers that sacrifice interpretability for performance. No existing work unifies physiologically organized concept encoders, cross-concept interaction modeling, and nonlinear tabular classification in a sleep EEG-based MCI detection framework. This study proposes Concept-guided Polynomial-transformed Tabular learning using Kolmogorov-Arnold Network (CPTabKAN), which maps heterogeneous EEG-derived features into domain-informed concept representations, expands them via degree-2 polynomial transformation to expose first- and second-order interactions, and applies a Fourier-parameterized TabKAN classifier to learn nonlinear decision boundaries. CPTabKAN was evaluated on the Study of Osteoporotic Fractures cohort (372 subjects, overnight polysomnography), using 1,379 features organized into ten physiologically motivated concept groups. Under 10-fold cross-validation, CPTabKAN-Second Order achieved a weighted F1-score of 0.9038 (SD 0.034), outperforming GradientBoosting by 5.65 percentage points (t(9)=1.934,p=0.043, one-sided paired test), with advantages persisting under SMOTE-based balancing. Ablation analysis confirmed independent contributions from each component. Concept importance analysis revealed that power spectral density, multi-scale entropy, and Hjorth parameters dominated first-order weights, while cross-concept interactions involving Lempel-Ziv-Welch complexity, statistics, demographics, and slow oscillations exceeded all first-order scores. These results demonstrate that concept-structured, interaction-aware tabular learning surfaces physiologically coherent reasoning, supporting clinical trust.

Problem

• mild cognitive impairment
• EEG
• interpretability
• concept representation
• tabular learning

Innovation

• Concept-guided learning
• Polynomial interaction modeling
• Tabular Kolmogorov-Arnold Network
• Interpretable EEG analysis
• Mild Cognitive Impairment detection