Self-Organized Explainable Symbolic EEG Feature Engineering Using Directed Lobish and Previous-Next Pattern: A Proof-of-Concept Study on Dementia Classification
10 affiliations
- 1Department of Neurology, School of Medicine, Firat University, 23119 Elazig, Türkiye
- 2Faculty of Engineering and Information Technology, University of Technology Sydney, Sydney, NSW 2007, Australia
- 3Australian International Institute of Higher Education, Brisbane, QLD 4000, Australia
- 4Department of Computer Engineering, College of Engineering, Erzurum Technical University, 25050 Erzurum, Türkiye
- 5School of Mathematics, Physics and Computing, University of Southern Queensland, Toowoomba, QLD 4350, Australia
- 6School of Computing and Information Technology, University of Wollongong, Wollongong, NSW 2522, Australia
- 7School of Population Health, UNSW Sydney, Kensington, NSW 2052, Australia
- 8Department of Digital Forensics Engineering, Technology Faculty, Firat University, 23119 Elazig, Türkiye
- 9Vocational School of Technical Sciences, Firat University, 23119 Elazig, Türkiye
- 10Faculty of Advanced Science and Technology, Kumamoto University, Kumamoto 860-8555, Japan
Abstract
Objective: Dementia is a prevalent neurological disorder, and accurate detection is essential for timely intervention. This study aims to design an interpretable symbolic feature engineering framework for EEG-based dementia classification and to examine how its performance depends on the validation protocol. Methods: We used a publicly available EEG dementia dataset including Alzheimer’s disease (AD), frontotemporal dementia (FD), and healthy controls. A self-organized explainable feature engineering (SOXFE) model was introduced with five phases: (i) feature extraction using Bipartite Channel Transformer (BCT) and Previous-Next Pattern (PNPat), (ii) feature selection via cumulative weight-based iterative neighborhood component analysis (CWINCA), (iii) explainable output generation using Directed Lobish (DLob), (iv) classification through an ensemble tkNN classifier, and (v) information fusion with combinational majority voting (CMV). Performance was evaluated with segment-wise 10-fold cross-validation, leave-one-subject-out (LOSO) cross-validation, and nested participant-grouped 10-fold cross-validation. Results: Segment-wise 10-fold cross-validation, which allowed segments of one participant in both training and test folds, yielded accuracies of 99.12–99.74%. Participant-independent accuracy was substantially lower: 51.19–63.81% for LOSO and 49.00–62.50% for nested participant-grouped validation at the segment level. Participant-level LOSO accuracy equaled the majority-class rate for dementia versus healthy controls (67.05%) and fell below it for AD versus FD (57.63% versus 61.02%). No exact McNemar comparison showed an accuracy advantage over the majority-class baseline. DLob outputs were produced as feature-level visualizations. Conclusions: The findings support the feasibility of interpretable, algorithm-derived symbolic feature representations. They do not establish reproducible disease-related EEG characteristics or clinical diagnostic utility. DLob outputs describe the channel transitions used by the model. Independent cohorts and stronger participant-independent performance are required before clinical application.
Related
- An Explainable Disease-Aware Intelligent Histopathological Representation Learning Framework for Generalizable Breast Cancer Classification
- Multimodal fusion of SpO₂ time series and clinical data for explainable classification of obstructive sleep apnea severity
- CERVEX: A Foundation-Model Framework With Built-In Explainable AI for Automated Cervical Cytology Classification From Pap Smear Images
- Engineering T cell-selective chemokine inhibitors through network-level directed evolution
- Parkinson’s disease assessment with selfie video via feature engineering and explainable AI
BibTeX
@article{balgetir2026selforganized,
title = {{Self-Organized Explainable Symbolic EEG Feature Engineering Using Directed Lobish and Previous-Next Pattern: A Proof-of-Concept Study on Dementia Classification}},
author = {Ferhat Balgetir and Prabal Datta Barua and Jun Shen and Padmanesan Narasimhan and Smitha Shivshankar and Sengul Dogan and Irem Tasci and Burak Tasci and Mehmet Baygin and Turker Tuncer and Makiko Kobayashi},
journal = {Journal of Clinical Medicine},
year = {2026},
month = nov,
doi = {10.3390/jcm15218572},
url = {https://www.mdpi.com/2077-0383/15/21/8572},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Self-Organized Explainable Symbolic EEG Feature Engineering Using Directed Lobish and Previous-Next Pattern: A Proof-of-Concept Study on Dementia Classification
AU - Ferhat Balgetir
AU - Prabal Datta Barua
AU - Jun Shen
AU - Padmanesan Narasimhan
AU - Smitha Shivshankar
AU - Sengul Dogan
AU - Irem Tasci
AU - Burak Tasci
AU - Mehmet Baygin
AU - Turker Tuncer
AU - Makiko Kobayashi
PY - 2026
DA - 2026/11/03
JO - Journal of Clinical Medicine
DO - 10.3390/jcm15218572
UR - https://www.mdpi.com/2077-0383/15/21/8572
AB - Objective: Dementia is a prevalent neurological disorder, and accurate detection is essential for timely intervention. This study aims to design an interpretable symbolic feature engineering framework for EEG-based dementia classification and to examine how its performance depends on the validation protocol. Methods: We used a publicly available EEG dementia dataset including Alzheimer’s disease (AD), frontotemporal dementia (FD), and healthy controls. A self-organized explainable feature engineering (SOXFE) model was introduced with five phases: (i) feature extraction using Bipartite Channel Transformer (BCT) and Previous-Next Pattern (PNPat), (ii) feature selection via cumulative weight-based iterative neighborhood component analysis (CWINCA), (iii) explainable output generation using Directed Lobish (DLob), (iv) classification through an ensemble tkNN classifier, and (v) information fusion with combinational majority voting (CMV). Performance was evaluated with segment-wise 10-fold cross-validation, leave-one-subject-out (LOSO) cross-validation, and nested participant-grouped 10-fold cross-validation. Results: Segment-wise 10-fold cross-validation, which allowed segments of one participant in both training and test folds, yielded accuracies of 99.12–99.74%. Participant-independent accuracy was substantially lower: 51.19–63.81% for LOSO and 49.00–62.50% for nested participant-grouped validation at the segment level. Participant-level LOSO accuracy equaled the majority-class rate for dementia versus healthy controls (67.05%) and fell below it for AD versus FD (57.63% versus 61.02%). No exact McNemar comparison showed an accuracy advantage over the majority-class baseline. DLob outputs were produced as feature-level visualizations. Conclusions: The findings support the feasibility of interpretable, algorithm-derived symbolic feature representations. They do not establish reproducible disease-related EEG characteristics or clinical diagnostic utility. DLob outputs describe the channel transitions used by the model. Independent cohorts and stronger participant-independent performance are required before clinical application.
SN - 2077-0383
ER -