OSCR

XGBoost Classification of Epileptic EEG Using Nonlinear Dynamical Features and SHAP.

Overview

Authors: Xiaojie Lu1,2,3, Hui Lou1, Xiaoyang Jin1, Bianmei Zhang1,2
  1. School of Medical Information, Wannan Medical University, Wuhu 241002, China; (X.L.)
  2. Anhui Province High-Quality Dataset Construction Base for Smart Healthcare, Wuhu 241002, China
  3. Institute of Medical Artificial Intelligence, Wannan Medical University, Wuhu 241002, China
Institutions: Wannan Medical College (China)
Journal: Entropy (Basel, Switzerland), volume 28, issue 8, article 920
Dates: received 14 July 2026; accepted 15 August 2026; published online 17 August 2026
Type: Research article · Language: English
License: CC BY
Identifiers: DOI 10.3390/e28080920 · PMID 42649714 · PMCID PMC13512526 · OpenAlex W7203597884
Open access: gold, a free copy (OpenAlex)
Status: code on request (how the code is found and verified)
Categories: EEG (modality), epilepsy (population), methods / tools (subfield)
Methods: Spectral & time-frequency, Statistics, Complexity, Preprocessing, Machine learning
Keywords: EEG, nonlinear dynamic, XGBoost, SHAP
Topic: EEG and Brain-Computer Interfaces (Cognitive Neuroscience, Neuroscience), according to OpenAlex
Funding: Anhui Provincial Department of Education (2023AH051757, 2025cjkc057, 2025zyxwjxalk363)
Citations: not cited yet (Europe PMC); 21 references in the paper

Abstract

To evaluate whether nonlinear descriptors of electroencephalogram (EEG) signals support interpretable XGBoost classification and to determine how analysis window duration affects performance. A secondary analysis of the public Bonn EEG dataset was performed. Nine nonlinear features were extracted from non-overlapping 1, 5, 10, and 20 s windows after an original-recording-level train/validation/test split, and a multiclass XGBoost model was interpreted with class-specific SHAP values. The model achieved 93.3% overall accuracy; the class-specific AUC values were 0.978 for Z/O, 0.978 for N/F, and 0.984 for S. Across the four fixed-split duration conditions, the 1 s condition had the lowest descriptive performance, whereas the 5, 10, and 20 s conditions were broadly comparable; no uniquely optimal duration was established. The nonlinear-feature/XGBoost framework provides interpretable benchmark segment classification evidence. Because EEG is modeled as a stochastic process and the dataset is small and heterogeneous, the SHAP attributions do not establish physiological causality or clinical diagnostic validity.

Reproduced under the paper's license (CC BY), from the paper cited above.

Code

The paper says that its authors' code is available on request: it was not published with the paper, so there is nothing to verify.

The paper's code and data availability statement is in the Data section.

Tracing map

A tracing map links a paper to the code its authors published: this paper has none (its code is available on request), so it has no map.

Data

No dataset and no data link were found in the paper.

Data Availability Statement

The EEG dataset analyzed in this study is publicly available at https://www.ukbonn.de/epileptologie/arbeitsgruppen/ag-lehnertz-neurophysik/downloads/ (accessed on 10 March 2020). The source code supporting the findings of this study is available from the corresponding author upon reasonable request.

Reproduced under the paper's license (CC BY), from the paper cited above.

Versions

The history of this record: each version stored by the harvester or made by a correction of its authors or of the maintainers of its code, and what changed in its facts. The texts of the paper (its abstract, its availability statements) are not part of it; versions that changed only those are not listed.

Version 1, 27 September 2026: the first record

Recorded: type, language, journal, volume, issue, pages, dates, 4 authors, 4 keywords, 1 funder, 20 references.

Cite

This paper

Lu, X., Lou, H., Jin, X., & Zhang, B. (2026). XGBoost Classification of Epileptic EEG Using Nonlinear Dynamical Features and SHAP. Entropy (Basel, Switzerland), 28(8), 920. https://doi.org/10.3390/e28080920

BibTeX

@article{lu2026xgboost,
author = {Lu, Xiaojie and Lou, Hui and Jin, Xiaoyang and Zhang, Bianmei},
title = {{XGBoost Classification of Epileptic EEG Using Nonlinear Dynamical Features and SHAP}},
journal = {Entropy (Basel, Switzerland)},
year = {2026},
month = aug,
volume = {28},
number = {8},
pages = {920},
publisher = {Multidisciplinary Digital Publishing Institute (MDPI)},
issn = {1099-4300},
doi = {10.3390/e28080920},
url = {https://doi.org/10.3390/e28080920},
pmid = {42649714},
pmcid = {PMC13512526}
}

RIS

TY - JOUR
AU - Lu, Xiaojie
AU - Lou, Hui
AU - Jin, Xiaoyang
AU - Zhang, Bianmei
TI - XGBoost Classification of Epileptic EEG Using Nonlinear Dynamical Features and SHAP
T2 - Entropy (Basel, Switzerland)
J2 - Entropy (Basel)
PY - 2026
DA - 2026/08/17
VL - 28
IS - 8
SP - 920
SN - 1099-4300
PB - Multidisciplinary Digital Publishing Institute (MDPI)
DO - 10.3390/e28080920
UR - https://doi.org/10.3390/e28080920
LA - en
ER -

CSL-JSON

{
"id": "10.3390/e28080920",
"type": "article-journal",
"title": "XGBoost Classification of Epileptic EEG Using Nonlinear Dynamical Features and SHAP",
"container-title": "Entropy (Basel, Switzerland)",
"author": [
{
"family": "Lu",
"given": "Xiaojie"
},
{
"family": "Lou",
"given": "Hui"
},
{
"family": "Jin",
"given": "Xiaoyang"
},
{
"family": "Zhang",
"given": "Bianmei"
}
],
"container-title-short": "Entropy (Basel)",
"volume": "28",
"issue": "8",
"page": "920",
"DOI": "10.3390/e28080920",
"PMID": "42649714",
"PMCID": "PMC13512526",
"ISSN": "1099-4300",
"publisher": "Multidisciplinary Digital Publishing Institute (MDPI)",
"URL": "https://doi.org/10.3390/e28080920",
"language": "en",
"issued": {
"date-parts": [
[
2026,
8,
17
]
]
}
}

Similar papers

The papers with a page that share the most with this one: the tools found in their code, their categories, datasets, cited references and authors, the rarest counting most.

[1] doi:10.3389/fnins.2025.1630664
Automatic detection and prediction of epileptic EEG signals based on nonlinear dynamics and deep learning: a review.
Journal: Frontiers in neuroscience
In common: epilepsy, EEG, 3 references
[2] doi:10.3390/diagnostics16162609
Automated Multimodal Sleep Staging Using DWT-Based Wavelet Decomposition and Explainable Machine Learning with Signal Sculpting Topographies.
Journal: Diagnostics (Basel, Switzerland)
In common: methods / tools, EEG, 3 references
[3] doi:10.1038/s42003-022-04331-7 [code]
Neural complexity is a common denominator of human consciousness across diverse regimes of cortical dynamics.
Journal: Communications biology
In common: EEG, 3 references
[4] doi:10.1038/s41598-026-55673-9
Patient-independent hybrid generative-discriminative modeling for seizure detection in long-term scalp EEG.
Journal: Scientific reports
In common: epilepsy, EEG, 2 references
[5] doi:10.3390/e24111540
Epileptic Seizure Detection Using Geometric Features Extracted from SODP Shape of EEG Signals and AsyLnCPSO-GA.
Journal: Entropy (Basel, Switzerland)
In common: epilepsy, EEG, 2 references
[6] doi:10.3390/s22083066
Evaluation of Feature Selection Methods for Classification of Epileptic Seizure EEG Signals.
Journal: Sensors (Basel, Switzerland)
In common: epilepsy, methods / tools, EEG, 2 references
[7] doi:10.3389/fnins.2026.1909680
A CNN-BiLSTM-GRU and attention-integrated hybrid network for epileptic seizure detection.
Journal: Frontiers in neuroscience
In common: epilepsy, EEG, 2 references
[8] doi:10.3390/neurosci5010004
Novel ML-Based Algorithm for Detecting Seizures from Single-Channel EEG.
Journal: NeuroSci
In common: epilepsy, EEG, 2 references
[9] doi:10.1002/hbm.26129
The complexity of spontaneous brain activity changes in schizophrenia, bipolar disorder, and ADHD was examined using different variations of entropy.
Journal: Human brain mapping
In common: methods / tools, 2 references
[10] doi:10.3390/bios15040240
Intra- and Inter-Regional Complexity in Multi-Channel Awake EEG Through Multivariate Multiscale Dispersion Entropy for Assessing Sleep Quality and Aging.
Journal: Biosensors
In common: EEG, 2 references

Contribute

The authors of this paper can claim it, correct its record and validate its tracing map, and the maintainers of its code (its owner, or a public member of its organization) correct what it says of their repository; anyone signed in can ask for its removal. Every request goes to OSCR's own machine, which answers it; your account page follows them.

Sign in with ORCID to claim this paper as one of its authors, correct its record or validate its tracing map: when the paper's metadata lists your ORCID iD, you are recognized at once. Maintainers of its code: sign in with GitHub, then claim the repository on your account page.

Request its removal

To ask OSCR to remove this record, the copies of its authors' scripts or its tracing map, use the removal request page: signed in, you say who you are, what to remove and why, then review and confirm the request. Published rules decide every request (how).

Discussion, reproductions, activity

Discussion: code errors and mismatches between the code and the paper about this paper and its code, from readers and its authors, beside the paper and its tracing map.

Reproductions: reports from readers who ran the authors' code and did not get the paper's result, what they ran, with which environment, commit and data.

Activity: what happens around this paper: new versions of its record, its map's validation, discussions and reproductions. It opens with sign-in.