Skip to content
DraconiticResearch
Search Sources API
197,768 studies31 sourcesUpdated 4 min ago
Peer-reviewedCC BYFrontiers in neurology

Electroencephalography-based Alzheimer's disease detection using multi-scale information retention transform: a subject-level comparison of machine learning classifiers

  • Peng Zhang1
  • Wei Tong2
  • Lijun Jiang1
  • Jing Kan2
  • Wenjie Zeng3
  • Bicheng Wu2
  • Haisheng Zhang4
4 affiliations
  1. 1Affiliated Xiaoshan Hospital, Hangzhou Normal University, Hangzhou , ,
  2. 2Laboratory of Intelligent Brain Neuroimaging, Advanced Institute of Information Technology (AIIT), Hangzhou , ,
  3. 3School of Statistics and Data Science, Zhejiang Gongshang University, Hangzhou , ,
  4. 4Department of Clinical Psychology, Affiliated Hangzhou First People's Hospital, Westlake University School of Medicine, Hangzhou , ,
Read full textPDFDOICite

Abstract

Introduction: This exploratory study evaluates Multi-Scale Information Retention Transform (MSIRT), a feature-fusion framework for electroencephalography (EEG)-based Alzheimer's disease (AD) classification.

Methods: Ratios between differential entropies were removed, and all EEG amplitudes were converted to volts before feature extraction because the transform is not invariant to arbitrary physical units. Revised MSIRT combines fuzzy entropy (FE), permutation entropy (PE), and differential entropy (DE) using training-fold normalization and cross-feature products. The framework was evaluated on OpenNeuro ds004504 (29 healthy controls and 36 participants with AD) using participant-level cross-validation.

Results: With Random Forest (RF) held fixed, direct concatenation achieved 84.62% accuracy and an area under the receiver operating characteristic curve (AUC) of 0.8956, whereas revised MSIRT achieved 86.15% accuracy and an AUC of 0.9004. An interaction-only control achieved 80.00% accuracy and an AUC of 0.8908. Because the no-product transform is feature-wise strictly monotonic, it was treated as an RF invariance control rather than a source of new predictive information. The paired AUC difference between revised MSIRT and concatenation was 0.0048 (bootstrap 95% confidence interval, −0.0067 to 0.0192; paired permutation p = 0.3780). In 20 sex-balanced resamples, revised MSIRT had a mean AUC of 0.8846.

Discussion: These results do not establish statistical superiority or clinical utility.

Related

  1. Decoding object categories from electroencephalography during free viewing reveals early information evolution compared to passive viewing.
    Carmel R Auerbach-Asch, Gal Vishne, Oded Wertheimer +1
    Journal of visionPeer-reviewedCC BY-NC-NDFull textNeuroscienceEurope PMC
  2. A comparison of artificial neural networks and partial least squares modelling for the rapid detection of the microbial spoilage of beef fillets based on Fourier transform infrared spectral fingerprints
    Efstathios Panagou, Fady Mohareb, Anthoula Argyri +2
    Food MicrobiologyPeer-reviewedLicence not statedFull text76 citedGeneral biologyStatisticsHAL
  3. Single molecule detection of Tau seeds in extracellular vesicles from Alzheimer's disease brains
    Dalya Gulseren, Anaïs Bécot, Bank Neuroceb Brain +2
    HALPreprintCC BY 4.0Full textGeneral biochemistryCell biologyHALbioRxiv
  4. Large-scale head-to-head comparison of on- and off-target profiles of [18F]Tauvid and [18F]MK6240 tau-PET in aging and Alzheimer’s disease
    Cécile Tissot, Hsin-Yeh Tsai, Ji Yeon Lee +24
    Research SquarePreprintCC BY 4.0Full textNeurosciencePMC
  5. Distinct electroencephalography alpha-band rhythms map macroscale brain gradients
    Cerebral cortex (New York, N.Y. : 1991)Peer-reviewedCC BY-NCFull textNeuroscienceEurope PMC

Cite

BibTeXRIS
BibTeX
@article{zhang2026electroencephalographybased,
  title = {{Electroencephalography-based Alzheimer's disease detection using multi-scale information retention transform: a subject-level comparison of machine learning classifiers}},
  author = {Peng Zhang and Wei Tong and Lijun Jiang and Jing Kan and Wenjie Zeng and Bicheng Wu and Haisheng Zhang},
  journal = {Frontiers in neurology},
  year = {2026},
  month = sep,
  doi = {10.3389/fneur.2026.1869211},
  url = {https://europepmc.org/article/PMC/PMC13661320},
  note = {CC BY},
}
RIS
TY  - JOUR
TI  - Electroencephalography-based Alzheimer's disease detection using multi-scale information retention transform: a subject-level comparison of machine learning classifiers
AU  - Peng Zhang
AU  - Wei Tong
AU  - Lijun Jiang
AU  - Jing Kan
AU  - Wenjie Zeng
AU  - Bicheng Wu
AU  - Haisheng Zhang
PY  - 2026
DA  - 2026/09/30
JO  - Frontiers in neurology
DO  - 10.3389/fneur.2026.1869211
UR  - https://europepmc.org/article/PMC/PMC13661320
AB  - Introduction: This exploratory study evaluates Multi-Scale Information Retention Transform (MSIRT), a feature-fusion framework for electroencephalography (EEG)-based Alzheimer's disease (AD) classification.  Methods: Ratios between differential entropies were removed, and all EEG amplitudes were converted to volts before feature extraction because the transform is not invariant to arbitrary physical units. Revised MSIRT combines fuzzy entropy (FE), permutation entropy (PE), and differential entropy (DE) using training-fold normalization and cross-feature products. The framework was evaluated on OpenNeuro ds004504 (29 healthy controls and 36 participants with AD) using participant-level cross-validation.  Results: With Random Forest (RF) held fixed, direct concatenation achieved 84.62% accuracy and an area under the receiver operating characteristic curve (AUC) of 0.8956, whereas revised MSIRT achieved 86.15% accuracy and an AUC of 0.9004. An interaction-only control achieved 80.00% accuracy and an AUC of 0.8908. Because the no-product transform is feature-wise strictly monotonic, it was treated as an RF invariance control rather than a source of new predictive information. The paired AUC difference between revised MSIRT and concatenation was 0.0048 (bootstrap 95% confidence interval, −0.0067 to 0.0192; paired permutation p = 0.3780). In 20 sex-balanced resamples, revised MSIRT had a mean AUC of 0.8846.  Discussion: These results do not establish statistical superiority or clinical utility.
SN  - 1664-2295
ER  -