Skip to content
DraconiticResearch
Search Sources API
201,683 studies31 sourcesUpdated 46 min ago
Peer-reviewedCC BY 4.0Translational Psychiatry

Information leakage and performance overestimation in EEG-based schizophrenia detection: evidence from literature and empirical analyses

  • Frigyes Samuel Racz1,2,3
  • Gabor Csukly4
4 affiliations
  1. 1Department of Physiology, Semmelweis University, Budapest, Hungary
  2. 2Department of Neurology, Dell Medical School, The University of Texas at Austin, Austin, TX 78712 USA
  3. 3Mulva Clinic for the Neurosciences, Dell Medical School, The University of Texas at Austin, Austin, TX 78712 USA
  4. 4Department of Psychiatry and Psychotherapy, Semmelweis University, Budapest, Hungary
Read full textPDFDOICite

Abstract

Detecting schizophrenia (SZ) from electroencephalography (EEG) signals using machine- and deep learning models gained traction lately due to potential utility in early disease detection and differential diagnosis. Classification performance reports in the range of 95% accuracy and above are common; however, review of state-of-the-art literature indicates that ~65% of published works involve erroneous practices in the evaluation pipeline such as epoch- instead of subject-based data splitting or ranking and selecting features before data partitioning. The consequent information leakage can result in an overestimation of SZ detection performance. Here we explicitly test this on three open SZ-EEG datasets using gold standard classification approaches in leaky and leakage-free implementations. Results indicate that information leakage can inflate SZ classification accuracy by up to ~30%. Accordingly, best practices regarding EEG-based SZ detection must be established and promoted before this technology can be further developed into a clinical decision-making tool.

Related

  1. Reaction time-related compression underlies overestimation of time-constrained reachability
    Keisuke Tani, Soma Okuuchi, Satoshi Hirose +1
    iSciencePeer-reviewedCC BY 4.0NeuroscienceNeurologyCell PressCrossref
  2. Detection of Carbapenemase-Producing Bacteria by Using an Ultra-Performance Liquid Chromatography-Tandem Mass Spectrometry Method
    Anne Carricajo, Paul Verhoeven, Salim Guezzou +2
    Antimicrobial Agents and ChemotherapyPeer-reviewedLicence not statedFull text18 citedCell biologyMicrobiologyHAL
  3. Counteracting the overestimation of effect size in mass-univariate brain-wide association studies using cross-validation
    Janik Goltermann, Nils Ralf Winter, Marius Gruber +33
    Translational PsychiatryPeer-reviewedCC BY 4.0Full text7 citedNeuroscienceCrossrefOpenAlexPMC
  4. Research on multimodal dense small object detection algorithms guided by cross-modal information
    Jiacheng Hu, Yumeng Ma, Yue Xing +5
    PLOS OnePeer-reviewedCC BY 4.0Full textNeurosciencePLOS
  5. Electroencephalography-based Alzheimer's disease detection using multi-scale information retention transform: a subject-level comparison of machine learning classifiers
    Peng Zhang, Wei Tong, Lijun Jiang +4
    Frontiers in neurologyPeer-reviewedCC BYFull textNeuroscienceEurope PMCCrossref

Cite

BibTeXRIS
BibTeX
@article{racz2026information,
  title = {{Information leakage and performance overestimation in EEG-based schizophrenia detection: evidence from literature and empirical analyses}},
  author = {Frigyes Samuel Racz and Gabor Csukly},
  journal = {Translational Psychiatry},
  year = {2026},
  month = aug,
  doi = {10.1038/s41398-026-04315-9},
  url = {https://doi.org/10.1038/s41398-026-04315-9},
  note = {CC BY 4.0},
}
RIS
TY  - JOUR
TI  - Information leakage and performance overestimation in EEG-based schizophrenia detection: evidence from literature and empirical analyses
AU  - Frigyes Samuel Racz
AU  - Gabor Csukly
PY  - 2026
DA  - 2026/08/01
JO  - Translational Psychiatry
DO  - 10.1038/s41398-026-04315-9
UR  - https://doi.org/10.1038/s41398-026-04315-9
AB  - Detecting schizophrenia (SZ) from electroencephalography (EEG) signals using machine- and deep learning models gained traction lately due to potential utility in early disease detection and differential diagnosis. Classification performance reports in the range of 95% accuracy and above are common; however, review of state-of-the-art literature indicates that ~65% of published works involve erroneous practices in the evaluation pipeline such as epoch- instead of subject-based data splitting or ranking and selecting features before data partitioning. The consequent information leakage can result in an overestimation of SZ detection performance. Here we explicitly test this on three open SZ-EEG datasets using gold standard classification approaches in leaky and leakage-free implementations. Results indicate that information leakage can inflate SZ classification accuracy by up to ~30%. Accordingly, best practices regarding EEG-based SZ detection must be established and promoted before this technology can be further developed into a clinical decision-making tool.
SN  - 2158-3188
ER  -