EUVIMEDExplore EUVIMED
Open access

Reducing Educational Bias in Cognitive Assessment via Dynamic Support Vector Machine Weighting: Validation Study on an Education-Stratified Dataset

Qing Liu, Chi Ma, Mengyuan Liu, Suhui Chen, Mengting Yu, Lijuan Xia, Qi Zhang, Ming Wu

JMIR Rehabilitation and Assistive Technologies · 2026-02-25 · 13 · e79841-e79841

DOI: 10.2196/79841 ↗

At a glance

Journal article in JMIR Rehabilitation and Assistive Technologies, published online on 2026-02-25. Written by Qing Liu and 7 co-authors. Institution: School of Humanities and Social Sciences, University of Science and Technology of China, Hefei, China. Cited 2 times so far (Crossref); 41 references. The full text is freely available in PubMed Central.

AbstractSubjectsBibliographic detailsCiteRelated studies

Abstract (original language)

Background

The Mini-Mental State Examination (MMSE) remains widely used for cognitive screening, yet its performance varies substantially across educational backgrounds. Linear education corrections fail to capture the nonlinear interference patterns among subitems.

Objective

This study aimed to analyze how educational level shapes MMSE subitem contributions and to develop an education-adaptive optimization strategy using support vector machine–based weighting.

Methods

MMSE data from 812 participants were stratified into 4 education groups. Subitem deletion experiments quantified each subitem’s contribution (Δ). Education-specific support vector machine models were then constructed to derive dynamic weighting coefficients. Performance improvements were assessed before and after weighting.

Results

The illiterate group relied heavily on spatial orientation and memory, whereas university-educated individuals depended more on executive and calculation functions. Several education-dependent interference items were identified (eg, visuospatial construction in the primary group and basic orientation tasks in the university group). Dynamic weighting improved accuracy in all cohorts, most notably among illiterate individuals (Δ = 7.25%; P =.06), followed by the primary school group (Δ=3.12%; P =.03).

Conclusions

Education-stratified weighting enhances the fairness and interpretability of MMSE-based screening. External validation confirmed generalizability, although multicenter studies are needed.

Subjects

Author keywords

MMSEMini-Mental State ExaminationSVMdynamic weighted modeleducational backgroundmachine learningsupport vector machine

Authors and institutions

  • School of Humanities and Social Sciences, University of Science and Technology of China, Hefei, China.
  • School of the Gifted Young, University of Science and Technology of China, Hefei, China.
  • Department of Rehabilitation Medicine, The First Affiliated Hospital of USTC, Division of Life Sciences and Medicine, University of Science and Technology of China, Tian'e Hu No.1, Hefei, 230000, China, 86 186 5510 6697.
  • Department of Rehabilitation Medicine, The Second People's Hospital, Wuhu, China.
  • Shuguang Hospital Anhui Branch Affiliated to Shanghai University of Traditional Chinese Medicine, Hefei, China.

Cite

APA

Liu, Q., Ma, C., Liu, M., Chen, S., Yu, M., Xia, L., Zhang, Q., & Wu, M. (2026). Reducing Educational Bias in Cognitive Assessment via Dynamic Support Vector Machine Weighting: Validation Study on an Education-Stratified Dataset. JMIR Rehabilitation and Assistive Technologies, 13, e79841-e79841. https://doi.org/10.2196/79841

Vancouver

Liu Q, Ma C, Liu M, Chen S, Yu M, Xia L, et al. Reducing Educational Bias in Cognitive Assessment via Dynamic Support Vector Machine Weighting: Validation Study on an Education-Stratified Dataset. JMIR Rehabilitation and Assistive Technologies. 2026;13:e79841-e79841. doi:10.2196/79841

BibTeX
@article{Liu2026,
  author = {Liu, Qing and Ma, Chi and Liu, Mengyuan and Chen, Suhui and Yu, Mengting and Xia, Lijuan and Zhang, Qi and Wu, Ming},
  title = {Reducing Educational Bias in Cognitive Assessment via Dynamic Support Vector Machine Weighting: Validation Study on an Education-Stratified Dataset},
  journal = {JMIR Rehabilitation and Assistive Technologies},
  year = {2026},
  volume = {13},
  pages = {e79841--e79841},
  doi = {10.2196/79841}
}
Start for free →