Interpretable machine learning for identifying mild cognitive impairment among Chinese patients with obstructive sleep apnea: a multicenter cross-sectional study
6 affiliations
- 1College of Management, Changchun University of Chinese Medicine, No. 1035 Boshuo Road, Changchun Jingyue National High-tech Industrial Development Zone, Changchun, Jilin Province 130117 China
- 2Liaoning University of Traditional Chinese Medicine, No. 79 Chongshan East Road, Huanggu District, Shenyang, Liaoning Province 110847 China
- 3Department of Accounting and Finance, Heilongjiang Polytechnic, No. 5 Xuefu Road, Nangang District, Harbin, Heilongjiang Province 150080 China
- 4College of Traditional Chinese Medicine, Changchun University of Chinese Medicine, No. 1035 Boshuo Road, Jingyue National High-tech Industrial Development Zone, Changchun, Jilin Province 130117 China
- 5Psychology and Health Management Center, Harbin Medical University, No. 157 Baojian Road, Nangang District, Harbin, Heilongjiang Province 150078 China
- 6Department of General Practice, The Affiliated Luohu Hospital of Shenzhen University Medical School, No. 47 Youyi Road, Luohu District, Shenzhen, Guangdong Province 518005 China
Abstract
Background: Mild cognitive impairment (MCI) is clinically relevant in obstructive sleep apnea (OSA). We developed and externally validated an interpretable machine-learning model for concurrent clinically defined MCI to support prioritization for formal cognitive assessment rather than diagnosis.
Methods: This multicenter cross-sectional diagnostic-prediction study included 400 eligible patients with OSA from Changchun, divided into a training set (n = 280) and an internal algorithm-selection hold-out (n = 120), plus an independent Shenzhen external-validation sample (n = 192). MCI classification integrated reported cognitive decline, objective cognitive evidence, largely preserved everyday independence, dementia exclusion, and assessment-team consensus with physician participation. Of 27 candidate predictors, redundancy review left 24 and LASSO-Boruta consensus retained nine. Feature selection was performed once on the full training set before repeated cross-validation and was not nested within resampling folds. Eight algorithms were tuned using repeated 10-fold cross-validation with training-fold-only SMOTE and PR-AUC as the primary metric. The RF threshold (0.3228) was derived from training out-of-fold probabilities and locked before validation.
Results: MCI was classified in 93/280 (33.2%), 38/120 (31.7%), and 57/192 (29.7%) participants. The nine RF predictors were coffee consumption, physical activity, AHI, MSaO2, PHQ-9, GAD-7, ESS, CPSS-14, and PSQI total scores. In the internal hold-out used for algorithm selection, RF achieved ROC-AUC 0.953, PR-AUC 0.936, and Brier score 0.060; these selection-conditioned estimates are not unbiased generalization estimates. In independent external validation, ROC-AUC was 0.957, PR-AUC 0.809, and Brier score 0.051. Calibration intercept − 0.616 and slope 0.701 indicated systematic probability overestimation despite strong discrimination.
Conclusions: The RF model discriminated strongly in Shenzhen, but external miscalibration and systematic probability overestimation mean that high AUC does not make raw probabilities transportable; local calibration assessment and, if needed, recalibration are required before clinical use. The MCI outcome cannot be reproduced from questionnaire/scale scores alone, and Supplementary Table S10 is a retrospective operational specification rather than the original CRF. Performance evidence comes almost entirely from male OSA patients; validity in women is unestablished, and the model should not be used for clinical assessment in females. It is a research screening-prioritization tool, not a diagnostic, causal, prognostic, or clinical-impact instrument.
Supplementary Information: The online version contains supplementary material available at https://doi.org/10.1186/s12888-026-08712-8.
Related
- Radiomics-based interpretable machine learning model for identifying high-risk patients with small bowel obstruction: a multicenter study.
- Identifying Mild Cognitive Impairment from Routinely Collected Physical, Functional and Psychosocial Measures in Community-Dwelling Older Adults: A Comparative Machine-Learning and Regression Modelling Study in the Northern Japanese ORANGE Registry.
- Qualitative clock-drawing errors across dementia etiologies and mild cognitive impairment: a clinically interpretable six-class framework
- Development and validation of a risk stratification model for mild cognitive impairment among older adults with possible sarcopenia integrating traditional Chinese medicine constitution: a multicenter cross-sectional study
- Development and validation of an interpretable machine learning model for early prediction of mortality in ICU patients with chronic obstructive pulmonary disease and heart failure
BibTeX
@article{li2026interpretable,
title = {{Interpretable machine learning for identifying mild cognitive impairment among Chinese patients with obstructive sleep apnea: a multicenter cross-sectional study}},
author = {Enguang Li and Fangzhu Ai and Peipei Cai and Kuo Wen and Botang Guo and Hongjuan Wen},
journal = {BMC Psychiatry},
year = {2026},
month = oct,
doi = {10.1186/s12888-026-08712-8},
url = {https://doi.org/10.1186/s12888-026-08712-8},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - JOUR
TI - Interpretable machine learning for identifying mild cognitive impairment among Chinese patients with obstructive sleep apnea: a multicenter cross-sectional study
AU - Enguang Li
AU - Fangzhu Ai
AU - Peipei Cai
AU - Kuo Wen
AU - Botang Guo
AU - Hongjuan Wen
PY - 2026
DA - 2026/10/11
JO - BMC Psychiatry
DO - 10.1186/s12888-026-08712-8
UR - https://doi.org/10.1186/s12888-026-08712-8
AB - Background: Mild cognitive impairment (MCI) is clinically relevant in obstructive sleep apnea (OSA). We developed and externally validated an interpretable machine-learning model for concurrent clinically defined MCI to support prioritization for formal cognitive assessment rather than diagnosis. Methods: This multicenter cross-sectional diagnostic-prediction study included 400 eligible patients with OSA from Changchun, divided into a training set (n = 280) and an internal algorithm-selection hold-out (n = 120), plus an independent Shenzhen external-validation sample (n = 192). MCI classification integrated reported cognitive decline, objective cognitive evidence, largely preserved everyday independence, dementia exclusion, and assessment-team consensus with physician participation. Of 27 candidate predictors, redundancy review left 24 and LASSO-Boruta consensus retained nine. Feature selection was performed once on the full training set before repeated cross-validation and was not nested within resampling folds. Eight algorithms were tuned using repeated 10-fold cross-validation with training-fold-only SMOTE and PR-AUC as the primary metric. The RF threshold (0.3228) was derived from training out-of-fold probabilities and locked before validation. Results: MCI was classified in 93/280 (33.2%), 38/120 (31.7%), and 57/192 (29.7%) participants. The nine RF predictors were coffee consumption, physical activity, AHI, MSaO2, PHQ-9, GAD-7, ESS, CPSS-14, and PSQI total scores. In the internal hold-out used for algorithm selection, RF achieved ROC-AUC 0.953, PR-AUC 0.936, and Brier score 0.060; these selection-conditioned estimates are not unbiased generalization estimates. In independent external validation, ROC-AUC was 0.957, PR-AUC 0.809, and Brier score 0.051. Calibration intercept − 0.616 and slope 0.701 indicated systematic probability overestimation despite strong discrimination. Conclusions: The RF model discriminated strongly in Shenzhen, but external miscalibration and systematic probability overestimation mean that high AUC does not make raw probabilities transportable; local calibration assessment and, if needed, recalibration are required before clinical use. The MCI outcome cannot be reproduced from questionnaire/scale scores alone, and Supplementary Table S10 is a retrospective operational specification rather than the original CRF. Performance evidence comes almost entirely from male OSA patients; validity in women is unestablished, and the model should not be used for clinical assessment in females. It is a research screening-prioritization tool, not a diagnostic, causal, prognostic, or clinical-impact instrument. Supplementary Information: The online version contains supplementary material available at https://doi.org/10.1186/s12888-026-08712-8.
SN - 1471-244X
ER -