Physiological markers of low-altitude exercise load for predicting acute mountain sickness: development of an interpretable machine learning model and research web calculator
8 affiliations
- 1School of Electronics and Information Engineering, Tiangong University
- 2Department of Quality Control, The General Hospital of Northern Theater Command
- 3Department of Oral and Maxillofacial Surgery, The General Hospital of Northern Theater Command
- 4Department of Cardiology, The 967th Hospital of Joint Logistics Support Force of Chinese PLA
- 5Department of Critical Care Medicine, The 967th Hospital of Joint Logistics Support Force of Chinese PLA
- 6Department of Emergency, The 967th Hospital of Joint Logistics Support Force of Chinese PLA
- 7Department of Infection Control, The 967th Hospital of Joint Logistics Support Force of Chinese PLA
- 8Department of Cardiac Surgery, The General Hospital of Northern Theater Command
Abstract
Acute mountain sickness (AMS) poses a significant threat to individuals ascending to high-altitude regions, yet early identification of susceptible populations remains challenging. This study aimed to develop an interpretable machine learning model for estimating AMS risk from exercise-induced physiological responses measured at low altitude. A total of 171 healthy participants were included, with 111 participants (51 AMS events) comprising the development cohort and 60 participants (27 AMS events) forming an external validation cohort. Model development was repeated within a nested cross-validation framework, including LASSO feature selection and algorithm selection within the resampling process. The final locked five-feature model included changes in post-exercise pulse oxygen saturation (ΔSpO ₂ ) and cardiac output (ΔCO), sex, smoking history, and drinking history, and combined logistic regression, support vector machine, and random forest by equal-weight soft voting. In the development cohort, pooled outer-fold predictions yielded an AUC of 0.866 (95% CI, 0.791–0.929), an AUPRC of 0.842 (95% CI, 0.748–0.932), and a Brier score of 0.148. The calibration intercept was −0.033 (95% CI, −0.504 to 0.527), the calibration slope was 0.931 (95% CI, 0.664–1.351) and the positive net benefit range was 0.078–0.840. In the external validation cohort, the locked model yielded an AUC of 0.834 (95% CI, 0.723–0.929), an AUPRC of 0.867 (95% CI, 0.758–0.946), and a Brier score of 0.152; the calibration intercept was −0.103 (95% CI, −0.834 to 0.531), the calibration slope was 1.013 (95% CI, 0.655–1.666) and a positive net benefit range of 0.200–0.953. Post-hoc SHAP analysis identified ΔSpO ₂ and ΔCO as the two most influential predictors. These findings support an association between low-altitude exercise responses and subsequent AMS. However, the prediction model requires further development and validation in substantially larger cohorts before clinical deployment.
Related
- Development and validation of an interpretable machine learning model for predicting in-hospital mortality among critically ill patients with liver cirrhosis
- Development and Validation of an Interpretable Machine‐Learning Model for Predicting Suboptimal Clinical Response After Sleeve Gastrectomy: A Multicenter Study
- An interpretable machine learning model for predicting intensive care unit mortality risk in elderly patients with concomitant cardiogenic shock and sepsis: Development and external validation using two large critical care databases
- Development and validation of interpretable machine learning models for predicting cumulative live birth rates in assisted reproductive technology: a SHAP-based analysis.
- Development and internal validation of an interpretable machine learning model for predicting vancomycin-induced nephrotoxicity in hospitalized children
BibTeX
@article{miao2026physiological,
title = {{Physiological markers of low-altitude exercise load for predicting acute mountain sickness: development of an interpretable machine learning model and research web calculator}},
author = {Fengbo Miao and Yi Yuan and Shuai Wang and Chengli Tao and Yupeng Liu and Qi Mao and Hualin Zhang and Ning Li and Renzheng Chen and Jianfu Sui},
journal = {Frontiers in Public Health},
year = {2026},
month = nov,
doi = {10.3389/fpubh.2026.1920845},
url = {https://www.frontiersin.org/articles/10.3389/fpubh.2026.1920845/full},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Physiological markers of low-altitude exercise load for predicting acute mountain sickness: development of an interpretable machine learning model and research web calculator
AU - Fengbo Miao
AU - Yi Yuan
AU - Shuai Wang
AU - Chengli Tao
AU - Yupeng Liu
AU - Qi Mao
AU - Hualin Zhang
AU - Ning Li
AU - Renzheng Chen
AU - Jianfu Sui
PY - 2026
DA - 2026/11/02
JO - Frontiers in Public Health
DO - 10.3389/fpubh.2026.1920845
UR - https://www.frontiersin.org/articles/10.3389/fpubh.2026.1920845/full
AB - Acute mountain sickness (AMS) poses a significant threat to individuals ascending to high-altitude regions, yet early identification of susceptible populations remains challenging. This study aimed to develop an interpretable machine learning model for estimating AMS risk from exercise-induced physiological responses measured at low altitude. A total of 171 healthy participants were included, with 111 participants (51 AMS events) comprising the development cohort and 60 participants (27 AMS events) forming an external validation cohort. Model development was repeated within a nested cross-validation framework, including LASSO feature selection and algorithm selection within the resampling process. The final locked five-feature model included changes in post-exercise pulse oxygen saturation (ΔSpO ₂ ) and cardiac output (ΔCO), sex, smoking history, and drinking history, and combined logistic regression, support vector machine, and random forest by equal-weight soft voting. In the development cohort, pooled outer-fold predictions yielded an AUC of 0.866 (95% CI, 0.791–0.929), an AUPRC of 0.842 (95% CI, 0.748–0.932), and a Brier score of 0.148. The calibration intercept was −0.033 (95% CI, −0.504 to 0.527), the calibration slope was 0.931 (95% CI, 0.664–1.351) and the positive net benefit range was 0.078–0.840. In the external validation cohort, the locked model yielded an AUC of 0.834 (95% CI, 0.723–0.929), an AUPRC of 0.867 (95% CI, 0.758–0.946), and a Brier score of 0.152; the calibration intercept was −0.103 (95% CI, −0.834 to 0.531), the calibration slope was 1.013 (95% CI, 0.655–1.666) and a positive net benefit range of 0.200–0.953. Post-hoc SHAP analysis identified ΔSpO ₂ and ΔCO as the two most influential predictors. These findings support an association between low-altitude exercise responses and subsequent AMS. However, the prediction model requires further development and validation in substantially larger cohorts before clinical deployment.
SN - 2296-2565
ER -