Predicting knee pain severity in professional distance runners using gaussian process regression
14 affiliations
- 1Pasinler Atatürk Regional Boarding Middle School, Erzurum Provincial Directorate of National Education, Ministry of National Education (Turkey)
- 2Department of Coaching Education, Faculty of Sports Sciences, Istanbul Rumeli University
- 3Department of Coaching Education, Faculty of Sports Sciences, Fırat University
- 4Department of Sport Management, Faculty of Sports Sciences, Hitit University
- 5Çorum Provincial Directorate of National Education, Ministry of National Education (Turkey)
- 6Department of Coaching Education, Faculty of Sport Sciences, Istanbul Esenyurt University
- 7Department of Movement and Training Sciences, Faculty of Sports Sciences, Düzce University
- 8Department of Exercise and Sport Sciences, Faculty of Sport Sciences, Avrasya University
- 9Red and River Digital Publishing and Trade Limited Company
- 10Department of Exercise and Sports Sciences for Individuals with Disabilities, Faculty of Sports Sciences, Istanbul Gelişim University
- 11Department of Sports Injuries and Corrective Exercises, Faculty of Sports Sciences, Shahid Bahonar University of Kerman
- 12HERC – Health, Exercise and Research Center
- 13Department of Physical Education, Technical and Vocational University (TVU)
- 14Institute of Human Movement Science, Sport and Health, University of Graz
Abstract
Background: Knee pain is common in runners and may precede injury; clinical labels are sparse, but repeated symptom monitoring is practical for training decisions. The purpose of this paper is to develop a probabilistic, interpretable model that predicts a continuous knee pain severity score in professional distance runners using Gaussian Process Regression (GPR), and to quantify the relative influence of routinely measurable training and anthropometric variables.
Methods: Fifty professional male distance runners aged 20–30 years were recruited from professional athletics in Ankara, Turkey. Six predictors were used: resistance training frequency, weekly running distance, training surface, running technique (foot-strike category), history of prior knee problems, and body mass index (BMI). The outcome was a continuous knee pain severity score normalized to 0–1 from a self-reported numerical rating scale. A GPR model with a radial basis function kernel was trained using a train-test split and evaluated with mean squared error (MSE).
Results: Prediction uncertainty was quantified using 95% credible intervals, and sensitivity analysis were performed to assess feature influence. The GPR model achieved low prediction error on the held-out test set [MSE = (0.02)] and stable performance under cross-validation [mean MSE = (0.028)].
Conclusion: Sensitivity analysis showed pronounced GPR response patterns for resistance training frequency and BMI, with smaller response patterns for weekly running distance and prior knee problems/injuries. Training surface and running technique showed comparatively limited sensitivity. GPR effectively models self-reported pain with uncertainty quantification and is scalable for symptom-based risk assessment, though not intended for clinical diagnosis.
Related
- Multivariate Analysis of the Spatial Distance Between the Corneal Thinnest Point and Maximum Keratometry in Keratoconic Corneas: Association with Disease Severity
- Predicting recovery trajectories and injury severity following partial crush spinal cord injury in mice
- Travel distance to tertiary healthcare reflects disease severity and medication access in Italian systemic sclerosis patients: a geographical study on adult-onset rare diseases from SPRING-SIR registry
- Can fluoxetine's diminished efficacy in pediatric depression be explained by study sites, baseline severity, age, and psychological interventions? A secondary exploratory meta-regression
- Symptom severity at first digital self-assessment is not associated with greater subsequent booking of professional care on a Saudi mental health platform: a cohort study of 5,524 users
BibTeX
@article{sadır2026predicting,
title = {{Predicting knee pain severity in professional distance runners using gaussian process regression}},
author = {Yunus Sadır and Mehmet İnan and Mustafa Karadağ and Levent Ceylan and Tülay Ceylan and Heidar Sajedi and Faik Öz and Yunus Şahinler and Mohammad Alimoradi and Mohammad Alghosi and Andreas Konrad},
journal = {Frontiers in Public Health},
year = {2026},
month = oct,
doi = {10.3389/fpubh.2026.1973629},
url = {https://www.frontiersin.org/articles/10.3389/fpubh.2026.1973629/full},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Predicting knee pain severity in professional distance runners using gaussian process regression
AU - Yunus Sadır
AU - Mehmet İnan
AU - Mustafa Karadağ
AU - Levent Ceylan
AU - Tülay Ceylan
AU - Heidar Sajedi
AU - Faik Öz
AU - Yunus Şahinler
AU - Mohammad Alimoradi
AU - Mohammad Alghosi
AU - Andreas Konrad
PY - 2026
DA - 2026/10/27
JO - Frontiers in Public Health
DO - 10.3389/fpubh.2026.1973629
UR - https://www.frontiersin.org/articles/10.3389/fpubh.2026.1973629/full
AB - Background: Knee pain is common in runners and may precede injury; clinical labels are sparse, but repeated symptom monitoring is practical for training decisions. The purpose of this paper is to develop a probabilistic, interpretable model that predicts a continuous knee pain severity score in professional distance runners using Gaussian Process Regression (GPR), and to quantify the relative influence of routinely measurable training and anthropometric variables. Methods: Fifty professional male distance runners aged 20–30 years were recruited from professional athletics in Ankara, Turkey. Six predictors were used: resistance training frequency, weekly running distance, training surface, running technique (foot-strike category), history of prior knee problems, and body mass index (BMI). The outcome was a continuous knee pain severity score normalized to 0–1 from a self-reported numerical rating scale. A GPR model with a radial basis function kernel was trained using a train-test split and evaluated with mean squared error (MSE). Results: Prediction uncertainty was quantified using 95% credible intervals, and sensitivity analysis were performed to assess feature influence. The GPR model achieved low prediction error on the held-out test set [MSE = (0.02)] and stable performance under cross-validation [mean MSE = (0.028)]. Conclusion: Sensitivity analysis showed pronounced GPR response patterns for resistance training frequency and BMI, with smaller response patterns for weekly running distance and prior knee problems/injuries. Training surface and running technique showed comparatively limited sensitivity. GPR effectively models self-reported pain with uncertainty quantification and is scalable for symptom-based risk assessment, though not intended for clinical diagnosis.
SN - 2296-2565
ER -