Explainable machine learning identifies predictors for postural instability and cognitive impairment in early Parkinsons Disease
- 1University of Tuebingen
Abstract
Parkinsons disease (PD) exhibits considerable variability in symptom onset, severity, and clinical trajectories, making prediction of individual disease progression a complex challenge. Leveraging large multimodal datasets from two independent cohorts of early PD patients, we examined predictive features for progression to mild cognitive impairment (MCI) and postural instability (PI) using Explainable Boosting Machines (EBM) - an interpretable machine learning (ML) approach. We found comparable predictive performances of EBM models using a minimal clinico-demographic dataset compared to an extended multimodal dataset including volumetric brain imaging and fluid biomarkers. Baseline age and Montreal Cognitive Assessment total score were identified as most important predictive features for both MCI and PI using the minimal dataset, while additional motor, cognitive and imaging features were identified using the extended dataset. Our study highlights the potential of ML predictions of cognitive and motor endpoints of PD progression at the individual level bringing personalized clinical decision-making within reach.
Related
- Biomechanical and movement-quality predictors of lower-limb load-related injury in collegiate athletes: a prospective cohort study using functional movement screening, wearable inertial sensing, and explainable machine learning
- Stage-Stratified Explainable Machine Learning Identifies a Reproducible Transcriptomic Classifier with Stage-Dependent Feature Importance in Colorectal Cancer: A Multi-Cohort Study with Saudi Population Contextualisation
- 482 Biomarker analysis from KESTREL identifies tumor cytotoxicity and immunogenic cell death as predictors of benefit with durvalumab plus tremelimumab in HNSCC
- Interpretable machine learning identifies predictors of IGF-1 during growth hormone therapy in boys
- A multi-omics analysis of advanced triple-negative breast cancer identifies predictors to chemo-immunotherapy and on-treatment dynamics
BibTeX
@misc{sailer2026explainable,
title = {{Explainable machine learning identifies predictors for postural instability and cognitive impairment in early Parkinsons Disease}},
author = {B. Sailer and B. Roeben and C. Kiefer and L.-C. Achauer and R. Kemmner and I. Wurster and S. Lerche and B. Bender and T. Lindig and C. Schulte and C. Deuschle and C. Erhardt and T. Gasser and K. Brockmann and O. Kohlbacher and S. Biergans and I. Liepelt-Scarfone},
howpublished = {medRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.30.26364120},
url = {https://www.medrxiv.org/content/10.64898/2026.09.30.26364120v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - Explainable machine learning identifies predictors for postural instability and cognitive impairment in early Parkinsons Disease
AU - B. Sailer
AU - B. Roeben
AU - C. Kiefer
AU - L.-C. Achauer
AU - R. Kemmner
AU - I. Wurster
AU - S. Lerche
AU - B. Bender
AU - T. Lindig
AU - C. Schulte
AU - C. Deuschle
AU - C. Erhardt
AU - T. Gasser
AU - K. Brockmann
AU - O. Kohlbacher
AU - S. Biergans
AU - I. Liepelt-Scarfone
PY - 2026
DA - 2026/09/30
PB - medRxiv
DO - 10.64898/2026.09.30.26364120
UR - https://www.medrxiv.org/content/10.64898/2026.09.30.26364120v1
AB - Parkinsons disease (PD) exhibits considerable variability in symptom onset, severity, and clinical trajectories, making prediction of individual disease progression a complex challenge. Leveraging large multimodal datasets from two independent cohorts of early PD patients, we examined predictive features for progression to mild cognitive impairment (MCI) and postural instability (PI) using Explainable Boosting Machines (EBM) - an interpretable machine learning (ML) approach. We found comparable predictive performances of EBM models using a minimal clinico-demographic dataset compared to an extended multimodal dataset including volumetric brain imaging and fluid biomarkers. Baseline age and Montreal Cognitive Assessment total score were identified as most important predictive features for both MCI and PI using the minimal dataset, while additional motor, cognitive and imaging features were identified using the extended dataset. Our study highlights the potential of ML predictions of cognitive and motor endpoints of PD progression at the individual level bringing personalized clinical decision-making within reach.
ER -