TY - JOUR
T1 - Multi-cohort machine learning identifies predictors of cognitive impairment in Parkinson’s disease
AU - Loo, Rebecca Ting Jiin
AU - Pavelka, Lukas
AU - Mangone, Graziella
AU - Khoury, Fouad
AU - Vidailhet, Marie
AU - Corvol, Jean Christophe
AU - Glaab, Enrico
AU - Yahia-Cherif, Lydia
AU - Weill, Caroline
AU - Vidailhet, Marie
AU - Valabregue, Romain
AU - Tenenhaus, Arthur
AU - Socha, Julie
AU - Sambin, Sara
AU - Rivaud-Péchoux, Sophie
AU - Pyatigorskaya, Nadya
AU - Pineau, Fanny
AU - Petrovska, Dijana
AU - Perlbarg, Vincent
AU - Mochel, Fanny
AU - Menon, Poornima
AU - Mangone, Graziella
AU - Maheo, Valentine
AU - Levy, Richard
AU - Semenescu, Smaranda Leu
AU - Lesage, Suzanne
AU - Lehéricy, Stéphane
AU - Lé, Mickaël
AU - Mariani, Louise Laure
AU - Laganot, Christelle
AU - Jeancolas, Laetitia
AU - Ihle, Jonas
AU - Ichou, Farid
AU - Hainque, Élodie
AU - Habert, Marie Odile
AU - Grabli, David
AU - Gomes, Manon
AU - Gaurav, Rahul
AU - Gallea, Cécile
AU - Dongmo-Kenfack, Carole
AU - Dodet, Pauline
AU - Degos, Bertrand
AU - Czernecki, Virginie
AU - Corvol, Jean Christophe
AU - Cormier-Dequaire, Florence
AU - Colsch, Benoit
AU - Chalançon, Alizé
AU - Brice, Alexis
AU - Benchetrit, Eve
AU - Bekadar, Samir
AU - NCER-PD Consortium
AU - Hanff, Anne-Marie
AU - ICEBERG study group
N1 - Funding Information:
We acknowledge support by the Luxembourg National Research Fund (FNR) as part of the projects RECAST (INTER/22/17104370/RECAST), PreDYT (INTER/EJP RD22/17027921/PreDYT), AD-PLCG2 (INTER/JPND23/17999421/AD-PLCG2), and EPI_T-ALL (INTER/TRANSCAN23/18333087/EPI_T-ALL). For the purpose of open access, and in fulfillment of the obligations arising from the grant agreement, the authors have applied a Creative Commons Attribution 4.0 International (CC BY 4.0) license to any Author Accepted Manuscript version arising from this submission. The ICEBERG cohort received funding and support from the Agence Nationale de la Recherche (ANR) under grant agreements ANR-10-IAIHU-06 (IHU ICM), association France Parkinson, and the Fondation d\u2019Entreprise EDF, and the Fondation Saint Michel, and Energipole. The machine learning predictions in this paper were partly performed using the HPC facilities of the University of Luxembourg (see http://hpc.uni.lu). We are grateful to Patrick May and Zied Landoulsi for providing the genetic data that was crucial for this work. We also acknowledge the valuable contributions of the pre-publication check (PPC) team for their pre-review process (see https://r3.lcsb.uni.lu/). We would like to thank all participants of the Luxembourg Parkinson\u2019s Study for their important support of our research. Furthermore, we acknowledge the joint effort of the members of the National Centre of Excellence in Research on Parkinson\u2019s Disease (NCER-PD) Consortium from the partner institutions Luxembourg Centre for Systems Biomedicine, Luxembourg Institute of Health, Centre Hospitalier de Luxembourg, and Laboratoire National de Sant\u00E9 generally contributing to the Luxembourg Parkinson\u2019s Study. Furthermore, we extend our gratitude to the ICEBERG study group for their contribution. The DIGIPD (INTER/ERAPerMed20/14599012) research project used data collected during the C13-74 ICEBERG study sponsored by Inserm. The study was granted approval by the local Ethics Committee (\u201CComit\u00E9 de Protection des Personnes Ile-de-France VI\u201D) on August 25, 2014, authorized by the French National Agency for Medicines and Health Products Safety (ANSM) on July 11, 2014, and by the Commission Nationale Informatique et Libert\u00E9s (CNIL) on January 25, 2021. The study was registered in a public trials registry (NCT02305147). All study participants provided their informed written consent in accordance with French legal guidelines.
Publisher Copyright:
© The Author(s) 2025.
PY - 2025/12/1
Y1 - 2025/12/1
N2 - Cognitive impairment is a frequent complication of Parkinson’s disease (PD), affecting up to half of newly diagnosed patients. To improve early detection and risk assessment, we developed machine learning models using clinical data from three independent PD cohorts, which are (LuxPARK, PPMI, ICEBERG). Models were trained to predict mild cognitive impairment (PD-MCI) and subjective cognitive decline (SCD) using Explainable Artificial Intelligence (XAI) for classification and time-to-event analysis. Multi-cohort models showed greater performance stability over single-cohort models, while retaining competitive average performance. Age at diagnosis and visuospatial ability were identified as key predictors. Significant sex differences observed highlight the importance of considering sex-specific factors in cognitive assessment. Men were more likely to report SCD. Our findings highlight the potential of multi-cohort machine learning for early identification and personalized management of cognitive decline in PD.
AB - Cognitive impairment is a frequent complication of Parkinson’s disease (PD), affecting up to half of newly diagnosed patients. To improve early detection and risk assessment, we developed machine learning models using clinical data from three independent PD cohorts, which are (LuxPARK, PPMI, ICEBERG). Models were trained to predict mild cognitive impairment (PD-MCI) and subjective cognitive decline (SCD) using Explainable Artificial Intelligence (XAI) for classification and time-to-event analysis. Multi-cohort models showed greater performance stability over single-cohort models, while retaining competitive average performance. Age at diagnosis and visuospatial ability were identified as key predictors. Significant sex differences observed highlight the importance of considering sex-specific factors in cognitive assessment. Men were more likely to report SCD. Our findings highlight the potential of multi-cohort machine learning for early identification and personalized management of cognitive decline in PD.
U2 - 10.1038/s41746-025-01862-1
DO - 10.1038/s41746-025-01862-1
M3 - Article
SN - 2398-6352
VL - 8
JO - npj Digital Medicine
JF - npj Digital Medicine
IS - 1
M1 - 482
ER -