Multi-cohort machine learning identifies predictors of cognitive impairment in Parkinson’s disease

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.

Authors

Institutions

Publication Details

Journal
npj Digital Medicine
Published
2025-07-26
DOI
https://doi.org/10.1038/s41746-025-01862-1
Citations
4
Primary Topic
Parkinson's Disease Mechanisms and Treatments
Type
article
Field-Weighted Citation Impact
5.56

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Multi-cohort machine learning identifies predictors of cognitive impairment in Parkinson’s disease

Tainá M. Marques, Armin Rauschenberger, Linda K. Hansen, Cécile Galléa et al.
4 citations
npj Digital Medicine
Parkinson's Disease Mechanisms and Treatments
5.56
article

Multi-cohort machine learning identifies predictors of cognitive impairment in Parkinson’s disease

Tainá M. Marques, Armin Rauschenberger, Linda K. Hansen, Cécile Galléa, Alexandre Bisdorff, Wim Ammerlann, Ibrahim Boussaad, Jean‐Christophe Corvol, Geeta Acharya, Olga Kofanova, Stefano Sapienza, Sonja Jónsdóttir, Joëlle V. Fritz, Deborah McIntyre, Myriam Alexandre, Rebecca Ting Jiin Loo, Smaranda Leu Semenescu, Hermann Thien, Marie Vidailhet, Myriam Menster, Maura Minelli, Chouaib Mediouni, Brian Dewitt, Clarissa P. C. Gomes, Ruxandra Soare, Michel Mittelbronn, Farid Ichou, Anne-Marie Hanff, Nadya Pyatigorskaya, Lucie Remark, Victoria Lorentz, Sarah Nickels, Nancy De Bremaeker, Jonas Ihle, Bertrand Degos, Nico J. Diederich, Jean-Paul Nicolay, Poornima Jayadev Menon, Michel Vaillant, Lukas Pavelka, Rajesh Rawal, Alexis Brice, Sylvia Herbrink, Sascha Herzinger, Christelle Laganot, Romain Valabrègue, Kate Sokolowska, Ulf Nehrbass, Piotr Gawron, Vincent Perlbarg, Sabine Schmitz, Magali Perquin, Katy Beaumont, Jérôme Graas, Achilleas Pexaras, Manon Gantenbein, Pauline Dodet, Rahul Gaurav, Patricia Martins Conde, Guy Berchem, Guilherme Fernandes Marques, Sinthuja Pachchek, Élodie Hainque, Gessica Contesotto, Elodie Thiry, Fanny Mochel, Katrin Frauenknecht, Muhammad Ali, Virginie Czernecki, Michele Bassis, Amir Sharify, Nadine Jacoby, Lorieza Castillo, David Bouvier, Roxane Batutu, Dheeraj Reddy Bobbili, Roseline Lentz, Mariella Graziano, Martine Goergen, Sophie Rivaud-Péchoux, David Grabli, Laura Georges, Patrick May, Zied Landoulsi, Enrico Glaab, Graziella Mangone, Valentin Grouès, Giuseppe Arena, Jens C. Schwamborn, Romain Nati, Françoise Meisch, Carlos Vega, Jochen Klucken, Nancy E. Ramia, Gaël Hammot, Isabelle Arnulf, Sibylle Béchet, Caroline Weill, Claire Pauly, Pauline Lambert, Laetitia Jeancolas, Rene Dondelinger, Louise‐Laure Mariani, Kirsten Roomp, Lydia Yahia‐Cherif, Paul Wilmes, Carole Dongmo-Kenfack, Michael T. Heneka, Anne Grünewald, Elisa Gómez De Lope, Benoît Colsch, Reinhard Schneider, Estelle Henry, Laure Pauly, Valentine Maheo, Evi Wollscheid-Lengeling, Olena Tsurkalenko, Alexander Hundt, Dijana Petrovska, Johanna Trouet, Olivia Roland, Alizé Chalançon, Venkata Satagopam, Gloria Aguayo, Maud Thérésine, Soumyabrata Ghosh, Liliana Vilas Boas, Julie Socha, Gelani Zelimkhanov, Marijus Giraitis, Maeva Munsch, Carlos Gamio, Alexia Mendibide, Noor Fatima, Fouad Khoury, Béatrice Nicolaı̈, Eduardo Rosales Jubal, Ana Festas Lopes, Sara Sambin, Fanny Pineau, Eve Benchetrit, Richard Lévy, Samir Bekadar, Ekaterina Soboleva, Arthur Tenenhaus, Suzanne Lesage, S Mtimet, Marie-Odile Habert, Stéphane Lehéricy, Margaux Henry, M. Uribe, Florence Cormier‐Dequaire, I Richard, Angelo Ferrari, Mickaël Lé, Manon Gomes, Rejko Krüger, Raquel Severino
article en
4 citations

Abstract

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.

npj Digital MedicineVol. 8(1)
Supélec (FR), Télécom Paris (FR), Centre National de la Recherche Scientifique (FR), Inserm (FR), Centre Hospitalier de Luxembourg (LU), University of Luxembourg (LU), Commissariat à l'Énergie Atomique et aux Énergies Alternatives (FR), Maastricht University Medical Centre (NL), CEA Paris-Saclay - Etablissement de Saclay (FR), Luxembourg School of Business (LU), Sorbonne Université (FR), Assistance Publique – Hôpitaux de Paris (FR), Luxembourg Institute of Health (LU), Pitié-Salpêtrière Hospital (FR), CEA Paris-Saclay (FR), Hôpital Avicenne (FR), Laboratoire National de Santé (LU), Luxembourg Institute of Socio-Economic Research (LU), Institut du Cerveau (FR), Hôpital de la Timone (FR)
Université du Luxembourg, EU Joint Programme – Neurodegenerative Disease Research, Agence Nationale de la Recherche, Institut National de la Santé et de la Recherche Médicale, Fonds National de la Recherche Luxembourg, Agence Nationale de Sécurité du Médicament et des Produits de Santé, Association France Parkinson
Openalex Percentile: Top 4%
Parkinson's Disease Mechanisms and Treatments
5.56
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.