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

Lukas Pavelka; Gelani Zelimkhanov; Carlos Vega; Michel Vaillant; Olena Tsurkalenko; Johanna Trouet; Hermann Thien; Maud Theresine; Kate Sokolowska; Amir Sharify; Stefano Sapienza; Olivia Roland; Ilsé Richard; Lucie Remark; Armin Rauschenberger; Achilleas Pexaras; Magali Perquin; Laure Pauly; Claire Pauly; Fozia Noor; Sarah Nickels; Ulf Nehrbass; Maeva Munsch; Saïda Mtimet; Michel Mittelbronn; Myriam Menster; Alexia Mendibide; Chouaib Mediouni; Deborah Mcintyre; Guilherme Marques; Tainá M. Marques; Victoria Lorentz; Zied Landoulsi; Pauline Lambert; Rejko Krüger; Olga Kofanova; Jochen Klucken; Sonja Jónsdóttir; Alexander Hundt; Margaux Henry; Estelle Henry; Anne Marie Hanff; Jérôme Graas; Marijus Giraitis; Laura Georges; Manon Gantenbein; Carlos Gamio; Joëlle Fritz; Ana Festas Lopes; Angelo Ferrari; Nancy De Bremaeker; Gessica Contesotto; David Bouvier; Ibrahim Boussaad; Guy Berchem; Sibylle Béchet; Katy Beaumont; Roxane Batutu; Myriam Alexandre; Gloria Aguayo; Geeta Acharya; NCER-PD Consortium; the ICEBERG study group

doi:10.1038/s41746-025-01862-1

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

Lukas Pavelka
, Gelani Zelimkhanov
, Carlos Vega
, Michel Vaillant
, Olena Tsurkalenko
, Johanna Trouet
, Hermann Thien
, Maud Theresine
, Kate Sokolowska
, Amir Sharify
, Stefano Sapienza
, Olivia Roland
, Ilsé Richard
, Lucie Remark
, Armin Rauschenberger
, Achilleas Pexaras
, Magali Perquin
, Laure Pauly
, Claire Pauly
, Fozia Noor

Sarah Nickels, Ulf Nehrbass, Maeva Munsch, Saïda Mtimet, Michel Mittelbronn, Myriam Menster, Alexia Mendibide, Chouaib Mediouni, Deborah Mcintyre, Guilherme Marques, Tainá M. Marques, Victoria Lorentz, Zied Landoulsi, Pauline Lambert, Rejko Krüger, Olga Kofanova, Jochen Klucken, Sonja Jónsdóttir, Alexander Hundt, Margaux Henry, Estelle Henry, Anne Marie Hanff, Jérôme Graas, Marijus Giraitis, Laura Georges, Manon Gantenbein, Carlos Gamio, Joëlle Fritz, Ana Festas Lopes, Angelo Ferrari, Nancy De Bremaeker, Gessica Contesotto, David Bouvier, Ibrahim Boussaad, Guy Berchem, Sibylle Béchet, Katy Beaumont, Roxane Batutu, Myriam Alexandre, Gloria Aguayo, Geeta Acharya, NCER-PD Consortium, the ICEBERG study group

Research output: Contribution to journal › Article › Research › peer-review

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.

Original language	English
Article number	482
Number of pages	12
Journal	npj Digital Medicine
Volume	8
Issue number	1
DOIs	https://doi.org/10.1038/s41746-025-01862-1
Publication status	Published - 26 Jul 2025

Access to Document

10.1038/s41746-025-01862-1Licence: CC BY

Cite this

Pavelka, L., Zelimkhanov, G., Vega, C., Vaillant, M., Tsurkalenko, O., Trouet, J., Thien, H., Theresine, M., Sokolowska, K., Sharify, A., Sapienza, S., Roland, O., Richard, I., Remark, L., Rauschenberger, A., Pexaras, A., Perquin, M., Pauly, L., Pauly, C., ... the ICEBERG study group (2025). Multi-cohort machine learning identifies predictors of cognitive impairment in Parkinson’s disease. npj Digital Medicine, 8(1), Article 482. https://doi.org/10.1038/s41746-025-01862-1

@article{581fac956c104ade8607649159f1209b,

title = "Multi-cohort machine learning identifies predictors of cognitive impairment in Parkinson{\textquoteright}s disease",

abstract = "Cognitive impairment is a frequent complication of Parkinson{\textquoteright}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.",

author = "Lukas Pavelka and Gelani Zelimkhanov and Carlos Vega and Michel Vaillant and Olena Tsurkalenko and Johanna Trouet and Hermann Thien and Maud Theresine and Kate Sokolowska and Amir Sharify and Stefano Sapienza and Olivia Roland and Ils{\'e} Richard and Lucie Remark and Armin Rauschenberger and Achilleas Pexaras and Magali Perquin and Laure Pauly and Claire Pauly and Fozia Noor and Sarah Nickels and Ulf Nehrbass and Maeva Munsch and Sa{\"i}da Mtimet and Michel Mittelbronn and Myriam Menster and Alexia Mendibide and Chouaib Mediouni and Deborah Mcintyre and Guilherme Marques and Marques, \{Tain{\'a} M.\} and Victoria Lorentz and Zied Landoulsi and Pauline Lambert and Rejko Kr{\"u}ger and Olga Kofanova and Jochen Klucken and Sonja J{\'o}nsd{\'o}ttir and Alexander Hundt and Margaux Henry and Estelle Henry and Hanff, \{Anne Marie\} and J{\'e}r{\^o}me Graas and Marijus Giraitis and Laura Georges and Manon Gantenbein and Carlos Gamio and Jo{\"e}lle Fritz and Lopes, \{Ana Festas\} and Angelo Ferrari and \{De Bremaeker\}, Nancy and Gessica Contesotto and David Bouvier and Ibrahim Boussaad and Guy Berchem and Sibylle B{\'e}chet and Katy Beaumont and Roxane Batutu and Myriam Alexandre and Gloria Aguayo and Geeta Acharya and \{NCER-PD Consortium\} and \{the ICEBERG study group\}",

note = "Funding: 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{\textquoteright}Entreprise EDF, and the Fondation Saint Michel, and Energipole. Publisher Copyright: {\textcopyright} The Author(s) 2025.",

year = "2025",

month = jul,

day = "26",

doi = "10.1038/s41746-025-01862-1",

language = "English",

volume = "8",

journal = "npj Digital Medicine",

issn = "2398-6352",

publisher = "Nature Publishing Group",

number = "1",

}

Pavelka, L, Zelimkhanov, G, Vega, C , Vaillant, M, Tsurkalenko, O, Trouet, J , Thien, H , Theresine, M, Sokolowska, K, Sharify, A, Sapienza, S, Roland, O, Richard, I, Remark, L , Rauschenberger, A , Pexaras, A , Perquin, M, Pauly, L, Pauly, C , Noor, F, Nickels, S, Nehrbass, U, Munsch, M , Mtimet, S, Mittelbronn, M, Menster, M, Mendibide, A, Mediouni, C, Mcintyre, D, Marques, G, Marques, TM, Lorentz, V , Landoulsi, Z , Lambert, P , Krüger, R , Kofanova, O, Klucken, J, Jónsdóttir, S , Hundt, A, Henry, M, Henry, E , Hanff, AM , Graas, J , Giraitis, M , Georges, L , Gantenbein, M , Gamio, C , Fritz, J , Lopes, AF , Ferrari, A, De Bremaeker, N, Contesotto, G , Bouvier, D , Boussaad, I , Berchem, G , Béchet, S , Beaumont, K, Batutu, R, Alexandre, M , Aguayo, G , Acharya, G, NCER-PD Consortium & the ICEBERG study group 2025, 'Multi-cohort machine learning identifies predictors of cognitive impairment in Parkinson’s disease', npj Digital Medicine, vol. 8, no. 1, 482. https://doi.org/10.1038/s41746-025-01862-1

TY - JOUR

T1 - Multi-cohort machine learning identifies predictors of cognitive impairment in Parkinson’s disease

AU - Pavelka, Lukas

AU - Zelimkhanov, Gelani

AU - Vega, Carlos

AU - Vaillant, Michel

AU - Tsurkalenko, Olena

AU - Trouet, Johanna

AU - Thien, Hermann

AU - Theresine, Maud

AU - Sokolowska, Kate

AU - Sharify, Amir

AU - Sapienza, Stefano

AU - Roland, Olivia

AU - Richard, Ilsé

AU - Remark, Lucie

AU - Rauschenberger, Armin

AU - Pexaras, Achilleas

AU - Perquin, Magali

AU - Pauly, Laure

AU - Pauly, Claire

AU - Noor, Fozia

AU - Nickels, Sarah

AU - Nehrbass, Ulf

AU - Munsch, Maeva

AU - Mtimet, Saïda

AU - Mittelbronn, Michel

AU - Menster, Myriam

AU - Mendibide, Alexia

AU - Mediouni, Chouaib

AU - Mcintyre, Deborah

AU - Marques, Guilherme

AU - Marques, Tainá M.

AU - Lorentz, Victoria

AU - Landoulsi, Zied

AU - Lambert, Pauline

AU - Krüger, Rejko

AU - Kofanova, Olga

AU - Klucken, Jochen

AU - Jónsdóttir, Sonja

AU - Hundt, Alexander

AU - Henry, Margaux

AU - Henry, Estelle

AU - Hanff, Anne Marie

AU - Graas, Jérôme

AU - Giraitis, Marijus

AU - Georges, Laura

AU - Gantenbein, Manon

AU - Gamio, Carlos

AU - Fritz, Joëlle

AU - Lopes, Ana Festas

AU - Ferrari, Angelo

AU - De Bremaeker, Nancy

AU - Contesotto, Gessica

AU - Bouvier, David

AU - Boussaad, Ibrahim

AU - Berchem, Guy

AU - Béchet, Sibylle

AU - Beaumont, Katy

AU - Batutu, Roxane

AU - Alexandre, Myriam

AU - Aguayo, Gloria

AU - Acharya, Geeta

AU - NCER-PD Consortium

AU - the ICEBERG study group

N1 - Funding: 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’Entreprise EDF, and the Fondation Saint Michel, and Energipole. Publisher Copyright: © The Author(s) 2025.

PY - 2025/7/26

Y1 - 2025/7/26

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.

UR - https://www.scopus.com/pages/publications/105022013634

UR - https://pubmed.ncbi.nlm.nih.gov/40715642/

U2 - 10.1038/s41746-025-01862-1

DO - 10.1038/s41746-025-01862-1

M3 - Article

C2 - 40715642

AN - SCOPUS:105022013634

SN - 2398-6352

VL - 8

JO - npj Digital Medicine

JF - npj Digital Medicine

IS - 1

M1 - 482

ER -

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

Abstract

Access to Document

Other files and links

Fingerprint

Cite this