Explainable machine learning reveals the challenges of predicting new-onset motor complications in Parkinson’s disease
Predicting new-onset motor fluctuations and levodopa-induced dyskinesias (LID) is crucial for optimizing Parkinson’s disease management. To establish a transparent prognostic framework, we applied explainable machine learning to real-world, multicentric clinical data from 247 patients to forecast the 3-year onset of these complications. Evaluated strictly on complication-free patients, the models achieved moderate predictive power (LID MCC = 0.28; fluctuations MCC = 0.32). SHAP-based interpretability confirmed predictions aligned accurately with established clinical knowledge, driven primarily by levodopa duration and Levodopa Equivalent Daily Dose, with risk increasing significantly above a 300–400 mg threshold. Crucially, an ablation study revealed that excluding patients with pre-existing complications from training caused model sensitivity to collapse, demonstrating that the full spectrum of disease severity is essential for robust risk stratification. Ultimately, this rigorous methodological stress-test demonstrates that baseline clinical features alone yield limited absolute sensitivity, highlighting the necessity of integrating dynamic, longitudinal data to achieve clinical-grade individualized prediction.
Authors
- Venet Osmani (ORCID: https://orcid.org/0000-0001-7306-2972)
- Federica Sabatini (ORCID: https://orcid.org/0000-0003-2271-2597)
- Antonio Verrico (ORCID: https://orcid.org/0000-0002-4648-4384)
- Domenico Tortora (ORCID: https://orcid.org/0000-0002-5621-4046)
- Isabella Cama (ORCID: https://orcid.org/0000-0002-2096-4793)
- Claudia Niccolai (ORCID: https://orcid.org/0000-0003-2746-7500)
- Matteo Betti (ORCID: https://orcid.org/0009-0008-8458-3747)
- Sara Garbarino (ORCID: https://orcid.org/0000-0002-3583-3630)
- Bruno Giometto (ORCID: https://orcid.org/0000-0003-2311-9752)
- Emilio Portaccio (ORCID: https://orcid.org/0000-0002-9662-1762)
- Antonio Uccelli (ORCID: https://orcid.org/0000-0002-2008-6038)
- Guido Pasquini (ORCID: https://orcid.org/0000-0002-1811-102X)
- Monica Moroni (ORCID: https://orcid.org/0000-0003-1852-7217)
- Raffaella Di Giacopo (ORCID: https://orcid.org/0009-0009-7652-0006)
- Nicole Campese (ORCID: https://orcid.org/0000-0002-9980-3086)
- Stefano Bovo (ORCID: https://orcid.org/0000-0002-0278-6786)
- Antonella Castellano (ORCID: https://orcid.org/0000-0002-4137-9016)
- Lorenzo Gios (ORCID: https://orcid.org/0000-0003-2981-0851)
- Flavio Ragni (ORCID: https://orcid.org/0000-0001-7767-7288)
- Andrea Falini (ORCID: https://orcid.org/0000-0002-1461-8755)
- Filippo Gerli (ORCID: https://orcid.org/0000-0002-0409-2518)
- Cristina Campi (ORCID: https://orcid.org/0000-0003-2105-8554)
- Maria Chiara Malaguti (ORCID: https://orcid.org/0000-0002-4807-4063)
- Walter Endrizzi
- Andrea Rossi (ORCID: https://orcid.org/0000-0001-8575-700X)
- A. Cirone (ORCID: https://orcid.org/0000-0003-4299-8609)
- Giuseppe Jurman (ORCID: https://orcid.org/0000-0002-2705-5728)
- Costanza Parodi
- Ruggero Bacchin
- Michele Piana
- the NeuroArtP3 Network
- Donatella Ottaviani
- Chiara Longo
Institutions
- Humanitas University (IT)
- Vita-Salute San Raffaele University (IT)
- Queen Mary University of London (GB)
- University of Trento (IT)
- Fondazione Bruno Kessler (IT)
- Istituto Giannina Gaslini (IT)
- Ospedale Santa Chiara (IT)
- Provincia Autonoma di Trento (IT)
- Ospedale Santa Maria (IT)
- Don Carlo Gnocchi Foundation (IT)
- IRCCS Ospedale San Raffaele (IT)
- Ospedale Policlinico San Martino (IT)
- Istituti di Ricovero e Cura a Carattere Scientifico (IT)
- University of Florence (IT)
- University of Genoa (IT)
Publication Details
- Journal
- npj Parkinson s Disease
- Published
- 2026-09-04
- DOI
- https://doi.org/10.1038/s41531-026-01550-1
- Primary Topic
- Parkinson's Disease Mechanisms and Treatments
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Ministero della Salute