A Longitudinal Stacked LSTM-Based Framework for Predicting Parkinson’s Disease Motor Progression through Clinical Assessments
Procedia Computer Science · 2026 · 283 · 1571-1580
DOI: 10.1016/j.procs.2026.06.234 ↗Kurz gefasst
Fachartikel in der Zeitschrift Procedia Computer Science, 2026. Verfasst von Munzir Hubiba Bashir und 1 weiteren Autor:innen. Die Arbeit nennt 28 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).
Abstract (Originalsprache)
Parkinson’s disease is a worsening neurodegenerative condition marked by diverse and evolving motor as well as non-motor symptoms, complicating the prediction of individual disease trajectories. Accurate forecasting of Parkinson’s progression remains an unmet clinical need due to substantial patient heterogeneity and inherent data imbalance across severity levels. In this study, a stacked Long Short-Term Memory network is developed to model longitudinal assessments acquired through the Parkinson’s Progression Markers Initiative dataset, capturing temporal dependencies and symptom dynamics across multiple patient visits. This dataset is a multi-centre and extensive observational study that collects comprehensive clinical, imaging, and bio-sample data to advance the discovery in Parkinson’s disease. On test data, the model achieved a 0.0805 mean absolute error, a 0.1045 root mean square error, and a 0.4323 coefficient of determination, demonstrating consistent performance and generalization between training and test sets. The outcomes highlight the promise of temporal deep learning approaches for generating reliable predictive insights into PD prognosis.
Zitieren
APA
Bashir, M. H., & Shaikh, T. A. (2026). A Longitudinal Stacked LSTM-Based Framework for Predicting Parkinson’s Disease Motor Progression through Clinical Assessments. Procedia Computer Science, 283, 1571-1580. https://doi.org/10.1016/j.procs.2026.06.234
Vancouver
Bashir MH, Shaikh TA. A Longitudinal Stacked LSTM-Based Framework for Predicting Parkinson’s Disease Motor Progression through Clinical Assessments. Procedia Computer Science. 2026;283:1571-1580. doi:10.1016/j.procs.2026.06.234
BibTeX
@article{Bashir2026,
author = {Bashir, Munzir Hubiba and Shaikh, Tawseef Ayoub},
title = {A Longitudinal Stacked LSTM-Based Framework for Predicting Parkinson’s Disease Motor Progression through Clinical Assessments},
journal = {Procedia Computer Science},
year = {2026},
volume = {283},
pages = {1571--1580},
doi = {10.1016/j.procs.2026.06.234}
}
EUVIMED