EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Machine Learning Prediction of Hoehn and Yahr Scores at 5‐Years Post‐ 123 I ‐Ioflupane SPECT Imaging in a Real‐World Parkinson's Disease Dataset

Xuehua Ye, Mette Niemann Johansen iD, Balazs Laczi iD, Mads Peter Horndrup iD, Michael Zaucha Sørensen iD, Rasmus Vestergård Madsen iD, Jan Wolber, Markus Lonsdale iD, Annemette Lokkegaard iD, Renee Miller iD

Movement Disorders Clinical Practice · 2026-11 · 13 (11) · 2843-2852

DOI: 10.1002/mdc3.70749 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Movement Disorders Clinical Practice, online erschienen am 25.07.2026. Verfasst von Xuehua Ye und 9 weiteren Autor:innen. Beteiligte Einrichtung: GE HealthCare, Chalfont Saint Giles, UK. Hauptthemen laut MEDLINE: Tomography, Emission-Computed, Single-Photon, Parkinson Disease, Nortropanes, Machine Learning. Die Arbeit nennt 38 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: GE Healthcare.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Parkinson's disease (PD) progression is highly heterogeneous, complicating clinical management and prognostication. While machine learning models have been developed using research datasets such as Parkinson's Precision Medicine Initiative (PPMI) and Parkinson's Disease Biomarkers Program (PDBP), their clinical translatability is limited due to differences in routinely collected data. The Hoehn and Yahr (H&Y) scale is commonly used in clinical practice to stage PD, yet most predictive models focus on less practical measures.

Objectives

This study developed and validated machine learning models to predict H&Y scores at 5 years post‐123I‐ioflupane single‐photon emission computed tomography (SPECT) imaging, leveraging both a real‐world dataset and a subset of the PPMI cohort. The goal was to assess the utility of routinely collected clinical and imaging data for prognostic modeling.

Methods

Data from medical records and imaging were harmonized from 343 real‐world patients and 134 PPMI patients, resulting in a merged dataset with 83 overlapping features. Random Forest and Gradient Boosting models were trained to predict 5‐year H&Y scores using varying amounts of longitudinal data and imaging features.

Results

Models using 2 years of clinical follow‐up data achieved the highest predictive accuracy. The most important predictors were early H&Y scores, gait symptom severity, and select imaging features.

Conclusions

Machine learning models can predict 5‐year H&Y scores in PD using real‐world clinical data, but imaging features add limited prognostic value. This study demonstrated that implementing machine learning models, when using real‐world data, did not significantly improve the already known gap between prognostic modeling and real‐world implementation. Improvement of models is, however, a promising prospect and further studies are encouraged.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Tomography, Emission-Computed, Single-PhotonParkinson DiseaseNortropanesMachine LearningHumansPredictive Learning ModelsFemaleMaleAgedDisease ProgressionSeverity of Illness IndexPrognosis

Stichwörter der Autor:innen

123I‐IoflupaneParkinson's diseaseclinical progressionmachine learning

Substanzen

ioflupaneNortropanes

Autor:innen und Institutionen

  • GE HealthCare, Chalfont Saint Giles, UK.
  • Department of Neurology, Bispebjerg-Frederiksberg Hospital, Copenhagen, Denmark.
  • Department of Clinical Medicine, Copenhagen University, Copenhagen, Denmark.

ORCID-Profile: Mette Niemann Johansen, Balazs Laczi, Mads Peter Horndrup, Michael Zaucha Sørensen, Rasmus Vestergård Madsen, Markus Lonsdale, Annemette Lokkegaard, Renee Miller

Förderung

  • GE Healthcare

Zitieren

APA

Ye, X., Johansen, M. N., Laczi, B., Horndrup, M. P., Sørensen, M. Z., Madsen, R. V., Wolber, J., Lonsdale, M., Lokkegaard, A., & Miller, R. (2026). Machine Learning Prediction of Hoehn and Yahr Scores at 5‐Years Post‐ 123 I ‐Ioflupane SPECT Imaging in a Real‐World Parkinson's Disease Dataset. Movement Disorders Clinical Practice, 13(11), 2843-2852. https://doi.org/10.1002/mdc3.70749

Vancouver

Ye X, Johansen MN, Laczi B, Horndrup MP, Sørensen MZ, Madsen RV, et al. Machine Learning Prediction of Hoehn and Yahr Scores at 5‐Years Post‐ 123 I ‐Ioflupane SPECT Imaging in a Real‐World Parkinson's Disease Dataset. Movement Disorders Clinical Practice. 2026;13(11):2843-2852. doi:10.1002/mdc3.70749

BibTeX
@article{Ye2026,
  author = {Ye, Xuehua and Johansen, Mette Niemann and Laczi, Balazs and Horndrup, Mads Peter and Sørensen, Michael Zaucha and Madsen, Rasmus Vestergård and Wolber, Jan and Lonsdale, Markus and Lokkegaard, Annemette and Miller, Renee},
  title = {Machine Learning Prediction of Hoehn and Yahr Scores at 5‐Years Post‐ 123 I ‐Ioflupane SPECT Imaging in a Real‐World Parkinson's Disease Dataset},
  journal = {Movement Disorders Clinical Practice},
  year = {2026},
  volume = {13},
  number = {11},
  pages = {2843--2852},
  doi = {10.1002/mdc3.70749}
}
Kostenlos starten →