Smartphone-based prediction of dopaminergic deficit in prodromal and manifest Parkinson’s disease - Summary - MDSpire

Smartphone-based prediction of dopaminergic deficit in prodromal and manifest Parkinson’s disease

  • By

  • Katarina M. Gunter

  • Karolien Groenewald

  • Timothee Aubourg

  • Christine Lo

  • Jessica Welch

  • Jamil Razzaque

  • Ludo van Hillegondsberg

  • Adriana Nastasa

  • Pietro-Luca Ratti

  • Beatrice Orso

  • Pietro Mattioli

  • Matteo Pardini

  • Stefano Raffa

  • Federico Massa

  • Daniel R. McGowan

  • Kevin M. Bradley

  • Dario Arnaldi

  • Johannes C. Klein

  • Siddharth Arora

  • Michele T. Hu

  • December 1, 2025

  • 0 min

Share

Objective:

To investigate whether smartphone-based motor assessments can predict DaT scan results as a scalable alternative to confirm dopaminergic deficiency in Parkinson's disease, specifically focusing on the accuracy of these assessments.

Key Findings:
  • The smartphone-only XGBoost model achieved an AUC of 0.80, improving to 0.82 when combined with MDS-UPDRS-III, indicating a significant enhancement in predictive capability.
  • A simpler logistic regression model performed better with MDS-UPDRS-III alone (AUC = 0.83) compared to smartphone features, with slightly higher performance when combined (AUC = 0.85).
  • Gait, tremor, and dexterity features were identified as the most predictive, highlighting key areas for assessment.
Interpretation:

Smartphone-based assessments may effectively complement clinical evaluations in predicting dopaminergic deficiency, but larger independent validation is necessary to confirm these findings.

Limitations:
  • The study's sample size was limited to 100 DaT scans, which may affect the generalizability of the results.
  • Further validation in larger, independent cohorts is essential to establish the reliability of the findings.
Conclusion:

Smartphone-derived assessments could provide a cost-effective and accessible method for screening dopaminergic deficiency in Parkinson’s disease.

Original Source(s)

Smartphone-based prediction of dopaminergic deficit in prodromal and manifest Parkinson’s disease

  • by Katarina M. Gunter, Karolien Groenewald, Timothee Aubourg, Christine Lo, Jessica Welch, Jamil Razzaque, Ludo van Hillegondsberg, Adriana Nastasa, Pietro-Luca Ratti, Beatrice Orso, Pietro Mattioli, Matteo Pardini, Stefano Raffa, Federico Massa, Daniel R. McGowan, Kevin M. Bradley, Dario Arnaldi, Johannes C. Klein, Siddharth Arora, Michele T. Hu

  • December 1, 2025

Related Content

Electrophysiological signatures predict the therapeutic window of deep brain stimulation electrode contacts

Japan Clears First iPSC Therapies for Clinical Use 

Two iPSC-derived cell therapies receive time-limited authorization following early clinical trials

  • March 25, 2026

  • 2 min
Clinical AI is Not (Yet) Trustworthy-But It Could Be