AI-based Progression and Medication Response Prediction Study in Parkinson's Disease

Not yet recruiting

• Conditions: PARKINSON DISEASE (Disorder)

• Registration Number: NCT07189468
• Lead Sponsor: University Hospital, Toulouse

Brief Summary

The study aims to provide initial PoC validation data of two AI models to predict disease progression and treatment side effects in PD patients using as input patients' demographic, clinical and genetic information, as well as digital biomarker measurements in daily living collected via a smartwatch and a mobile application.

Detailed Description

Not available

Study Timeline

• Status Verified: 2025-09
• Study First Submitted: 2025-09-16
• Study First Submitted QC: 2025-09-16
• Last Update Submitted: 2025-09-16
• Study First Posted: 2025-09-24
• Last Update Posted: 2025-09-24
• Study Start: 2025-10-06
• Primary Completion: 2027-04-01
• Study Completion: 2027-04-01

Recruitment & Eligibility

• Status: NOT_YET_RECRUITING
• Sex: All
• Target Recruitment: 100

Inclusion Criteria

• Parkinson's disease diagnosis according to MDS criteria (Postuma et al., 2015)
• Disease duration ranging from 5 to 10 years
• Age 40-80
• Patients in stages 2 and 3 of the Hoehn and Yahr (a functional disability scale) in the ON condition
• The participant is using a compatible smartphone
• Written informed consent

Exclusion Criteria

• Atypical Parkinsonian Syndrome
• Second-line device-aided treatments
• Patients with >4 daily doses of L-DOPA
• Daily levodopa equivalent dose > 1500 mg
• Ongoing hallucinations requiring short-term treatment adjustment
• Inability to provide informed consent or participate in the study
• Inability to use the smartwatch and/or the mAI-Care app - as judged by investigator
• Lacking motivation to participate in study procedures - as judged by investigator
• Under adult autonomy protection system, legal guardianship or incapacitation
• Pregnant and breast-feeding women

Study & Design

• Study Type: OBSERVATIONAL
• Study Design: Not specified

Primary Outcome Measures

Name	Time	Method
Classification performance of the Parkinson's disease (PD) progression prediction model	Between baseline and 12 months	Classification performance of the model in predicting observed motor disease progression, defined as a binary outcome: the condition of a patient is classified as worsened if either the Movement Disorder Society Unified Parkinson's Disease (MDS-UPDRS) rating scale part III score in ON-state increases by more than 3 points (\>3), or the Levodopa Equivalent Daily Dose (LEDD) increases by more than 10%.

Secondary Outcome Measures

Name	Time	Method
Usability of the mAI-Care app	12 month visit	System Usability Scale score of end-users (PD patients)
Usability of the mAI-Insights app	12 month visit	System Usability Scale score of end-users (clinicians)

Trial Locations

Locations (4)

University Hospital of Toulouse; 🇫🇷 Toulouse, France

Technische Universität Dresden; 🇩🇪 Dresden, Germany

Hospital Ruber Internacional; 🇪🇸 Madrid, Spain

Queen Mary University of London; 🇬🇧 London, United Kingdom