Idiopathic Parkinson's disease (PD) is a neurodegenerative disease that progressively causes both motor and non-motor symptoms. As the second most common neurodegenerative disease and most common movement disorder, it affects over 8.5 million people worldwide and 13,000 people in Hong Kong. The most classical symptoms of PD are resting tremors, rigidity of the muscles, bradykinesia (slowing of movement), and gait difficulty. Other symptoms include sleep disorders, psychiatric symptoms, cognitive impairment, and autonomic dysfunction. Its pathophysiology is marked by the loss of dopaminergic neurons and the accumulation of aggregates called Lewy bodies.
The severity of PD-related motor symptoms is usually semi-quantitatively ("normal", "slight", "mild", "moderate", and "severe") evaluated by expert physicians and physiotherapists according to the Movement Disorder Society-sponsored revision of the Unified Parkinson's Disease Rating Scale Part III (MDS-UPDRS III). However, the MDS-UPDRS III is semiquantitative and subjective, which might mask mild treatment effects or even provide false-positive results. Moreover, it takes significant time and effort for assessment with expected inter-observer variations.
To address these issues, various artificial intelligence (AI) technologies and telemedicine approaches have been investigated for patient evaluation. However, previous studies did not incorporate items assessing rigidity and postural stability, which require physical contact as per the MDS-UPDRS III instructions. Zhu et al. explored a motor symptom machine-rating system for the complete MDS-UPDRS III. Nevertheless, they employed a depth camera and conducted the tests within a strictly controlled ideal laboratory environment. For the widespread implementation of AI-assisted rating, the RGB camera is a more accessible alternative.
Age ≥18 years
Diagnosis of "Clinically Established PD" as defined by the Movement Disorder Society Clinical Diagnostic Criteria for Parkinson's disease (MDS-PD criteria) [12]
Able to provide informed consent and willing to participate in video-recorded MDS-UPDRS Part III assessments
No significant visual, auditory, or musculoskeletal impairments that would interfere with video-based motor assessments
Unwillingness to be video recorded for study purposes
History of neurodevelopmental disorder, neurodegenerative disease other than PD, CNS infection, neuroinflammatory disease (e.g. multiple sclerosis, CNS lupus), malignancy within the last 10 years, cerebrovascular accident, HIV infection, systemic autoimmune disease, alcohol dependence or other substance use
patients with Parkinson's disease
Cohort
Prospective
Observational
Other interventionclinical profile, MDS-UPDRS III, Video Recording
AI-based motor assessment tool
AI-based motor assessment tool utilizing RGB video for reliable and objective ratings of MDS-UPDRS III motor symptoms, including rigidity and postural stability.
Feasibility of implementing RGB camera-based assessments
Feasibility of implementing RGB camera-based assessments in routine clinical settings will be assessed by the proportion of assessments in which the AI system is able to generate an estimated MDS-UPDRS Part III total score based on RGB video that can be directly compared with clinician-rated MDS-UPDRS Part III scores. Patients perform standardized motor tasks under physician guidance while RGB video is recorded using a smartphone. Clinician-rated MDS-UPDRS Part III scores are used as the ground truth. Feasibility outcomes will be reported as the percentage (%) of assessments with valid AI-generated scores over a 3-year study period.
System's effectiveness
System effectiveness in estimating motor symptom severity will be measured by the agreement between AI-predicted and clinician-rated MDS-UPDRS Part III scores. RGB video is recorded using a smartphone while patients perform standardized motor tasks under physician supervision. Clinician-rated MDS-UPDRS Part III scores are used as the ground truth. The AI system generates predicted MDS-UPDRS Part III total scores (range 0-108, with higher scores indicating more severe motor impairment). System effectiveness will be evaluated using F1 score, correlation between predicted and clinician-rated scores, and sensitivity and specificity for the detection of clinically significant motor impairment based on predefined score thresholds. Effectiveness outcomes will be reported over a 3-year study period.
Patient and clinician satisfaction
Patient satisfaction with the AI-assisted assessment system will be assessed using a study-specific questionnaire administered after completion of RGB camera-based motor tasks. Questionnaire items are rated on a Likert scale, with higher scores indicating greater satisfaction. Clinician evaluation of the AI-assisted assessment system will be assessed using a study-specific questionnaire evaluating perceived credibility, perceived effectiveness, and overall satisfaction with the system. Questionnaire items are rated on a Likert scale, with higher scores indicating more positive evaluations. Patient and clinician evaluation outcomes will be reported as mean ± standard deviation over the study period.
System's performance
System performance will be measured by the accuracy and mean absolute error (MAE) of AI-predicted MDS-UPDRS Part III total scores compared with clinician-rated scores. Patients perform standardized motor tasks under physician supervision while RGB video is recorded using a smartphone. Clinician-rated MDS-UPDRS Part III scores are used as the ground truth. System performance is defined as the accuracy and MAE of AI predictions. Performance outcomes will be summarized over the study period.
Click on the lead sponsor to view all of their trials.
Hong Kong University of Science and Technology
Lead sponsor
The Queen Elizabeth Hospital
Collaborator
