Objective: To develop and assess the feasibility of a clinical decision support system (CDSS) to assist clinicians in interpreting multimodal remote monitoring data in Parkinson’s disease.

Background: Parkinson’s disease is a complex and fluctuating condition requiring continuous assessment and personalised management. Traditional care pathways often struggle to capture symptom variability and meet the needs of people with Parkinson’s (PwP). To address this, we developed Home-Based Care (HBC), a digitally enabled pathway integrating remote monitoring technologies to support clinical decision-making and continuous assessment. However, interpreting large volumes of patient-generated data remains a barrier to implementation at scale. We are developing SMaRT-PD, a CDSS designed to integrate multimodal data within the HBC pathway, including patient-reported outcomes, wearable sensor data, and clinical records, to support structured interpretation and clinical insights.

Method: SMaRT-PD was developed using a rule-based architecture modelling symptom management pathways in Parkinson’s disease. 58 decision-tree wireframes were created to represent clinical reasoning processes. Clinical priorities were defined using the MoSCoW prioritisation framework, informed by a survey of 80 Parkinson’s clinicians. For feasibility testing, seven prioritised decision trees were implemented in a cloud-based environment using Azure Function Apps. Data was initially sourced from an Excel repository stored in Azure Blob Storage. Iterative implementation cycles evaluated feasibility and identified architectural limitations.

Results: Initial implementation demonstrated the feasibility of encoding Parkinson’s management pathways within a rule-based CDSS. Early testing identified limitations including duplicated rule logic, reliance on file-based data storage, and challenges maintaining scalable decision logic. The architecture was redesigned using a modular structure supported by a central PostgreSQL database, improving data validation, version control, and integration across decision pathways.

Conclusion: SMaRT-PD demonstrates the feasibility of a rule-based CDSS to support interpretation of multimodal remote monitoring data in Parkinson’s disease and highlights the potential of decision support tools to enable digitally supported models of Parkinson’s care.

« Back to 2026 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/development-of-a-clinical-decision-support-system-to-support-parkinsons-disease-management/