Home
Identifying and utilizing suitable cohort studies is challenging. PDataViewer is a web application designed to streamline this process. It ranks cohorts based on the variables required for specific research questions, provides summary statistics, visualizes biomarker distributions and drop-out rates.
PDataViewer is designed to:
- List Cohort Studies: Provide a comprehensive list of Parkinson's disease cohort studies, including summary statistics (total participants, diagnosis groups, location), references, and data access links.
- Visualize Distributions: Display box plots showing the distribution of specific variables across selected cohorts and diagnosis types.
- Rank Cohorts: Identify relevant cohorts based on user-provided variables and showcase the distribution of those variables within each study.
- Map Features: Demonstrate semantic feature mappings across different studies.
- Track Attrition: Showcase participant drop-out rates in longitudinal studies for specified variables.
Accessing patient-level data is often delayed by data security and privacy protocols. Furthermore, variables listed in study documentation are not always present in the actual dataset; for example, clinicians may opt out of invasive measurements, or participants in longitudinal studies may drop out, reducing data availability for follow-up visits.
Additionally, cohort studies lack standard naming conventions. A participant's age might be labeled as "Age", "age", "age_of_participant", or "participant_age". Before data-driven approaches can be applied, these variables must be semantically harmonized -- a time-consuming process when dealing with diverse cohorts.
PDataViewer increases data transparency, findability, and accessibility while reducing the time required for semantic harmonization. By providing insights into patient-level data and direct application links, PDataViewer enhances the FAIRness (Findability, Accessibility, Interoperability, and Reusability) of data in Parkinson's disease research.
Note: The data harmonization process has been moved to Kitsune