TrialView: An AI-powered Visual Analytics System for Temporal Event Data in Clinical Trials

Area of Science:

• Clinical Informatics
• Data Visualization
• Artificial Intelligence in Medicine

Background:

• Randomized controlled trials (RCTs) are crucial for evaluating interventions, but analyzing individual patient experiences over time remains challenging.
• Existing analytical tools often lack the capability to effectively summarize and visualize temporal patient data from clinical trials.
• Understanding the time course of interventions is vital for comprehensive efficacy and safety assessments.

Purpose of the Study:

• To introduce TrialView, an innovative information system designed to enhance the dissemination and analysis of clinical trial data.
• To provide tools for interactive exploration of individual and group-level temporal event patterns in clinical trials.
• To leverage visual analytics and graph artificial intelligence for deeper insights into patient journeys.

Main Methods:

• Development of TrialView, an information system integrating graph AI and visual analytics.
• Implementation of four interconnected views: Individual, Cohort, Progression, and Statistics.
• Utilizing graph AI, knowledge-guided clustering, explanatory modeling, and graph-based agglomeration algorithms.

Main Results:

• TrialView enables interactive exploration of temporal event data for both individual participants and cohorts.
• The system facilitates a comprehensive examination of patient experience timelines within clinical trials.
• A case study demonstrated the system's effectiveness in analyzing temporal event data.

Conclusions:

• TrialView offers a novel approach to analyzing and visualizing clinical trial data, focusing on patient experience over time.
• The system's integration of graph AI and visual analytics provides valuable insights for healthcare professionals.
• TrialView is a versatile analytical tool applicable to a wide range of clinical trial designs.