An Entity-Based Visual Analytics System Enhancing Medical Expertise Acquisition: Development and Verification Study

Abstract Background Acquiring medical expertise from the vast body of medical text is a critical component of medical education. However, the majority of medical knowledge resides in unstructured texts. Data heterogeneity across institutions and strict privacy regulations hinder the use of general-purpose analysis tools. This creates a substantial barrier to the efficient acquisition of expertise for learners. Objective This study aimed to design, develop, and evaluate MExplore, an interactive visual analytics system to facilitate the acquisition of medical expertise from unstructured clinical texts. Methods We propose a localized, cost-effective workflow for the automatic extraction of medical entities. Building on this workflow, MExplore provides a multilevel visual framework featuring coordinated views for progressive, entity-centered exploration. The system was evaluated through case studies conducted with domain experts, a user study, and semistructured interviews. Results The evaluation demonstrated that MExplore significantly enhances the medical expertise acquisition process. Our findings confirm that MExplore provides an effective and interactive approach for structuring complex knowledge, facilitating the construction of illness scripts, and strengthening knowledge retention. Conclusions MExplore provides an intuitive and powerful approach for acquiring medical expertise. The results suggest that it effectively supports medical learners in conducting in-depth data exploration and developing robust clinical reasoning skills.