Distributed, intelligent, interactive visualization and exploration of time-oriented clinical data and their...

Area of Science:

• Clinical Informatics
• Artificial Intelligence
• Data Science

Background:

• Managing large volumes of distributed, time-oriented clinical data presents significant challenges for interpretation and exploration.
• Existing methods like paper charts and spreadsheets are often inefficient for complex clinical data analysis.

Purpose of the Study:

• To introduce KNAVE-II, an intelligent interface designed for querying, knowledge-based interpretation, summarization, visualization, and interactive exploration of clinical time-oriented data.
• To evaluate KNAVE-II's effectiveness and usability compared to traditional methods in a clinical oncology setting.

Main Methods:

• KNAVE-II was developed based on user-centered evaluation and incorporates a computational engine (ALMA) and a distributed architecture (IDAN) for temporal reasoning and semantic data navigation.
• A cross-over study design was employed, comparing KNAVE-II against paper charts and Excel with clinicians querying a dataset of 1000 oncology patients.

Main Results:

• Users achieved faster and more accurate results for medium to hard difficulty queries using KNAVE-II compared to paper charts and Excel.
• KNAVE-II was universally preferred by users across all usability dimensions.

Conclusions:

• KNAVE-II demonstrates highly encouraging results in initial evaluations for managing and analyzing complex clinical time-oriented data.
• The knowledge-based temporal-mediation architecture supporting KNAVE-II shows promise for improving clinical data analysis, particularly in long-term patient monitoring scenarios like bone marrow transplantation.