Health-Terrain: Visualizing Large Scale Health Data
Abstract
The promise of the benefits of fully integrated electronic health care systems can only be realized if the quality of emerging large medical databases can be characterized and the meaning of the data understood. For this purpose, the effective visualization of large and complex health data for timely decision making is critical. Our long-term goal is to improve the usability of emerging large scale clinical data sets by developing effective and efficient open-source systems for health data analytics and visualization tools for clinicians, healthcare professionals, administrators, and patients. The objective of this application is to develop a prototype system to test the effectiveness of this approach on a large scale health care database that is currently available at Regenstrief Institute. We have reached this objective with the following specific accomplishments: Built a relational database as the representation of a health concept space, extracted from the NCD dataset, Natural Language Processing techniques were carried out to process 325791 clinical notes to extract new terms including diseases, symptoms, and mental and risky behaviors, Data mining techniques were applied to extract associations between terms in the concept space, and to discover new cluster terms, Designed and implemented a suite of novel visualization algorithms that allows the users to interactively explore the data based on the user selected terms and filters, Designed and implemented a web based graphical user interface for the prototype system, and Designed and tested an evaluation procedure for health data visualization system. This visualization framework offers a real time and web-based solution for the effective use of large scale military electronic health record systems by allowing system level integration of the human' visual capabilities into the overall health data based decision making system.
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 01, 2014
- Accession Number
- ADA624344
Entities
People
- Jennifer L. Williams
- Mathew Palakal
- Shaun J. Grannis
- Shiaofen Fang
- Yuni Xia
Organizations
- Indiana University