Related Experiment Videos
Framework for characterizing data and identifying anomalies in health care databases.
1Division of Medical Informatics and Outcomes Research, Oregon Health Sciences University, Portland, USA.
Proceedings. AMIA Symposium
|November 24, 1999
Summary
This study introduces a framework using metadata to characterize healthcare databases and identify data anomalies. Further research could make this framework a predictive tool for anomaly detection in healthcare data.
Area of Science:
- Health Informatics
- Data Science
- Database Management
Background:
- Healthcare databases are increasingly vital for health data.
- Data anomalies in these databases are a growing concern.
- Understanding data generation processes is crucial.
Purpose of the Study:
- To propose a framework for characterizing healthcare database data.
- To identify and discover anomalies within healthcare data.
- To understand the element structures and processes influencing data generation.
Main Methods:
- Utilizing metadata for data characterization.
- Developing a framework to analyze data structures and generation processes.
- Focusing on anomaly identification and discovery techniques.
Main Results:
- A novel framework for healthcare data characterization has been proposed.
- The framework aids in understanding data generation.
- The framework facilitates anomaly identification.
Conclusions:
- The proposed framework offers a method to characterize healthcare data.
- It aids in understanding data anomalies and their origins.
- Future research can refine this into a predictive anomaly detection tool.