Related Experiment Videos
Analysis of hospitalised patient flows using data-mining.
Thierry Dart1, Yongmei Cui, Gilles Chatellier
1Santé Publique et Informatique Médicale (SPIM), University of Paris 6, France. thierry.dart@spim.jussieu.fr
Studies in Health Technology and Informatics
|December 11, 2003
Summary
Data-mining techniques analyzed patient pathways within a hospital information system. This analysis created a predictive model to forecast patient movements between medical units, improving healthcare management.
Area of Science:
- Health Informatics
- Data Mining
- Hospital Management
Background:
- The implementation of hospital information systems (HIS) facilitates medical data analysis.
- Computerized patient records enhance the ease of analyzing complex healthcare data.
Purpose of the Study:
- To analyze intra-hospital patient pathways using data-mining technology.
- To develop a predictive model for patient movements within a hospital setting.
Main Methods:
- Sequential patterns mining was employed to identify frequent patient pathways.
- An integrated framework combining association rules mining and classification rule mining was utilized to build prediction models.
Main Results:
- A rule-based prediction model was successfully constructed.
- The model effectively predicts patient flow tendencies between different medical units.
Conclusions:
- Data-mining offers a powerful approach for understanding and predicting patient trajectories in hospitals.
- Predictive models can optimize resource allocation and patient care coordination within healthcare facilities.