Related Experiment Videos
Active subgroup mining: a case study in coronary heart disease risk group detection
Dragan Gamberger1, Nada Lavrac, Goran Krstacić
1Rudjer Bosković Institute, Zagreb, Croatia. dragan.gamberger@irb.hr
Artificial Intelligence in Medicine
|July 10, 2003
Summary
This study introduces an active mining approach for patient records to identify groups at high risk for coronary heart disease (CHD). Expert involvement in data analysis enhances the early detection of at-risk populations.
Area of Science:
- Medical Informatics
- Public Health
- Cardiology
Background:
- Coronary heart disease (CHD) poses a significant public health challenge.
- Effective patient screening methods are crucial for early risk identification.
Purpose of the Study:
- To develop and present an active mining methodology for discovering patient subgroups at high risk for CHD.
- To leverage expert involvement in the knowledge discovery process for improved accuracy.
Main Methods:
- Active mining of patient records with expert participation.
- Data gathering, cleaning, transformation, and subgroup discovery.
- Statistical characterization and interpretation of identified risk subgroups.
Main Results:
- The methodology successfully identifies patient groups at high risk for CHD.
- Key risk factors contributing to CHD are explicitly characterized.
- The approach facilitates the statistical validation of risk subgroups.
Conclusions:
- The proposed active mining approach is effective for identifying CHD risk.
- Expert involvement enhances the reliability of subgroup discovery and characterization.
- This methodology shows high potential for patient screening and early CHD detection.