Related Experiment Video
Updated: Feb 24, 2026
Hydra, a Computer-Based Platform for Aiding Clinicians in Cardiovascular Analysis and Diagnosis
Published on: September 26, 2018
A comparative study of different methods for automatic identification of clopidogrel-induced bleedings in electronic
Hee-Jin Lee1, Min Jiang1, Yonghui Wu1
1University of Texas Health Science Center at Houston, Houston, TX.
Insights
Automated methods can effectively identify adverse drug reactions (ADRs) from electronic health records (EHRs). Scoring and machine learning approaches show promise for pharmacogenomic research, improving patient safety and drug discovery.
Area of Science:
- Bioinformatics
- Pharmacogenomics
- Health Informatics
Background:
- Electronic health records (EHRs) linked with biobanks are crucial for pharmacogenomic studies.
- Identifying patients with adverse drug reactions (ADRs) from large EHR datasets is challenging.
- Manual chart review for ADR identification is time-consuming and expensive.
Purpose of the Study:
- To develop and compare informatics approaches for automatic ADR identification from EHRs.
- To utilize clopidogrel-induced bleeding as a case study for ADR detection.
- To advance EHR-based pharmacogenomic research through efficient ADR identification.
Main Methods:
- Investigated three methods: rule-based, machine learning-based, and scoring function-based.
- Applied these methods to identify clopidogrel-induced bleeding events in EHRs.
- Analyzed feature contributions, including temporality and physician assertions.
Main Results:
- Both machine learning and scoring methods demonstrated effectiveness in identifying ADRs.
- The scoring method achieved high precision with reasonable recall.
- Temporality of drug-event and textual evidence from physicians significantly aided identification.
Conclusions:
- Automated informatics approaches are valuable for identifying ADRs in EHRs.
- Scoring and machine learning methods offer efficient alternatives to manual review.
- Findings support the advancement of EHR-based pharmacogenomic studies and patient safety.
Abstract:
Electronic health records (EHRs) linked with biobanks have been recognized as valuable data sources for pharmacogenomic studies, which require identification of patients with certain adverse drug reactions (ADRs) from a large population. Since manual chart review is costly and time-consuming, automatic methods to accurately identify patients with ADRs have been called for. In this study, we developed and compared different informatics approaches to identify ADRs from EHRs, using clopidogrel-induced bleeding as our case study. Three different types of methods were investigated: 1) rule-based methods; 2) machine learning-based methods; and 3) scoring function-based methods. Our results show that both machine learning and scoring methods are effective and the scoring method can achieve a high precision with a reasonable recall. We also analyzed the contributions of different types of features and found that the temporality information between clopidogrel and bleeding events, as well as textual evidence from physicians' assertion of the adverse events are helpful. We believe that our findings are valuable in advancing EHR-based pharmacogenomic studies.
More Related Videos
05:03Patient Directed Recording of a Bipolar Three-Lead Electrocardiogram using a Smartwatch with ECG Function
Published on: December 11, 2019
18:11A Research Method For Detecting Transient Myocardial Ischemia In Patients With Suspected Acute Coronary Syndrome Using Continuous ST-segment Analysis
Published on: December 28, 2012