OHPEN Lab: Optimization of Health ProcEsses and Networks Laboratory

OHPENLab is directed by You Chen, PhD, Associate Professor

OHPENLab is dedicated to advancing healthcare by leveraging cutting-edge technologies within a transformative informatics framework. Our research leverages large-scale, multimodal health data and advanced computational models to enhance care delivery and ensure drug safety. Our lab integrates cloud computing platforms (e.g., Databricks, Google Cloud, Terra), large language models, data mining, machine learning, temporal and network analysis, and advanced statistical methods to analyze large-scale multimodal data—including electronic health records (EHRs), genotypes, and EHR access logs—and multi-scale knowledge from sources such as PubMed literature, FAERS, MAUDE, DrugBank, and the OHDSI phenotype library. Our goal is to uncover meaningful patterns in healthcare processes, human diseases, genomics, and pharmacology, ultimately improving patient outcomes.


Our specific research interests include:


1. Optimizing Clinical Pathways: Measuring and refining clinical pathways to enhance patient care journeys, optimize resource allocation, and improve patient outcomes.


2. Digital Information Flow Monitoring: Enhancing the efficiency and safety of digital technology use in healthcare by monitoring and analyzing information flow in modern care coordination.


3. Teamwork Efficiency in Patient Care: Analyzing teamwork patterns to boost efficiency in collaborative and coordinated patient care.


4. Adverse Drug-Drug Interactions: Discovering and mitigating adverse drug-drug interactions by incorporating human genotype data.


5. Healthcare Disparities: Quantifying and addressing racial and socioeconomic disparities in healthcare processes and patient outcomes.


6. Neurodegeneration Following ICU Delirium: Integrating multi-scale knowledge and multi-modal data to deepen the understanding of neurodegeneration in patients who have experienced ICU delirium.


News

Graduate student and post-doc positions available

Job Description:The target project for these positions is the development of network analysis, data mining and machine learning algorithms to derive enhanced predictive models and actionable knowledge from data in electronic health records, social media, community health status indicators, mobile health and human genomics.

Requirements: Candidates should have sufficient experience in network analysis, data mining, machine learning, large-scale data analysis or statistics. He/she should have discipline and high motivation to pursue independent research in computer science or biomedical informatics.

How to Apply: To apply, send a CV, a research statement and three reference letters to you.chen@vanderbilt.edu

Publications

People

Projects

R01LM014199: Machine learning drives translational research from drug interactions to pharmacogenetics.

Genetics of Drug-Drug Interactions: This study aims to explore the genetic factors behind bad side effects of DDIs using BioVU resources. We will compare those who had bad reactions because of DDIs with those who did not. Our team wants to focus on genetic differences that may make someone more likely to experience these bad reactions. Some genetic differences can make someone even more likely to have heart or kidney issues because of DDIs.

  • February 2023
  • July 2028

1R01LM012854: Discovering Care Coordination Practice Patterns in the EMR.

The overarching goal of this project is to learn care coordination patterns through the data stored in electronic medical record systems, assess their influence on care effectiveness (in the form of LOS and readmission), and evaluate the extent to which they are ready for adoption by healthcare organizations through comprehensive surveys and interviews with knowledgeable healthcare experts.

  • August 2019
  • July 2024

K99/R00LM011933: Learning Patterns of Collaboration to Optimize the Management of Care Providers

Our research aims to uncover collaboration patterns among healthcare providers that can enhance patient care by analyzing EHR utilization data. Specifically, we seek to identify optimal care teams tailored to specific diseases, understand dependencies between providers to improve resource allocation, and model disease-specific treatment workflows to pinpoint the most efficient and effective sequences of care.

  • May 2015
  • April 2019
More projects

Course