Predicting Drug Concentrations
Title: Predicting Drug Concentration
June 2017 - March 2018
It can be quite difficult for clinicians to determine the correct dose and how quickly a patient processes a drug can vary depending on demographic, treatment, and genetic information. Most of the algorithms for predicting drug dose are based on relatively small clinical population and the predictive accuracy is not well known. Insufficient or overexposure dosing can have a serious consequences for patients' health, and daily monitoring and maintaining target concentration is necessary. Therefore, there is an increasing need to develop improved strategies for prescribing the appropriate dose. The main goal of this project is to build a better predictive model to estimate the drug concentration in patients undergoing lung transplants. Eventually, we would have access to ~500 patients time series data.
If you are interested in this project, please contact Arya (email@example.com). Please, also read a recently published paper that looked at predicting the optimal dose of a drug in renal transplant patients. Our problem is slightly different, but it's a good place to get started.
This project need IRB approval and the dataset must not be shared outside of the group, however the algorithms, codes, and results can be published. You also need to complete "Human Subjects Biomedical & Health Sciences Module & Certification Test", "Foundations of Good Research Practices Module & Certification Test". It takes 1.5 hour to do both. You can use this link to complete these certification tests.
Leader(s): Sean Ma (slack: @seanma, email: firstname.lastname@example.org)
MDST Participants: TBD
Looking for: -
External Collaborators: Dan Hertz and Amy Pasternak (from the UM College of Pharmacy)
Link to the Codes: N/A
Link to the Dataset: N/A