Challenge

Patients are prescribed with different drugs. The tests and measurement responses of different patients with respect to different medications is a topic of current research for PRA healthcare.

Approach

Built several machine learning models, including classification and regression. A classifier was used to predict the recovery of a patient given hundreds of tests measurements, drug name, age and gender. Regression models were used to predict the change of patient’s measurements (for example, white cell blood) after taking a drug.

Industry

  • Healthcare

Tools Used

  • Data Science Experience

Data Science Techniques

  • Visualization, regression, classification

Benefits:

By finding out the predicted changes on specific patient tests, specialists can better understand the effects of drugs on different patients. In addition, predicting a patient’s recovery and explaining what factors impact such recovery help PRA health with its clinical research mission.

Need help with a similar case?

Contact us to find out how we can help!