Office of Technology Transfer – University of Michigan

A Random Forest Based Risk Model for Reliable and Accurate Prediction Of Risk of In-Hospital Mortality in Patients Undergoing PCI

Technology #6204

Questions about this technology? Ask a Technology Manager

Download Printable PDF

Researchers
Hitinder S. Gurm
Managed By
Jessica Soulliere

Random forest based risk model for reliable and accurate prediction of risk of in-hospital mortality in patients undergoing PCI. This easy to use, validated model predicts post procedural in-hospital mortality following PCI. It is useful for benchmarking and guiding quality improvement.

The predictive model was developed using 22 pre-procedural clinical and laboratory variables to estimate the risk of in-hosptial mortality in patients undergoing percutaneous coronary intervention. This model was estimated using data from 120,090 PCI procedures occurring in the state of Michigan between January 1, 2010 and September 30, 2013.

Applications • Online calculator • Smart phone application • EMR integrated application

Advantages • High level of discrimination • Easy of Use • Availability