Dr. Thomas Byrd received pilot funding from the Clinical and Translational Science Institute (CTSI) alongside three other awardees to advance solutions that improve the process of translation from research to practice.

Dr. Byrd will be funded $40,000 over the course of 12 months to conduct an in-depth case study on the implementation of a predictive model within the UMN and M Health Fairview system that is designed to predict when a patient’s condition will likely worsen.

Implementing a validated model into a clinical workflow is one of the most difficult and resource-intensive parts of integrating a new machine learning model into practice. By documenting the entire process in detail and analyzing barriers encountered, Dr. Byrd hopes to create a detailed playbook that can be used to guide the implementation of other models.

Dr. Byrd has previously done extensive work on clinical deterioration, where a patient’s condition suddenly and unexpectedly worsens while hospitalized. The algorithm he plans to implement was developed during his time as a MN-LHS scholar to improve on an existing model used by Fairview that predicts deterioration before it occurs.

This project will lead to both the implementation of a more effective predictive model and generalizable learnings that can be used to accelerate the integration of future models into practice.