*********************************
There is now a CONTENT FREEZE for Mercury while we switch to a new platform. It began on Friday, March 10 at 6pm and will end on Wednesday, March 15 at noon. No new content can be created during this time, but all material in the system as of the beginning of the freeze will be migrated to the new platform, including users and groups. Functionally the new site is identical to the old one. webteam@gatech.edu
*********************************
Felipe Giuste
BME PhD Proposal Presentation
Date:2022-02-17
Time: 1:15 PM - 2:30 PM
Location / Meeting Link: https://bluejeans.com/4043855059
Committee Members:
May D. Wang, PhD (Advisor) Blake Anderson, MD Robert Gross, MD/PhD Shriprasad Deshpande, MD David Wright, MD
Title: Advancing clinical practice using translational clinical informatics
Abstract: My goal is to develop and integrate AI-based solutions into clinical workflows. There are three major informatics challenges which must be solved for successfully accomplishing this goal: data harmonization, model explanation, and decision support automation. This proposal aims to establish solutions to these three informatics challenges and illustrate their value in real-world clinical cases. First, I propose a solution to the lack of healthcare information harmonization through the creation of a web-based application to standardize pediatric bone diseases data at Shriners Children’s Hospital using the Fast Health Interoperability Resource (FHIR) standard. Next, I pursue better model interpretability by leveraging recent explainable AI solutions, including Shapley Additive exPlanations (SHAP) for feature ranking, to predict COVID-19 positive patient outcomes. Finally, I aim to implement insights gained from the first two solutions to generate a data-driven decision support tool for pediatric heart transplant rejection detection. Specifically, I will generate a user interface to support expert detection of rejection. Together, these aims provide solutions to major challenges in clinical informatics and showcase their value in three real clinical situations. This work will advance healthcare informatics by generating practical solutions to the three major challenges of data harmonization, prediction explanation, and effective decision support.
