Please review below our ICHI 2018 Tutorials presenters. We think this remarkable set of speakers will inspire you in your respective work!
An interactive tutorial on simulated pharmacogenomic clinical trials
Author: Kourosh Ravvaz, MD, PhD, MPH
Date/Time: June 4th, 1:30pm to 5:00pm
Tutorial Description and Scope
During this presentation, we will elaborate and demonstrate a new approach to enhance clinical trial design using electronic health records combined with computer simulation. Topics covered will include, data mining from electronic health records (EHRs), the limitations of simulation results, testing and validation and lastly dissemination of results through living papers.
Our primary audience is clinical researchers looking for ways to translate their scientific discoveries more rapidly to clinical practice. A second audience is for bioinformatic students and young professionals who are looking for innovative ways to connect big data discoveries to clinical care. Some interactive portions will require novice technology level skills. Other interactive portions will require expert level technology skills, but will be optional.
New Knowledge from the Tutorial
We will help the audience replicate the study design from our recent simulated clinical trial experiment (PMID: 29237680), using open source software, public data and the interactive results visualization tool, viewed online at https://ari-cds-ar.shinyapps.io/circcvg_paper_app/. The primary thesis of our presentation will elucidate what we call backwards-design for simulated clinical trials. We hope that the lessons we learned will help other students of informatics more wisely design and craft similar simulation studies, mainstreaming simulation into clinical research.
**NEW**During this tutorial, participants will execute sample components of a clinical trial simulation platform. Each user is asked to bring their individual laptop to follow along with the tutorial. A smart phone or iPad will not be sufficient to participate in the tutorial. Publicly available software tools will be demonstrated during this tutorial and it is highly recommended that all participants download and test launch the following programs on their laptops prior to attending.
In this tutorial, the following online tool will also be used. It is an interactive view only tool: https://ari-cds-ar.shinyapps.io/circcvg_paper_app/
Is the best model also the right solution?
Author: Mark Kanner, Sylvie Lardeux and Foruhar Shiva
Time: June 4th, 9:00am to 12:30pm
This tutorial will provide participants hands-on experience in building models using medical claims data to answer a business question. The aim is for the participant to complete the entire cycle of forming a problem statement, exploring the data and building a model to answer a real world business question, utilizing a tool of their choosing. This will be accomplished by focusing on a real world case study, providing the relevant data for the participants and scaffolding a solution as they go through the steps outlined above. The participants will first brainstorm and share their own ideas before hearing how instructors have handled these challenges. Modeling and validation will be addressed, however special attention will be paid to formulating a solution that is tailored to address the business problem and choosing the appropriate analytics approach to accomplish this; Is the information provided by the model actionable? Can a business partner use these results to make decisions? We believe this emphasis on real world application will provide a unique experience to the participants and equip them with techniques to use in practice. Healthcare informatics practitioners would have the opportunity to learn specific analytic approaches relevant to them. Students who may understand the process of modeling would gain experience in applying those skills in an open ended task.
Interpretable Machine Learning in Healthcare
Author: Muhammad Ahmad
Time: June 4th, 9:00am to 12:30pm
This tutorial extensively covers the definitions, nuances, challenges, and requirements for the design of interpretable and explainable machine learning models and systems in healthcare. We discuss many uses in which interpretable machine learning models are needed in healthcare and how they should be deployed. Additionally, we explore the landscape of recent advances to address the challenges model interpretability in healthcare and also describe how one would go about choosing the right interpretable machine learnig algorithm for a given problem in healthcare.