Conference Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 537 - Recent Developments in Causal Inference with Real World Evidence in Drug Development
Type: Invited
Date/Time: Thursday, August 11, 2022 : 10:30 AM to 12:20 PM
Sponsor: Biopharmaceutical Section
Abstract #320708
Title: Regulatory Considerations on the Use of Machine Learning for Real-World Evidence Studies
Author(s): Hana Lee*
Companies: FDA
Keywords:
Abstract:

Given unprecedented advances in health care technologies, interest in incorporating machine learning (ML) functionality in medical product development has increased dramatically in recent years. Over the past decades, FDA has received and reviewed a growing number of new applications of medical products utilizing ML functionality from the vast amount of data generated from the delivery of routine health care, referred to as real world data (RWD). This talk plans to overview regulatory considerations regarding challenges of extracting substantial evidence for regulatory approval (i.e., real world evidence) based on ML and RWD. This talk will start from a brief overview regarding regulatory considerations on the use of ML, focusing on the adequacy of the use of ML for design and analysis of real world evidence (RWE) studies for medical product development. Then it will discuss issues related to reproducibility and transparency. One of FDA-funded projects will be presented to illustrate how ML-based approach can be completely pre-specified to ensure reproducibility and transparency while still remaining data-adaptive.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2022 program