BROADCASTS.com

Dr. Jon Morrow, Senior VP of Medical Affairs and Informatics, and Daniel Blumenthal, Vice President of Strategy at MDClone, shed light on the meaning of synthetic data for drug discovery and patient care. Using both original data and statistical models of the data, researchers and clinicians can quickly gain insights that can impact patient care, facilitate collaborations and partnerships, and dramatically shorten the time required for product development.

Daniel explains, "Actually, the system is able to produce synthetic data on the fly on demand. It reacts to the end user's request for data. So, if I'm an end user and I want to build a population, I want to look at a population of patients with diabetes and understand the medications they were on and lab tests that were drawn about them. To understand their disease trajectory over time. I'm able to define that, and the way our engine works is it actually can take that original population. But without sharing that population, as John articulated at the beginning, without actually sharing that original population. Just learn the statistics of that population and then build this brand new set of synthetic data. That synthetic data, every time it's generated, is compared to the original."

Jon elaborates, "So rather than exposing my patient's information or using my patient's information with approval, how about if I ask a system like MDClone ADAMS to generate a synthetic dataset that looks statistically exactly like my population of patients with gestational diabetes? And it's pulling the data from the electronic medical record at my hospital. It's running this process behind the firewall of my hospital, but it's providing me with data on a population that doesn't exist but behaves exactly like my patients at the population level."

@MDClone_ #SyntheticData #LifeScience #ResearchData #RareDiseases

mdclone.com

Download the transcript here