Addressing Data Management Challenges With Automation

ACT: How can data management challenges be addressed? Specifically, is there any opportunity to address them with artificial intelligence (AI)?

Lacroix: There is a lot of opportunity across the life cycle to assist with this and it's interesting because historically it was always: we want to throw people at some of these problems. I think it's important to mention that people are not always the answer. Specifically in data management, there's a shortage of resources. There's a shortage of resources with the appropriate skill sets to manage the complexity, so we really do need to look at innovative processes and technologies to assist with managing this complexity; AI is one of those. When we're talking about all of the data and the high volume of data coming in, and specifically these adaptive trial designs, AI can help us with processing those large volumes of data to be able to automate and we need to build in automation for ingestion of these continual data flows, and utilize artificial intelligence to help us interrogate that data to be able to gain those insights, specifically around these adaptive trials, because of the ongoing decision making that happens throughout the course of these trials.

Utilizing automation is what we do at eClinical Solutions, as far as automating all those data sources into a clinical data management platform, elluminate, where the tool and the technology helps us to not only ingest, but to standardize and harmonize that data in a way that unifies that data so that we can look at that data more holistically and cohesively for that decision making for these adaptive trials. The data cleaning part of it is specifically for data management, it’s a big area that this can help with. As these large volumes of data continue to grow, we can't continue to throw edit checks at it or line listings at it. We need to use more smart and innovative tools to be able to process and interrogate that data.