Icon and Informa team on AI-based study feasibility technology
Icon has tapped industry study data to train the artificial intelligence (AI) based clinical trial feasibility service being developed with Informa Pharma Intelligence.
The Ireland-headquartered contract research organization announced its partnership with Informa this week, explaining the idea is to set up an AI-based clinical trial feasibility and planning service.
The service – known as CiteLine Study Feasibility – is designed to help sponsors assess the feasibility of a planned clinical trial, accelerate proposed timelines and – when the program is running – identify and reduce non- and low-enrolling sites.
The service uses a type of artificial intelligence technology called machine learning, which is a section of code able to make predictions based on the analysis of data it has been “trained” to interpret over time.
The algorithm will be trained on data from Informa Pharma Intelligence’s clinical trial and real-world data sets as well as information from Icon clinical trial activation, cycle time and performance databases.
Icon claimed predictions made through the Citeline Study Feasibility service will help reduce cost and manual effort in the clinical trial planning process, citing optimized enrolment timelines as one of the likely benefits.
Chief information officer Tom O’Leary stressed the importance of fully assessing a proposed clinical program before it begins, explaining “An optimal study feasibility process is one of the cornerstones of successful clinical drug delivery.”
He also lauded the quality of the data that will be used to train the CiteLine Study Feasibility software.
“Informa’s access to industry data and their data science expertise, make it an ideal partner for Icon to work with to continue providing cutting edge, innovative and best in class data driven solutions to our customers.”
Nicky Marlin, chief product and technology officer at Informa Pharma Intelligence, made a similar point, commenting that “A strong foundation of credible data is essential to refining any machine learning model.”
Image: Stock Photo Secrets