AI and Machine Learning literature monitoring: improve quality and save time!

Assessing scientific literature for patient safety is a highly regulated activity. Changes to processes and work instructions are required to be documented which is contrary to the fundamentals of a machine learning system. This coupled with the implications of false negatives – the relevant references that the machine deems to be irrelevant – has prevented AI from becoming a mainstream replacement for human review teams reviewing medical literature for patient safety issues.
However, machine learning does lend itself well to certain aspects of the review process.
In this presentation, ProQuest will present:
1 - The results of our research into complementing and improving the efficiency of human reviewers by reliably solving key steps of the medical literature monitoring process using deep learning and other machine learning techniques.
2 – Our approach to introduce machine learning the Literature Review workflow that complements the human review helping improve assessment quality and reduce the overall amount of time required to assess references, without running the risk of missing relevant literature cases.

About the speaker
Mark Drinkwater

Mark Drinkwater was co-founder of Pi2 Solutions at ProQuest. Pi2 Solutions provides cutting-edge product literature databases and workflow solutions to the pharmaceutical industry, and Mark has led Pi2’s initiatives in the drug safety arena, particularly with regard to literature screening process efficiency to aid compliance. A chemist by training, Mark has worked for several pharmaceutical information companies, and has advised global organizations through to small biotechs. Mark obtained his BSc (Hons) in Applied Chemistry from De Montford University in Leicester, UK.

Cookies help us deliver our services. By using our services, you agree to our use of cookies. Learn more