AI-aided evidence-based medical guidelines
Medical specialist guidelines serve as the foundation of medical practice. They support clinical decision-making and contribute to increased transparency, reduced unwanted variations in practice, and therewith provide better healthcare. These guidelines consist of evidence-based recommendations.
For ensuring evidence-based practices, it is vital to systematically review the existing literature. However, in today's rapidly evolving medical landscape, the volume of available literature is growing exponentially with each passing day. This ever-expanding medical knowledge presents both opportunities and challenges. On the one hand, it allows for a more comprehensive understanding of medical conditions, treatments, and outcomes. On the other hand, the magnitude of information can make the task of developing guidelines by conducting thorough and up-to-date systematic reviews an increasingly complex endeavor.
To address this challenge, the AI-aided Knowledge Discovery Lab, the Knowledge Institute of the Dutch Federation of Medical Specialists, Cochrane Netherlands and the Julius Center are collaborating to develop a pipeline aimed to accelerate the development of guidelines supported by AI. It is examined how to implement AI within the screening phase of this pipeline. Furthermore, the potential of utilizing AI in a different stage of the process is assessed, specifically, to give a weighted assessment of the relevant literature. With these implementations, medical guidelines can be developed more rapidly and will be more grounded in a robust evidence-based approach.
These projects are being funded by ZonMw and SKMS.
People involved
- Laura Hofstee, PhD student
The Knowledge Institute of the Dutch Federation of Medical Specialists, Utrecht University - Janke de Groot
The Knowledge Institute of the Dutch Federation of Medical Specialists - Tim Christen
The Knowledge Institute of the Dutch Federation of Medical Specialists - Lotty Hooft
Cochrane Netherlands and the Julius Center - Rens van de Schoot
Utrecht University