Artificial Intelligence in Medical Education and Simulation

​One of our core ambitions at CAMES is being the world leader in research within Artificial Intelligence (AI) application in medical education and simulation.

​At CAMES, we aim to reduce medical errors by using decision-support systems to improve health care professionals’ learning and clinical performances. Instead of using AI to replace tasks usually performed by clinicians, we seek to develop AI systems that support learning processes and clinical decision-making steps to make clinicians better at what they do. We will achieve this through a close collaboration between data scientists, medical education scientists, and clinicians.


A unique feature of the AI research at CAMES is utilizing medical simulation training labs (Sim-Labs) to link AI system development with proposed clinical setting benefits. A Sim-Lab should be considered a laboratory, where clinicians’ behavior and interaction with systems and clinical equipment can be observed. We use Sim-Labs for rapid cycles of development, iterative testing, evaluation, and revision of decision-support systems with no risk of patient harm and thereby no limiting regulatory restrictions. Clinical end-users of the AI-based decision support systems are closely involved in this process and trained to use the system before implementing clinical practice. This improves the efficacy of translational efforts from basic science to bedside and the overall quality, usability, and acceptability of the systems introduced into clinical practice.

In short, we use Sim-Labs to bridge the gap between data scientists, medical education experts, and clinicians when developing decision-support systems that aim to support clinicians’ learning and decision-making with the ultimate purpose of improving patient care and safety. 

Current projects

  • Decision-support for obstetric ultrasound. This project aims to develop decision-support systems for improving clinicians' learning and performance when performing obstetric ultrasound scans. Scientists: Martin G. Tolsgaard, Anders Nymark Christensen, and Morten Bo S Svendsen.

  • AI-augmented training of doctors - skin cancer diagnostics. A project that aims to use AI for improving doctors' ability to diagnose skin cancer. Scientists: Niels Ternov, Gustav Nervild, and the AISC group.

  • AI-based decision-support for hip fracture surgery. This project aims to develop decision-support systems that help guide surgeons in making the best choice when selecting surgical methods and placement of ostesynthesis. Scientists: Amandus Gustafsson, Morten Bo S Svendsen, Martin Tolsgaard, Mads Nielsen.

  • Performance assessment of invasive procedures in fetal medicine. This project aims to develop AI-based automatized performance assessments of invasive procedures in fetal medicine. Scientists: Vilma Johnsson, Morten Bo S Svendsen, Olav B Pedersen, Karin Sundberg, Martin Tolsgaard.

  • Team factors contributing to quality of care during Emergency Caesarean Sections. This project uses machine learning techniques for analyzing complex datasets involving social ties between team members in emergency caesarean sections. Scientists: Betina Ristorp Andersen, Ellen Løkkegaard, Sune Lehman, Martin Tolsgaard.

Responsible for the CAMES AI group

Responsible editor