Artificial Intelligence in Health(care)

ICAI

The Innovation Center for Artificial Intelligence is a national network aimed at technology and talent development between knowledge institutes, industry, and government in the area of artificial intelligence.

Radboud AI for Health and Thira Lab are proud members of ICAI.

Visit the ICAI website.

Radboud AI

Radboud AI is an initiative to drive public-private collaboration and start new AI-projects in Nijmegen and surroundings. Researchers and students from all faculties work together with industry to create solutions that allow for a future where privacy, security and understanding of AI is valued.

Visit the Radboud AI website.

What is Radboud AI for Health?

Radboud AI for Health is a collaboration between Radboud university medical center and Radboud University, aimed at developing innovations based on Artificial Intelligence that solve clinical problems. The innovations will be applied in Radboudumc and other healthcare institutes. We perform PhD research projects and BSc and MSc research projects. We also offer courses for Radboudumc personnel on the use of AI in healthcare.

Healthcare is seen as one of the areas where AI will have a profound impact. Data is ubiquitous in hospitals, and is available in a many forms: patient records, clinical signs and measurements, genetic information, text reports, and images. This data forms the basis for decisions, the diagnosis of a disease, detection of acute and long-term risks, treatment plans, treatment monitoring, and real-time support for example during surgery and interventions.

Artificial intelligence can support clinical decisions, assist in numerous ways, improve healthcare by finding better ways to extract clinically useful information from data, and automate tasks and thereby reduce costs and keep healthcare affordable. By carrying out projects within hospitals, the vast amounts of data available in healthcare can be unlocked for research, of course always in compliance with privacy regulations and ethical standards, and this makes healthcare an interesting domain for AI researchers.

The Radboud AI for Health ICAI Lab received an initial funding from Radboudumc's innovation board and started in September 2019 with 6 PhD research projects and funding to support around 30 Master or Bachelor research projects every year. We aim to organize two courses every year for Radboudumc employees interested in AI. The lab profits from the AI expertise within Radboudumc and Radboud University. The lab is housed in the Radboudumc Innovation Space, collocated with iLab and REshape and the Radboudumc Technology Center Deep Learning to implement the lab's innovations into the clinic. The initiative is embedded within the national ICAI network and is actively looking to partner with others to grow further in the near future.

Research and education

Courses

Radboud AI for Health plans to run two courses every year for a group of about 25 professionals from Radboudumc interested to learn more about AI applications in healthcare. The course program will consist of lectures and practical exercises, and will conclude with a practical project.

More information can be found on the Radboud AI for Health website.

BSc and MSc projects

Each year, approximately 35 bachelor’s and/or master’s students will perform a project in the field of AI in healtcare.

More information can be found on the Radboud AI for Health website.

PhD projects

Six PhD research projects started in September. Teams of researchers will be working on developing several AI-applications in healthcare, which will be tested in a clinical setting. In short, the projects consist of developing:

• a real-time warning system for infection risk of central venous cathethers for patients in the intensive care unit.
• an AI system to automatically diagnose patients with genetic diseases based on their exome and whole genome sequencing data.
• an AI-based MRI analysis for treatment decision support in patients with chronic degenerative low back pain.
• a system that can automatically create annotated radiology reports that a patient can understand.
• an AI algorithm that can help predict vascular function of the brain from short, non-invasive ten-minute recordings of blood pressure, heart rate, and cerebral perfusion.
• A deep-learning system that can predict, on the basis of clinical information, panoramic radiographs, and possibly cone-beam CT, if wisdom tooth removal is recommended.

More information can be found on the Radboud AI for Health website.

What is Thira Lab?

Thira Lab is a collaboration between Radboud university medical center and Thirona, a spin-out company from Radboudumc and Delft Imaging Systems that develops healthcare solutions for the specific needs of vulnerable communities around the world. In Thira Lab, nine PhD candidates and postdocs from Radboudumc work on deep learning image analysis of CT scans, radiographs and retinal images. It is Thira Lab’s mission to perform world-class research to strengthen healthcare with innovative imaging solutions.

More information can be found on Thira Lab's website.