ICAI Interview with Rianne Fijten: Tightening the relationship between medical clinics and commercial parties
In order to implement new AI technology in medical clinics in a sustainable way, close collaboration between the clinic and commercial parties is crucial, argues Rianne Fijten. ‘You need to make sure that if the grant money runs out, which it always does, the product that you built is not just lost.’
Rianne Fijten is one of the scientific directors of Brightlands Smart Health Lab, assistant professor and senior scientist of clinical data science at Maastro clinic.
Brightlands Smart Health Lab is a collaboration between Maastricht University, Brightlands Institute for Smart Society, Zuyd University of Applied Sciences, Maastro Clinic, Maastricht UMC+, ilionx and Netherlands Comprehensive Cancer Organization.
Could you tell me about the research happening in the lab? What makes this research unique?
‘What is interesting about our lab, is that we really go from technology to the clinic. That’s a concept I’ve not seen anywhere else. Usually a research group focuses on a very specific part of a pipeline, problem or societal issue. Within the lab we have three pillars: data infrastructure, data science and clinical implementation. It is a pipeline from start to finish: we set up the infrastructures to get the data out of the hospitals, extract the data, build AI models and then implement it into the clinic.’
‘Another important thing is that we are close to business. Getting data science into a medical clinic is difficult, but getting it into the clinic without a commercial party involved, is even more difficult. To make sure that the new techniques are supported and maintained, it is crucial to connect the clinic to commercial parties, because researchers will not sustain it after their research is done. They have other research to do.’
What is your personal mission within this lab?
‘My main focus is on the last pillar. Since AI is booming business, so many AI-models have been built. But what you see in healthcare is that implementing those in the clinic is the difficult part. So we try to implement clinically relevant tools, but also find out why research doesn’t end up in the clinic, and what the problems and issues are in that process.’
What kind of clinical needs are you addressing?
‘A good example is a decision aid for prostate cancer patients that we built with the company Patient Plus. As every treatment has different side effects, this tool gives patients the option to find their personal risks of getting side effects, based on their personal characteristics. Prostate cancer is an interesting choice for a decision aid tool, because this disease has a very high survival rate, which makes it possible for patients to choose between different treatments. Patients answer questions like ‘what is your age?’, ‘do you smoke?’ or ‘are you a diabetic?’ Those are all risk factors for incontinence for example. At the end the patient will get a visualization of their personal risks and learns about the disease along the ride. For this tool we have set up a collaboration with urologists that we know very well. And we then offered it to a company, under certain conditions of course, so that they can make sure it will be used in the clinic in the future.’
The lab collaborates with seven different partners. What is it like to work with so many partners?
‘It gives us a lot of flexibility. Working with this big pool of collaborators allows us to set up different alliances that are suited to answer a specific question or solve a specific problem.’
All nine PhD students of the lab are located physically at the partners and mentored by senior scientists at the partners. Why did you choose that approach?
‘In order to keep the collaborations alive and to keep the relationships good, it is important to work together, even if you don’t have a specific project that you are working on that very moment. I think it is very important to establish long-term relationships and by working together in supervision of these PhD students you achieve that.’
What do you want to have achieved in four years?
‘If anything comes out of our ICAI lab, I hope that it is raising more awareness about closer collaboration with the clinics and industrial partners. What we see a lot within the projects is that at first the people at the clinic don’t really see the need for or are a bit anxious to involve industrial parties. I don’t know why, I think it’s the non-profit versus for-profit problem. I hope that with the projects we are going to do within the ICAI lab, that this is one of the take-home messages that we can deliver. We are currently forming the bridge, and hopefully in the future they can keep finding each other without our help.’
On April 21, 2022, the Brightlands Smart Health Lab will talk about their current work during the lunch Meetup of ‘ICAI: The Labs’ on AI for Radiation Treatment in the Netherlands. Want to join? Sign up!