In our series of articles about artificial intelligence (AI) we discuss the biggest opportunities and challenges for scaling and implementing AI into life sciences and healthcare. In this blog we discuss diagnostic AI tools with Bram van Ginneken, Professor of Medical Image Analysis at the Radboud University Medical Center.
AI solutions in healthcare
As discussed in our AI whitepaper two important challenges in expanding the use of AI in the clinical setting concern data (gathering, regulations and security) and finding ways to integrate AI into existing medical workflows. Professor van Ginneken likes to highlight another huge burden for bringing AI into the medical practice: the aspect of reimbursement.
“The question is: Who is going to pay for the newly available technological solutions?”
– BRAM VAN GINNEKEN – PROFESSOR OF MEDICAL IMAGE ANALYSIS, RADBOUD UNIVERSITY MEDICAL CENTER
We currently seem to be at the peak of the hype cycle of AI: expectations are high but many new AI tools don’t make it to the market. “The question is: Who is going to pay for the newly available technological solutions? How are AI companies going to achieve a return on investment? How are hospitals going to pay for it?”, poses van Ginneken.
In the current healthcare system, each “approved” medical action receives a “code” for which insurance companies pay a reimbursement. It differs per continent and country how reimbursement is organized. Optimally navigating and using this system is a big challenge for developers of new AI-based diagnostic tools, according to Professor van Ginneken: “The healthcare system is extremely complicated and many ways in which it is organized are problematic. The potential of new diagnostic tools are huge, and there is a significant willingness of medical professionals to use them, but the current reimbursement system slows down progression. It requires an extensive lobby of companies and patient organisations to have new diagnostic AI tools (currently categorized as “advanced processing” operations) included into new codes, and thus into the reimbursement system.”
The healthcare sytem itself is limiting the potential of AI
Because of the complicated reimbursement system, the cost-benefit analysis of hospital departments for AI investments become quite peculiar. Van Ginneken explains: “If a new application lowers expenses directly, for example due to a reduction in the amount of required manual labour in case fewer radiologist need be hired, it is attracting to invest in it. But if an AI tool can lower the number of MRI scans needed by predicting its usefulness in individual cases, the revenue coming in from reimbursed MRI scans can be lowered and then hospital managers are unwilling to introduce such applications.”
The patient is the victim
Life science companies need to come up with strategies to have their technology implemented. They try to make deals with national healthcare systems, for example. But this does not necessarily lead to better care. “Some national healthcare providers, for example the NHS in the UK, are suggesting in their reports that they believe data should stay within a country and could only be made available to companies active in that country. This may result in limiting the rights to share data. The result could be that in a particular country, only a subset of the available software could be used or reimbursed. In the US it is easier to patent algorithms and software than in Europa and companies with patents can prevent competitors, who may have developed better algorithms for a specific task, from entering the market. The result is that the clinics are running suboptimal or needlessly expensive solutions.” The current healthcare system and its financial structure limit the development of new diagnostic tools and thereby the quality of care. “Ultimately this policy negatively affects the outcome for the patient. That is a problem of the healthcare system.”
Co-diagnostics and deal making as a solution
Next to lecturing activities, managing his academic research group, working for Fraunhofer MEVIS in Bremen (Germany), professor van Ginneken also has founded Thirona, a company that develops software and provides services for medical image analysis.
Van Ginneken elucidates how Thirona tackled the reimbursement issue for one of their solutions “We developed a so called “co-diagnostics” system. Simply put, we earn money by providing AI driven analysis. Our clients, medical centres, load their data onto our cloud and we run our algorithms on these data sets. We subsequently provide the medical specialists with our insights and advise. We are reimbursed by the suppliers of biotech products such as artificial heart valves. It is in their interest that the right patient is selected for the treatment they offer, and therefore they are willing to pay for our service.”
With the developments in AI and the wave of new products and services being launched, all eyes are on AI experts like Professor van Ginneken. Scientific researchers, developers and AI entrepreneurs receive a lot of attention from multiple parties from many different fields. “Not a day goes by where I do not receive a request for a new collaboration, helping organisations with lobbying, giving lectures, or setting up new studies.”
The mission of ttopstart is to transform healthcare and life sciences by accelerating the introduction of impactful innovations like AI in healthcare. Our way to realize this is by co-creating competitive strategies with and for people like van Ginneken. “It is very hard for an expert like me to decide on the right strategy, to define our long-term goals, how to plan for it and who to work with. ttopstart has been a great partner for me in this process”.
Our whitepaper will provide you an overview of AI solutions in healthcare. We discuss the opportunities and challenges of implementing AI in healthcare and give actionable insights to help you overcome barriers and develop smarter strategies that accelerate the introduction of such AI innovations.
Download whitepaper AI
In this whitepaper ttopstart will give you insights on accelerating Artificial Intelligence for healthcare and life sciences. This paper will provide an overview of the opportunities and challenges of implementing AI in healthcare and give actionable insights to help the different stakeholders to overcome barriers and develop smarter strategies that accelerate the introduction of such AI innovations.