Bringing AI to daily clinical care
Tapping into the innovations of artificial intelligence for the benefit of patients is a hot topic these days. But how close are we to actually applying these innovations to daily clinical care? Few people want to commit to a specific timeline. Amsterdam UMC is making significant steps forward – and it’s clear that it’s not a question of ‘if’ but of ‘when’.
Restructuring for a medical data future
At the 11th Medical Data plus Pizza meeting on Tuesday 18 June 2019, the focus is on the location itself: the Amsterdam University Medical Center (Amsterdam UMC). The centre arose last year from the fusion of the city’s two main university hospitals. This restructuring came with new opportunities.
For example, the consolidated Amsterdam Cancer Center is now one of the largest in Europe and thereby better capable of attracting new partners and increased funding. Plus, the development of a shared IT structure could simultaneously confront the challenges in streamlining research using data science and AI.
However, when it comes to applying data science to a clinical setting, the key word remains ‘challenging’. In his introduction for the evening, Dr Lucas Fleuren recommends that all attendees dive deep into the US’s Food and Drug Administration (FDA) recent discussion paper on how to regulate medical devices that use artificial intelligence. Essentially, the question is: how do you determine if a product is effective and safe, if, by its very nature, it’s expected to continue to change?
To algorithm is to not to err
During the first presentation, ‘Deep learning for tumour response evaluation’, Dr Geert Kazemier notes how the expression “to err is human” needs to be updated. “To err is now also non-human. And while we find it acceptable that humans make mistakes 1 in 250 times, we can’t accept this from an algorithm.”
Dr Kazemier is a surgeon and interim clinical director at the Cancer Center Amsterdam. He is specialised in liver and pancreas tumours. Two years ago, he was approached by a global leader in analytics software, SAS, and asked how they could help. A research project was born.
“And it’s been very fruitful: I’m now asking that all our cancer researchers use it,” says Kazemier. “And to think we came to our agreement two years ago in Boston, on the night Trump became the chosen one – or at least that’s how he likes to call himself.”
Seeking cures through analytics
The collaboration shows how computer vision and predictive analytics can be applied to improve the treatment and care of patients with colorectal cancer – the third most common cancer worldwide. Almost half of all cases metastasise; a third of which metastasise to the liver.
A developed application can now capture highly detailed 3D geometries of liver lesions and tumours of patients that can help tailor treatments. In particular, it can help find those patients who respond well to chemotherapy and can thereby become more likely candidates for life-saving surgery.
While the project has only started, first findings seem to show that treatment response assessments can be done automatically, more quickly and more accurately than by humans alone.
“And it also means we can get more information to help determine the best further treatment,” says one of the researchers, Dr Nina Wesdorp. And with more information, more questions arise that can inspire the development of new models.
The current model only tests a few variables – such as size and grey-scale intensity of the lesions and tumours. But eventually they’ll be able to do it for hundreds of variables.
“We know that all sorts of variables change during treatment,” adds Kazemier. “But we still have to figure out how these corelate to clinical outcome – and survival. Then it will get very interesting. Plus, this approach can also be applied in the assessment of many types of solid tumours.”
AI is coming
“We are here to facilitate all these beautiful projects,” says Arno Sinjewel, UMC’s ICT manager of research, education and innovation. In his talk, ‘Why you should use our research infrastructure’, he outlines all the innovative – and powerful – features of his department’s new integrated multi-cloud service: the Amsterdam UMC Research Cloud.
“So please, research away!” says Sinjewel as his conclusion.
But when will all this research crossover to the real-time of the daily clinical setting? During an informal polling during the pizza segment of the evening, no one was willing to commit to a specific date. These are scientists after all.
But one thing is clear: “It’s inevitable. It just has too much potential.”
Please join our Amsterdam Medical Data Science group if you want to be on board!
About Amsterdam Medical Data Science
The Medical Data plus Pizza meeting aims to bridge the gap between health professionals and data scientists by bringing both together in an informal setting for presentations and pizza. At Amsterdam UMC data scientists, medical professionals and researchers discuss how AI (Artificial Intelligence) and medical data can be used to benefit patients and improve medical practices. Since it’s launch in August 2018 it has steadily grown in popularity with more and more people attending to hear the latest news and innovations in medical data and spark new collaborations.
For more reports from previous Medical Data plus Pizza Meet-ups, click here.
The Amsterdam Medical Data Science Group meetings are supported by The Right Data Right Now consortium, which includes Amsterdam UMC, OLVG, Vrije Universiteit, Pacmed, and the Amsterdam Economic Board.
Want to read more articles like this one?
Follow us daily on LinkedIn, Twitter and sign up for our Board Update.
Two pioneering companies, Agendia and IQVIA, are applying AI and ...
The merger of our Smart Health Amsterdam initiative with Amsterdam AI became ...
Smart Health Amsterdam is merging with Amsterdam AI. Geert ten ...