Speed dating for Doctors and Data Scientists

Calling all doctors. Calling all data scientists

On 21 May 2019, the monthly Data plus Pizza event enjoyed its largest crowd to date as it filled the Rijnzaal at Amsterdam UMC. Half the attendees were data scientists and the other half was split between medical professionals and stakeholders from the commercial sector.

Whether the pitch was related to developing a ‘Netflix Experience for hospitals’, decreasing the number of in-hospital falls, the better prediction of complications after cancer surgery, alternatives to pupillometry for detecting listening effort in the hearing impaired, or scaling a HoloLens that overcomes the freezing of gait in Parkinson’s, all presentations were greeted with knowing nods from specific members of the audience. Nods that said: I think I can help.

Gracious sharing through computational thinking

Prof. dr. Guus Schreiber, dean of the VU’s Faculty of Science, gave a short introductory speech. He stated his belief that data science will have a profound economic and social impact in the coming years. And for this reason, whether one studies medicine or the humanities, all should be educated in the potential of computational thinking. Schreiber regards the current rise of data science as akin to the internet returning to its roots. The world wide web was originally formulated by inventor Tim Berners-Lee to be open to all – “humble and unclaimable by any single entity”.

“This is why Facebook is disliked by many in the field – it’s creating its own closed space,” observes Schreiber. “So, it’s very exciting to see that the web’s original spirit lives on in this emerging new science. Yes, privacy is important. But so is the gracious sharing with others.”

Game changer for intensive care

In a later speech, Prof. dr. Armand Girbes, head of Amsterdam UMC’s Intensive Care, expanded on these thoughts by describing his own ‘wow moment’ when it came to combining medicine and data science. “Thank you all for inspiring me. I’ve long been telling the world that randomized control trials in intensive care do not work. They simply don’t make you wiser,” says Dr Girbes.

“And since hearing about this new field only a couple of years ago, I find it inspiring what data scientists and doctors are doing together. It’s fantastic to see these connections developing. We need each other. And I believe in the future such collaborations will largely come to replace randomized control trials. So, thank you.”

Doing good with data

Annanina Koster from Ynformed began her pitch with a personal story: how her once physically active mother received surgery for her osteoarthritis only to experience a slow recovery and a disappointing outcome. So, Koster thought about how real data could help better manage expectations during the shared decision-making process.

Her company produced a proof of concept using patient-reported outcome measures (PROMs) of those having undergone a hip replacement. The model’s numbers look promising. And by expanding on their model, Ynformed hope to test different treatments and thereby provide more tailored choices to fit the needs of an individual patient. One example is being able to opt for increased pain if it means increased mobility. And that’s why they need doctors: to help access more PROMs, suggest other influencing variables, and bring their model to the clinic setting for scaling.

During the question period, Dr Girbes gave some free advice: include the surgeon in the model. “After all, the quality of the painting is only as good as the quality of the painter.”

In another pitch, Brian Doelkahar, a data scientist at ABN AMRO, explained how he developed an effective model that could search for anomalies in millions of daily financial transactions. He then applied it to identify medical anomalies. His tests on photos of blood cells infected with malaria proved to be very promising. So, were there doctors here that could help take his tests to the next level? Once again, several heads in the auditorium nodded affirmative.

We have a match

Patrick Thoral (Amsterdam UMC) named his pitch ‘Save a Life: Predicting Serious Adverse Events’. Simply put, he points to a “goldmine of data” in the “free text” notes of attending nurses that might be the key to identifying potential patterns that could be a sign of impending cardiac arrest, respiratory failure or septic shock.

But yes, “free notes” are notoriously difficult to translate to usable data. So, it was an inspired choice by the organizers to follow this pitch with Martijn Schut’s ‘Mining Text with Meaning’. As a university professor and researcher at the Medical Informatics department at Amsterdam UMC, Schut is a specialist in the exact services that Thoral requires. An appointment was surely made during the pizza that followed.

In fact, small crowds formed around each of the 10 pitchers during the networking portion of the evening. Now that’s the magic of pizza…

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.