18:30 Welcome drinks & food
19:00 Presentations (see program below)
20:30 Drinks & networking
Artificial Intelligence is probably the biggest buzz word of 2017 and 2018, and with good reason! Unfortunately, many developments that can be categorized under the moniker ‘AI’ haven’t matured enough to leave the research and innovation labs. Healthcare however seems to be an industry that, as an early adopter, is already reaping the benefits of applying AI as part of their core processes. Machines turn out to be often better at distinguishing between benign and malignant tumors, or at diagnosing rare genetic conditions using facial recognition algorithms. This doesn’t mean your oncologist or radiologist will be out of a job anytime soon; it only means their work will be of higher quality due to intelligent assistants that help them to obtain better diagnoses. During this meetup we’ll highlight the most recent developments in AI assisted healthcare and illustrate this with experiences from the field.
Program: - Welcome and Opening: Jos van Dongen, SAS
- Speaker 1: Joost Huiskens, SAS – “From research to day to day care”
Within the Cancer Center Amsterdam, multiple projects are generating large amounts of data that can result in promising leads to novel oncological diagnostics and therapeutics. In 2017 the first use case for Artificial Intelligence on Image Analytics was born in collaboration with the SAS Research and Development Labs. CCA and SAS started with the ‘CEASAR-project’ with the goal to only select patients with colorectal liver metastases for a treatment that will benefit from it. In the first phase of the CAESAR-project image analytics is used to assess the response to chemotherapy.
- Speaker 2: Daniel Vijlbrief, UMC Utrecht – “Big data for small babies”
Is it possible to proactively treat an infection in premature babies or even prevent it with the use of advanced analytics? This was the question that the Neonatology Department (NICU) of UMC Utrecht wanted to answer. Daniel Vijlbrief will explain how they developed a smart algorithm that can confirm or deny the suspicion of an infection in premature babies. Can unnecessary use of antibiotics in premature babies be prevented?
- Speaker 3: Arjan Sammani & Nathan van der Lei, UMC Utrecht & Finaps –“Medic Miner, text mining in healthcare”
Registration regulations are a big burden in healthcare. The Medic Miner is a novel text mining tool, which automates diagnosis registration whilst keeping the doctor at control. The algorithms help to discover patterns between the presence of certain terms and properties of the text, such as medical history of a patient. By analyzing texts from the electronic health records, suggestions are given for standardization of diagnosis registration. This helps to speed up and improve the quality of the performed registration. The development, as well as the future of the algorithms and application will be elaborated and discussed.