What happens if you let data-scientists, hackers and coders play around with random medical data? Probably a lot. Probably a whole lot. You would see chaos arise within a couple of minutes!
But what if you would zoom in to a very specific medical data-set? For example DNA-samples of people who suffer from a specific type of leukemia… then you would end up with a briefing close to this:
We will give you a USB drive with datasets about sick (leukemia) and healthy people’s cells and hope you can tell the difference at the genomic level. We hope your data analysis can tell us more about why one patient responds better to treatment than other patients.
Date: 9 – 11 June 2017
Location: Sugar City (Halfweg)
Expert 1: Peter van der Spek (Erasmus Medical Center)
Expert 2: Jules Meijerink (Prinses Maxima Centrum voor kinderoncologie)
Leading up to the event Van der Spek and Meijerink gathered the data of 270 children with leukemia. Because these children have participated in medical research in the past, their DNA data could be used. These datasets were labeled with information of which therapy they have been treated and how effective this has been. There were close to 54.000 datapoints per patient. Needless to say, the data was anonymized before it was shared with the hackathon participants.
The participants were able to find a number of predictive indicators. They even managed to locate a “gene cluster” that can predict whether or not the treatment will be effective. This was something that (traditional) medical researchers weren’t able to find up to now.
Disclaimer: the findings of the hackathon still need to be scientifically validated before they can be used in practice.
- Space for Good - May 8, 2019
- Chimpface - February 5, 2019
- AI tegen stropers, illegale houtkap, en andere bedreigingen - February 4, 2019
Category: EnglishDiscover more: hackathon > health > medicine