FruitPunch Challenge
Share your expertise and innovative ideas and contribute to improving care for premature babies. Dutch student teams were given this amazing opportunity this month in the FruitPunch AI Challenge.
Lees artikel in het Nederlands >
Non-profit organization FruitPunch AI, dedicated to the use of artificial intelligence in healthcare, develops challenges for students to learn about Artificial Intelligence (AI) and further develop their knowledge and skills in it. Together with Utrecht University, Eindhoven University of Technology and Wageningen University, we are encouraging the use of AI at UMC Utrecht by supporting and guiding these challenges.
Is day-night rhythm for premature babies essential?
Where last year students from UMC Utrecht worked on an AI model to predict sepsis in premature babies, this year's Challenge revolved around an AI model to map the day-night rhythm for premature babies in incubators, with the question: does the Neonatology department of UMC Utrecht have a day-night rhythm? The students also looked at the impact of the presence or absence of such a day-night rhythm on the growth of premature babies. Furthermore, they looked at infection rates and brain development using MRI scan results.
Result: AI model that automatically maintains day-night rhythm
To determine the optimal day-night rhythm for premature babies in incubators, the students analyzed data from many babies and combined it with advanced machine learning technologies. Thus, the students managed to develop a machine learning model that automatically represents the day-night rhythm. In the future, this AI model can hopefully help improve the care of premature babies and reduce the risks of health problems by giving us insights into the impact of day-night rhythms on the room.
On to the next challenge
All data, which the students used, could be securely shared via Digital Research Environment (DRE), UMC Utrecht's secure digital research environment. Thus, for the second time, an AI question was successfully tackled in a FruitPunch challenge. Neonatologist Jeroen Dudink of UMC Utrecht: "AI has the potential to support medical professionals and improve patient care. Collaborating with data scientists and physicians during the FruitPunch Challenge to achieve this was a very valuable experience. I am looking forward to the next project."
About Digital Health
The UMC Utrecht is strongly committed to healthcare innovation. We do this by working with patients, general practitioners and other healthcare providers to create a seamless care pathway for patients and give patients more control. In doing so, we use ICT and mobile eHealth technology, such as apps and home monitoring devices. This gives patients more self-direction at home and is supported by care when it is really needed or wanted with more person-centered advice and better coordination between care providers.
View the most recent Data Science and eHealth solutions from UMC Utrecht.
Translated with DeepL