The challenge

Nowadays we have many advanced devices for medical imaging. However, the waiting time to get diagnosed can still take up to a couple of weeks, which leads to a lot of worrying for the patient and takes too much time away from the medical experts. Luckily, by using machine learning algorithms, valuable time is saved. The only problem is, for this to work, the algorithm requires a lot of data. And that’s where we step in.

The solution

In this project, we use crowdsourcing as a tool to annotate images of lungs and provide training data for machine learning. The application we made is a game called NoDe which translates the actions of the user to data for the computer. NoDe provides not only a fun game but is also educational, simplistic and guiding. Therefore crowdsourcers with or without prior medical expertise can play this game. Our goal is to provide larger datasets of annotated lung images for machine learning. Therefore, with this knowledge, patients can be diagnosed at an early stage. This is beneficial for the patient because an early treatment increases the survival rate.

This project is being coached by