What is Classifeye?
Simply put, can we find a way to accelerate the work being done by scientists to analyse these samples in order to identify positive treatments.
Think of it as 2 scientists spending 2 weeks in a lab, or 50 members of the public spending just 5 minutes on the tube one day. Mindblowing.
How does it work?
The basic idea is to take tumour micro array spots like this:
Split them down into small high resolution samples like this:
Adjust the images to make them easier for the general public to read/classify like this:
Then ask the public some questions about the sample. Their responses can then be used in huge-scale aggregate form to provide statistically meaningful data.
How Far Did You Get?
So far we have explored
- Web App using sliders for classification
- Web app using buttons for classification
- iPhone/Android HTML App
- iPad Web App which can be found here (Looks good on iPads and pretty bad on everything else. There's only so much you can do in 2 days...)
The next steps will be to take everything we learned from hacking something together in 48 hours, extract the positives and build something solid and scalable with slightly more care and attention than the original 48 hours allowed!
We would love it if Cancer Research, Good For Nothing and Citizen Science wanted to take this forward. Trust me when I say that we really do get that passionate about this kind of thing all the time, especially in this case and given what's at stake, if you want this, we will make it happen.