Code for diagnosing glaucoma from Lumos lens
- set up network for classification
- 1070 glaucoma images
- 180 non-glaucoma images
- data sources
- larger dataset
- use actual lumos lens
- categorize different features (like cup size, etc.) and classify with those
- review: https://dspace.lboro.ac.uk/dspace-jspui/bitstream/2134/20268/3/Muh%2BCMIG-Review2014-accepted.pdf
- previous NN paper: http://journals.lww.com/glaucomajournal/Abstract/2007/01000/Trained_Artificial_Neural_Network_for_Glaucoma.4.aspx
- poster - http://www.oic.it/~egscopenaghen2012/posters/june19/P4.58/poster.pdf
Ensure you a version of Python in the 2.7-3.6 range installed, then run:
pip install -r requirements.txt