Big abstract task right now, will be split into smaller tasks as we work on them. Should probably only be done after we figure out how to segment the cervix.
Some ideas right now:
Determine % cervix covered by cancer, possibly through pre-CNN color/texture filters, with the % being fed in as an input to the CNN.
Find the entrance to the cervical canal; determine if it is "blocked" with cancer
Big abstract task right now, will be split into smaller tasks as we work on them. Should probably only be done after we figure out how to segment the cervix.
Some ideas right now:
Determine % cervix covered by cancer, possibly through pre-CNN color/texture filters, with the % being fed in as an input to the CNN.
Find the entrance to the cervical canal; determine if it is "blocked" with cancer