Research what CV models to use for xrays, and how we should approach fracture classification

[hargunmujral]

Perhaps this research paper can be of help: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8152712/

Likely, we will use ResNET or a similar model. Refer to the slides I made for more info on datasets and other models: https://docs.google.com/presentation/d/1-2Il6Uh0vAhcOTZlWOqakd6s56UVBoragmW_-3HW9Kk/edit#slide=id.g289a6c7339a_0_133

Then, decide on an action plan for steps 1 and 2 from the project outline. How should we classify between injury and not? How should we differentiate limbs? Should we fine tune a pretrained model from ImageNET? Should we perhaps combine steps 1 and 2? Are there resources that can help us with this challenge?

The goal of this ticket is to begin answering these questions, by researching and playing around. Then, create a few slides or a document outlining an action plan.