Medical MVP: Choose a dataset for the demo

[bcebere]

We have to choose a dataset that can help us demo the following:

• MONAI support/any other established medical framework.
• Usable from a mobile device, for demonstrating the ease of use.
• Relevant information in the interpretability benchmarks, for overcoming the medical skepticism to AI.

[tudorcebere]

I'll look into this.

[andreimano]

I found a VQA dataset, it's rather small (5k samples), but we should be able to do some kind of transfer learning on it: https://www.imageclef.org/2020/medical/vqa

EDIT: Also, an overview on medical VQA - https://lhncbc.nlm.nih.gov/system/files/pub9992.pdf