Increasing the relevance of MONAI to CAI (computer assisted intervention).

MONAI has become widely used in the field of medical imaging, thanks primarily to its ability to help process 3D imaging modalities common/unique to the field.

However, a large part of the medical engineering community doesn't deal with such modalities. For instance, contributors to the field of computer assisted intervention work far more commonly with 2D natural images such as in endoscopic computer vision, or with other modalities more specific to medical application such as ultrasound. While it is possible to use MONAI with these modalities, there is currently little value added over, for example, using torchvision for 2D natural images.

I believe there is significant scope for improved support in these areas which would help broaden MONAI’s adoption across the community and accelerate development in these fields. This is an exciting opportunity to identify pain points within the broader medical engineering community that can best be solved by a centralised movement such as MONAI.

After discussion with @ericspod, he suggested I capture some ideas on here. I will start us off with a few ideas below but I wanted to leave things open to other perspectives so please do chime in if you have any ideas.

The issue names no files, tests, or entry points and presents broad opportunities rather than a defined change. Start by identifying concrete CAI pain points and the MONAI components they affect; completion criteria would need to be agreed before implementation begins.