Classic Computer Vision Based CT Bone Segmentation Operator

User Story
As a developer for a clinical application that uses CT Images, I want to isolate bone structures to facilitate rest of the application processing. I want the bone localization algorithm to be GPU accelerated and done in a way so that it can be used in a MONAI Deploy application

Background
The segmentation of bones in CT scans is a crucial step for helping physicians in several medical tasks. For example, it is widely used in orthopedic surgery, in locating fractures, diagnosing bone diseases and to support planning for therapies. Although most of the bones can be visually identified in CT images without difficulties, a precise automated segmentation is still very challenging. CT scans present often a low signal-to-noise ratio, insufficient space resolution and several artifacts that make it a difficult task. Manual or semiautomatic bone segmentation is tedious, however, and often not practical.

Success Criteria

• A set of low level primitives that perform upstreamed to cuCIM
• A set of higher level 3D processing operators
• A bone segmentation operator that conforms to MONAI Deploy App SDK's operator interface and makes use of the intermediate operators
• A bone segmentation example application
• Updated API documentation

** Related Issues **

• #198