annotations indexed by opensearch

[TanyaStere42]

Feature description and context

Now that we have a foundation for data processing in opensearch, we can start indexing across the biological data of the platform. We will start with annotations as they are the foundation to all CV AI in the platform.

Primary use of search on Annotations is for matching set building; therefore, care should be given to include fields needed for this task (e.g. coming from Encounter, MediaAsset).

Feature sign-off requirements

Annotation data is indexed in opensearch across the following fields:

• Encounter-sourced data (for matching set, see above):
  • taxonomy
  • locationID
  • ownership (is this still used in matching set?)
  • (others? e.g. date? tbd)
• MediaAsset: (unsure if needed for matching set, but might be desirable)
  • id
  • metadata: major mime type, exif timestamp perhaps
• id
• acmId
• viewpoint
• iaClass
• ia status?
• matchAgainst?
• perhaps bounding box?