Hope Valley Hospital (a fictional institution) wants to train AI diagnostic
models and also deliver AI based diagnostics based on those models.  The on higher quality and more diverse data to reduce population bias.
They
However they cannot aggregate data into a single location, as is done
traditionally, due to various privacy laws differing in strictness and breadth.

This demo shows how Hope Valley can train a model without ever moving the data
or sending the whole model to any of their partner hospitals.  The result is
a robust model which can be commercialized privately.

The trained model (aka "algorithm") is used in two ways.  First, it can be
executed as a "federated" model within the network of Hope Valley Hospital
clinics.  These organizations have the highest level of trust and can gain
some operational speed by running in this mode.

Secondly, the model is shared with one of the partners, but using SMPC.  This
gives an additional layer of protection to the intellectual property of
Hope Valley while allowing their partner to benefit from the full power
of the model.
