Talking with a friend of mine, Machine Learning specialist @Renato Banzai , he brought one of the biggest challenges faced by companies nowadays: deploying ML/AI in live environments.
InterSystems IRIS offers IntegratedML. IntegratedML it's a great feature to train, test, and make deploys of ML/AI models.
The hardest part in creating ML/AI is to make the data treatment, clean up, and make them trusted.
This is where we can take advantage of FHIR powerful standard!
The project idea shows how we can create/train/validate ML/AI models with FHIR and utilize them with data from different sources.
We believe that this project has great potential and a few ideas that can be explored:
- Reuse/extend DTL transformations in other FHIR databases for custom ML models
- Use DTL transformations to normalize FHIR messages and publish ML models as services
- Create a kind of models + transformations rules repository for use within any FHIR dataset
Exploring new possibilities with this project, we can imagine data from different sources.
In the image above, the FHIR Resource, consuming the REST API, can be used with an FHIRaaS.
nd not only using FHIRaaS on AWS, but we can also make use of the new service HealthShare Message Transformation Services, that automates the conversion of HL7v2 to FHIR® to populate Amazon HealthLake, where you can extract more value from your data.
With these small demonstrations, I see these resources being used very well in larger scenarios, enabling and delivering more easily deploys in production in truly innovative environments, like the AWS Healthlake. Why not?! 😃
If you liked the idea, enjoying what we are doing in the community, please consider voting for fhir-integratedml-example and help us in this journey!