Skip to content

Project-Herophilus/IntegrationFramework-HDF

Repository files navigation

Integration Framework for HDF (Healthcare Data Foundation) Background

the Integration Framework for HDF (Healthcare Data Foundation) is intended to be a tier of reusable and extensible capabilities. As such, it is all about enabling a consistent design pattern/accelerator based mindset to help healthcare organizations innovate in a new and comprehensive manner. Our focus has been, and will continue to be, how we simplify data access within healthcare and focus on ensuring data is the asset.

Here are a few key things to know as you read this:

  • All the Integration Framework for HDF assets, like all assets within Project-Herophilus, are available and provided "as-is" under the Apache 2 license model.
  • The basis of all the efforts for Intelligent DaaS is that Data is the Asset and this set of repositories are intended to help anyone connect and build innovative platforms for usage with a variety of data: healthcare and beyond.
  • Intelligent DaaS (Data as a Service), or any of its components, ARE PROVIDED UNDER Apache-2 Open Source licenses AS IS. THEY ARE NOT PRODUCTS AND WILL NOT BECOME A PRODUCTS. We work very hard to ensure they continue to drive enabling capabilities in a consitent manner, as a design pattern/accelerator. This specific repository is focused around providing comprehensive healthcare connectivity and routing of data.
  • Integration Framework for HDF is all about enabling connectivity to data. For ease of use and scale we have isolated each specific type of connectivity into its own specific solution. This repository is all about showcasing capabilities through software and is a value add.
  • This platform has everything of other integration platforms such as management interfaces. There is also the capability to quickly and easily customize where you want logs to end up with numerous industry leading technologies such as Data Dog, Splunk, etc. In addition, the platform enables implementations to also build highly customized visualizd dashboards as well simply with Grafana.

Integration Framework for HDF (Healthcare Data Foundation) - A Proven Reference Architecture/Design Pattern Based Platform

Just to provide a few visuals to how to visualize it as we talk about each submodule within each part of the platform. Below please find a visual that does visualize the entire iDaaS capabilities set. The key thing to note is while each specific iDaaS capability is purpose built and designed for any type of customer, without regard for implementation type, to ensure our focus is on meeting data where it is securely and at scale.

Processors and Converters

Each Integration Framework for HDF component/module has a consistent pattern within it, this approach consists of having processes and converters. By doing this you can fully customize the experience needed for the most common tasks. This design enables the simplist to most complex implementations to be delivered consistently.

Processors

Processors, as hopefully depicted by the name, are intended to be for specific processing needs for data enablement capabilities.

Processor Description
Cloud Processing for public cloud components
Deidentification Processing for deidentification
EMPI Processing for EMPI (Enteprise Master Person Index)
HEDA (Healthcare Event Driven Architecture) Processing for built a specific healthcare event driven architcture
SDOH Processing for SDOH (Social Determinants) data
Terminologies Processing for terminologies

Converters

Converters are designed with ONLY one function, to convert data from one format to another. As of Integration Framework for HDF v4 there is ONLY code based conversions available. However, there has been some work to start enabling mapping based conversions.

Processor Description
CCDAtoFHIR Converting from CCDA Data format to FHIR
HL7toFHIR Converting from HL7v2 to FHIR

Statistics

Stats are now included for all components that are leveraged. Statistics have a naming convention for ease of use.

Processor Description Sample Naming Convention
Errors Any exception/error handling (module name)_exception
Processing Components Common processing components (Procesing Component Name)_Inbd_ProcessedEvent
Converters Conversion components (converter)Conversion_(Direction)_ProcessedEvent
General Any component within any modules (Protocol)_(TransactionType)_Inbd_ProcessedEvent

Integration Framework for HDF is all about Connecting and Data Integration

Like most of the repositories provided for usage, The iDaaS Connect repository consists of numerous modules within it. The reason for this approach is because iDaaS Connect focus is about providing a wide variety of connectivity options based on specific needs or industry centric specific connectivity needs.

Here is a detailed and Cloud Agnostic visual:

iDaaS Cloud Agnostic Data Flow.png

Here are the modules that are in Integration Framework for HDF.

Identifier Description
Blue Button BlueButton is intended to be a very specific implementation to support puling of data to support several defined and specific government initiatives.
Cloud Cloud is intended to be a very specific set of connectivity for the three major public cloud vendors - AWS, Azure and GCP.
CMS Interoperability CMS Interoperability is intended to be a very specific project to implement the wide variety CMS Interoperability use cases.
EDI EDI has been a standard around for decades, this repository does not introduce capabilities that compete with capabilities vailable for claims processing or other EDI very specific needs. The intent of this repository it to enable the processing of EDI data such as claims and Supply chain.
FHIR FHIR is a modern based integration standard that has been adopted by the government to assist them in addressing new federal mandates such as the Interoperability and Patient Access Rule. The Integration Framework for HDF-FHIR component fully supports integrating to multiple external vendor FHIR servers in a consistent design pattern manner.
HL7 HL7 v2 is a very legacy based client/server socket protocol that has been out for decades and has thousands of unique implementations across healthcare vendors. Additionally, CCDA was developed for web services based XML based document exchanges of clinical data.
NCPDP This iDaaS Connect accelerator is all about providing the capability to process NCPDP data. This accelerator apart from handling the needed specific connectivity also does a minimal initial routing of data and has complete auditing integrated.
SAP This iDaaS Connect accelerator is specifically designed to receive data from the various SAP supported connectors. This accelerator apart from handling the needed specific connectivity also does a minimal initial routing of data and has complete auditing integrated.
Third Party This iDaaS Connect accelerator is specifically designed to receive data from several dozens connectors. The connectorsinclude JDBC (any jdbc compliant data source with a jar), Kafka, FTP/sFTP and sFTP, AS400, HTTP(s), REST and many more. Since this accelerator is built atop the upstream of Apache Camel this accelerator can leverage any supported components. This accelerator apart from handling the needed specific connectivity also does a minimal initial routing of data and has complete auditing integrated.

Community Add-Ons

To support any iDaaS branded artifact there are a subset of assets avaiable to assist in the implementation :

Add-On Description
Project Herophilus Technical Content Variety of content specific to iDaaS - PowerPoints, Word Docs, and Draw.IO diagrams
Test Data Test data for all healthcare industry standards work
Synthetic Data When it comes to building or testing assets having lots of data to resemble you production data is critical, it also helps drive innovation. We have open sourced a synthetic data platform and continue to enahnce it based on feedback. It currently contains over 18 billion specific data attributes that can be used
Data Simulators Having data is one aspect, the other is having tooling that can quickly be configured and used to test components. We have developed a data-simulators for our core components, Simulators to help in implementation and testing

Additional Add-Ons

To support any iDaaS branded artifact there are a subset of assets avaiable to assist in the implementation :

Add-On Description
API - Testing Leverage the published PostmanCollections, they are named by community and specific reference architecture they support.

Administering the Platform - Management and Insight of Components

Administering and seeing what the plaform is doing is also critical. Within each specific repository there is an administrative user interface that allows for monitoring and insight into the connectivity of any endpoint. Additionally, there is also the implementation to enable implementations to build there own by exposing the metadata. The data is exposed and can be used in numerous very common tools like Data Dog, Prometheus and so forth. This capability to enable would require a few additional properties to be set.

Below is a generic visual of how this looks (the visual below is specific to iDaaS Connect HL7):

Integration Framework for HDF Data Flow - Detailed

For more specific details around the management and administrative capabilities you can go here.

Platform General Pre-Requisites

For all iDaaS design patterns it should be assumed that you will either install as part of this effort or need the following on any environment you choose to implement the platfom on.