AWS Machine Learning Blog

Announcing the InterSystems HealthShare Message Transformation Service for Amazon HealthLake

Amazon HealthLake is a new HIPAA-eligible service designed to store, transform, query, and analyze health data at scale. Amazon HealthLake removes the heavy lifting of organizing, indexing, and structuring patient information to provide a complete view of the health of individual patients and entire patient populations in a secure, compliant, and auditable manner. With the Amazon HealthLake APIs, organizations can easily store health data already in the Fast Healthcare Interoperability Resources (HL7 FHIR) industry standard to a secure data lake in the cloud.

For those organizations who don’t have their data in FHIR format, Amazon has partnered with industry leaders in healthcare interoperability to develop connectors, which help you with data transformations. This post highlights one of those organizations, InterSystems, and their solution validated for use with Amazon HealthLake, called the HealthShare Message Transformation Service.

About InterSystems

Established in 1978, InterSystems provides innovative data solutions for organizations with critical information needs in the healthcare, finance, and logistics sectors, and beyond. Their cloud-first data platforms solve interoperability, speed, and scalability problems for organizations around the globe. InterSystems was awarded the 2020 and 2021 KLAS Category Leader for Interoperability Platforms and was named a Visionary in the first Gartner Magic Quadrant for Cloud Database Management Systems (DBMS).

HealthShare Message Transformation Service

The HealthShare Message Transformation Service is a software as a service (SaaS)-based cloud service that provides a fully automated single interface method for healthcare message transformation into HL7 FHIR. The service enables automated conversion of existing data standards starting with HL7v2 to FHIR for use in Amazon HealthLake.

How it works

The InterSystems HealthShare Message Transformation Service enables you to upload an HL7v2 or InterSystems defined CSV formatted file into an Amazon Simple Storage Service (Amazon S3) bucket of your choosing. The service configures secure access to the S3 bucket and your Amazon HealthLake data store (in the your AWS account). After it’s configured, you can write HL7v2 messages or CSV files into your S3 bucket, and InterSystems processes the file, transforms it to FHIR R4, and populates Amazon HealthLake within seconds.

To better illustrate this, let’s look at an example HL7v2 MDM message for a clinical note put into an S3 bucket with a .txt extension:

MSH|^~\&||HC6|||||MDM^T02|||2.5
EVN|T02|20210104094500
PID|||LD572046^^^HC6^MR||Smith^John||19301019|M|||1 Memorial Drive^^Cambridge^MA^02142||||||||063070516
PV1||O|||||ISCGP001^Moore^James|||||||EO|||||HSVN00006|||||||||||||||||||||||||20210104094500|20210104094500
TXA||Progress note||202101040945|JJ021^James^John||||ISCGP001^Moore^James|||19815952^TRANS
OBX||FT|RTF^TRANS||Patient complaining of pain in right big toe. Toe red in color. To commence on antibiotics. ||||||R

The HealthShare Message Transformation Service from InterSystems converts the message into a properly formatted FHIR R4 DocumentReference and writes this to Amazon HealthLake, where Amazon HealthLake automatically applies natural language processing to the clinical note. The preceding HL7 message results in Amazon HealthLake suggesting an ICD-10 code of M79.674 (pain in right toes), condition of pain, as well as organ site of “big toe” and generic antibiotics for medication.

The following diagram illustrates the overall architecture of the HealthShare Message Transformation Service.

 

CureMatch to use the HealthShare Message Transformation Service with Amazon HealthLake

CureMatch™, Inc. is a San Diego-based digital health company focused on personalized medicine and combination therapy in oncology. CureMatch’s Decision Support System guides oncologists in the selection of cancer drugs that are customized for individual patients based on their molecular tumor profile. This provides them with actionable intelligence towards advanced cancer treatment options.

CureMatch plans to use these two solutions to improve efficiency and scale faster by automating data transformations and using the features within Amazon HealthLake to move to an even more AI-empowered and personalized approach.

“With the use of the HealthShare Message Transformation Service by InterSystems and Amazon HealthLake, we will be able to access and transform molecular profile data from EHR into FHIR to run advanced analytics and algorithms, providing clinical decision support and guidance on treatments to ultimately improve patient outcomes,” says Philippe Faurie, Vice President of Professional Services at CureMatch.

Conclusion

Amazon HealthLake transforms unstructured data using specialized machine learning models, like natural language processing, to automatically extract meaningful medical information from data, and provides powerful query and search capabilities. InterSystems, with their ability to quickly ingest and convert HL7v2 messages into FHIR and populate Amazon HealthLake, provides a quick and easy way to use Amazon HealthLake to unlock the full potential of your healthcare data.

Amazon HealthLake with the HealthShare Message Transformation Service from InterSystems can help extract meaningful information from raw, disparate, unstructured data across your organization. To learn more, visit the InterSystems Amazon Marketplace listing or the InterSystems product page.


About the Authors

Todd Sylvester is Director of Cloud Strategy at InterSystems.

 

Brian Warwick is a Solutions Architect supporting global AWS Partners who build healthcare solutions on AWS. Brian is passionate about helping customers leverage the latest in technology in order to transform the healthcare industry.