FIRE under the Electronic Health Record (FHIR – Google HL7 Opensource)

By Ram Kadambi

Posted on March 5, 2018

With good amount of technology penetration and consumer / patient adoption of the medical technology, the Electronic health record is getting more acceptance and usage. With multiple player playing role on the structure and communication protocol, the common standard for both medical device manufacturers and software developers is the need of the hour. HL7 has been there for some time, the structure and protocol covered is very exhaustive(Which may not be applicable for all players).
HL7 came up with FHIR (pronounced as Fire) as a combination of HL7 multiple version with CDA and focus on implementation and adaptation by consumers (and providers).
Some of the key features of the FHIR is quick kick start guide based on examples, full support to REST services, OAuth security implementations and ontology based analysis.

HL7 partnered with Google cloud in early last year to support FHIR development. Google team partnered with HL7 and have the opensource version released. The Google cloud enables Machine learning and pattern recognition to be leveraged for apps. The clinical data that are applicable in FHIR format allows predictive analysis to be performed and improves accuracy of early detection.

Google conducted full version usage of Electronic Health Record with University of California San Francisco (UCSF), Stanford University, and University of Chicago Medicine (UCM) found representations of a comprehensive patient EHR using Fast Healthcare Interoperability Resources (FHIR) can be used for more accurate predictive analytics.

The ability to bring in multiple types of healthcare under the FHIR also enables deep learning on the images and voice based medical data. The future of the Google patnership in Health record space will enable leveraging TensorFlow and Chatbot integration. We foresee health record authenticated consumer to talk to the bots and get responses for the health record. The aggregate sanitized health records will prove as a big assets for clinical trial and getting training data sets.

The FHIR will also create ecosystem for technology team to customize applications for each speciality. The medical history is sequenced over the time and predictive analysis based on Snomed Code (diagnosis and Finding code) with ICD can help in creating recommendations for the patient. Since, each consumer is unique, the ability to apply multiple combinations in realtime will help the care giver to focus on right approach.

The startup community in tech and healthcare space should look at adopting and performing analytics using FHIR soon.

FIRE under the Electronic Health Record (FHIR – Google HL7 Opensource)