• AMIA 2018 Annual Symposium

    Data, Technology, and Innovation for Better Health

AMIA/HL7 FHIR® Applications Showcase

AMIA and HL7 have partnered to bring an exciting application showcase to attendees of the AMIA 2018 Annual Symposium. 

The HL7 Fast Healthcare Interoperability Resources (FHIR®) Specification has become one of the most rapidly adopted methods for accessing information in health information technology systems. As patients move around the healthcare ecosystem, their electronic health records must be available, discoverable and understandable. Further, to support automated clinical decision support and other machine-based processing, the data must also be structured and standardized.

HL7 has been addressing these challenges by producing healthcare data exchange and information modeling standards for over 30 years. FHIR® is a new specification based on emerging industry approaches, but informed by years of lessons around requirements, successes and challenges gained through defining and implementing HL7 v2, HL7 v3 and the RIM, and CDA. The health IT community has rapidly adopted the FHIR® Specification, and the AMIA/HL7 FHIR® Applications Showcase it is intended to showcase the best and most innovative FHIR® Apps.

FHIR® is the registered trademark of Health Level Seven International.

S76: AMIA/HL7 FHIR® Applications Showcase
Tuesday, November 6
10:30 a.m. – 12:00 p.m.

SMART on FHIR Shared Decision Making System for Diabetes Pharmacotherapy

Shinji Tarumi; Hitachi America, Ltd.

Predictive analytics can help enable individualized care optimized for a given patient. Together with the University of Utah, Hitachi has developed a machine learning based pharmacotherapy decision support system (PDSS) for Type-2 Diabetes Mellitus (T2DM) that is seamlessly integrated with the electronic health record (EHR) through the SMART on FHIR interoperability framework. The PDSS compares alternative pharmacotherapy options as well as weight loss, providing personalized guidance on predicted efficacy, risk of side effects, and costs. To facilitate shared decision making, the information is provided in a way that is easy for both clinicians and patients to understand. The system has been integrated with the Epic EHR at the University of Utah through the SMART on FHIR framework, with a goal of scaling the solution across different EHR platforms and healthcare systems. In this session, it will be shown how the PDSS helps guide both clinicians and patients make informed decisions for T2DM.

Pain Management Summary: A SMART on FHIR Dashboard for Managing Pain

Chris Moesel; The MITRE Corporation\

The Agency for Healthcare Research and Quality (AHRQ), in partnership with the MITRE Corporation and OCHIN, developed a SMART on FHIR app that serves as a dashboard or summary for managing chronic pain and pain-related treatment within the electronic health record (EHR). The app was developed as part of an ongoing project called CDS Connect, which provides a platform for authoring and sharing publicly-available clinical decision support (CDS) knowledge resources. The SMART on FHIR app summarizes pain-related information in a single view, such as pain-related diagnoses, medications, lab results, and pain assessment scores. Visual cues point the user to potential issues, e.g., when current medications include both benzodiazepines and opioids. The app has been piloted through OCHIN, a health information and innovation network serving safety net clinics, where clinicians found it “simple and intuitive.” The app and detailed implementation guidance will be available in September 2018 on CDS Connect and on GitHub as open source code.

NLP2FHIR: A FHIR-based Clinical Data Normalization Pipeline and Its Application on Electronic Health Records (EHR)-Driven Phenotyping

Na Hong; Mayo Clinic

The next generation phenotyping of Electronic Health Record (EHR) features the identification of true patient state in an accurate and high-throughput manner. To realize this vision, there is an urgent need to improve the reproducibility and interpretability of the underlying phenotype models and algorithms through a standards-based framework. The Fast Healthcare Interoperability Resources (FHIR) standard was developed to meet a variety of clinical interoperability needs. In this showcase session, we will demonstrate a FHIR-based clinical data normalization pipeline known as NLP2FHIR developed at the Mayo Clinic and an application of this pipeline to perform EHR-based phenotyping. As a case study, we applied the NLP2FHIR to the task of identifying the presence of obesity and 15 of its comorbidities in a given patient from their semi-structured discharge summary. We demonstrate the value of the NLP2FHIR in enabling precise clinical data capturing and making EHR-based patient identifications more reproducible and interpretable.

Meducation® - A SMART on FHIR Integrated Solution Enabling Improved Patient Understanding of Medication Instructions and Regimens

Daniel Fritsch; First Databank, Inc.

Prescribed medications can only be effective when taken properly. Yet, one in three patients struggle to understand the medication instructions provided to them. Meducation® is a FHIR®-enabled solution – accessible to health care providers within their clinical workflow via their EMR system – that delivers medication instructions that especially target individuals considered to be “high risk” due to low health literacy, impaired vision, and language barriers. It enables health care providers to create personalized patient medication instructions, in more than 20 languages — written at a 5th to 8th grade reading level with large font sizes, pictograms, and videos — to make understanding of how to take medications intuitive and simple.

MedTrue™ – Geisinger/Merck Collaboration

Julie Rockey; Geisinger

Inaccurate medication lists can lead to preventable medication errors that annually attribute to a $21 billion loss, 7.1 million serious medical events, and over 7,000 deaths. To help address this national problem, Geisinger and Merck collaborated to develop MedTrue™, a web-based, EHR-agnostic workflow solution that will use SMART on FHIR technology to present a provider with a patient-validated medication list that includes self-reported adherence data as well as medication fill data from a national, leading prescription data aggregator. Pilot testing of MedTrue™ at several Geisinger clinics confirmed the need for direct embedment of the solution within the provider’s EHR workflow. With enhancements completed from the pilot testing, MedTrue™ is now being further evaluated as part of an exploratory evaluation study with results anticipated to be submitted for publication while the collaboration also explores commercialization of MedTrue™ with other organizations in the United States.

YouScript Leveraging the Power of Precision Medicine Through Smart on FHIR

Kristine Ashcraft; YouScript

YouScript Precision Prescribing Software identifies at-risk individuals within patient populations, helps medical professionals interpret pharmacogenetic testing results, and provides expert support for medication management. The YouScript Precision Prescribing Software is the first and only analytics software tool available to prescribers that predicts potential drug-drug and drug-gene, and cumulative drug interactions based on both cytochrome P450 metabolism and genetic testing. Our software is available in all major electronic health records covering over 60% of the US population with the capacity to impact internationally.

Beginning with a Healthy Start: Reducing the Need for Non-Preventive Healthcare Utilization in the First Month of Life

Andrew Douglas Robinson; UCSF

Feeding is often difficult for breastfed newborns, and feeding difficulty can lead to hyperbilirubinemia and dehydration. Together, these three interrelated diagnoses (feeding difficulty, hyperbilirubinemia, and dehydration) are the most frequent barriers to a healthy first month of life and cause the large majority of neonatal readmissions. Further, monitoring and treating these diagnoses is a common cause of both extended length of stay during the birth hospitalization and of increased outpatient utilization after the birth hospitalization. Earlier identification of infants with feeding difficulty might allow effective interventions to address this problem and prevent hyperbilirubinemia and dehydration, improving the length of stay during the birth hospitalization, readmission following discharge from the birth hospitalization, and the number of outpatient visits in the first month. In order to accomplish this, we have implemented a SMART on FHIR integration of the Newborn Weight Tool (NEWT).

DeepAISE on FHIR — An Interoperable Real-Time Predictive Analytic Platform for Early Prediction of Sepsis

Ashish Sharma; Emory University

Sepsis, a dysregulated immune-mediated host response to infection, is lethal, prevalent, and costly. Its early detection has the potential to drastically reduce morbidity/mortality. We have developed a real-time cloud-based application that predicts onset-time of sepsis based on live ICU data and provides clinicians with actionable visual alerts. Clinicians and nurses can examine these alerts and initiate appropriate interventions. The prediction engine (DeepAISE) is a Deep Learning-based algorithm trained to reliably predict sepsis 4-6 hours in advance of clinical recognition. A scalable, cloud-based, system continuously streams bedside data and uses the prediction engine to generate hourly scores and displays these to clinicians. Interoperability is achieved through the use of FHIR resources and APIs. This system is monitoring ~100 patients on a daily basis at the Emory Tele-ICU center and has been shown to reliably predict onset of sepsis with an AUC of 0.9.

CentricityTM Healthcare's FHIR application enables ‘Blue Button’ integration between Beth Israel Deaconess Medical Center and Fenway Health.

Kedar Ganta; VVC Holding Corp. (formerly GE Healthcare VBC)

CentricityTM FHIR application enables ‘Blue Button’ integration with Beth Israel Deaconess Medical Center (BIDMC) and Fenway Health. Fenway Health and Beth Israel Deaconess Medical Center (BIDMC) use two different EHR systems. The providers at BIDMC often see patients who are treated at Fenway Health and the providers need access to patient records stored at Fenway Health to make better care decisions on time. CentricityTM FHIR application is integrated into BIDMC EHR so that the providers at BIDMC can query for patient records without leaving their EMR screens. The providers at BIDMC can click on the button inside their EMR to launch a query containing Name, Gender, Date of Birth against Fenway Health’s CentricityTM EMR. The CentricityTM FHIR application establishes a trust between BIDMC and Fenway Health systems to authenticate and get appropriate patient records.


AMIA 2018 Annual Symposium Sponsors

AMIA First Look

AMIA Working Group Supporters (NIWG)


Student Paper Prize