• 2017 Joint Summits

    Informatics Implementation Track

    Apple ResearchKit | CDISC | CTSA | CTSI | EMERGe | FHIR | HealthLink  | HL7 | i2b2 | LOINC | MD2K/OpenMhealth | Mole Mapper | MyHeart Counts | OCHIN | PCORI-CDM | PCORNet | PRIDE Research | PROMIS | REDCap | SMART

2017 Joint Summits Informatics Implementation Track

Creating Informatics Tools, Resources, Training Opportunities, and Organizational Models to Support and Transform Translational Research

The AMIA 2017 Joint Summits Informatics Implementation Track is dedicated to exchange of information and ideas, report-outs from best-of-breed software teams, gap analyses, requirements gathering, professional development opportunities, networking sessions and mentoring opportunities geared specifically to support existing or new CRI implementation experts. 

Content for the four-day event consists of: interactive working sessions designed to promote collaboration and group-wide discussions; invited talks and panel discussions from leaders in the field; and exercises designed to promote networking and partnership.

The lead designer of the Informatics Implementation Track is Bernie LaSalle, Director of Operations Biomedical Informatics Core, Center for Clinical and Translational Science, University of Utah University of Utah


Who Should Attend?

Individuals whose role within their organization primarily focuses upon the practical development, deployment, and management of clinical research information management tools and platforms should attend the Informatics Implementation Track at the 2017 Joint Summits on Translational Science.

These professionals could reside in academic or private-sector organizations and examples include:

  • software implementers
  • functional informatics managers from NCI Cancer Centers
  • CTSA or CTSI affiliated programs
  • individuals working for clinical research management organizations, or bio-technology/pharmaceutical companies

Content for the Informatics Implementation Track was curated by the Track Leads.

The IIT Planning Committee:

Bernard A. LaSalle, Chair, University of Utah Health Science Center
William R. Hogan, MD, University of Florida
Leslie D. McIntosh, PhD, MPH, Washington University
Jihad Obeid, MD, Medical University of South Carolina
Adam B. Wilcox, PhD, University of Washington


Day 1: Setting the Stage
Day 2: Structure
Day 3: Methods
Day 4: Application

DAY 1:

Monday, March 27

10:30 a.m. – 12:00 p.m.

IIT01: Panel - i2b2 and REDCap Dynamic Data Pull (DDP) and REDCap: An Auto Instantiated Data Mart

Diane Keogh, i2b2 Foundation
Paul Harris, Vanderbilt University
Robert Taylor, Vanderbilt University

The panel will discuss existing linkages (from community) and prospectively work to formally define a plan for solving two use cases (i2b2 à REDCap DDP; REDCap as an auto-instantiated Data Mart)

About REDCap: REDCap is a secure web application for building and managing online surveys and databases. While REDCap can be used to collect virtually any type of data (including 21 CFR Part 11, FISMA, and HIPAA-compliant environments), it is specifically geared to support online or offline data capture for research studies and operations. The REDCap Consortium, a vast support network of collaborators, is composed of thousands of active institutional partners in over one hundred countries who utilize and support REDCap in various ways.

The REDCap Consortium has 2,189 active partners in 108 countries. REDCap software has generated over 363,000 projects from over 464,000 users generating 2,942 published journal articles

About i2b2 Foundation: i2b2 (Informatics for Integrating Biology and the Bedside) is an NIH-funded National Center for Biomedical Computing based at Partners HealthCare System. The i2b2 Center is developing a scalable informatics framework that will enable clinical researchers to use existing clinical data for discovery research and, when combined with IRB-approved genomic data, facilitate the design of targeted therapies for individual patients with diseases having genetic origins. This platform currently enjoys wide US and international adoption including the CTSA networks, academic health centers, and industry. i2b2 is funded as a cooperative agreement with the National Institutes of Health.

i2b2 is now a non-profit foundation that was established in January of 2016. The mission of the foundation is to provide on-going support to the i2b2 community and continued development of the platform.

1:30 p.m. – 3:00 p.m.

IIT02: Tools for Accessible and Reproducible Computing in Microbiome Research

Alexander Alekseyenko, Medical University of South Carolina
Paul McMurdie, Wholebiome, LLC
Susan Holmes, Stanford University
Gail Rosen, Drexel University

Human microbiome, the collection of all microorganisms cohabitating the human body, is making an impact on a broad range of biomedical fields. Actionable associations of the host microbiota with health conditions span many health areas, such as oral and dental health, inflammatory bowel disease, metabolic syndrome, and other systemic conditions, and cancers. The underlying data enabling microbiome research are high-throughput, complex, and heterogeneous. However, relatively few informatics efforts are dedicated to microbiome research at present. This leaves the investigators with a need for user accessible means to process their data. In response to this need, several platforms have been developed to ease analysis of the data. These platforms range from systems implementing specialized analysis programing languages, to libraries for popular programming environments, such as R and python. This panel presents several state-of-the-art approaches, developed to enable reproducible computing for microbiome research. The practices employed by these approaches are generalizable to other high-throughput informatics domain and present interesting use cases and design patterns for the field. 

3:30 p.m. – 5:00 p.m.

IIT03: REDCap as a Platform – Innovating Through Implementation

Paul Harris, Vanderbilt University Medical Center
Robert Taylor, Vanderbilt University Medical Center

REDCap is a secure, web-based application designed originally to support data capture for research studies so that researchers had ready access to centralized methods for planning, capturing, storing, managing, and exporting research study data. REDCap was further developed for widespread applicability for the national research community and eventually led to the development of the REDCap Consortium, now supporting ~2,200 academic, non-profit and government organizations in 108 countries (with 475,000 end users, and 3,100 publications).  We have and will continue to employ a cooperative development methodology in which users are engaged in idea generation, design, beta testing, and refinement.

The REDCap cooperative has benefitted from strong innovation for diverse use cases catalyzed by end-users across the world (study coordinators, data managers, statisticians, investigators, etc), as well as from the informatics community.  To encourage and facilitate further innovation, we have built into the REDCap software extensive capacity for plug-ins and a rich library of web services.  As a result, we and others are increasingly using REDCap as a platform to solve complex use cases and support systems-based interoperability.

This session will begin with a brief overview of REDCap and a description of programmable features/functions that should be of interest to and enabling for implementation technologists.   Next, we will review a number of example ‘power use cases’ leveraging REDCap as a platform.   Finally, we will provide specifications for a driving clinical research informatics use case and issue a challenge to attendees to create a REDCap-based solution over the course of 2-weeks.  A judging panel will convene to select a winner following the two-week submission deadline with cash prize of $2,000 awarded along with an invitation to provide a webinar demonstration to the REDCap consortium.  Cash prize will be provided by Vanderbilt University Medical Center (this is not an AMIA award).

DAY 2:

Tuesday, March 28

8:30 a.m. – 10:00 a.m.

Informatics Implementation Keynote Panel

Informaticians who Got the Job Done: So if You Don't Know, Now you Know

Bernard LaSalle, University of Utah Health Science Center

Leslie McIntosh, Washington University in St. Louis

The AMIA Joint Summits Implementation Track is designed for individuals whose role within their organization focuses upon the practical development, deployment, and management of clinical research information management tools and platforms and the growing role of managing the data collected within these systems.

Like a ‘year-in-review’, we will present recent work highlighting tools, platforms, and data through noted publications. Looking beyond publications, though, we will also feature news coverage, blogs/social media, websites, and applications. While aimed at clinical informaticians, topics outside this domain may be included as well to showcase salient work from which we could learn.

10:30 a.m. – 12:00 p.m.

IIT04 – DataEd Hackathon

The Open Data for Discovery Science Education Hackathon (DataEd Hackathon) seeks to encourage multi-disciplinary teams to develop educational modules that will enable to trainees to develop the practical skills needed for the analysis of publicly available, large-scale personalized “–omics” data in a reproducible and rigorous manner.

Reproducibility and rigor in biomedical research that leverages open data is an important topic that has been a subject of intense recent interest, and further, is not often addressed in Biomedical Informatics training curricula.

In response to this gap in knowledge, the goals for the DataEd Hackathon are:

  • To highlight differences in approach to the problem of reproducing and/or replicating findings from published literature and;
  • To extend the analysis of the dataset to other publicly available resources towards generating novel hypotheses.

1:30 p.m. – 3:00 p.m.

IIT05: Panel Structures in Implementation: Experience of the EDM Forum

Margo Edmunds, Academy Health
Michael Kahn, University of Colorado
Philip Payne, Washington University
Adam Wilcox, University of Washington

The EDM Forum was created in 2010 with funding from AHRQ to coordinate lessons learned regarding implementation among multiple large informatics grants designed to advance the use of data for comparative effectiveness research. Since that time, the EDM Forum has been actively coordinating and disseminating methods for implementation in the fields of health services research and informatics.

As the funding from AHRQ is ending, this presents an opportunity to review both the lessons learned regarding coordination of large data initiatives, as well as insight in new directions anticipated for the EDM Forum. The EDM Forum has been a unique structure facilitating a broader understanding of clinical informatics implementation.

3:30 p.m. – 5:00 p.m.

IIT06: Panel: Workforce Structures for Implementing a Learning Health System

Umberto Tachinardi, University of Wisconsin-Madison
Genevieve Melton-Meaux, University of Minnesota
David Vawdrey, Columbia University
Tara Payne, Lumeris

This panel is made up of individuals who have specific leadership roles in advancing the use of data and analytics for their organizations. In a post-Meaningful Use world, the challenge moves from EHR implementation to analytics and reporting implementation. In the same way that the CMIO role emerged before, the Chief Analytics Officer or Chief Data Officer is now emerging. Each will discuss their current analytics role within the health system, and describe their organizations’ analytics strategy. Discussion will then be about common threads in the strategies among the organizations, that will hopefully help other institutions who may be pursuing similar implementation challenges.

DAY 3:

Wednesday, March 29

8:30 a.m. – 10:00 a.m.

IIT07: Panel: Implementing Record Linkage Solutions in the PCORnet Clinical Data Research Networks

Jiang Bian, University of Florida
Abel N. Kho, Northwestern University
Russ Waitman, University of Kansas
Xiaoqian Jian, University of California San Diego
Shaun Grannis, Regenstrief Institute/Indiana University
Toan Ong, University of Colorado Denver

Record linkage is a fundamental problem in combining data from different sources to provide a unified view of the data. Nevertheless, record linkage in healthcare is not a trivial task. The common approach in healthcare settings uses personal identifiable information collected from different sources to perform either a deterministic or probabilistic matching, with manual review to resolve uncertainty.

However, in building the National Patient-Centered Clinical Research Network (PCORnet), different organizations are developing different record linkage strategies customized due to many factors including the availability of these personal identifiable data, local technology infrastructures, data use agreements and data sharing polices, security and privacy considerations, institutional review board (IRB) concerns, and relationships with various stakeholders. Each of the panelists represents a Clinical Data Research Network (CDRN) within the national PCORnet. Each of us will describe and share our experiences with implementing record linkage solutions within our respective CDRNs, and then discuss the pragmatic implications, insights, and lessons learned.

10:30 a.m. – 12:00 p.m.

IIT08: Panel: Rigor and Reproducibility in Biomedical Informatics

Elaine Collier, NCATS/NIH
Lucila Ohno-Machado, University of California San Diego
Leon Rosenblit, Prometheus Research, LLC
Casey Overby, Johns Hopkins University

Research reproducibility and verifiability are essential components of the scientific method. Quality documentation and transparency throughout the research data lifecycle enhance the reliability of findings aimed at understanding and improving health outcomes. Sharing biomedical health and clinical data is complex given the sensitive nature of the information, the large size of the data set, the constantly evolving nature of the data, and the varying health data formats. Yet, given the increased volume of electronic health records data and secondary research data available for reuse - criteria, tools, and recommendations for reproducibility within these fields are of significant importance.

This panel will provide attendees with an overview of the current state of rigor and reproducibility in biomedical informatics particularly with the following: funding mechanisms in this nascent domain; tools to support rigor and reproducibility; improving rigor through data provenance practices; and, improving rigorous workflow practices in biomedical informatics research.

1:30 p.m. – 3:00 p.m.

IIT09: Data Science Training Workshop: Using Jupyter Notebook and R (with a little Spark)

Connie Zabarovskaya, Washington University in St. Louis
Lorinette Wirth, Saint Louis University
Cynthia Vitale, Washington University in St. Louis
Leslie McIntosh, Washington University in St. Louis

Improving skills and updating knowledge are a lifelong challenge in a dynamic field such as informatics. Implementing rigor and reproducibility in biomedical informatics research includes employing tools and practices for a robust research workflow. One cutting-edge tool is Jupyter Notebook, which “support[s] interactive data science and scientific computing across all programming languages (http://jupyter.org).” Jupyter Notebook offers an open-source, interactive environment to combine data code, text, and output in one application. While many programming languages can be used, we will practice with the popular analytics software R (https://www.r-project.org) to highlight many of the features in Jupyter. Additionally, we will demonstrate Apache Spark (http://spark.apache.org) so attendees can understand the versatility of the notebook using this powerful, open-source cluster computing framework.

This workshop will include:
A basic background on Jupyter Notebook and its components
Benefits of using Jupyter Notebook and comparison to other tools.
Hands-on Exercises in Running R in Jupyter Notebook
A demonstration of combining the power of Jupyter Notebook on top of Apache Spark


Thursday, March 30

8:30 a.m. – 10:00 a.m.

IIT10: Panel - Data Extraction for National Data Sharing Initiatives: Process Pragmatics and Comparison/Commonality of Pipelines

Patrick Ryan, OHDSI/Janssen Research and Development
Roy Pardee, Group Health Cooperative
Jon Puro, OCHIN

This panel will focus on how well do people extraction pipelines work as they’ve gotten involved in these national sharing efforts like OHDSI, i2b2, PCORnet. The goal of the panel is to present an honest discussion of the challenges with this type of ETL process, what works and where are the ‘pain points’.

10:30 a.m. – 12:00 p.m.

IIT11: Panel-  Research Networking

Griffin Weber, Harvard Medical School
Julia Trimmer, Duke University
Alex Viggio, University of Colorado Boulder
Eric Meeks, University of California San Francisco

Research Networking Systems (RNS), originally described as ‘expertise mining‘ applications, have become more common in the biomedical community. But there are challenges in launching and supporting an enterprise RNS as well as values and opportunities, and we are just now starting to understand how these systems influence research and researcher behavior in an increasingly online world. Four panelist with experience in establishing Research Networking Systems at their institutions will share the ups and downs, both anticipated and often otherwise, in charting a path to success with their RNS installations.

1:30 p.m. – 3:00 p.m.

IIT12: Panel - Optimizing Data Extraction: Performance and Maintainability - Lessons Learned

Michael Matheny, Vanderbilt University
Jack London, Thomas Jefferson University
Umberto Tachinardi, University of Wisconsin Madison

This panel will focus is on actual performance at scale: how long it takes to do the actual ETL into the destination data-mart?

Maintainability: Is the process transparent and maintainable for the team?
Reproducible: Do results differ from one release to another?
Shareable: Can you share it with others?
Portable: Will it work with other database technologies?