• Mar 21-24, San Francisco

    2016 Joint Summits on Translational Science

2016 Joint Summits Panels

Monday, March 21, 2016

10:30 a.m. – 12:00 p.m.

S01: Panel - Precision Medicine: Leveraging Genomics in Diverse Indications

A. Butte, University of California, San Francisco; D. McGovern, Cedars-Sinai Medical Center; O. Morozova, University of California, Santa Cruz; C. Chiu, University of California, San Francisco

Precision medicine holds tremendous potential to improve health outcomes in many different disease areas across broad socioeconomic strata. By continually expanding our understanding of the molecular and other determinants of disease, diagnoses will become more accurate and prevention and treatment strategies more targeted toward the mechanisms driving disease in different individuals. This panel focuses on nucleic acid sequence-based precision medicine advances, covering diverse approaches and diverse disease indications. Panelists will address the use of computational approaches, employing existing large data sets, for drug repositioning, for comparative analyses that enable discovery of clinical leads based on tumor genomics and for rapid and accurate identification of pathogens. They will also highlight how molecular characterization and stratification of patient populations has the potential to improve clinical management of debilitating diseases. In addition, panelists will discuss the challenges associated with integrating large and diverse data sets and how outcomes research is an integral part of advancing precision medicine into clinical practice.

S02: Panel - The Many Meanings of Precision Medicine

A. Solomonides, NorthShore University HealthSystem; R. Altman, Stanford University; J. Denny, Vanderbilt University; L. Ozeran, Clinical Informatics, Inc.; J. Tenenbaum, Duke University

Some interpret “Precision Medicine” as synonymous with Molecular Medicine, while others treat it as an improvement over the term Personalized Medicine. The concept of Translational Medicine is also linked in some way. How do these concepts interrelate? How should AMIA and its constituent Working Groups, with the different foci, view Precision Medicine – as an extension of what we have already been doing, or as a radical change of direction? Why should we consider the President’s initiative to be part of our mission? How do we interpret our own efforts in Ethics, in Genomics, in Clinical Research Informatics in the context of Precision Medicine? Speakers at this panel will assert personal positions to provoke fresh thought and a lively debate, both among themselves and with participants from the floor. There are probably as many points of view as there are possible discussants, so this will be a great opportunity to air opinions among our peers.

1:30 p.m. – 3:00 p.m.

S05: Panel - Funding Opportunities and Government-Funded Initiatives in the Areas of Translational Bioinformatics and Clinical Research Data Infrastructure

J. Skipper, HHS; J. Lee, NIH; E. Collier, NCATS; J. Larkin, ADDS; J. Rutter, NIH; M. Rocca, FDA

This panel will bring leaders across the Department of Health and Human Services to discuss key funding opportunities and government-funded initiatives in the areas of translational bioinformatics and clinical research data infrastructure.

S06: Panel - Career Opportunities to the Many Paths to Informatics

L. Wiley, Vanderbilt University; C. Overby, University of Maryland; L. Rozenblit, Prometheus Research LLC; M. Sirota, University of California San Francisco

Translational Bioinformatics (TBI) and Clinical Research Informatics (CRI) are young, diverse fields with increasing workforce demands. However due to the youth and breadth of the fields, the diversity of career opportunities are not always clear. Following the successful tradition of AMIA Student Working Group organized “career panels” at the Fall Symposium, we propose a TBI/CRI focused career panel to offer perspectives and advice for students on career opportunities and professional development. This year, the panelists include an interdisciplinary academician with work in both the TBI and CRI space, an CRI entrepreneur and a TBI academician who has also spent significant time in the pharmaceutical industry. They will share their career and educational experiences and discuss upcoming trends in informatics careers. This panel will help future and current TBI and CRI students and other early-career professionals to better prepare for and develop their careers.

3:30 p.m. – 5:00 p.m.

II03: Panel - Efficiently and Sustainably Delivering Integrated Research Registries in Academic Health Centers: Experiences from the Trenches

L. Rozenblit, Prometheus Research, LLC; I. Brooks, The University of Tennessee Health Science Center; P. Embi, The Ohio State University; M. Ahuja, T. Campion, Weill Cornell Medical College

Integrated research registries are a rapidly growing class of clinical research informatics systems that support investigators by bringing together data collected primarily for research purposes with clinical data and other institutional data. Unfortunately, many existing tools and approaches for delivering research registries are inadequate because they are either very expensive or unacceptably ad hoc. Research activities undertaken inside academic medical centers place considerable demands on informaticists supporting those efforts. On the one hand, there is tremendous demand for registries and data repositories that support research activities. On the other hand, priority, support and budgets for creating such systems are small to nonexistent, limiting their supply. In the background, institutional IT is desperate to control access to PHI and to avoid a proliferation of ill-designed, hard-to-maintain ad hoc solutions.
This panel will discuss various approaches for closing the gap between the demand and the supply for integrated research registries by focusing on highly efficient strategies for delivery. Panelists will describe both organizational and technical strategies for reducing the cost of delivery to a number that a typical research department can afford. Our ultimate goal is to establish a common approach to meeting the growing need for integrated research registries.

Tuesday, March 22, 2016

8:30 a.m. - 10:00 a.m.

S13: Panel - BD2K (Big Data to Knowledge) Program Driving Clinical Genomic and Disease Discovery

S. Madhavan, Georgetown University; J. Tenenbaum, Duke University; M. Haendel, OHSU; J. Cherry, M. Musen, Stanford University

In June 2012, a Working Group of the Advisory Committee to the NIH Director (ACD) on Data and Informatics delivered a report1 that addressed the rapid growth in biomedical data. The report contained a series of recommendations for bringing NIH and its total investigator community into the era of Big Data. The NIH Director Francis Collins accepted the recommendations and initiated the Big Data to Knowledge (BD2K) extramural program under the leadership of Dr. Phil Bourne who was appointed as Associate Director of Data Science in early 2014. On this panel, we present four BD2K funded initiatives including Standards coordination center and CEDAR.

10:30 a.m. – 12:00 p.m.

S19: Panel - Insight Generation from Real-world and Real-time Treatment Pathways

Z. Cai, Celgene; D. Gotz, University of North Carolina at Chapel Hill; A. Kamauu, Anolinx LLC; G. Petratos, Hiteks Solutions Inc.

Understanding of patient outcomes in relation to patient treatment pathways is critical for Comparative Effectiveness Research (CER) and healthcare quality improvement. The recent growing prevalence of longitudinal healthcare data from Electronic Health Records (EHR), health insurance claims, and disease/patient registries has made it possible to retrospectively reconstruct patient-level treatment pathways based on real-world data. Real-time decision support at the point of care for treatment pathways is also becoming available as the EHR systems are incorporating alerting and recommendation functionality into their user interface for physicians. The panel will focus on such retrospective and prospective pathway reconstructions and temporal visualization and analysis for better understanding treatment patters in real world. From a medical informatics perspective, this panel will address the needs to: 1) reconstruct real-patient treatment pathways by introducing high performance platforms or tools to extract relevant data elements and information out of structured and unstructured healthcare data, 2) identify real-world treatment patterns using visual analytics, and to associate patient outcomes with treatment pathways, and 3) enable answering “what if” questions by predicting outcomes of new interventions in the context of prospective treatment pathways.

1:30 p.m. – 3:00 p.m.

S21: Panel - ClinGen Informatics Resources for Researchers, Clinicians, and Laboratories

J. Cherry, Stanford University; A. Milosavljevic, Baylor College of Medicine; S. Prabhu, Stanford University; M. Williams, Geisinger Health System

ClinGen, the Clinical Genome Resource, is a National Institutes of Health (NIH)-funded project to create a centralized repository and interconnected resources for clinically annotated genes and variants that can be leveraged to improve our understanding of genomic variation and optimize its use in clinical care. In this panel, we will highlight the tools, resources, and standards developed by the ClinGen consortium to improve patient care in the information age of genomics.

Wednesday, March 23

8:30 a.m. – 10:00 a.m.

S24: Panel - Practical Implementation of Genomic Sequencing in Healthcare Settings

C. Overby, University of Maryland; M. Williams, Geisinger Health System; D. Crosslin, University of Washington; Z. Zhao, Vanderbilt University; W. Chung, Columbia University

Implementing whole genome sequencing and exome sequencing in a healthcare setting will require biomedical informatics approaches that facilitate feasible and appropriate management, analysis and return of results over an individual’s lifetime. Our panel will describe experiences with and lessons learned from current processes of receiving and using genomic sequencing results for the diagnosis and treatment of genetic disease, establishing infrastructure for reporting genome and exome sequencing results to providers and patients in a large integrated health care delivery system, current processes for managing genomic sequencing data across electronic medical records and genomics (eMERGE) network institutions, and reporting actionable results from the targeted Pharmacogenomics Research Network sequence platform.

10:30 a.m. – 12:00 p.m.

II08: Panel - I2b2: Challenges and Solutions to Integrate FHIR and PCORI CDM

K. Wagholikar, S. Murphy, J. Klann, Massachusetts General Hospital/Harvard University; J. Mandel, Boston Children’s Hospital

This panel will introduce the challenges posed in integrating FHIR and CDM into the platform for Informatics for Integrating Biology and the Bedside (i2b2). Panelists will share the experience on projects involving these models and standards. They will address the topics of impact of adoption of the standards within the current ecosystem of collaborative, distributed research, and on innovations to develop health care applications. They will also discuss ways to resolve the challenges and will share their views on evolution of the i2b2 framework and ongoing projects. The panel will engage the audience to discuss the challenges faced by the audience to adopt their informatics infrastructure to the FHIR and PCORI- CDM and how they can leverage the work done by i2b2 and SMART to address the challenges.

Thursday, March 24

8:30 a.m. – 10:00 a.m.

S33: Panel - Observational Health Data Sciences and Informatics (OHDSI): A Rapidly Growing International Network for Open Science and Data Analytics in Healthcare

J. Duke, Regenstrief Institute; G. Hripcsak, Columbia University; N. Shah, Stanford University; P. Ryan, Janssen Research and Development; V. Huser, NIH

Observational Health Data Sciences and Informatics (OHDSI) is an international collaborative creating open-source solutions for performing large-scale analytics using observational health data. OHDSI facilitates collaborative research by establishing a worldwide network of observational databases and providing a community where advanced analytic methods and interoperable software tools can be easily shared. This panel will discuss recent advances and opportunities for participation in the OHDSI collaborative, focusing on network science, data quality, evidence dissemination, and phenotype development. Ample time will be allotted for interactive discussion based on questions from the audience.

10:30 a.m. – 12:00 p.m.

II11: Panel - Creating Collaborative Opportunities in Research Data Management: Implementation and Impact of REDCAP in U.S. Department of Veterans Affairs

E. Whittier, Department of Veterans Affairs; P. Harris, Vanderbilt University School of Medicine and School of Engineering; C. Franciscus, Iowa City VA Health Care System; P. Addy, Connecticut Healthcare System/Department of Veterans Affairs; D. Hynes, Department of Veterans Affairs/University of Illinois at Chicago

Collaboration across university and US Department of Veterans Affairs (VA) affiliated medical centers is common in many multisite studies and large program projects, however use of comparable data management systems have long been a challenge. With the recent centralized and national deployment of REDCAP in the VA, new and innovative opportunities to collaborate using REDCAP are now available. This panel focuses on implementation of REDCAP, Research Electronic Data Capture, in the VA as a web-based tool to collect and manage research and quality improvement data. REDCAP provides functionality that enables multisite collaboration across VA sites and with university-based affiliates. With parallel design, data collection using REDCAP at VA and non-VA sites can be accomplished using the same schema and forms. The tool enables systematic clinical and research data collection, integration of external data sources, and export of data to statistical analysis packages.

1:30 a.m. – 3:00 p.m.

S38: Panel - Advanced Machine Learning for Healthcare

J. Sun, Georgia Institute of Technology; J. Rehg, Georgia Institute of Technology; S. Saria, Johns Hopkins University; H. Xu, UT Health; E. Xing, Carnegie Mellon

Due to the explosion of health-related “big data,” ranging from electronic health records and genomic sequencing to environmental and wearable sensor data, the healthcare industry is in desperate need of effective ways to utilize these complex and diverse data sources in order to simultaneously improve quality of care and reduce cost. Machine learning is playing a central role in data-driven healthcare applications such as clinical risk prediction, computational phenotyping, treatment recommendation, and disease progression modeling. However, most clinical informaticians have limited knowledge of or exposure to the latest developments in machine learning. This panel will discuss the latest advances in machine learning methods and models along with their applications to healthcare. In particular, we will describe scalable methods for clustering, mining, modeling, and predicting health states and treatment outcomes from diverse multimodal health datasets. The goal of this panel is to enable the exchange and collaboration between researchers and practitioners in the clinical informatics and machine learning communities.

S39: Panel - Completing the Learning Health Care System Cycle: Developing and Testing eCQM in pSCANNER, a PCORNet Research Network

L. Schilling, University of Colorado School of Medicine; E. Holve, AcademyHealth; J. Goldwater, National Quality Forum; P. Payne, The Ohio State University; D. Meeker, University of Southern California

The goal of this panel is to discuss the ways in which two historically independent data-driven communities key to the Learning Health System, researchers and quality measure developers, have come together to advance the work of both. Members of this panel are collaborating on a prototype infrastructure to advance the work of both communities. Measure developers often test measure feasibility with select health care organizations, but broad feasibility testing and the ability to assess a measure’s true predictive correlation with meaningful outcomes is difficult. The clinical research informatics community has made progress in technologies that support the reuse of EHR data however, both communities face challenges of EHR data quality, heterogeneity, timeliness, accessibility, and concomitant regulatory “hurdles”.