Pre-conference workshops are included in the registration fee and require sign-up to secure your seat.
Please select your preferred workshops during the registration process (limit one per time slot).
Tuesday, May 2
8:00 a.m. – 10:00 a.m.
WS01: Workshop – Clinical Knowledge Management
Dominik Aronsky, Vanderbilt University/Semedy AG
Dirk Wenke, Semedy AG
Asli Ozdas-Weitkamp, Vanderbilt University
Health care institutions build increasingly large amounts of clinical knowledge assets. They are responsible for the accuracy, transparency and updating of the content. Inconsistent, incomplete, and outdated clinical knowledge assets represent unnecessary patient safety risks. Unfortunately, many organizations limit their knowledge management activities to a reactive and an ad-hoc approach and lack a formal content review and maintenance process, or an appropriate system that supports an institution-wide knowledge management strategy. This workshop will provide an introduction to clinical knowledge management topics that include the cataloguing of knowledge assets, authoring and modeling of metadata, managing relationship and dependencies among data, supporting the collaboration among knowledge teams and clinicians, importing and exporting knowledge assets from and to other clinical applications, guaranteeing structural and semantic integrity when knowledge assets change, data governance process, lifecycle process, and team organization. The workshop includes practical experiences, challenges and lessons learned.
WS02: Workshop – Understanding and Communicating Risk: Critical Competencies for the Healthcare Workforce
Greg Nelson, ThotWave Healthcare Analytics Academy
Monica Horvath, ThotWave Healthcare Analytics Academy
The popularity of concepts such as “Big Data” and the elusive “Data Scientist” is at an all-time high. These concepts and the application of advanced analytic techniques brings the promise of data-driven approaches closer to reality. As the velocity, variety, and volume of health data increase, so do the intellectual demands of its workforce. Healthcare organizations are being asked to stratify patients by risk as part of value-based payer contracts. To be successful, the risk assessment concepts must be connected to care management strategies and patient outcome metrics. Data literacy, numeracy, and statistical literacy are burgeoning competencies that essential for the modern healthcare enterprise. Without a firm understanding of these methods and their interpretation, data-driven strategies to improve healthcare will fall short of their potential impact. The goals of this workshop are to empower informaticists to: interpret risk, make better decisions about the viability of using risk scores, and explain risk concepts to others.
We will cover topics that include: How and why risk is commonly misunderstood; the methodologies behind how risk is calculated; techniques for interpreting and evaluating risk as produced by others; best practices for visualizing risk; and how to work with the data scientists who create the models to ensure their relevance and applicability. A better understanding of risk scores as applied to individual patient decisions stands to improve medical decision making. It will be important for clinicians to master these concepts as to think both as caregivers and business people since these scores will greatly impact reimbursement in the coming years.
10:30 a.m. – 12:30 p.m.
WS03: Workshop – Building an Innovation Ecosystem: Strategies and Lessons Learned in Spreading IT Innovations using the Diffusion of Excellence Model
Elizabeth Floto, VA San Diego Healthcare System
Vanessa Coronel, VA Boston Healthcare System
Shereef Elnahal, Department of Veterans Affairs
Andrea Ippolito, Department of Veterans Affairs
Thalia Sirjue, Atlas Research
The Department of Veterans Affairs (VA) is transforming the way it operationalizes and institutionalizes information technology (IT) solutions to meet the needs of Veterans and their families. VA is building an innovation ecosystem that empowers employees to develop creative solutions (both IT and non-IT) that address access to care, quality and safety, and care coordination in a team-based integrated healthcare setting. The innovation ecosystem empowers staff to innovate while ensuring transparency and consistency by spreading best practices to medical centers across VA. Workshop participants will learn about VA initiatives that drive performance and improve quality of care for Veterans. Participants will form teams to discuss strategies for engaging stakeholders while applying evidence-based methods to disseminate practices. Frontline VA innovators who have designed, prototyped, tested, implemented and scaled innovative projects will share case studies (eScreening and Flu Self-Reporting Desktop Icon) developed through VA’s innovation ecosystem. Participants will gain an understanding of what organizational factors, facility characteristics, and other variables lead to successful implementation. eScreening is a mobile technology developed by VA San Diego Healthcare System that can improve care coordination by offering Veteran-directed screening, real-time scoring, patient feedback, instant medical record clinical documentation, clinical alerts for evaluation/triage, and monitoring of treatment outcomes. The Flu Self-Reporting Desktop Icon allows employees to self-report flu vaccinations received outside the VA quickly, increasing reporting compliance. This practice, developed at VA Boston Healthcare System, has been replicated at 45 VA medical centers, increasing employee reporting of flu vaccinations. Participants will discuss lessons learned and formulate strategies from the VA model to spread an IT innovation across their own healthcare organization.
WS04: Workshop – Leveraging Electronic Health Records Using Recurrent Neural Networks
David Ledbetter, Children's Hospital Los Angeles
Long Ho, Children's Hospital Los Angeles
Abel Brown, NVIDIA
This tutorial session will explore how to leverage deep learning methods and Electronic Health Records (EHR) to predict severity of illness of patients in a Pediatric Intensive Care Unit (PICU). This lab will use the deep learning framework Keras to build a recurrent neural network (RNN). The resulting model provides medical professionals the capability to generate risk of mortality predictions in a temporally dynamic fashion.
1:30 p.m. – 3:30 p.m.
WS05: Workshop – Predictive Analytics using Open Source Machine Learning WEKA Software
Robert Hoyt, University of West Florida
Dallas Snider, University of West Florida
The goal of this workshop is to provide a meaningful overview of predictive analytics using the open source machine learning software, known as WEKA. We plan to discuss machine learning and predictive analytics in the first hour, with hands-on exercises using WEKA in the second hour. In the didactic section, we will discuss the history of machine learning and how it fits into the data science schema. Additionally, we will compare and contrast predictive analytics using machine learning with predictive analytics based on statistics. The common algorithms utilized in both supervised and unsupervised learning will be discussed; specifically, classification, regression, and clustering algorithms. We will recommend that attendees bring a laptop computer and download the software program WEKA ahead of time.
WEKA, is an open source program created by the University of Waikato in New Zealand. This free comprehensive package is available for Windows, Mac and Linux operating systems. The WEKA machine learning program was selected because it is accompanied by a free MOOC course and an affordable print textbook. Alternately, two affordable e-books are highly recommended for attendees. In the second hour several validated health-related public data sets will be analyzed by learners. The hands on exercises will use classification algorithms, such as Naïve Bayes, k-nearest neighbor, logistic regression and decision trees (CART). The algorithm results will be compared using standard outcome measures: recall (sensitivity), precision (positive predictive value) and area under the curve (AUC). Text mining will not be part of this presentation.
Given the user-friendly nature of the open-source software, we believe it is possible for the average clinical informaticist to use machine learning to perform basic predictive analytics.
WS06: Workshop - Agile Clinical Decision Support
DuWayne Willett, University of Texas Southwestern Health System
Vaishnavi Kannan, University of Texas Southwestern Health System
Mujeeb Basit, University of Texas Southwestern Health System
Designing effective Clinical Decision Support (CDS) tools in an Electronic Health Record (EHR) can prove challenging, due to complex real-world scenarios and newly-discovered requirements. As such, deploying new CDS EHR tools shares much in common with new product development, where “agile” principles and practices consistently prove more effective than traditional project management. Typical agile principles and practices can thus prove helpful on CDS projects, including time-boxed “sprints” and lightweight requirements gathering with User Stories and acceptance criteria. Modeling CDS behavior removes ambiguity and promotes shared understanding of desired behavior, but risks analysis paralysis: an Agile Modeling approach can foster effective rapid-cycle CDS design and optimization. The agile practice of automated testing for test-driven design and regression testing can be applied to CDS development in EHRs using open-source tools. Ongoing monitoring of CDS behavior once released to production can identify anomalies and prompt rapid-cycle redesign to further enhance CDS effectiveness. The workshop participant will learn about these topics in interactive didactic sessions, with time for practicing the techniques taught.
4:00 p.m. – 6:00 p.m.
WS07: Workshop - Fast Healthcare Interoperability Resources (FHIR): A Workshop for Implementers
Charles Jaffe, Health Level Seven (HL7)
Grahame Grieve, Health Level Seven (HL7)
Alistair Erskine, Geisinger
Viet Nguyen, Leidos
Russ Leftwich, Intersystems
Stan Huff, Intermountain Healthcare
Since its creation in HL7 more than 5 years ago, the FHIR (Fast Healthcare Interoperability Resources) platform has emerged as the best hope for achieving interoperability. FHIR was created for implementers, focused upon ease of use and rapid, cost-effective solution development. Since its creation by the Joint Standards and Policy Committees in October 2014, the Argonaut Project has fostered a unique environment for and focus upon implementation. Since the release of FHIR 2.0, there has been truly global adoption and implementation among over 300 organizations, including academia, EHR vendors, healthcare systems, BioPharma and genomics companies, the Payer community, clinical professional societies, and government agencies. FHIR 2.0 also provided the requisite stack for true intersystem authentication and security. To date, CMS, FDA, CDC, VHA, NIH, and DoD have committed to supporting FHIR implementation. In addition, FHIR 3.0 will incorporate resources for workflow as well as other enhancements.
This workshop will allow a broad cross-section of the medical informatics community to experience the promise of FHIR. Emphasis will not be placed on FHIR standards technical development, but rather on the business and process fundamental to implementation. Specific examples and use cases will be described that highlight some of the unique applications made possible by a wide range of clinical and research uses.
WS08: Workshop - Application of Model-based Computer Simulation in Medical Informatics
Duane Steward, Texas A&M Health Science Center
Model-based computer simulation is a powerful tool for investigating the behavior of complex systems. The use of computer programs to imitate the behavior of real world systems based on assumptions regarding the structure, dynamics and various attributes enables hypothesis testing, performance analysis, evaluation of alternatives and validation studies with the economy of digital modeling before physical implementation. This tutorial provides an introduction to the nature and discipline of that technology and its use in problem solving.
The aim of this workshop is to provide beginners with an introduction the fundamentals and provide a first experience with essential concepts. Although it is not possible to master the technology or the full breadth of possible application in two hours, participants can anticipate leaving with a desire to do more, learn more and be willing to invest what is required while appreciating why. Time will be devoted to a split of didactic presentation and the other half spent exercising software tools to finish partially constructed simulation projects that illustrate the concepts and practices described. Participants will be expected to install an evaluation copy on personal laptops to complete the exercises and fulfill the learning objectives.
Although this technology was first popularized before 1980 and despite its abundant use in manufacturing and business, it has not been widely used in healthcare despite persistent use of hospital emergency departments as illustrative examples since its inception. Nonetheless recent literature describes the best practice and merits of the discipline applied in healthcare. It is hoped that this introduction will endow participants with sufficient insight to better map out a migration path from quick applications at high levels of abstraction to the sequential levels of detail that will lower cost, increase demand, promote adoption and proliferate benefits.