Full Tutorials

T01: Primer Series

An Introduction to Clinical Natural Language Processing

Leonard D'Avolio, VA Boston Healthcare System, Harvard School of Medicine, Dina Demner-Fushman, National Library of Medicine, Wendy Chapman, University of Pittsburgh, John Pestian, Cincinnati Children's Hospital Medical Center, University of Cincinnati

Natural language processing is the umbrella term used to describe the automated structuring and extraction of information formatted as free text. The demand for natural language processing technologies in medicine will grow significantly in the coming years. This growth will be fueled by the continuing adoption of the electronic medical record, increasing emphasis on quality measurement and improvement initiatives, and the growing need for evidence to be used as part of evidence-based medicine. This half-day tutorial is designed to introduce clinicians and informaticians to the practice, tools, techniques, and science of clinical natural language processing.

Instruction will be hands on, inter-active, and case driven. The tutorial will focus primarily on clinical NLP, although related uses and methods such as literature-based NLP and text mining will be discussed to lend context. Topics covered include: an overview of clinical NLP and its uses in medicine; a brief history of clinical NLP and the evolution of NLP methods; the challenges to NLP; the number of approaches used to process natural language and the strengths and weaknesses of each; implementation considerations, creating annotated corpora as training / test sets, evaluation of NLP, and a review of open source tools for natural language processing. Demonstrations and in-class exercises will be used to help tie the theory of NLP to everyday research problems addressed by these technologies. The tutorial will be taught by four instructors experienced as researchers, developers, and users of a variety of tools and approaches to clinical NLP. Users will also be exposed to several open source technologies for clinical NLP including the Unstructured Information Management Architecture (UIMA) and Knowtator for manual annotation. They will also experience, first hand, the challenges of clinical NLP through manual annotation of de-identified patient records.

Outline:

• Overview: What is NLP and how is it being used in medicine?
  • Literature
  • Clinical reports
  • Applied to bioinformatics
• What makes clinical NLP so difficult?
  • Overview of policies affecting NLP
  • Characteristics of the clinical documentation environment
• Different approaches to clinical NLP
  • Simple rules-based (information extraction)
  • Statistical
  • Symbolic or grammatical
  • Hybrid approaches
• The clinical NLP process
  • The various components of clinical NLP
  • Annotated corpora for training \ testing
  • The pipeline for clinical NLP software
  • Evaluation (its role in the process)
• Available open source tools and components
• Implementation considerations
• Evaluating clinical NLP (in greater detail)

By the end of this tutorial, attendees should be capable of:

• Describing the current uses of clinical NLP
• Describe the relationship between clinical NLP and related techniques such as text and data mining
• Understand the challenges to clinical NLP
• Describe the various approaches to clinical NLP and their strengths and weaknesses
• Understand the process of clinical NLP and its various components
• Find available open source clinical NLP components, frameworks, and packages
• Identifying potential implementation concerns and challenges
• Understand the process of creating and using annotated corpora
• Interpret the performance of published clinical NLP research

Intended Audience: Any clinician or medical informatician with an interest in learning more about clinical NLP.

Content Level: 70% Basic, 30% Intermediate

T02: Primer Series

Using Social Media for Healthcare Institutions

Faculty: Cynthia Manley, Shana Bresnahan, Betsy Brandes, Gretchen Purcell Jackson, Vanderbilt University, and Michèle Ledgerwood, Global Health Policy Center at the Center for Strategic and International Studies (CSIS)

Social media are Internet-based communication tools that allow the creation and dissemination of user-generated content. These emerging technologies offer healthcare institutions exciting ways to interact with the community and to broadcast information about their clinical programs and research discoveries. They also pose daunting new challenges as patients and families can rapidly share both positive and negative experiences with the world.

This tutorial provides a general introduction to social media, with an emphasis on popular social networking applications such as Facebook and Twitter. Participants will learn the basic functions of these tools and see examples of how healthcare and public health organizations have used them both to gather and disseminate information. Instructors will outline guidelines for developing social media policies for individuals and groups and discuss the time, staffing, and skills needed to manage social media interactions for a healthcare institution.

By the end of the tutorial, participants will be able to:

• Identify the general types of social media applications
• Describe the functions of the most popular social media applications
• Enumerate the potential uses of social media by health-related organizations
• Discuss issues of privacy and security in use of social media
• List important topics and issues that social media policies should address

Outline of Topics:

• Social media platforms (e.g., Facebook, Twitter)
• Social media applications
• Using social media for healthcare delivery, research, and teaching
• Privacy, security, and medical-legal concerns
• Strategies and guidelines for social media use
• Policies for online personal and professional behavior
• Global public health awareness and advocacy through social media
• Social media in pandemic and humanitarian emergency response

Intended Audience: Healthcare administrators, clinicians, biomedical scientists, policy makers, computer scientists, and system developers who are interested in using social media or informatics professionals who may conduct research about these technologies.

Content level: 40% Basic; 50% Intermediate; 10% Advanced

T03: E.H.R. Series

Evaluating Health IT Projects: A Practical Approach

Caitlin M. Cusack, Insight Informatics, Eric G. Poon, Brigham and Women's Hospital/Harvard Medical School, and Alana Knudson, NORC

This tutorial provides real-world tools to assist in choosing measures and constructing evaluation plans in HIT from the theory of evaluation to practical approaches to planning and executing evaluations.

With the influx of money into the health information technology (HIT) industry from the Health Information Technology for Economic and Clinical Health Act (HITECH) under the American Recovery and Reinvestment Act (ARRA), the number of organizations implementing technology into their healthcare delivery processes is anticipated to grow rapidly. Historically, evaluation of health IT has been conducted by large organizations and academic centers, many with home-grown systems whose findings may not be applicable to those embarking on implementation today. These new implementations represent a fresh chance to study health IT and to evaluate its impacts on safety, quality, efficiency and effectiveness on healthcare from outside these traditional venues. This tutorial will review the importance of evaluating the impacts of health IT, present tools to assist in designing an evaluation plan, review common pitfalls in evaluation, and present a practical case study. In addition, the tutorial will give some focus on how to disseminate findings from one’s evaluation, a critical piece to assist others in their journey to adopt and use health IT.

By the end of the tutorial, participants will be able to:

• Compile metrics for a health informatics project
• Determine which metrics are practical to measure
• Formulate a plan around the chosen metrics
• Know where tools and resources are located
• Understand common challenges of evaluation
• Understand the basics of disseminating findings

Outline of Topics:

• Motivations Behind Health Informatics Evaluations: Why Do We Care About Evaluation?
• Choosing Evaluation Measures: A Practical Approach
• Study Design Considerations
• Evaluation Tools
• Case Study: Evaluation on a Shoestring
• Dissemination
• Common Pitfalls and Challenges

Intended Audience: Evaluators, researchers, physicians, nurses, and other healthcare professionals. The tutorial is intended for those individuals seeking guidance around evaluating health IT projects. Basic knowledge of medical informatics is assumed.

Content level: 50% Basic; 45% Intermediate; 5% Advanced

T04: E.H.R. Series

Practical Modeling Issues Representing Coded and Structured Patient Data in EHR Systems

Stanley Huff, Intermountain Healthcare

This tutorial provides models for flexible representation of patient data; the proper roles for standard terminologies like LOINC, SNOMED CT, First DataBank, and RxNORM; approaches to handling pertinent negative findings and negation; support for precoordinated data entry while storing the data in a post coordinated database; and storage of data that belongs to another patient in the patient record.

The tutorial describes the need for formal data models for the EHR and how standard terminologies are used in the models. Starting with use cases encountered while developing EHR systems at Intermountain Healthcare, the instructors will discuss the basic name-value pair paradigm for flexible representation of patient data; the proper roles for standard terminologies like LOINC, SNOMED CT, First Data Bank, and RxNORM; approaches to handling pertinent negative findings and negation; support for precoordinated data entry while storing the data in a post coordinated database; and storage of data that belongs to another patient (baby or donor) in the patient record.

There are no absolute prerequisites for this tutorial. However, those who have experience in designing, developing, configuring, and implementing EHR systems will find the tutorial more meaningful. Experience in modeling of medical data and knowledge of standard coded terminologies like SNOMED CT, LOINC, and RxNORM will also be very helpful.

By the end of the tutorial, participants will be able to understand:

• The assumptions and motivation for formal definitions of detailed clinical models
• How standard coded terminologies are referenced by detailed clinical models, and the different roles that SNOMED CT and LOINC play in the models
• The various alternative logical models for implementing clinical models related to diagnoses, allergies, problems, procedures, orders, and observations.
• The importance of adhering to terminology and modeling standards in developing or purchasing interoperable EHR systems
• National and international activities for sharing models that enable interoperability of EHR systems

Outline of Topics:

• What are detailed clinical models?
• Why are detailed clinical models important?
• What are the requirements for defining and using detailed clinical models?
• Name-value pair (NVP) and entity-attribute-value (EAV) strategies for representing clinical data
• What are the proper roles for use of LOINC, SNOMED CT, drug codes (First Data Bank, RxNorm) and classifications in the models
• The necessity of supporting both pre and post coordinated models in a clinical system
• Approaches to the representation of negation and pertinent negative findings
• Storing data that belongs to another person (relative, family member, donor) in the patient record
• Specific alternatives for modeling including observations, diagnoses, problems, procedures, allergies
• Open candid discussion of ideas that the participants have about ways that the modeling issues can be addressed
• Importance of supporting open consensus standards for EHR systems that are purchased or developed
• Brief discussion of various national and international activities related to formal clinical data models

Intended Audience: Designers, developers, implementers of EHR systems, scientists, educators, researchers, and biomedical engineers interested in clinical data modeling and interoperability of EHR systems.

Content level: 50% Intermediate, 50% Advanced

T05: E.H.R. Series

Human-centered Design and Evaluation of Health Information Systems

Jiajie Zhang and Vimla L. Patel, University of Texas Health Science Center at Houston

EHR usability and human-centered design is a major barrier to EHR adoption and achievement of meaningful use. A current and significant challenge in the design and implementation of health information technology (HIT) is to deal with the high failure rate of HIT projects. A large number of HIT projects fail. Most of these failures are not due to flawed technology, but rather due to the lack of systematic considerations of human and other non-technological issues in the design and implementation processes. In other words, designing and implementing HIT is not so much an IT project as a human project about human-centered computing such as human-computer interaction, workflow, organizational change, and process reengineering. Due to the complexity and unique features of healthcare, human-centered methods and techniques specifically developed for healthcare are necessary for the successful development of health information systems that would increase efficiency and productivity, increase ease of use and ease of learning, increase user adoption, retention, and satisfaction, and decrease medical errors, decrease development time and cost, and decrease support and training cost.

In this tutorial we will teach one such method that is specifically developed for Electronic Health Records (EHR) systems. After the half-day tutorial, the attendees should have a basic understanding of the usability issues in health IT and should be able to use the learned method to evaluate the usability of EHR and related products.

By the end of the tutorial, participants will be able to:

• Understand the principles of human-centered design
• Understand the significance of human-centered design and the consequences of technology-driven development
• Use human-centered methodologies and techniques to evaluate health information systems
• Use human-centered processes to design health information systems that have good usability

Outline of Topics:

• The human side of human-computer interaction
• Basic frameworks and methodologies of human-centered design
• Usability evaluation techniques
• UFuRT: a new framework and process for human-centered design and evaluation
• Case studies: usability evaluation of EHR
• Case studies: usability evaluation of medical devices
• Patient safety and human-centered design

Intended Audience: Scientists, researchers, physicians, nurses, and other healthcare professionals; computer scientists, system developers, and programmers; graduate students and postdoctoral fellows.

Content level: 40% Basic, 50% Intermediate, 10% Advanced

T06: Methods Series

A Gentle Introduction to Support Vector Machines in Biomedicine

Alexander Statnikov, Constantin F. Aliferis, New York University; Douglas Hardin, Vanderbilt University; Isabelle Guyon, Clopinet

This half-day tutorial will introduce support vector machines (SVMs) and their applications in biomedicine. SVMs are among the most important recent developments of machine learning and pattern recognition and have extensive applications in biomedicine and other fields. Unlike other approaches, these techniques are robust in data analysis with high variable-to-sample ratios and large number of irrelevant variables, they can learn efficiently very complex functions, and they employ powerful regularization principles to avoid overfitting.

A common obstacle in understanding and using SVMs is that they are mathematically challenging especially for biomedical researchers lacking extensive technical backgrounds. The tutorial is designed to enable all interested researchers grasp SVM fundamentals regardless of prior mathematical training. First, we will introduce basic principles behind SVMs in an intuitive manner. Then we will describe SVM-based algorithms for classification, regression, clustering, novelty detection, and variable selection. These algorithms are widely used and/or gaining popularity in biomedical applications. Throughout the tutorial we will provide case studies for each class of methods and give pointers to software implementations and additional literature. The knowledge gained in this tutorial will allow researchers to break the barriers of classical statistics and older pattern recognition and be able to conduct complex and high-dimensional analyses easily and efficiently.

By the end of the tutorial, participants will be able to:

• Execute complex machine learning and pattern recognition tasks in high-dimensional biomedical datasets by following the protocols and principles presented
• Efficiently perform variable selection using state-of-the-art SVM-based methods
• Understand advantages and limitations of SVM-based algorithms compared to classical statistics and other pattern recognition techniques

Outline of Topics:

• Part I
  • Introduction
  • Necessary mathematical concepts
  • Support vector machines (SVMs) for binary classification: classical formulation
  • Basic principles of statistical machine learning


• Part II
  • Model selection for SVMs
  • Extensions to the basic SVM model:
    • SVMs for multicategory classification
    • Support vector regression
    • Novelty detection with SVM-based methods
    • Support vector clustering
    • SVM-based variable selection
    • Computing posterior class probabilities for SVM classifiers


• Part III
  • Case study 1: Application and comparison of multicategory SVMs and state-of-the art classifiers in cancer microarray gene expression data
  • Case study 2: Application of SVM classifiers to a variety of text categorization tasks in biomedicine
  • Case study 3: Prediction of clinical laboratory values using SVM classifiers
  • Case study 4: Using SVMs for modeling clinical judgment and implicit guideline compliance in the diagnosis of melanomas
  • Case study 5: Using SVMs for feature selection
  • Case study 6: Comparison of SVM-based variable selection methods with Markov-blanket methods. Pitfalls of using SVM-based methods for causal feature selection
  • Case study 7: Outlier detection with SVM-based methods in ovarian cancer mass-spectrometry proteomics data


• Part IV
  • Online demonstrations
  • Software implementations
  • Additional literature
  • Conclusions.

Intended Audience: Biomedical informatics scientists and researchers, clinicians, bioinformaticians, biostatisticians, data analysts, programmers, students.

Content level: 25% Basic; 50% Intermediate; 25% Advanced

T07: Methods Series

Knowledge-based Decision-support Systems for Implementing Clinical Practice Guidelines

Samson Tu, Stanford University, Mor Peleg, University of Haifa

This tutorial gives an overview of the issues involved in, methods for, and examples of implementing clinical decision support systems for guideline-based care.

Clinical practice guidelines, or more generally clinical recommendations, are summaries of evidence-based best practices. In recent years, there has been an explosion of published guidelines. Computer-based decision-support tools can enhance the implementation of these guidelines by bringing focused recommendations to care providers at the point of decision-making. This tutorial will first give a general introduction to the history and practice of clinical decision-support systems, and then look at alternative methods for representing and delivering clinical recommendations.

We will use examples of guideline-based knowledge-based systems that have been implemented to illustrate both technical aspects of system development and organizational aspects of deployment and integration into clinical workflow. The technical aspects will include the steps and problems involved in formalizing guideline recommendations in computable format, system architecture, and technical challenges of integrating an external application into a continually-changing IT environment. The organizational aspect will address successful strategies for implementing the system into multiple medical centers of a large health care network. Presentation will be mixed with exercises and demonstrations of actual clinical systems that can provide decision support to primary care clinicians.

By the end of the tutorial, participants will be able to:

• Discuss characteristics of CDSS that help or hinder their acceptance
• Describe alternative methods for representing computable CPGs
• Understand the steps and issues involved in encoding guideline knowledge
• Outline the issues involved in deploying, integrating, and maintaining a knowledge-based DSS for implementing guidelines

Outline of Topics:

• Introduction to CDSS
• Alert and reminder systems
• Alternative computable models of CPG
• The roles of standards
• Knowledge acquisition and maintenance for guideline-based CDSS
• Sociotechnical issues of deploying CDSS

Intended Audience: researchers; physicians, nurses, and other healthcare professionals; computer scientists, system developers, and programmers.

Content level: 20% Basic, 60% Intermediate, 20% Advanced

T08: Selected Topics Series

Transforming & Visualizing Clinical Data for Research

Shawn Murphy, Massachusetts General Hospital

Using data collected in the clinical care domain for clinical research poses many challenges. Clinical data is diverse in structure and reliability. It is generated in truly massive quantities, but for consumption mostly by human eyes, not by machines. To be useful for clinical research, the data must usually be transformed to a machine readable format. Considerations of sensitivity and specificity must be considered when performing these transformations. One then needs a systematic approach to organizing the data such that queries can be generated against seemingly disparate data. This usually translates into finding a suitable "atomic fact" and organizing the data into these discrete pieces. Finally, the data must be shown to clinical researchers and allowed to be queried in a format that provides insights into and hypothesis testing of the data.

By the end of the tutorial, participants will be able to:

• Understand the complexities of using clinical data for research and be given methods and approaches for which to solve local problems.
• Understand the concept and value of structuring clinical data such that queries against disparate data can be performed.
• Understand the various solutions that currently exist for querying and visualizing clinical data.

Outline of Topics:

• Principles and methods for transforming clinical data for use in clinical research
• Organizing clinical data into databases for use in clinical research
• Visualizing and querying clinical data to test hypothesis
• Insights and solutions from querying and visualizing clinical data
• Limitations in using this data for clinical research

Intended Audience: Scientists, clinical researchers, people working in safety and quality research, biomedical engineers and workers in bioinformatics, and programmers.

Content level: 100% Intermediate

8:30 AM - 4:30 PM Affiliate Events

People and Organizational Issues WG Doctoral Consortium (by invitation)

The 4th Annual Doctoral Consortium on Sociotechnical Issues in Medical Informatics is a National Science Foundation (NSF) sponsored one day workshop that will bring together doctoral students from different disciplines to discuss their research on the design, development, and adoption of health information technologies. The goal of the Doctoral Consortium is to support the progress of doctoral students through constructive remarks and feedback on their research from prominent researchers in the field and interaction with other students. Through the Doctoral Consortium, students will be encouraged to examine health-related technologies from multiple perspectives and understand how technical design is affected by organizational/social considerations and vice-versa.

T09: Selected Topics Series

Clinical Classifications and Biomedical Ontologies: Terminology Evolution, Principles, and Practicalities

Christopher G. Ghute, Mayo Clinic and James J. Cimino, National Library of Medicine

Standardized terminologies and classification systems are an essential component of the information infrastructure that supports healthcare delivery and evaluation. Despite significant advances and increased motivation for the use of terminology systems, widespread integration of standardized terminologies into computer-based systems has not yet occurred. In this tutorial, we provide an overview of the state of the science related to terminologies and classification systems and demonstrate application of selected terminologies to a patient case study to highlight the strengths and weaknesses of various terminologies. Standardized terminologies alone are insufficient to achieve semantic interoperability. Consequently, the tutorial will include content designed to elucidate the relationships among standards for terminologies, information models, messages, and document and record structures. In addition, we will demonstrate the use of advanced terminology tools that facilitate the use of standardized terms in computer-based systems and provide an overview of significant international and national initiatives related to terminology systems.

By the end of the tutorial, participants will be able to:

• Understand the origins and evolution of terminologies and ontologies
• Appreciate the present state of the art in health terminology development and deployment
• Articulate the dependencies within Meaningful Use standards on ontologies and vocabularies
• Demonstrate practical access methods to many terminologies and ontologies

Outline of Topics:

• Historical appreciation of the evolution and development of terminology and ontologies
• An overview of current terminologies and ontologies used in clinical practice
• The relationship of terminologies and ontologies to Meaningful Use Standards
• Speculations about the future trends and current developments in health related terminologies and ontologies
• A laboratory practicum on accessing and using clinical terminologies
• Desiderata for practical and effective terminologies and ontologies
• Practical examples and demonstrations about current terminologies and ontologies

Intended Audience: Scientists; researchers; physicians, nurses, and other healthcare professionals; computer scientists, system developers, and programmers.

Content level: 75% Basic; 25% Intermediate

T10: Selected Topics Series

Meaningful Use of HIT in Nursing Practice

This tutorial presents a critical analysis of meaningful use and lays the groundwork for discussion of the possible impact of meaningful use on nursing informatics and on the nursing profession. (Additional registration required).

Speakers & Topics in order of presentation:

Moderator: Anita Ground

• Meaningful Use in Nursing Informatics: Judy Murphy
• Current system certification requirements: Anita Ground
• Meaningful use Education to New Nurses and Workforce: Judy Warren
• Discussion re: Meaningful use and Nursing Practice: Patti Dykes
• Structuring Nursing Data for Meaningful Use: Norma Lang, and Ellen Harper
• Meaningful Use International Perspective – Norway: Anne Moen
• Meaningful Use International Perspective – UK: Peter Murray

By the end of the tutorial, participants will be able to:

Intended Audience: Scientists; researchers; physicians, nurses, and other healthcare professionals; computer scientists, system developers, and programmers, etc.

Content level: 15% basic; 70% intermediate; 15% advanced

T11: Primer Series

Introduction to Biomedical and Health Informatics

Dominic Covvey, University of Waterloo; John Holmes, University of Pennsylvania; Christopher Cimino, Albert Einstein College of Medicine This tutorial gives the 30,000-foot view of healthcare informatics through a combination of presentations and audience discussions. Experts in the field will describe the general principles, jargon, and major problems in each of a half dozen healthcare informatics domains. The audience will be given a chance to struggle with some of these problems to gain a sense of the underlying intricacies.

The session will orient participants as to the content of the major healthcare informatics domains and how they interact. While participants should not expect to be able to start solving informatics problems based on this tutorial, they should have an understanding of what the problems are, which ones are attractive to them, and how they can acquire more knowledge and training to enter into the domain.

By the end of the tutorial, participants will be able to:

• Attend any of the AMIA scientific sessions and have a basic understanding of what is being presented and why it is important,
• Make an informed choice of a healthcare informatics domain which they would like to learn more about,
• Know about several options for training opportunities to acquire that learning.

Outline of Topics:

• Human and Social Aspects of Systems and Usability
• Evaluation
• Nature and Structure of Health Information
• Bioinformatics
• Information Retrieval
• Public Health Informatics
• The Electronic Health Record
• Workflow Analysis

Intended Audience: Anyone new to healthcare informatics who is looking for a broad overview of the field. Informatics workers who are familiar with one domain but are looking to become familiar with other domains.

Content level: 90% Basic; 9% Intermediate; 1% Advanced

T12: Primer Series

Clinical Research Informatics: Theory, Methods, and Best Practices

Philip Payne, The Ohio State University; Peter Embi, University of Cincinnati

This tutorial provides participants with a unique opportunity to increase their knowledge and understanding of core Clinical Research Informatics (CRI) theories and methods, as well as recent policy and funding developments, in the context of a rapidly growing and increasingly high-demand informatics practice domain.

Clinical research is critical to the advancement of medical science and public health. Conducting such research is a complex, resource intensive endeavor comprised of a multitude of actors, workflows, processes, and information resources. Recent national-level research and policy efforts have explicitly focused on increasing the clinical research capacity of the biomedical sector, largely through fostering improvements in both workflow and information management infrastructure. These efforts have served to increase attention on clinical research throughout the governmental, academic, and private sectors. In the specific context of the intersection between biomedical informatics and clinical research, the emergence of both a notable body of literature and a set of targeted funding mechanisms such as the National Cancer Institute’s (NCI) Cancer Biomedical Informatics Grid (caBIG) and National Center for Research Resources (NCRR) Clinical and Translational Science Award (CTSA) programs have served as significant catalysts for the emergence of a robust sub-discipline of informatics focusing on clinical research applications, known as Clinical Research Informatics (CRI).

In this tutorial, we provide researchers, technical leaders, and technical staff with an overview of the core definitions and informatics theory that collectively contribute to the successful practice of CRI. We use a set of research vignettes to illustrate common challenges and opportunities in the CRI space and best-practice approaches to such scenarios, including: 1) the design and implementation of integrative clinical research information management systems; 2) the query of disparate enterprise and research information systems to support clinical research activities and information dissemination/reporting; and 3) the identification and recruitment of clinical research participants via multiple retrospective and prospective modalities.

By the end of the tutorial, participants will be able to:

• Define clinical research informatics and understand its relationship with other biomedical informatics sub-disciplines as well as the clinical/translational sciences
• Synthesize the role of current healthcare and informatics policy and standards setting developments relative to the practice of CRI
• Identify and apply core informatics theories, methods, and best practices in order to analyze and plan solutions to common clinical research information management challenges
• Plan for and implement complex information management architectures in order to support a full spectrum of clinical research activities

Outline of Topics:

• Clinical Research Informatics (CRI) definition and relationship(s) to other biomedical informatics sub-disciplines and the clinical/translational sciences
• Current CRI-relevant funding, policy, and standards-setting efforts or initiatives
• Enterprise architectures for integrative clinical research information management
• Semantic interoperability and knowledge engineering in the clinical research domain
• The relationship of human factors and/or workflow optimization to the effective deployment and utilization of clinical research information management systems
• Common CRI challenges and their solutions:
  • Integrating enterprise systems and CRI platforms
  • Querying disparate data sources in support of clinical research related analyses and/or information dissemination
  • Clinical research participant identification and recruitment

Intended Audience: Scientists; researchers; physicians, nurses, and other healthcare professionals; computer scientists, system developers, and programmers.

Content Level: 20% Basic; 50% Intermediate; 30% Advanced

T13: Primer Series

Role of Social Networks in Teaching and Advancing Health Informatics: What Can Second Life Contribute to This Process?

Constance Johnson, Duke University; Chris Paton, University of Auckland; W. Ed Hammond, Duke University Medical Center, and Parvati Dev, Stanford University

Social Networking is creating quite a following these days. Is it just a passing fad, or will it provide new and innovative opportunities for health informatics? The Office of the National Coordinator and others have placed a new interest on the need to train informaticians and IT persons. AMIA has responded with the 10 x 10 program. What can social networking offer to make distance learning more accessible and better? With the costs and time of face –to face encounters, can social networking offer more reality in virtual meetings? This tutorial introduces social networking, from the simple to the complex, to the AMIA audience: what it is, what tools are available, how to get started, and how we can use it. The leaders will provide examples and opportunities from experiences in using this new domain.

Prior to the introduction of the Web 2.0, distance education was strictly unidirectional and flat. These evolving technologies are promoting collaborative work, social networking, and social interaction in distance education. For example, virtual environments (VE) such as Second Life® (SL) and VenueGen, promote immersion and presence and allow real-time teaching and discussion. Unlike traditional distance and online asynchronous education, multimodal spaces such as SL allow synchronous communication.

These multimodal spaces allow the development of simulated environments and the development of 3-D models. The promotion of social interaction between the users of Web 2.0 encourages active learning and enables direct feedback from mentors. There has been some evidence that this type of interaction between the mentor and mentee increases student motivation. With virtual environments, we now have the option to create social and professional relationships which have been the missing component of distance learning. Unlike traditional learning, there may be more potential for learning in virtual environments because of the level of interactions on the part of the user. Virtual environments now exist to give the participant the ability to feel as if they are experiencing real-life, in real-time, and in multi-participant simulations. Because of this realistic feeling, learning environments offer hypothetical scenarios to teach real clinical situations, thus students can learn and be tested on procedural knowledge as well as critical thinking skills. VE can be used to assess learning not only for the individual but for a group as well. This is particularly helpful when evaluating responses of multiple people such as in crisis scenarios for disaster preparedness. This type of learning offers immediate feedback that is necessary for effective learning.

The Duke School of Nursing has built a Second Life® infrastructure that allows faculty and students experiential learning, role-playing, and promotes social interaction, in a collaborative environment in our distance education program. It furthermore supports immersion and presence, allowing real-time teaching and discussion that enhances the mentor/mentee relationship in a simulated learning environment. In the School of Nursing, the use of virtual environments for distance education has been piloted for one year. This virtual classroom allowed faculty to test the feasibility of using this medium in distance-based education with graduate nursing informatics students.

The experiences and lessons learned will be shared with the tutorial class. The University of Auckland has created a virtual medical centre complete with wards, examination rooms, ambulances and medical equipment. Educators are able to take on the role of patients or health professionals and lead students through a variety of simulations. Resuscitation equipment has been constructed in the virtual environment that allows the educator to control pulse, blood pressure, oxygen saturations and cardiac rhythm. By viewing the readings on a large screen next to the patient, the student is able to complete simulations of multiple critical care scenarios. The UoA island also has a virtual lecture theatre which will be used for group discussions and student presentations for distance learning students on the UoA Health Informatics course. The tutorial will show participants how to create interactive objects in Second Life and examine the pros and cons of using interactive 'scripted’ equipment verses using role playing actors for controlling medical scenarios.

Innovations in Learning Inc. has introduced a number of innovative virtual worlds to teach clinical procedures, business processes, and coursework. Innovations in Learning is also using virtual worlds to capture activities in real worlds. This tutorial will provide great value to students in sharing real experiences, how to get started, and how to use these powerful tools for a new world.

Outline of topics:

• What is Social Networking
  • What is Web 2.0 all about?
  • What are some of the tools and applications available in this domain?
  • The role of social networking in teaching
  • The role of social networking in simulation
• What is Second Life
  • Become acquainted to what Second Life is all about and what it offers
  • Role playing in a 3-D world
  • Interactive learning
  • Avatars, building, communication, scripting, sharing, community
  • Capability and flexibility
  • The Second Life Community
• Teaching in a virtual world
  • Tools
    • Content authoring tools such as Moodle or Sloodle, Blackboard
    • Webinars such as Eluminate, Adobe Connect
    • Screen recorders such as Camtasia
    • More social networking tools such as Facebook and Twitter
    • Second Life teaching tools such as online tests, whiteboards, blackboards, interactive workstations and equipment
    • Videos such as Youtube
    • Others
  • Practical experience in using Second Life and other tools in Distance Learning in the Duke School of Nursing
  • Limitations of using virtual worlds compared to face to face teaching and 'gold-standard’ e-learning tools.
• Simulation
  • Role and opportunity for Second Life in simulation and building 3-D models
  • Comparison of using scripted objects versus using role playing with real people controlling characters
  • Practical experience in the Duke School of Medicine and Duke School of Nursing
  • Practical Experience in the National Institute for Health Innovation and the University of Auckland<

Educational Goals:

• Understanding social networking and its scope of potential use
• Understanding of Second Life as a major component for social networking
• Developing innovative concepts for the future built around social networking

Intended audience: Individuals interested in learning how to use social engineering in distance learning simulation, and new opportunities for personalized communication.

Content level: 60% Basic; 40% Intermediate

T14: E.H.R. Series

Standards for Storing and Exchanging Clinical Data in Electronic Health Record Systems

Walter Sujansky, Sujansky & Associates, LLC

This tutorial provides a thorough, practical, and understandable overview of the role of standards in EHRs and in health information exchange.The absence of widely implemented clinical data standards has been cited as a major barrier to the adoption of EHRs in the United States. This tutorial discusses the practical importance of data standards in ambulatory EHRs and describes some of the most important standards for storing and exchanging clinical data in real-world systems. Specific standards that will be covered include HL7, NCPDP Script, CCR, LOINC, SNOMED, and DOQ-IT. The tutorial will also address certain issues that have hindered the adoption and effective use of clinical data standards in the U.S., and will describe ongoing efforts to address these issues within the government and private sector. A recently developed HL7 standard for reporting laboratory results to ambulatory EHRs (ELINCS) will be presented as a case study of the process for developing, disseminating, and implementing a clinical data standard to meet a practical need.

By the end of the tutorial, participants will be able to:

• Understand how clinical data standards influence the functionality and usability of EHRs
• Evaluate the appropriate roles and the strengths/limitations of several prominent data standards
• Appreciate the practical difficulties of applying existing data standards effectively within EHRs
• Understand a number of current initiatives striving to increase the value of standards for EHRs, including the ELINCS lab-reporting project

Outline of Topics:

• Purpose/role of data standards in ambulatory EHRs
• No EHR is an island
• Semantic interoperability
• Important data standards for EHRs
• Messaging standards (HL7, NCPDP SCRIPT)
• Document standards (CCR, CDA)
• Terminology standards (LOINC, SNOMED-CT, NDC, RxNorm, other drug terminologies)
• Issues/barriers in the practical adoption and use of data standards
• Current initiatives to address/overcome barriers
• Work of private entities (HL7, IHE, Connecting For Health)
• Work of federal government (HITSP, CCHIT, NLM)
• Case Study: ELINCS (EHR-Laboratory Interoperability and Connectivity Standards)

Intended Audience: Developers of ambulatory EHR systems, including systems for decision support, reporting, analysis, and interfacing; Purchasers and users of ambulatory EHR systems; Developers of clinical data standards; Students interested in the practical uses and technical features of existing clinical data standards.

Content level: 40% Basic, 40% Intermediate, 20% Advanced

T15 - E.H.R. Series [DETAILED DESCRIPTION TO COME]

Privacy and Security in the New HITECH World

S. Sengupta, Columbia University

Continuous changes in regulations, new information technologies, and new types of threats in Privacy and Security of electronic data must be countered with tempered approaches and calm efforts that are balanced, cost-effective, and constant for good practice of care. We discuss the methods and principles of techniques helping to keep patient data private and secure.

T16: Methods Series

Designing, Conducting, and Analyzing Data from the Qualitative Research Interview for Health Informatics

Faculty: Martha Gaie, Strategic Consumer Health Informatics

This tutorial introduces a data-gathering technique beneficial when used to discover new processes, concepts, ideas, and/or explanations; to better understand how known variables or concepts may interact with each other; and to gain insight into the thought processes that result in measurable actions.

The continuing issue of high failure rates in health information technology (HIT) projects due to human factors suggests that there are fundamental weaknesses in the methods traditionally used to assess user-oriented variables. Part of the problem is a failure to understand the context in which HIT systems are to be used. Contextual factors may include existing work processes, what currently works and what doesn’t, what barriers and facilitators may be present, and psychological factors present in users that may affect how these systems will be received.

Perhaps the most direct and effective way to understand these influences is to conduct one-on-one interviews with prospective users, using a structured but flexible approach that maximizes exploratory data collection and promotes understanding of the user in the work context. Qualitative research interviews can provide rich descriptions of user perspectives, including attitudes, beliefs, behaviors, motivations, and processes, which help us better understand what constitutes "meaningful use" to various user groups and contexts. These interviews can also provide insights that we might never have considered, and thus never thought to ask about in a survey or evaluation, which may mean the difference between project success and failure. By exploring the user’s subjective "mental world," we can gain understanding of their daily experience, as they see it, including how they categorize data, as well as their logic, habits, and values. We can then use these insights to improve HIT design and implementation.

In this tutorial, I teach attendees how to design, conduct, and analyze data from the qualitative research interview specifically intended to understand the HIT user in context. After the half-day tutorial, attendees should have a basic understanding of how, when, and why to use the qualitative research interview to improve HIT project design and implementation.

By the end of the tutorial, participants will be able to:

• Understand the significance and principles of the qualitative research interview and the user in context
• Understand how to design a practical and systematic interview tool for strategic data collection
• Understand how to conduct an effective qualitative research interview to maximize data collection
• Understand how to analyze basic qualitative data for themes that can be used to direct HIT design and/or implementation

Outline of Topics:

• The HIT user in context
• The value, uses, and limitations of the qualitative research interview
• Defining the scope of your qualitative research interview
• Designing and constructing the qualitative research interview tool
• Conducting the qualitative research interview
• Techniques used by interviewers to draw useful responses from interviewees and maximize data collection
• The basics of how to code and categorize qualitative data
• How to apply qualitative data to HIT project design and implementation

Intended Audience: Scientists; researchers; physicians, nurses, and other healthcare professionals; computer scientists, system developers, and programmers; graduate students and postdoctoral fellows.

Content level: 50% Basic, 40% Intermediate, 10% Advanced

T17: Methods Series

NLM Terminology Resources in Practice

Olivier Bodenreider, James Case, Kin Wah Fung, John Kilbourne, Suresh Srinivasan, Jan Willis, National Library of Medicine

The NLM produces and distributes a growing set of terminology resources designed to enable more effective and interoperable biomedical information systems including electronic health records (EHR). Recent focus is given to products that support achievement of meaningful use of EHRs as defined by HHS. Central to these resources is the Unified Medical Language System® (UMLS®), which integrates and distributes key terminology, classification, and coding standards to support interoperability of information exchange across clinical and research platforms.

The course will provide an overview of the three UMLS Knowledge Sources (Metathesaurus, Semantic Network, SPECIALIST Lexicon and Lexical tools) and how they are used. Use of the Metathesaurus and its clinical terminologies designated by HHS as standards (LOINC®, RxNorm, and SNOMED CT®) will be explored. More recent resources to be discussed will include: (1) the CORE (Clinical Observations Recording and Encoding) Problem List Subset of SNOMED CT®, which facilitates use of SNOMED CT for coding problem list data in EHRs, and (2) the Newborn Screening Coding and Terminology Guide, used to record and transmit newborn screening test results. The course will conclude with a tour of the new UMLS Terminology Services (UTS) Web site, allowing access to NLM terminology resources including downloads, browsers, and documentation. New features such as the SNOMED CT browser and the UMLS user profile will be highlighted.

By the end of the tutorial, participants will be able to:

• Explain the content, structure and practical use of the UMLS Knowledge Sources;
• Access NLM terminology resources including downloads, browsers, and documentation on the UMLS Terminology Services Web site;
• Employ MetamorphoSys software to create useful subsets of the Metathesaurus that are fit for purpose in clinical and research applications;
• Utilize NLM terminology resources to enhance local systems in support of achieving meaningful use as defined by HHS; includes use of the CORE Problem List Subset of SNOMED CT and Newborn Screening Coding and Terminology Guide.

Outline of Topics:

• Overview of the Unified Medical Language System (UMLS)
• Using MetamorphoSys to install and customize the Metathesaurus
• Highlights and use of key clinical terminologies and resources:
  • RxNorm
  • SNOMED CT®
  • CORE Problem List Subset of SNOMED CT
  • Newborn Screening Coding and Terminology Guide
• UMLS Terminology Services Web site

Intended Audience: Professionals involved in creating and using interoperable healthcare systems including EHRs; researchers, physicians, nurses and other healthcare professionals; system developers.

Content level: 50% Basic; 50% Intermediate

T18: Methods Series

Ontology-oriented resources from the National Center for Biomedical Ontology

Nigam H. Shah and Mark A. Musen, Stanford University

The National Center for Biomedical Ontology (NCBO) offers a range of Web services that allow users to access biomedical terminologies and ontologies, to use ontology terms to create pick lists and lexicons, to identify terms from controlled terminologies and ontologies that can describe and index the contents of online data sets (data annotation), and to recommend particular terminologies and ontologies that would be appropriate for data-annotation tasks. An ontology repository, known as BioPortal, provides a Web-based interface that allows users to visualize ontologies, to map the terms in ontologies to one another, and to provide public comments on ontologies that can guide ontology developers and that can offer assistance to ontology users. This tutorial will provide hands-on experience in using the NCBO's resources, and will offer participants in-depth understanding of how ontologies and terminologies are used to solve problems in biomedical informatics. The tutorial will demonstrate the use of NCBO resources to facilitate tasks such as semantic data integration, information retrieval, structured data entry, and knowledge management.

By the end of the tutorial, participants will be able to:

• Understand the biomedical ontology landscape
• Understand the national infrastructure available for data annotation and knowledge management
• Conceive workflows that utilize NCBO Web Services to solve their own data entry and integration problems.

Outline of Topics:

• Overview of NCBO key activities
• Ontologies available in Biomedicine
• Web-based tools for Ontology search, visualization and review
• Tools and Web Services for data annotation and semantic integration
• Design of custom workflows to utilize national ontology-resources

Intended Audience: Scientists and researchers seeking to understand how to optimally use ontologies for problem solving. Health IT System developers and CIOs seeking to understand how to leverage NIH-funded infrastructure for using Ontologies.

Content level: 20% Basic; 60% Intermediate; 20% Advanced

T19: Selected Topics Series

Personal Health Records, Patient Portals & Consumer-Facing Health IT

Patricia Flatley Brennan, University of Wisconsin-Madison; Jonathan Wald, Partners HealthCare System; Stephen Ross, University of Colorado Denver As the HIT landscape continues to differentiate, consumer facing health information technology solutions are assuming increasing importance in engaging people in self care and disease management. Personal health information tools provide lay people with access to subsets of their clinical records and with the health information management tools needed for self-care and effective health care utilization. Taking on many forms, including PHRs, iPhone apps, patient portals, stand-alone applications, and Web 2.0 services, these innovative IT tools may also enable better access to the health care systems resources, including health information, appointment scheduling and provider communication, and personal health tracking. Through case studies this tutorial will introduce clinicians, systems administrators, and IT developers to critical issues regarding the design and deployment of PHRs and other personal health information management tools. During the tutorial, participants will have an opportunity to examine and critically evaluate existing tools & applications, explore patient portals, and discuss technical, ethical and policy considerations related to the deployment of personal health records tools. An update of the national environment and trends enabling (or interfering with) deploying IT tools for direct-to-consumer will be provided: meaningful use, privacy policies, payment schemes and health reform. Participants are encouraged to appraise their institutions' current plans for deploying consumer facing HIT and to come prepared to engage in discussions regarding implementation challenges and anticipated benefits.

By the end of the tutorial, participants will be able to:

• Determine how consumer-facing health information technologies, including PHRs and patient portals could achieve practice and agency goals
• Pose solutions to the clinical and usability challenges of effective consumer-facing health information technologies
• Evaluate the technical requirements, ethical considerations and social value of personal health records

Outline of Topics:

• Personal health information management tools
• Personal Health Records, patient portals, iPhone apps for health
  • Web 2.0, social media & health 2.0 – what does it offer personal health information management
  • Microsoft HealthVault and Google Health - Where do they fit in?
  • Clinical Uses
    • Self-management—observations in daily living
    • Care coordination
    • Life-long records
• Design
  • Technical Considerations and Challenges
  • Platforms and Devices
  • Integration with clinical information systems
  • Human-computer interaction
  • Clinical considerations
  • Fostering health goals with PHRs and patient portals
  • Personal health monitoring
• Health Education
  • Communication with professionals
• Policy and the View from the National Scene
  • View from the National Scene
  • Meaningful Use
  • Privacy
  • Policy: Payment schemes and health care reform
  • Ethical Issues
  • Social benefits & challenges of consumer-facing health IT
• Public Health
  • Surveillance
  • Public Health Education
• Health Services Research

Intended Audience: Clinicians, managers of patient portals, health educators, public health practitioners, engineers and computer scientists who work with distributed information systems, integration of disparate data bases or network-level authorization, authentication, or privacy policies.

Content level: 50% Basic, 30% Intermediate, 20% Advanced

T20: Selected Topics Series

Practical Clinical Knowledge Management

Roberto A. Rocha, Tonya M. Hongsermeier, Saverio M. Maviglia, Partners Healthcare

Clinical decision support includes a variety of activities, tactics, and technologies aimed at systematically providing to clinicians the necessary information and knowledge for making the right decisions for each patient. Current healthcare decision support programs frequently fail to consider the equally important aspects of knowledge asset management. Similar to what has been observed in other industries, healthcare must create processes to identify the "best" knowledge, ensure knowledge currency, align the application of knowledge to organizational goals, and implement the full range of IT-based and non-IT-based approaches.

This tutorial describes strategies and processes to create and maintain the clinical knowledge assets necessary for effective and sustainable decision support interventions. Attendees will learn about the different modalities of clinical knowledge assets and their application to support evidence-based medical practice and research. Knowledge asset authoring, validation, and maintenance tools and processes will be presented through practical and illustrative examples from real clinical settings. The instructors will explain practical approaches to collaborative knowledge asset management, leveraging their experience within large and integrated healthcare delivery organizations.

By the end of the tutorial, participants will be able to:

• Describe the importance of well defined knowledge management practices within healthcare organizations;
• Describe the complete lifecycle of clinical knowledge assets, emphasizing collaborative and systematic development and maintenance;
• Describe and categorize the knowledge assets required by the different types of computerized clinical decisions support interventions;
• Explain the most common clinical knowledge management processes and tools within healthcare organizations.

Outline of Topics:

• Clinical Knowledge Management Program
  • Strategic, Governance, and Business Issues
  • Standards and Regulatory Issues
  • Implementation and Staffing Considerations
• Clinical Knowledge Content Lifecycle
  • Collaborative Development and Long-term Maintenance
  • Implications for Disease, Trend, and Risk Management
• Clinical Knowledge Management Infrastructure:
  • Managing and Delivering Reference and Executable Knowledge
  • Information and Knowledge Representation Models

Intended Audience: Physicians, nurses, and other healthcare professionals involved with the configuration and maintenance of clinical knowledge assets and computerized clinical decision support systems; Knowledge engineers, informaticians, and business analysts involved with the development and deployment of clinical knowledge assets and computerized clinical decision support systems; Computer scientists, system developers, and programmers involved with the development of clinical information systems and electronic health records; Decision makers seeking to understand the rationale for implementing clinical knowledge management programs.

Content level: 50% Basic, 30% Intermediate, and 20% Advanced.

T21: Primer Series

Theme: Data Mining, NLP, Information Extraction

An Introduction to Data Mining Principles and Practice

Faculty: John H. Holmes, University of Pennsylvania

This interactive tutorial introduces attendees to the theory, tools, and techniques for discovering knowledge in biomedical data. Using a well-known data mining life cycle as a conceptual framework, attendees will experience first-hand, thorough demonstration and direct participation, the techniques of mining clinical data. These techniques include data preparation, description and visualization, feature selection, mining association, classification, prediction rules, and clustering. A variety of mining algorithms will be explored with each technique. The capstone of the tutorial will be the application of mined data to informing traditional statistical analysis. The tutorial will include hands-on experience in using Weka, a well-known open-source data mining software suite. Although not required, attendees will get the most out of the tutorial if they bring a laptop to the session to participate in the hands-on sessions. Instructions for downloading and installing Weka will be sent to registrants approximately one week prior to the tutorial.

By the end of the tutorial, participants will be able to:

• Understand the basic principles of data mining
• Apply appropriate data mining techniques in clinical research
• Interpret data mining results and how they inform statistical analysis

Outline of Topics:

• Biomedical databases
• Data visualization
• Intelligent data analysis
• Explanatory data mining
• Predictive data mining
• Data mining software
• Applications of data mining in biomedical domains
• Data preparation: methods for cleaning, reduction, and coding
• Association rule discovery
• Classification and prediction
• Clustering and visualization
• When to data mine
• Evaluating data mining software
• Ethical concerns in data mining

Intended Audience: Scientists; researchers; physicians, nurses, and other healthcare professionals; computer scientists, system developers, and programmers.

Content level: 70% Basic, 30% Intermediate

T22: Primer Series

What Is Public Policy and How to Help Shape It

Julie McGowan, Regenstrief Institute and Indiana University, Meryl Bloomrosen, American Medical Informatics Association, Doug Peddicord, Washington Health Strategies Group/Oldaker, Belair & Wittie

The field of medical informatics has never before been faced with such opportunities and issues. These arise from a growing national awareness of the potential benefits of health IT to improve the quality of care while reducing adverse events and medical errors. The Federal government, through legislation and rule making is shaping the direction in which our field will be moving for decades. But are they getting the right advice?

This tutorial provides information and hands-on activities focusing on health policy development and implementation. It will give an overview of federal and state regulatory programs affecting the health care industry in general and biomedical and health informatics in particular. Participants will gain an understanding of the purpose of policy advocacy and AMIA's role in educating and influencing policy makers.

By the end of the tutorial, participants will be able to:

• Appreciate the complexities of federal policy making as it affects biomedical and health informatics
• Impact decision-making about biomedical and health informatics policy
• Utilize congressional visits to push both AMIA's and personal policy agendas
• Effectively respond to requests for comments about rules impacting biomedical and health informatics

Outline of Topics:

• The Legislative Process: How Congress Works
• The Executive Branch and Rule Making
• The Federal Budgetary Process
• Information Sources for Policymaking
• Anatomy of a Bill, Or How I Get my Ideas into Law
• Making the Most of Your Congressional Visit
• Preparing a Response to Request for Comments

Intended Audience: Scientists; researchers; physicians, nurses, and other healthcare professionals; computer scientists, system developers, and programmers, etc. - anyone who is interested in how to effectively influence Federal Policy as it relates to Biomedical and Health Informatics.

Content level: 80% Basic; 20% Intermediate.

T23: Primer Series

Introduction to Translational Bioinformatics

Atul Butte, Stanford University

In 2005, Dr. Elias Zerhouni, Director of the National Institutes of Health (NIH), wrote "It is the responsibility of those of us involved in today’s biomedical research enterprise to translate the remarkable scientific innovations we are witnessing into health gains for the nation... At no other time has the need for a robust, bidirectional information flow between basic and translational scientists been so necessary." Clearly evident in Dr. Zerhouni’s quote is the role biomedical informatics needs to play in facilitating translational medicine. American Medical Informatics Association (AMIA) now hosts the Joint Summits on Translational Science of which the Summit on Translational Bioinformatics is one of the two components. This tutorial is designed around the successful curriculum used in Stanford's course in Translational Bioinformatics, one of the first courses to be offered in this field. This tutorial is designed to teach the basics of the various types of molecular data and methodologies currently used in bioinformatics and genomics research, and how these can interface with clinical data. This tutorial will address the hypotheses one can start with by integrating molecular biological data with clinical data, and will show how to implement systems to address these hypotheses. The tutorial will cover real-world case-studies of how genetic, genomics, and proteomic data has been integrated with clinical data.

By the end of the tutorial, participants will be able to:

• Understand why biologists and clinicians use each measurement technology, and the advantages of each.
• Be able to explain which genomic and genetic methods are most appropriate for studying diseases.
• Be able to list high-level requirements for an infrastructure relating research and clinical genetic and genomic data.

Outline of Topics:

• Basic understanding of various genome-scale measurement modalities: sequencing, polymorphisms, haplotypes, proteomics, gene expression, metabolomics, and others
• Crucial difference between genetic and genomic data
• Nature and format of expression, polymorphism, proteomics, and sequencing data
• Overview of the most commonly used structured vocabularies, taxonomies, and ontologies used in genomics research
• Description of the most frequently used analysis and clustering techniques
• How the genetic predisposition to disease is studied
• Use of genetic information across medical specialties
• How to find clinical genetic tests
• Genomic and clinical data to study patient disease-free status and survival
• How informatics can be used to identify potential drug targets
• Types of biomarkers
• Parallels between research methods in medical informatics and bioinformatics
• Relating clinical measurements with molecular measurements

Intended Audience: Academic faculty or professionals setting up bioinformatics facilities and/or relating these to clinical data repositories, or to data from General Clinical Research Centers or Clinical and Translational Science Awards; health information professionals responsible for clinical databases or data warehouses and tying these to researchers; informaticians, clinicians, and scientists interested in genetics, functional genomics, and microarray analysis; physicians interested in how medicine is advancing through the use of genomics and genetics; and students.

Content level: 20% Basic, 50% Intermediate, 30% Advanced

T24: Primer Series

Embracing Healthcare IT Standards in the World of Meaningful Use

Charles Jaffe, HL7, Rebecca Kush, CDISC, Dixie Baker, SAIC, Blackford Middleton, Harvard-Partners; Chris Chute, Mayo Clinic, Stanley Huff, Intermountan Healthcare, and Robert Dolin, Semantically Yours, LLC

Healthcare IT Standards have often been viewed with only passing interest in the medical informatics community. The HITECH provisions of the ARRA legislation have brought their importance to the fore. These standards are essential if we are to achieve any of the regulatory provisions that require reuse of healthcare data. At the top of anyone's list, the essential elements would include decision support, vocabulary binding, privacy and security, quality measurement, and clinical research integration. In the end, successful implementation of any solution is predicated on collaboration across standards developers, realizing both quality improvement and cost effectiveness.

By the end of the tutorial, participants will be able to:

• Identify the key standards specified in the meaningful use final rule.
• Define the standards requirements for security and privacy
• Understand the specific vocabulary standards and their binding to transmission specifications
• Establish metrics underlying the business case of standards requirements
• Recognize the value of standards development for the integration of patient care and clinical research data
• Leverage the underlying HL7 standards for messaging and the key components of Clinical Document Architecture
• Appreciate the value basis of international standards development
• Integrate the key components of the National Health Information Network
• Develop a comprehensive view of the standards required for (semantic) interoperability

Outline of Topics:

• Standards for healthcare as keystone for interoperability
• Standards for security and privacy
• Vocabulary standards
• Return on investment for standards adoption and deployment
• Integration of standards for clinical research and patient care
• HL7 messaging and Clinical Document Architecture (CDA)
• NHIN Architecture and the evolution of the NHIN Direct
• ISO and the international community of standards development

Intended Audience: Scientists; researchers; physicians, nurses, and other healthcare professionals; computer scientists, system developers, programmers, and CFOs

Content level: 30% Basic; 40% Intermediate; 30% Advanced

T25: E.H.R. Series

Clinical Decision Support: A Practical Guide to Developing Your Program to Improve Outcomes

Robert A. Jenders, University of California, Los Angeles; Jerome A. Osheroff, Thomson Reuters and University of Pennsylvania Health System; Jonathan M. Teich, Elsevier Health Sciences and Harvard University; Dean F. Sittig, University of Texas Health Science Center at Houston; and Robert E. Murphy, Memorial Hermann Healthcare System and University of Texas Health Science Center at Houston

This tutorial presents a practical, systematic approach to the implementation of clinical decision support (CDS) in an interactive, case-oriented fashion that brings together the latest developments in CDS technology and national programs.

This tutorial provides attendees with a practical approach to developing and deploying clinical decision support (CDS) interventions that measurably improve outcomes of interest to a health care delivery organization. The following key steps, including overcoming barriers, will be examined in detail: initiating an overall CDS program, including selecting appropriate CDS goals and enhancing organizational structures needed for CDS success in the context of current regulatory and safety initiatives; selectively implementing CDS technology to achieve a specific goal, with a focus on stakeholder and process analysis; and following up and monitoring CDS interventions. The role of national programs relevant to CDS, including knowledge sharing, structured guidelines and meaningful use, also will be explored. The systematic approach to CDS implementation will be presented in an interactive, case-oriented fashion, incorporating examples provided by tutorial leaders and participant's experiences. The course content is drawn from the tutorial leaders' popular and award-winning guidebook series on improving outcomes with clinical decision support.

By the end of the tutorial, participants will be able to:

• Follow a systematic process for developing, implementing and analyzing the effect of a clinical decision support program.
• Understand the types of CDS technology available for realizing desired outcomes.
• Detail factors both external and internal to a health care organization that drive CDS initiatives.

Outline of Topics:

• Selection of CDS goals.
• National programs relevant to CDS, including knowledge sharing, clinical guidelines and meaningful use.
• Developing organizational structures for implementing a CDS program.
• Selective implementation of goal-directed CDS interventions.
• Monitoring CDS interventions.

Intended Audience: Clinicians and administrators interested in quality improvement and patient safety; physicians, nurses and other health care professionals; and computer scientists, system developers and programmers interested in understanding applications of health information technology to clinical decision support.

Content level: 60% Basic; 40% Intermediate

T26: E.H.R. Series

Paradigm Shift – A Fresh Approach to the EHR

William W. Stead, Vanderbilt University and W. Edward Hammond, Duke University

Interoperable health information is essential to engineering a high quality cost effective system of healthcare delivery. However, a recent National Research Council (NRC) committee found "current efforts aimed at nationwide deployment of healthcare information technology will not be sufficient to achieve the vision of 21st century healthcare, and may even set back the cause" (Stead & Lin, 2009).

This tutorial introduces a fresh approach to achieving interoperable health information. The key shift is from thinking of the electronic health record as a by-product of automating practice, to thinking of it as a visualization of signals accumulated about patients and populations across scales of biology, time and geography. Participants will learn how to apply this paradigm shift to decrease the time and cost to deploy today’s EHR technologies and to increase the benefit of using the technology.

By the end of the tutorial, the participants will be able to:

• List expectations for EHRs as the foundation of an engineered system of healthcare that are not met by today’s deployed systems.
• Describe reasons for this gap
• Appreciate how a shift in the paradigm might close the gap
• Make better decisions about how to deploy today’s EHR technologies and when to look for alternative approaches.

Outline of topics:

• The problem
  • Requirements for an engineered system of health care’s information foundation
  • Reality of today’s deployed EHR systems
  • Root causes of the gap between what we have and what we need
• The paradigm shift
  • Working with information at multiple scales to manage complexity
  • Matching computational technique, effort and governance to the scale of the task
  • Examples from different perspectives: enterprise, regional, personal
• From idea to action
  • Shift from certification to measurement and reporting
  • Steps for health care enterprises & vendors
  • Research challenges

Intended Audience: Individuals responsible for quality/cost improvement in a health system or responsible for implementing or developing information technology to enable those improvements.

Content level: 10% Basic, 80% Intermediate, 10% Advanced

T27: E.H.R. Series

Approaches to Clinical Computer-based Documentation

S. Trent Rosenbloom, Kevin Johnson, Vanderbilt University

This tutorial outlines different methods for documenting clinical care, will review existing literature covering the history and evaluations of CBD systems, will discuss factors contributing to documentation content, will address how templates can impact documentation, and will conclude with summary recommendations for developers and adopters. Health care professionals have increasingly adopted electronic health record (EHR) systems, especially in larger practices or healthcare settings. Many currently available EHR systems include modules to support direct clinical computer-based documentation (CBD), and traditional documentation methods can be made to support EHR system adoption and use. The tutorial will outline different methods for documenting clinical care, will review existing literature covering the history and evaluations of CBD systems, will discuss factors contributing to documentation content, will address how templates can impact documentation, and will conclude with summary recommendations for developers and adopters.

Different documentation methods have varying strengths and weaknesses related to quality, usability, efficiency and data availability. It is likely that each note-capture mechanism will find a clinical niche, with different clinicians and different sites each using the type that best fits the practice situation. The proposed tutorial will outline different methods for documenting clinical care, will review existing literature covering the history and evaluations of CBD systems, will discuss factors contributing to documentation content, will address how templates can impact documentation, and will conclude with summary recommendations for developers and adopters.

By the end of the tutorial, participants will be able to:

• Understand the different methods for clinical documentation and their rationale
• Understand the history and near-term future of clinical documentation tools
• Characterize functional considerations that are important when selecting a clinical documentation tool

Outline of Topics:

• Types of Clinical Documentation
• External Drivers of Clinical Documentation Content
• History of Computer Based Documentation Systems
• Different Approaches for Templating Notes
• Discussion of related topics
• An Approach for Selecting Documentation Systems

Intended Audience: Computer-based documentation system developers and evaluators, healthcare providers, electronic health record system users, evaluators, and purchasers.

Content level: 30% Basic, 50% Intermediate, 20% Advanced

T28: Selected Topics Series

User Centered Design for Public Health & Consumer Health Information Systems

Barbara L. Massoudi, RTI International and Rupa S. Valdez, University of Wisconsin-Madison

User-centered design (UCD) is a system development approach that puts the user, rather than the technologist at the center of the design effort. Traditional systems development is often technologist driven, prioritizing the expertise and assumptions of the designer. Consequently, it often fails to adequately consider users’ needs and preferences. User acceptance of these systems is often challenging and redesign and development efforts can be prohibitively costly and time consuming.

Although UCD has demonstrated success in involving users and better meeting their needs, it is infrequently used in public health/consumer health information systems (PH/CHIS) development efforts. Using both didactic and hands-on small group case studies, we will introduce the concept of UCD; present the benefits and implications of UCD in PH/CHIS development efforts; and, introduce specific UCD methods/techniques. Additionally, we will discuss how UCD fits within the context of iterative and prototyping system development life cycles, business process reengineering, and usability testing, as these methods are often used in PH/CHIS development efforts. The techniques/methods presented will be relevant to multiple levels of design including the user interface, functionality, content, and integration with other elements of the larger system with which the PH/CHIS is to be integrated.

By the end of the tutorial, participants will be able to:

• Explain the rationale for incorporating user centered design in public health and consumer health information systems development projects.
• Formulate a user centered design work plan for an upcoming systems development project.
• Evaluate the appropriateness of specific user centered design methods and techniques to a design challenge.

Outline of Topics:

• Introductions, experience and expectation statements
• Rationale for user centered design
• User centered design methods and techniques
• Case study presentation
• Developing user centered design work plans: Small group activity
• Group presentations of user centered design work plan
• User centered design caveats

Intended Audience: Informatics professionals, project sponsors, project managers, and business analysts, working within public or consumer health settings.

Content level: 50% Basic; 50% Intermediate

T29: Methods Series

UMLS Concept Identification Using the MetaMap System

Alan R. Aronson, Dina Demner-Fushman, Francois-Michel Lang, James G. Mork, National Library of Medicine

Analyzing free text in order to identify concepts drawn from a controlled vocabulary is an important problem and ongoing research issue in medical informatics, and draws on both natural-language processing (NLP) and information retrieval (IR). Indeed many such concept-identifications systems have been built. One such well respected system is MetaMap, developed at the U.S. National Library of Medicine’s Lister Hill National Center for Biomedical Communication at the National Institutes of Health. MetaMap is a sophisticated application used by many researchers worldwide to map free biomedical text to concepts contained in the Unified Medical Language System (UMLSR) Metathesaurus.

This half-day tutorial is designed to introduce clinicians, researchers, and informaticians to the MetaMap system, to present numerous examples of how to use MetaMap, and to discuss several real life research projects that use MetaMap. Topics covered will include the importance and difficulty of concept-identification; the basic processing provided by MetaMap; an overview of MetaMap’s many processing options; how best to use (and not use) MetaMap; MetaMap’s limitations and future directions; and finally how to obtain MetaMap, run it, and customize the system for your own needs. Some basic knowledge of natural-language processing (NLP) and information retrieval (IR) will be helpful, but is not required. Knowledge of the UMLS Metathesaurus will also be helpful.

By the end of this tutorial, attendees will have a working understanding of:

• The difficulty, importance, and benefits of automated concept identification
• How MetaMap identifies UMLS Metathesaurus concepts referred to in biomedical text
• The various modules and components of MetaMap
• The basics of operating MetaMap, and how to modify its behavior
• How MetaMap can help analyze data

Outline of Topics:

• Introduction: Why concept identification is important, and why it’s difficult
• High-level components of MetaMap: tokenization, part-of-speech tagging, lexical lookup, acronym/abbreviation identification, syntactic analysis, variant generation, candidate identification, mapping construction, and word-sense disambiguation
• Expected input formats, available output formats
• Overview of MetaMap’s wide array of data, output, and processing options
• Processing modes: narrow focus vs. casting a wide net
• Misuse and abuse of MetaMap
• Research projects using MetaMap
• Limitations of current MetaMap and future directions
• Ways of obtaining and running MetaMap: Web interface, code and binary downloads, MetaMap Java API, SKR API, UIMA wrapper
• Customizing MetaMap to your own needs: Datafile Builder

Intended audience: Clinicians, researchers, and informaticians interested in learning about concept identification in general, and MetaMap in particular.

Content level: 50% Basic, 50% Intermediate

T30: Selected Topics Series

Imaging Informatics: Foundations and Clinical Applications

Daniel Rubin and David Channin, Stanford University

Medical imaging is a vital component of patient information. The rapidly growing image-related data in clinical records and research studies provides enormous opportunities for discovery and personalization of patient care, analogous to similar advances from biomedical informatics. The number of imaging modalities is exploding, providing richer characterizations of the anatomic, functional, cellular and molecular aspects of disease. Image-based characterization of "imaging phenotypes" is emerging with tremendous potential to diagnose disease, tailor optimum treatment, track disease response, and predict outcomes. Making these advances will require large-scale data-driven analyses of rapidly expanding image databases in hospitals and research institutions. The complexity of the information latent in images, however, thwarts direct application of current biomedical informatics methods to these essential data. Imaging informatics methods are being developed to access and extract quantitative information and semantic biomedical features from images, to integrate imaging and clinical/molecular data, and to enable data mining and discovery of image biomarkers. The translation of these methods into practice will improve the diagnostic accuracy of imaging and the clinical effectiveness of radiologists. A new era of "quantitative imaging" is emerging to improve accuracy and reproducibility of image interpretation, to reduce variation in practice, and to improve quality.

This tutorial provides an overview of the foundational methods and emerging clinical applications in imaging informatics. It will first review the major imaging modalities in radiology and the role of imaging data in the context of informing and guiding medical care. It will describe information systems involved in handling imaging data, and the major imaging-related standards (DICOM, RadLex, HL7, AIM, and IHE). The tutorial will describe the key methodologies of imaging informatics while pointing to synergies with the broader biomedical informatics techniques, including knowledge representation and ontologies, semantic image annotation, imaging information models and tools, natural language and image processing, statistical modeling of image information, and computer assisted detection/diagnosis (CAD). We will conclude by describing emerging imaging informatics applications: image similarity, content based image retrieval, semantic image analysis, structured reporting, and decision support of image interpretation. During the tutorial, we will recognize the generalizability of imaging informatics methods to other imaging domains such as pathology as well as to the biomedical informatics field in general. At the conclusion of the tutorial, attendees will have a deeper understanding of the exciting developments driving the imaging informatics field, the use of informatics methods to images, the new algorithms/techniques, and the forthcoming applications that will enable discovery in research and improvement in healthcare quality.

By the end of the tutorial, participants will be able to:

• Have an overview of the key methodologies and overview of some key applications of biomedical imaging informatics
• Recognize the generalizability of imaging informatics methods to other imaging domains such as pathology as well as to the biomedical informatics field in general
• Have a deeper understanding of the exciting developments driving the imaging informatics field
• Understand the use of informatics methods to images, the new algorithms/techniques, and the forthcoming applications that will enable discovery in research and improvement in healthcare quality

Outline of Topics:

• Introduction and motivation
• Overview of imaging modalities
• The Radiology Workflow
• Major imaging-related standards: DICOM, RadLex, caBIG AIM, HL7CDA, IHE
• Semantic image analysis (ontology-based image annotation; structured reporting)
• Quantitative image analysis (image processing; segmentation; quantitative features; image vectors
• Applications (iPAD tool, CBIR, automated treatment response assessment, image-based decision support)
• Research Directions (linking images to other data: pathology and molecular data)

Intended Audience: Scientists; researchers; physicians, nurses, and other healthcare professionals; computer scientists, system developers, and programmers who work with or are interested in learning about using medical images in their work.

Content level: 40% Basic; 30% Intermediate; 30% Advanced