The rapid change in computational capabilities has made Big Data a major modern statistical challenge.  Less well understood is the rise of Complex Data as a perhaps greater challenge.  Object Oriented Data Analysis (OODA) is a framework for addressing this, in particular providing a general approach to the definition, representation, visualization and analysis of Complex Data.  The notion of OODA generally guides data analysis, through providing a useful terminology for interdisciplinary discussion of the many choices typically needed in modern complex data analyses.  The main ideas are illustrated via a survey of a number of approaches which integrate differential geometry and Bayesian statistics, yielding powerful image segmentations. 

Dr. Marron is an elected Fellow of the American Statistical Institute and the Institute for Mathematical Statistics, and an elected Member of the International Statistical Institute.  Dr. Marron has served as Associate Editor for the Annals of Statistics, the Journal of the American Statistical Association, the Journal of Nonparametric Statistics, Computational Statistics, Test and the Electronic Journal of Statistics.  Dr. Marron has presented the Theory and Methods Invited Paper for the Journal of the American Statistical Association, been the Institute of Mathematical Statistics Medallion Lecturer, and presented the S. N. Roy Memorial Lecture at the University of Calcutta.  He has delivered the Bradley Lecture at the University of Georgia, and the Information Science and Technology Center Distinguished Lecture at Colorado State University.

The overall goal of our group is to leverage data-driven approaches to help improve patient outcomes. This talk will demonstrate examples of how we are working toward this goal by leveraging large clinical datasets, data analytics, machine learning and deep learning. Specific examples include: 1) using 46,583 clinically-acquired 3D computed tomography images of the brain to develop and implement a deep learning model to efficiently reprioritize radiology worklists for quicker diagnosis of intracranial hemorrhage; 2) using deep learning to analyze 723,754 echocardiographic videos of the heart to accurately predict patient mortality; 3) analyzing 2 million 12-lead electrocardiographic tracings from the heart to predict clinically relevant future events and 4) optimizing evidence-based care delivery for a population of >10,000 patients with heart failure using machine learning.

2 data projects will be reviewed

• Early Identification of Sepsis
• Geospatial Infection Surveillanc

As CMS Administrator Seema Verma states, “Interoperability imagines a world in which medical decisions are fully informed by medical history; health history follows consumers wherever they go; third parties can leverage Electronic Health Record (EHR) systems; data is used not only to treat, but to prevent illnesses; and researchers are using the information to develop cures.”  This presentation provides an introduction to the world of interoperability, an overview of the Electronic Medical Documentation Interoperability (EMDI) initiative, and how the program aligns with CMS’ goals.

As CMS Administrator Seema Verma states, “Interoperability imagines a world in which medical decisions are fully informed by medical history; health history follows consumers wherever they go; third parties can leverage Electronic Health Record (EHR) systems; data is used not only to treat, but to prevent illnesses; and researchers are using the information to develop cures.”  This presentation provides an introduction to the world of interoperability, an overview of the Electronic Medical Documentation Interoperability (EMDI) initiative, and how the program aligns with CMS’ goals.

Pallavi has worked on product development of face-to-face provider assessments and supplemental member encounter tool, with the goal of achieving a continuum of care. Pallavi has strong technical skills from her IT qualifications. She has extensive experience in spearheading a worldwide project to build a business intelligence system for generating analytical reports for an international organization using big data from diverse sources.

Pallavi is passionate about healthcare IT and interoperability. She finds intuitive ways to blend her expertise in business intelligence, data analysis, decision support systems, stakeholder management, product development, and software development to build nationwide interoperable solutions. 

There is little known about how academic medical centers (AMCs) in the US develop, implement, and maintain predictive modeling and machine learning (PM & ML) models.  We conducted semi-structured interviews with leaders from AMCs to assess their use of PM &ML in clinical care, understand associated challenges, and determine recommended best practices. Each transcribed interview was iteratively coded and reconciled by a minimum of two investigators to identify key barriers to and facilitators of PM & ML adoption and implementation in clinical care.  Interviews were conducted with 33 individuals from 19 AMCs nationally. AMCs varied greatly in the use of PM & ML within clinical care, from some just beginning to explore their utility to others with multiple models integrated into clinical care. Informants identified five key barriers to the adoption and implementation of PM & ML in clinical care: (1) Culture and personnel, (2) Clinical utility of the PM & ML tool, (3) Financing, (4) Technology, and (5) Data.  Recommendation to the informatics community to overcome these barriers included: (1) Development of robust evaluation methodologies, (2) Partnership with vendors, and (3) Development and dissemination of best practices. For institutions developing clinical PM & ML applications, they are advised to: (1) Develop appropriate governance, (2) Strengthen data access, integrity, and provenance, and (3) Adhere to the 5 rights of Clinical Decision Support.

His research interests have revolved around the use of health information technology to improve the quality of care and patient safety in both the ambulatory and hospital settings.  His work in the ambulatory setting has focused on the efficient delivery of decision support to clinicians to prevent errors of omission and commission during diagnostic test ordering and review of test results. He has also worked to use information technology, including secure on-line patient portals, to improve the communication between clinicians and patients around health maintenance and the follow-up of abnormal test results.  In the inpatient setting, he has conducted several studies to delineate the barriers to and facilitators of the wide-spread diffusion of computerized physician order entry and has led many studies evaluating the safety, financial and socio-technical impact of barcode technology in the hospital pharmacy and nursing units.

Dr. Poon also mentors residents, fellows and junior faculty members as they developed HIT intervention and evaluation strategies. He currently serves as Duke’s inaugural Program Director for its ACGME-accredited Clinical Informatics Fellowship Program.  He has also served as an Associate Editor of the Journal of the American Medical Informatics Association (JAMIA) since 2017. He was elected as a Fellow to the American College of Medical Informatics in 2018.

Currently, Ted’s interest are focused primarily on Artificial Intelligence and Agent-Based Systems as they are applied within Distributed Computing.

From startups to being responsible for C-Level P&L at publicly traded corporations Ted has held architect positions at both Apple and Microsoft and held instrumental roles in several start-ups, including most recently where he was Co-Founder and CTO of PokitDok Inc. acquired by Change Health in 2018. He was also with digidesign (IPO and acquired by Avid), Crystal River Engineering (acquired by Aureal Seminconductor), VP of R&D at MongoMusic (acquired by Microsoft) and Co-Founder and CTO of BeliefNetworks (acquired by Benefitfocus). He was also the CTO of Spatializer Audio Labs (NASDAQ: SPAZ), a company specializing in digital signal processing solutions. He is on the IAB for the University of South Carolina Computer Science Department as well as the Center for Intelligent Systems and Machine Learning at the University of Tennessee. He also serves on the Clemson University Restoration Institute Executive Advisory Board for the Zucker Family Graduate Education Center. Ted has published numerous articles in leading technical magazines and holds several patents in the areas of blockchain, semantics, machine learning, signal processing and signal protection. He has also held a Top Secret Security Clearance. He is also a member of the ACM and IEEE.

The growing availability of electronic health records data has
generated considerable research activity related to predicting
patient events and outcomes and investigating questions about
the causal effects of treatments and interventions.  Combined
with the accessibility of modern machine learning tools, the
possibilities for new, empirically-driven discoveries seem almost
unlimited.  Drawing from collaborations in HIV research, I will
describe several case studies where both causal and predictive
inference methods are used to investigate important clinical
and policy-related questions using electronic health records data.
I will highlight key differences between predictive and causal
inference, and will emphasize the importance of sampling design
and principled statistical reasoning in coping with the complexities
inherent in using electronic health records.  These include nonrandom
selection of patients, nonrandom allocation of treatments and
interventions, irregular observation patterns, and missing data.

Dr. Shalkham is a Diplomate in Clinical Informatics from the American Board of Preventive Medicine, a Fellow of the American College of Emergency Physicians and a member of the American Medical Informatics Association. She holds a B.A. in Environmental Science from Duke University and a M.D. from the Medical University of South Carolina. Dr. Shalkham completed her residency training in Emergency Medicine at NYU/Bellevue Hospital Center where she also served as chief medical resident.  She holds a Master’s in Public Health from the Arnold School of Public Health at the University of South Carolina. She continues to practice Emergency Medicine at Lexington Medical Center.

 How data-driven conversations create opportunities to address quality, cost and delivery of care … one state’s experience.

As CEO, Bennett oversees NJHA, the Health Research and Educational Trust of New Jersey, a not-for-profit affiliate that promotes continuing education, patient safety, quality improvement and research; and the for-profit Healthcare Business Solutions, which provides group purchasing and other business solutions for healthcare providers. She co-founded Q³HIP, LLC in 2018 with a focus on improving quality of care, patient safety and education and leveraging the collective capabilities of the Hospital Associations of New Jersey, Pennsylvania and Ohio.

Prior to being selected as CEO of NJHA, Bennett served as New Jersey’s 20th Health Commissioner.

During this interactive discussion please join Dr. Steve Kearney Medical Director at SAS as he takes you on a patient journey that demonstrates the operationalization of Big Data and Analytics.  Innovative stops along the way will include Electronic Health Records, Claims Data, Genomics and Social media. The focus will be on transforming a world of data into a world of intelligence.

An innovator in health outcomes and digital medicine, Dr. Kearney co-developed and implemented one of the first electronic disease registries at Duke Health where he practiced in both inpatient cardiology and outpatient internal medicine during his joint faculty appointment at Duke and UNC. Dr. Kearney then joined the Medical Outcomes group at Pfizer where he continued his practical actionable approach to data and the patient journey.  He focused on health outcome insights from electronic medical record migrations, early personal digital assistants, novel health software programs and the first large patient claims databases.  Prior to joining SAS in 2016, he led Pfizer’s ACO and PCMH initiatives while also working with national thought leaders on health policy for state and federal agencies. 

The field of machine learning has been advancing with many shallow algorithms in the 20th century, and surge of interest in deep learning in the 21st century with the development of a practical backpropagation algorithm and off-the-shelf architectures for parallel processing of tensor operations. However, use of artificial intelligence and machine learning in particular has lagged in healthcare compared to many other applications. This talk will address two key issues with a focus on the questions in healthcare that they can be used to address. The first will be a focus on the historical relationship between healthcare and quantitative models. We will then discuss issues of matching data volume and hypothesis space.  

Paul spent a year as faculty at Harvard Medical School and then moved to Maine to start a private practice, during which time he held local hospital leadership positions and began to use data to address strategic healthcare challenges in a consulting role. Paul also co-authored a peer-reviewed book on applications of the R language shortly after he started in private practice. In 2018, Paul was hired by Palmetto GBA as a medical director where he is involved in making coverage and pricing decisions on for new technology in health care as well as the use of data to understand broad healthcare utilization patterns among Medicare beneficiaries.

In August of 2017, Health Sciences South Carolina (HSSC) named Mr. Deans as President & Chief Executive Officer (CEO) of the nation’s first statewide biomedical and health research organization. HSSC was established in 2004 and today its member organizations include the state’s largest research-intensive universities, as well as, its largest healthcare systems. This unique collaboration was formed with the vision of improving South Carolina’s public health and economic wellbeing through research and innovation.

Prior to joining HSSC, Mr. Deans served in the capacity of Vice President and Chief Operating Officer of Chesapeake Regional Healthcare (CRH), a not-for-profit healthcare system located in Chesapeake, Virginia. He served as their Chief Information Officer and Chief Compliance Officer, before serving as their COO. Prior to CRH, he served as the Vice President and CIO of Bassett Healthcare, an integrated delivery network of 6 hospitals and nearly 40 health centers in central New York state. Before Bassett Healthcare, he served in various technical management and leadership capacities, including Director of Technology at Greenwich Hospital, a member of the Yale-New Haven Health System, positioned in Greenwich, Connecticut.

Mr. Deans received his MBA (Beta Gamma Sigma) from the College of William & Mary in Williamsburg, VA, and Bachelor of Science from Excelsior College in Albany, NY. He is also a graduate of the Executive Nonprofit Leadership program at Harvard Kennedy School, John F. Kennedy School of Government in Cambridge, MA. He enjoys boating, camping, hiking, cycling and various other outdoor adventures, especially those that involve travel.