Predictive Healthcare Speaker Series

Sharing lessons and perspectives on successfully implementing and integrating predictive healthcare solutions

As interest in applying the tools of data science to clinical and operational healthcare problems increases, the success of the solutions will depend on much more than the performance of a particular algorithm.

The Penn Medicine Predictive Healthcare speaker series aims to bring together a wide array of players from across the institution who have a stake in the successful operationalization of data science solutions in health care. The topics will encompass separate disciplines that come together to deliver these solutions, including:

  • Computer science
  • AI/machine learning
  • Human factors and human computer interaction
  • Real-time and big data architectures
  • Process improvement
  • Clinical/medical research
  • We hope that this speaker series and discussion forum will foster collaboration and an exchange of ideas to deliver effective predictive solutions in healthcare.

    Check back here for more information on upcoming events.

    Speaker/Topic Suggestions? Email Us


    System Pulse: Enabling a Learning Health System in the Heart and Vascular Service Line

    Jan 28th, 2022, 12:00-1:00pm Add to Calendar

    Speakers
    Abstract

    The Heart and Vascular Service Line (HVSL) is moving towards managing quality of care with a Learning Health System (LHS) framework. Currently, the HVSL obtains quality metrics that are more than six months old and are not specific to Penn Medicine care. The lack of access to this information in a timely manner limits the impact of quality improvement, research, innovation, and publishing, affecting care for patients and HVSL program evaluations.

    A key piece of the Learning Health System framework is a process and tool to identify unexpected patient outcomes based on patient characteristics and clinical care. With the pilot project in Cardiac Surgery, dubbed System Pulse, risk-adjusted patient outcomes are visualized to identify statistically significant trends, and quality metrics are derived 6-to-12-months sooner than report releases. Dr. Nimesh Desai will detail the design and methods that were used to develop System Pulse.

    Location

    Virtual



    Leveraging Electronic Health Records to Improve Disease Diagnosis: Study Design and Analysis Considerations

    Dec 17th, 2021, 11:00-12:00pm Add to Calendar

    Speaker

    Jinbo Chen, Ph.D., Professor of Biostatistics; Director, Statistical Center for Translational Research in Medicine; Department of Biostatistics, Epidemiology & Informatics, University of Pennsylvania Perelman School of Medicine

    Abstract

    In the context of several predictive healthcare projects based on Penn medicine EHR, Dr. Chen will highlight challenges in using EHR data for research towards improved disease diagnosis and clinical decision making, including EHR phenotyping, data under-reporting, data incompleteness, and data drifting. Dr. Chen will discuss study designs and statistical methods that they have developed to address these challenges.

    Location

    Virtual



    Melding Digital Twin and Live Experimentation to Optimize ED Staffing Alignment

    March 26th, 2021, 12:00-1:00pm Add to Calendar

    The ability to dynamically optimize staffing is crucial to an efficient Emergency Department; the COVID19 pandemic has highlighted even more clearly the need for agile responses to rapidly changing circumstances. We will present a digital simulation that enables ED Operations Leadership to model multiple variables rapidly and predict changes to key ED metrics based on variations in productivity and throughput.

    Speakers/Discussion
    Location

    Virtual


    Successful integration of deep learning into clinical care at Duke to improve sepsis care: an interdisciplinary discussion

    November 12th, 2020, 12:00-1:00pm Add to Calendar

    Speakers
    Moderators:
    Location

    Virtual


    Predictive Healthcare Speaker Series - Jon Pomeroy

    Title: Physician Insight on Predictive Models
    Speaker

    Jon Pomeroy, DO, MSHI, Senior Provider Clinical Application Analyst

    Dr. Pomeroy will discuss the physician perspective on what it takes to make a predictive healthcare project operationally successful.

    Location

    Flyers/76ers Surgery Theatre

    February 28th, 2020, 11:00-12:00pm

    Add to Calendar


    Inaugural Predictive Healthcare Speaker Series Talk

    Title: Precision palliative care: Leveraging big data to improve serious illness care
    Speaker

    Katherine Courtright, MD, Assistant Professor of Medicine

    Dr. Courtright will discuss the development and implementation of Palliative-Connect, a predictive healthcare initiative to improve patient access to specialty palliative care services.

    With introduction from Dr. Bill Hanson, CMIO.

    Location

    Flyers/76ers Surgery Theatre

    November 8th, 2019, 12:00-1:00pm

    Add to Calendar