Ethical and Synthetic Data Use for Exponential Technologies

Abstract: In part one, we will review a concept of synthetic healthcare data that preserves individual privacy and thus enables a much wider effort to conduct research and develop applications to improve consumers’ health and wellbeing than would be possible by using real patient data. We will discuss various approaches to generate synthetic data and present a machine learning model that is capable of generating synthetic medical claims which comprise the main form of data used in health insurance industry.

Part two will offer an introduction to AI Ethics in healthcare. We will review some of the sources of medical ethics, vulnerable populations, and the problem of bias in training data. We will discuss appropriate use of sensitive variables, techniques for fairly training supervised learning algorithms, and selecting the best fairness metric.

Bios: 
Mark Shapiro, PhD is a Senior Director at Anthem Exponential Technologies COE where he leads the effort to develop AI and Machine Learning capabilities for digital healthcare. Prior to joining Anthem, he conducted experimental research in Neuroscience, executed Data Science projects in the industry, and worked on the development of an automated Machine Learning platform at a successful startup.

A. Sasha Gutfraind, PhD is a Principal Data Scientist at Anthem’s Exponential Technologies. In his role, develops and builds algorithms that use AI to enable personal digital healthcare services. Sasha holds a PhD in Applied Mathematics and enjoys philosophy and sailing.

September 1st, 2021 – Wednesday @ 12:00 CT

Data Standards & Quality

Abstract: Join Dr. Nguyen and Dr. Vreeman in the discussion of how health data standards can improve health by making data portable and understandable.

Dr. Nguyen will describe how the adoption of the HL7 FHIR interoperability standard, spurred by federal regulations, is setting the stage for transformation of the health IT ecosystem. Dr. Vreeman will explore how standards that enable interoperable, reusable data can serve as rocket fuel for advanced computational methods. He and Dr. Vreeman will engage in discussion on the real-world implementation of standards such as LOINC and FHIR. They are the leading practitioners in this space and can provide the good, the bad and the ugly.

Bios: 
Dr. Viet Nguyen is an internist, pediatrician, clinical informaticist and consultant to government and commercial organizations in developing interoperable workflows and technologies. He has two decades of experience in Health IT focused on interoperability standards and product development. Formerly the Chief Medical Officer for Lockheed-Martin and Leidos Corporation, Dr. Nguyen is a nationally recognized FHIR educator, an HL7 Board Member, a FHIR Foundation Board Member, CMO for Logica, and Technical Director for the HL7 Da Vinci Project.

Dr. Daniel Vreeman is a physical therapist, biomedical informatician, and expert in health data standards. His work aims to create a global health ecosystem where data is available with open standards that unlock the potential for information systems and applications to improve health decision-making and care. Dr. Vreeman is the Senior Clinical Data Standards Lead at RTI International where he advances health data standards and interoperability.

August 20th, 2021 – Friday @ 12:00 CT

Is this a system? Making Sense of Health Care in the United States

Abstract: Establish the core principles and mechanisms of the US healthcare system, and also set the stage for a deeper dive into disparities.

Bio: I am general pediatrician/internist with deep interest and multifaceted activities in population-centered, timely health services and policy research within the context of community needs. I also have strong interests in interdisciplinary research that stretches over the lifecourse, and translational research that connects medical advances with community impact. My central areas of methods expertise are survey research and policy-focused analyses of health services data, and my areas of subject. I am general pediatrician/internist with deep interest and multifaceted activities in population-centered, timely health services and policy research within the context of community needs. I also have strong interests in interdisciplinary research that stretches over the lifecourse, and translational research that connects medical advances with community impact. My central areas of methods expertise are survey research and policy-focused analyses of health services data, and my areas of subject matter expertise include vaccines and health insurance.

Pre-recorded webinar to view at your convenience.

Augmented Ethical Intelligence for Deep Medicine

with guest panelist Dr. Kelly Michelson

Abstract: Augmented Intelligence presents tremendous opportunities to enhance diagnostic and other health care delivery activities. Nonetheless, it also has the potential of exacerbating the impact of human biases that can further disadvantage various populations. This presentation provides an overview of salient ethical considerations in using augmented intelligence to design health technologies, with particular attention to enhancing our ability to identify, address, and correct health disparity.

Bio: Anita Ho is a bioethicist and health services researcher with a unique combined academic training and experience in philosophy/bioethics, public health, and business. She is currently an Associate Professor at the Centre for Applied Ethics at the University of British Columbia and in the Bioethics Program at the University of California, San Francisco (UCSF), a Scientist at the Centre for Health Evaluation and Outcome Sciences (CHÉOS), and the Regional Director of Ethics in Northern California for Providence Health. An international scholar and author of more than 70 publications, Anita’s research focuses on the ethical dimensions of utilizing innovative and artificial intelligence technologies in health care, domestic and global health disparity, supportive decision making, and end-of-life care decisions. She is currently completing a book manuscript on AI health monitoring ethics, to be published by Oxford University Press. She is also working on a project as a Fellow at Emerson Collective on the use of digital monitoring during the COVID pandemic.

July 14th, 2021 – Wednesday @ 12:00PM CT

Bringing Machine Learning Models to the Bedside at Scale

Abstract: Early warning systems — machine learning (ML) models that run every few minutes on hospitalized patients — have the potential to play a large role in supporting patient care. By estimating the risk of a myriad of clinical outcomes using real-time data, such systems identify patients in need of attention even when clinicians are not in the room. When linked to interventions, these models bring care to patients when they need it most. However, deploying model-driven systems in a hospital environment at scale comes with several challenges related to software infrastructure and governance in addition to model-related issues (e.g., miscalibration, selection of relevant thresholds, and use of proprietary models) and intervention-related issues (e.g., alert fatigue, assessing efficacy). In this talk, I will share several examples of real-world issues faced by health systems related to deploying ML models at scale and in bringing them to the bedside. I will draw from my own experience, where I oversee the operationalization of ML models at Michigan Medicine, as well as that from colleagues across the country.

March 30th, 2021 – Tuesday @ 12:00pm CT