Analytics Track Keynote
Explainable Deep Tabular Learning Models for Biomedical and Healthcare Applications
University of Virginia
Professor of Computer Science, Biomedical Engineering, and Data Science
Abstract: Neural networks have achieved great success in many tasks. However, their black-box nature limits their use in sensitive areas such as medical applications in which both performance in accuracy and intelligibility are critically important. In this talk, I will discuss our recent research on explainable deep tabular learning, a specific kind of deep learning models for tabular data which is very common in medical and healthcare applications. In particular, I will show how the concept-based learning models and example-based learning models can be designed for explainable deep tabular learning. I will also discuss their applications in biomedicine and healthcare.
Bio:Dr. Aidong Zhang is a Professor of Computer Science, Data Science, and Biomedical Engineering at University of Virginia (UVA). Prof. Zhang’s research interests include machine learning, data science, bioinformatics and computational biology, and health informatics. Prof. Zhang was the Editor-in-Chief of the IEEE Transactions on Computational Biology and Bioinformatics (TCBB) from 2017 to 2021. She served as the founding Chair of ACM Special Interest Group on Bioinformatics and Computational Biology (SIGBio) from 2011 to 2015 and also served as the Chair of its advisory board from 2015 to 2018. She was also the founding and steering chair of ACM international conference on Bioinformatics, Computational Biology and Health Informatics (ACM-BCB) from 2010 to 2019. Prof. Zhang is a fellow of ACM and IEEE. She is also a fellow of the American Institute for Medical and Biological Engineering (AIMBE).
System Track Keynote
Challenges and Opportunities for Improving Patient Safety Through Data Science and Informatics
Harvard Medical School
Program Director Research, Center for Patient Safety, Research, and Practice, Brigham and Women’s Hospital, Associate Professor of Medicine
Abstract: Patient safety informatics is an important area of research because despite our good intentions, medical harms are a principal cause of preventable injury and remain a leading cause of death in the Unites States and globally. Data science and informatics offer significant opportunities for improving patient safety, including the ability to develop predictive models that can identify patients at risk of adverse events and interventions including clinical decision support that can mitigate those risks. Additionally, advanced analytics can help healthcare providers identify patterns and trends that may indicate safety concerns, allowing for proactive interventions to prevent harm. This presentation will highlight data science and informatics tools and approaches that have the potential to significantly improve patient safety and reduce the risk of adverse events in healthcare settings.
Bio: Patricia Dykes is Associate Professor of Medicine at Harvard Medical School and Research Program Director for the Center for Patient Safety, Research and Practice at Brigham and Women’s Hospital in Boston. Her research aims to improve quality and safety through patient engagement and clinical decision support (CDS). Dr. Dykes is currently leading patient safety informatics clinical trials to improve fall prevention and care transitions in primary care for patients with multiple chronic illness. She is also leading development of CDS and an electronic clinical quality measure to prevent and quantify delayed diagnosis of venous thromboembolism. In addition, Dr. Dykes is the site PI for the CONCERN study which uses data science and machine learning approaches to identify hospitalized patients at risk for deterioration. Dr. Dykes is author of 2 books, over 175 peer reviewed publications, and has presented her work nationally and internationally. She is immediate past President and Board Chair of the American Medical Informatics Association, an elected fellow of American Academy of Nursing, the American College of Medical Informatics, and the International Academy of Health Sciences Informatics.
Industry Track Keynote
Data and System Harmonization: Reflections and observations from 25 years of research in disparate fields
Abstract: Drawing on 25 years of personal and professional research experience in the fields of acoustics, psychology, diabetes, and data science, this presentation will share some of the general patterns (and problems) that exist across multiple and disparate disciplines. Perhaps these general patterns, or universal principles, hold the keys to solving complex system problems? This talk will cover some general topics such as the importance of starting with why, fusing human expertise and AI technology, and creating trusted interdependent systems. Real-world examples from the field of diabetes will be given with a focus on patient-centered care using just-in-time-adaptive interventions and clinician facing precision engagements.
Bio: Ed Nykaza, PhD, is Glooko’s VP, Data Science & Clinical Research. His 25-year research career spans the academic, government, non-profit, do-it-yourself (DIY) and industry sectors, including 100+ presentations and 35 published papers. Ed is passionate about building and leading high-performance teams, creating psychologically safe working environments, and using data and science to improve the quality of people’s lives.