By Rabail Baig
The COVID-19 pandemic has produced a staggering array of challenges that clinicians, public health experts, and policy makers are struggling to meet. Data scientists and quantitative experts across the globe have gone into overdrive as they work to analyze a flood of information, seeking not only to better understand, track, and predict the disease, but also to help guide the response to it and ensure that timely, accurate, and trustworthy information is readily available for everyone from scientists and clinicians to communities and members of the public.
This urgent need to bridge the worlds of data science, clinical research, and public health was the driving force behind this summer’s COVID + Data Science Virtual Seminars. Sponsored by Duke Plus Data Science (+DS), the 8-week series in summer 2020 was devoted to exploring data science methods with direct applications to the COVID-19 pandemic.
The series of 12 lectures attracted more than 1,500 virtual attendances, with many participants joining across multiple weeks and topics. The series, which wrapped up in late August, provided the opportunity for audiences both within Duke and beyond to hear directly from experts on topics spanning data analysis and visualization, deep learning, statistical methods, natural language processing, molecular methodology, and more.
The Duke Molecular Physiology Institute’s Matthew Hirschey, PhD, delivered the first session of the series, providing an introduction to the emerging field of data science with a special focus on its utility for gaining insights into the COVID-19 pandemic.
In the second talk of the series, David Carlson, PhD, assistant professor at Duke Civil and Environmental Engineering, Duke Biostatistics and Bioinformatics, and the Duke Department of Electrical and Computer Engineering, touched upon natural language processing (NLP) and understanding the evolving COVID literature. Read more about how a research scientist at Duke applied methods from Carlson’s talk to an important ongoing project.
In his talk on molecular methodology connected to COVID-19 data, Ricardo Henao, PhD, assistant professor at Duke Biostatistics and Bioinformatics and the Duke Department of Electrical and Computer Engineering, highlighted use cases related to COVID-19 molecular analysis being done at Duke and other institutions.
In a week that focused on foundational concepts, Vice President for Research Lawrence Carin provided a simple introduction to deep learning, paired with talks on the PyTorch computational platform delivered by Matthew Kenney, MFA, assistant research professor for Duke Computational Media Arts and Culture, and Duke Forge Scholar and Duke Neurosurgery assistant professor Timothy Dunn, PhD.
In a talk that took place at the end of July, Rachel Lea Ballantyne Draelos, MD, a computer science doctoral candidate in the Duke Medical Scientist Training Program, introduced a variety of machine learning models for automated interpretation of chest computed tomography (CT) scans, and reviewed recent reports in the published literature that described methods and tools proposed or used for this application.
The +DS program partnered with Duke Science and Society to pair 2 technical sessions with panel sessions as part of the Coronavirus Conversations to discuss the ethical and policy implications of the novel use of data science to address the pandemic. Duke Law professor and Founding Director of Duke Science and Society Nita Farahany, PhD, JD, moderated the panels for Ethics of AI and Image Analysis on July 30, and Ethics of AI and Wearables on August 20.
Duke Biostatistics & Bioinformatics’ Benjamin Alan Goldstein, PhD, kicked off the August seminars by providing an overview of approaches for using data science to optimize scheduling elective procedures around complications imposed by the response to COVID-19. This was followed by a talk on causal inference for quantifying the efficacy of potential COVID-19 medications and vaccines by Department of Statistical Science professor Fan Li, PhD.
In the final lecture of the series, which focused on using wearables for early COVID-19 detection, Duke Biomedical Engineering assistant professor Jessilyn Dunn, PhD, discussed how her Big Ideas Lab at Duke is using data from smart phones, smart watches, fitness trackers and other wearable devices to create a screening system for COVID-19. Read more about Dr. Dunn’s lecture and her work on the Duke CovIdentify Study.
“For this series, what we really wanted to do was introduce a diverse audience to the broad spectrum of COVID-19 research and expertise at Duke, and to offer them a closer look at some of the key data science methods supporting or even driving this research,” said Lawrence Carin, PhD. “We were really pleased to see the levels of interest and enthusiasm for these sessions, and we’re looking forward to presenting more virtual sessions this fall.”