Emily Fox, Distinguished Engineer at Apple and Professor at the University of Washington gave a tech talk during the WiDS conference on The joys and perils of leveraging mechanistic models in health machine learning. In her talk, she explores combining the domain knowledge of mechanistic models with the flexibility and expressivity of machine learning methods through two use cases: Modeling the relationship between mobility and transmission in the COVID-19 pandemic, and glucose forecasting in Type 1 diabetes.
Geetha Manjinath, Founder and CEO of AI startup Niramai, gave a talk on AI in Healthcare: Cancer to COVID. Her company, Niramai, developed a novel software-based medical device that uses machine learning to detect breast cancer at a much earlier stage than traditional methods or self-examination. This technology is deployed in several hospitals in India today and helps enhance productivity, provide better accuracy and improve patient outcomes. She also discusses how Niramai is using AI in the Niramai FeverTest technology to screen for COVID symptoms.
Haley Hedlin and Yingjie (Isabel) Weng
Two scientists from the Stanford University’s School of Medicine, Haley Hedlin, Associate Director of Clinical Trials and Yingjie (Isabel) Weng, Senior Biostatistician, hosted a workshop on Data Analysis for Health, with a Focus on COVID. They discussed the challenges and lessons learned from ongoing COVID clinical trials. These trials typically move very slowly but with COVID everyone needed to move fast. They described how it was challenging to be constantly learning new things about the disease in the midst of clinical trials. You make assumptions in the beginning of the trial but have to constantly adapt to a firehose of new information like new drugs and changes in the standard of care.
Francesca Dominici and Rachel Nethery
In their WiDS podcast, Using Data Science to Study Air Pollution Effect on COVID-19 Outcome, Francesca Dominici and Rachel Nethery, from Harvard University’s T.H. Chan School of Public Health, talk about their work to inform public health policy through research at the intersection of environmental health science, data science, climate change and health policy. With the onset of the COVID-19 pandemic, they saw a way to connect the research they were doing on air pollution and health with the pandemic. They are studying the effects of air pollution exposure on different causes of hospitalization to see if pollution could increase a person’s vulnerability to COVID-19. While the research is at a preliminary stage, there is a lot of information that points towards the possibility that long-term exposure to air pollution could increase the mortality risk for COVID-19.
In another podcast, Applying Data Assimilation Tools to COVID Forecasting Models, Femke Vossepoel, Professor in Geoscience and Engineering at Delft University of Technology in the Netherlands, explains how data assimilation tools can be used to improve COVID-19 forecasting models. Femke explains that data assimilation originated in weather forecasting, where a model is updated with the current day’s weather observations to provide a more accurate forecast for the next day, and this concept can be applied to COVID forecasting.
We also learn from Manisha Desai, professor of medicine and biomedical data science at Stanford University, who is an expert in the design and analysis of clinical trials and epidemiologic studies across multiple diseases. In her WiDS podcast, The Importance of Data Integrity in COVID-19 Clinical Trials, she shares insights about the challenges and progress of current COVID-19 clinical trials. As data scientists, it’s critical to ensure rigor, that studies are designed well, that limitations are understood and that there's an appropriate interpretation of the findings. She says one of the most important foundations to the success of the research is ensuring that the data has integrity. “There will be many lessons learned, not just for COVID, but there will be other pandemics, and we will be able to learn from this experience.”