ADS Webinar | NLP for Health Research
12:00 Introduction & Welcome
12:05 Talk #1 NLP 4 COVID: A Deep Active Learning Covid-19 Relevancy Algorithm to Identify Core Scientific Articles
12:30 Talk #2 TBA
Talk #1 by Zubair Afzal
Zubair Afzal is a Principal NLP Scientist at Elsevier. He has a PhD in Clinical Text Mining and a Professional Doctorate in Engineering. He has been working in Elsevier for over 5 years. His main interests are in the fields of NLP and NLU using deep learning methods.
Ever since the COVID-19 pandemic broke out, the academic and scientific research community, as well as industry and governments around the world, have joined forces in an unprecedented manner to fight the threat. In this combat against the virus responsible for the pandemic, key for any advancements is the timely, accurate, peer-reviewed, and efficient communication of any novel research findings. They present a novel framework to address the information need of filtering efficiently the scientific bibliography for relevant literature around COVID-19. The contributions of their work are summarized in the following: they define and describe the information need that encompasses the major requirements for COVID-19 articles’ relevancy, we present and release an expert-curated benchmark set for the task, and we introduce CORA (Covid Relevancy Algorithm), a deep active learning model to identify core scientific articles.
Talk #2 by Roser Morante, Stan Frinking and Edwin Geleijn
Roser Morante is assistant professor at the VU Amsterdam, Faculty of Humanities, where she teaches in the text mining master. She has worked on biomedical natural language processing mostly applying text mining techniques to extract clinical and biomedical information from scientific texts. She has worked extensively on processing extra-propositional linguistics aspects of meaning, such as modality and negation.
Stan Frinking has a background in linguistics, but made the switch to Artificial Intelligence/NLP. During his master’s study Text Mining, he wrote his thesis about the use of Text Mining Techniques to classify medical patient notes.
Edwin Geleijn is a physiotherapist and healthcare innovator working at the Amsterdam UMC. As a healthcare innovator he aims to explore new ways of improving healthcare which more and more involve ehealth such as activity tracking, development of smartphone apps and data science.
Falls that happen in hospitals pose serious clinical and legal problems, with regulatory consequences. To effectively use fall prevention techniques, hospitals need to have information about the falls, like location and time, but since falls are often underreported or misclassified in incident reporting systems, the best source of information are medical patient notes. Reading all these notes manually is very time-consuming, so Text Mining techniques can be applied instead. We present our design for a system architecture, discussing primarily its implementation and results at the AUMC hospital in Amsterdam, while also reflecting on challenges that we will have to overcome in the continuation of this project.
The event is hosted on Hopin, you will have to get a ticket to access the event: