DASH Webinar: Artificial Intelligence in the diagnosis and prognosis of Covid-19

News
During the covid pandemic, AI prediction models were urgently needed to support medical decision making. In theory, AI could make predictions based on data about the course of the disease or about the chance that someone will die from the virus. In practice, however, it is a lot more unruly, research now shows. During a DASH webinar on October 6 we welcome Laure Wynants, assistant professor Epidemiology at the universities of Maastricht and Leuven, to tell more about her research into the prediction models for diagnosis and prognosis of covid-19.

The research that Laure's team conducted shows that several things go wrong when using AI during the pandemic. In some cases, incorrect training data was used and/or data was of too low quality. Furthermore, issues such as lack of time and incompetence also play a role in this. How come so much goes wrong with these prediction models? And does this mean that AI cannot contribute to the provision of good healthcare? In this DASH Webinar we will discuss these matters and Laure will present the results, conclusions and future visions on the use of AI in the diagnosis of Covid-19.

Laure Wynants

Laure Wynants is an assistant professor of Epidemiology at the universities of Maastricht and Leuven. Her research focuses on the development, validation, and impact of clinical predictive models for diagnosis and prognosis. She is interested in methods to deal with heterogeneity between populations, and in the clinical utility of prediction models. Her applied work deals with models for gynecological cancers, bloodstream infections, and Covid-19.

This webinar has already taken place. You can watch the recording here.