THE Event

"AI, Stats & COVID-19" is an online event organized by the Artificial Intelligence Student Society to promote the efforts undertaken by researchers of our local community in studying the COVID-19 pandemic by the means of statistical and AI-related approaches, ranging from Bayesian modelling to natural language processing.

Alongside the work of local researchers, we propose a thematic introduction to relevant notions in virology held by Dr. Roberto Luzzati and an international perspective in the study of the pandemic by Prof. Lorenzo Pellis.

Interaction between participants and speakers will happen during Q&A sessions. Questions can be posed by participants through the Sli.do platform during the talks, and will be read by hosts in the Q&A sessions.

General Info

Date 📅 Friday, May 8th 2020 at 17:00

Location📍 Zoom

Q/A Answers❓ AI, Stats & COVID-19 Q&A


Speakers

Roberto Luzzati

Associate Professor and Dirctor of SC Infective Disease of Azienda Ospedaliero-Universitaria of Trieste


COVID-19: A Virological Perspective

Lorenzo Pellis

Sir Henry Dale Fellow, Department of Mathematics, University of Manchester

Challenges in control of Covid-19: short doubling time and long delay to effect of interventions

Early assessments of the spreading rate of COVID-19 were subject to significant uncertainty, as expected with limited data and difficulties in case ascertainment, but more reliable inferences can now be made. Here, we estimate from European data that COVID-19 cases are expected to double initially every three days, until social distancing interventions slow this growth, and that the impact of such measures is typically only seen nine days - i.e. three doubling times - after their implementation. We argue that such temporal patterns are more critical than precise estimates of the basic reproduction number for initiating interventions. This observation has particular implications for the low- and middle-income countries currently in the early stages of their local epidemics.

Luca Bortolussi

Associate Professor and Coordinator of the Master in Data Science and Scientific Computing, University of Trieste

The COVID-19 FVG data science initiative

We will briefly present ongoing activity of a group of scientists in the region in terms of modelling and statistical analysis of data related to the COVID-19 ongoing epidemics.

Leonardo Egidi

Postdoctoral Researcher Dept. of Business, Economics, Mathematics and Statistics, University of Trieste

Bayesian Hierarchical Regression for COVID-19 Intensive Care Units (ICU)

We developed some hierarchical Bayesian models to predict in each Italian region the daily number of total hospitalized and intensive care units (ICU) due to the first phase of the covid-19 spreading outbreak. We included some covariates accounting for the lockdown measures, the daily medical checks and the regional membership. Model's goodness of fit is checked via posterior predictive checks, whereas predictive accuracy has been computed in terms of mean absolute percentage (MAPE) error and Bayesian coverage. Further extensions could consider the inclusion of geospatial covariates.

Guido Sanguinetti

Professor at the School of Informatics, University of Edinburgh and Faculty member at SISSA Data Science Excellence Dept.

The COVID-19 FVG data science initiative

We will briefly present ongoing activity of a group of scientists in the region in terms of modelling and statistical analysis of data related to the COVID-19 ongoing epidemics.

Marco Franzon and Tommaso Rodani

Students at the Master in Data Science and Scientific Computing, University of Trieste

COVID-19 Semantic Browser: Natural Language Processing Joins the Fight against the Pandemic

As of today, more than 57 000 scientific papers have been published worldwide on coronavirus-related research, with more than 5000 publications on the topics of SARS-CoV-2 and COVID-19 in the last few months. It is very likely that in such a large quantity of text a lot of useful information is lost, making our knowledge on the subject too sparse to be exploited to its full potential. Our COVID-19 Semantic Browser allows users to browse those articles on a dedicated web platform, matching paper abstracts with user queries formulated in natural language and allowing researchers to delve deeper in our current knowledge of the subject.

SCHEDULE

  • 17:00 - Greetings by the AI2S Directive Committee

  • 17:05 - Prof. Roberto Luzzati - COVID-19: A Virological Perspective

  • 17:25 - Prof. Lorenzo Pellis - Challenges in control of Covid-19: short doubling time and long delay to effect of interventions

  • 17:45 - Q&A Session 1

  • 17:55 - Prof. Luca Bortolussi & Prof. Guido Sanguinetti - The COVID-19 FVG data science initiative

  • 18:15 - Prof. Leonardo Egidi - Bayesian Hierarchical Regression for COVID-19 Intensive Care Units (ICU)

  • 18:35 - Tommaso Rodani & Marco Franzon - COVID-19 Semantic Browser: NLP Joins the Fight against the Pandemic

  • 18:50 - Q&A Session 2

POST-EVENT Q&A

Thanks to the incredible participation to our event, we were not able to encompass all of the questions that were asked to our awesome speakers on the Q&A platform.

We are putting together a solution to allow our speakers to answer the questions in a written fashion. You can find the offline Q&A session here: AI, Stats & COVID-19 Q&A

Questions?

For any question concerning the event, please contact us at events@ai2s.it