Luca Romeo, Università Politecnica delle Marche, Ancona, Italy.
October 27, 2020. 12:45 – 1:45 pm.
A Virtual Seminar on Blackboard Collaborate.
Several countries are experiencing sustained local transmission of coronavirus disease (named COVID-19) in people of all ages. Although some people with COVID-19 have mild to moderate symptoms the disease can cause severe medical complications and lead to death in a relevant percentage of people. Recent years have witnessed an increasing amount of available Electronic Health Record (EHR) data and Machine Learning (ML) techniques have been considerably evolving. The data stored in EHRs can be analyzed through ML algorithms to discover complex patterns and set-up powerful ML models that can be integrated into a Clinical Decision Support System to predict the pathological risk condition related to COVID-19. However, managing and modeling this amount of information may lead to several challenges such as overfitting, model interpretability, sparse observation over time and the natural unbalance of the predictive task. Starting from these motivations, in this talk, we will present our recent works related to the design and application of sparse machine learning, multiple instance learning and multi-task learning methodologies for predicting the risk profiles associated with different diseases. The focus will be on how the proposed methodologies are extended and applied for early-stage prediction of complications and risk stratification of COVID-19 patients in Intensive Care Units (ICUs). The research activity is currently performed as part of the “Collaborative international ICU registry for critically ill COVID-19 patients – RISC-19-ICU” (https://sites.google.com/view/risc-19-icu) joining together the efforts on EHRs collection of 97 ICU centers from 16 countries.