Pictures copyrighted

J. Ramon - Learning predictive models from patient records, with application to intensive care

Recently, in many medical disciplines and especially in departments such as intensive care, electronic patient records containing high-resolution monitoring data have become available. In this talk, we will discuss the advantages and shortcomings of this type of data, and the opportunities it offers for improving care by building predictive models.

After a general discussion, we will focus on an application in the intensive care, outlining how predictive models were developed and validated in collaboration with medical researchers.