The COnfVID19 initiative is developing a support system in reliable, interpretable models to optimise healthcare resources
A method to create reliable early prognosis models to predict the progress of COVID-19 in patients is being developed by TECNALIA and the Computer Science and Artificial Intelligence Department of the Computer Science Faculty of the University of the Basque Country, (UPV-EHU), working with the Basque Center for Applied Mathematics (BCAM).
COnfVID19 will provide a medical prognosis model to be used in clinical practice by Basque public healthcare system (Osakidetza) professionals. To do this, artificial intelligence explainability and reliability models will use data supplied by HM Hospitales, using vital signs, analytical tests, treatments and radiography-based imaging as reference points.
The research will testify to the practical value of the reliability methodology of artificial intelligence models. Work is already being carried out in this area in the recently approved 3KIA project (“An Integrated, Transversal Proposal for the Design and Implantation of Reliable Artificial Intelligence-based Systems, ELKARTEK 2020”), incorporating the three institutions involved in the COnfVID19 project.
