CompBioMed’s 21st e-Seminar took place on 22 February 2022 focused on Use of Gaussian Process Emulators in Cardiovascular and Musculoskeletal Biomechanics
Sensitivity analysis can be used for the identification of surrogate variables in a computational model. It also finds application in the extraction of the most important features of a model in order to develop a less computational expensive, but still accurate, version of it. This however requires several evaluations of the model, which can be unfeasible for large models.
Gaussian Process Emulators (GPEs) are statistical tools that are able toemulate the behaviour of complex nonlinear systems using only a limited number of model runs for training, and thus are ideal for performing sensitivity analysis.
This seminar introduces the basic concepts behind GPEs and presents two applications of them in conjunction with sensitivity analysis: biomarker identification in cardiovascular modelling and model
Presentation given by Dr Ivan Benemerito (University of Sheffield).
He has a background in mathematics and a PhD in modelling of the ankle joint. Part of the INSIGNEO Institute for insilico medicine since 2015, he joined the CompBioMed project in 2018 to develop methodologies for improving the diagnosis and treatment of ischaemic stroke using a combination of deterministic and statistical modelling.
This webinar series is run in collaboration with the VPH Institute