Calina Ciuhu

With a background in theoretical physics, she carried out extensive research at Philips Research, focusing on data-driven solutions in the health-tech domain, ranging from video and image processing, medical image analysis, biomedical signal processing (e.g. measuring photoplethysmography, camera-based blood perfusion monitoring, core body temperature), and 3D parametric body modelling (prediction of internal anthropomorphic measurements from geometrical 3d-body measurements). 

In electromechanics, her approach follows the principles of physics, emphasizing problem formulation and solution interpretation as integral components of the methodology.  This approach combines domain knowledge as priors, and it is often perceived as a hybrid approach, in comparison with a black-box approach, in which the model is purely driven from data. 

Physics is governed by symmetries. I believe observed data contain the symmetries of the physical models that are discovered through these data.  (The action for the classical electromagnetism)