#MondaYmaging: seminari d'imatge mèdica · Automatic segmentation of cardiovascular MRI. Is the problem solved thanks to deep learning ?

Ponent:

• Dr. Alain Lalande, Université de Bourgogne

Resum

Magnetic Resonance Imaging (MRI) has become the technique of choice for the evaluation of cardiovascular diseases. However, the manual processing of the data is highly time-consuming and the reliability of the diagnostics largely depends on the experts' experience and skills. With the widespread application of cardiovascular MRI, researches about the automatic segmentation of the images are being carried out. Today, methods based on deep learning outperfom the state-of-the-art methods. This presentation will introduce the conventional deep learning approaches used in cardiovascular MRI (and more broadly in medical imaging), what we can expect and the specific difficulties and limitations, as the future works.

Si desitges obtenir l'enllaç de connexió contacta amb frecerca.comunicacio(ELIMINAR)@sjd.es

Recordem que els seminaris no es graven i, per tant, no hi ha possibilitat de tornar-los a visualitzar.