Structural and kinematic analysis of the arterial wall in ultrasound image sequences using deep learning

The project deals with automated analysis of ultrasound-image sequences representing the carotid-artery wall, to assess the tissue motion in an attempt to detect pathologic changes in the biomechanical properties. The first goal of the project is to develop a segmentation method capable of delineating the arterial wall throughout image sequences, so as to estimate its cyclic compression. To do so, a deep-learning approach will be explored, based on available data annotated by experts.

The candidate is expected to have skills in image processing, deep learning, and programming techniques including code versioning. Her/his professional project should include a PhD. Knowledge in ultrasound-image physics and computation, as well as in mechanical modelling, will be appreciated.

A detailed description of the whole project can be found using this link.