Supervisors: Sarah Leclerc, PhD, Alain Lalande, PhD
Starting date: 1st February 2021
One crucial parameter to evaluate the state of the heart after myocardial infarction (MI) is the viability of the myocardial segment, i.e. if the segment recovers its functionality upon revascularization. MRI performed several minutes after the injection of a contrast agent (delayed enhancement-MRI or DE-MRI) is a method of choice to evaluate the extent of MI, and by extension, to assess viable tissues after an injury. The knowledge of the myocardial borders is mandatory to evaluate the extent of the MI. However, as shown by the results of the Emidec challenge organized during the MICCAI conference (http://emidec.com/ or http://stacom2020.cardiacatlas.org/accepted-papers/) the results from deep learning approaches are very encouraging, but not still acceptable to a current use in clinical practice.
During a conventional cardiovascular exam, cine-MRI are also acquired in addition to DE-MRI. And we have shown [ber 18] that the use of deep learning approaches is very efficient to detect the myocardial border on cine-MRI. The objective of this internship is to propose an automatic segmentation of the myocardium on DE-MRI using also cine-MRI. Indeed these images are acquired at the same position. There is two options, the first one is to detect firstly the myocardium on cine-MRI and apply on DE-MRI (then some registration are necessary because the shape of the myocardium could be not exactly the same). The second one is to develop a network that take as input both modalities to segment DE-MRI.
To achieve this goal, the student will exploit the complete following dataset (with expertise): Dataset used for ACDC challenge and dataset used for Emidec challenge. Indeed, both dataset come from our structure. Moreover, student will have access to all data from University Hospital of Dijon (more than 15 new cases each week).
This work can be continued by a PhD position in the same project and thematic.
Contact: email@example.com and firstname.lastname@example.org
Closing date: 02th March 2020