Scar Quantification from Cardiac MRI

Project Code: 10337298

Faculty: Faculty of Medical and Health Sciences

Department: Anatomy

Main Supervisor: Professor Alistair Young

Co supervisors: Avan Suinesiaputra

Application open date: 15 Oct 2015

Application deadline:

Enrolment information: NZ Citizens, NZ Permanent Residents, International

Introduction

Heart attack is a common cause of death. In this disease, parts of the heart become scarred, which hinders activation of the heartbeat. This project seeks to identify and quantify regions of scar in MRI images of the heart. Although scar tissues appear bright in the images, variations in the contrast due to tissue composition and imaging artefacts are still causing the difficulty to quantify scar tissue accurately. Scar quantification from MRI plays an important role in the diagnosis of cardiac disease, which includes assessment of myocardial viability, predicting improvement in cardiac resynchronisation therapy, and management of patients with tachycardia.

What we are looking for in a successful applicant

We want a student with a strong interest in developing a computer-assisted diagnostic tool, which can accurately segment and quantify scar tissue from MRI examinations.  Ideally, this would be a computer science, electrical engineering or bioengineering student.

Objective
  • Develop an algorithm to delineate scar tissues either semi or fully automatically.
  • Develop an algorithm to quantify the degree of scar tissue
  • Test and validate the algorithm.
  • Optionally, to do benchmarking with an open-access challenge data available from (R. Karim et al., JCMR 2013).
  • Prepare a report.

 

Other information
Related opportunities