Automated cardiac-tissue identification in composite strain-encoded (C-SECN) images using fuzzy K-means and bayesian classifier

Composite Strain Encoding (C-SENC) is an MRI acquisition technique for simultaneous acquisition of cardiac tissue viability and contractility images. It combines the use of black-blood delayed-enhancement imaging to identify the infracted (dead) tissue inside the heart wall muscle and the ability to image myocardial deformation (MI) from the strain-encoding (SENC) imaging technique. In this work, we propose an automatic image processing technique to identify the different heart tissues. This provides physicians with a better clinical decision-making tool in patients with myocardial infarction. The technique is based on using Bayesian classifier to identify the background regions in the C-SENC images, and fuzzy clustering technique to identify the different types of the heart tissues. The proposed method is tested using numerical simulations of the heart C-SENC images with MI and real images of patients. The results show that the proposed technique is able to identify the different components of the image with a high accuracy.