Towards An Automatic Segmentation for Assessment of Cardiac Left Ventricle Function

Research on detecting, recognising and interpreting Cardiac MRI has started since the 1980s. The problem with manual tracing efforts hampering the adoption of cardiac MRI as routine investigation. Manual tracing is also dependent on image quality, and there is no one-size-fitsall MRI setting for...

Full description

Bibliographic Details
Main Authors: D. N. F, Awang Iskandar, Amjad, Khan
Format: Article
Language:English
Published: Universiti Sains Malaysia 2016
Subjects:
Online Access:http://ir.unimas.my/id/eprint/12434/
http://ir.unimas.my/id/eprint/12434/1/Towards%20An%20Automatic%20Segmentation%20for%20Assessment%20of%20Cardiac%20Left%20Ventricle%20Function%20%28abstract%29.pdf
Description
Summary:Research on detecting, recognising and interpreting Cardiac MRI has started since the 1980s. The problem with manual tracing efforts hampering the adoption of cardiac MRI as routine investigation. Manual tracing is also dependent on image quality, and there is no one-size-fitsall MRI setting for the optimum image result. In this paper, we present a proposed approach to automatically detect the left ventricle (LV) wall in the effort to automatically assist the assessment of cardiac function. Using a standard bechmark dataset, our experiments have shown that our proposed method had effectively improve the automatic detection of the epicardium and endocardium.