Meeting Banner
Abstract #1441

Spectrum-Based Automatic Localization of Left Ventricle from Cardiac Magnetic Resonance Imaging

Liang Zhong1, Min Wan1, Jun Mei Zhang1, Ru San Tan1

1National Heart Centre Singapore, Singapore, Singapore

A novel method for automatic localization of left ventricle (LV) via spectrum analysis in magnetic resonance imaging is proposed. Two temporal approaches based on the variance image and the first harmonic image are applied and reformulated in a unified framework to reveal their intrinsic connections. The first harmonic image and some higher harmonic image are used in an anisotropic weighted circle Hough transform. The first harmonic image votes with positive weights while the higher harmonic image votes with negative weights to suppress the right ventricular interruption. The results in 17 patients show the capability for automatic localization of LV.

Keywords

accuracy agency anisotropic assisted assumption auto automatic boundary canny cardiac cine circle combination component components computation computational computer computing concerning conducted contains council critically cropped cycle decrease defined dependent detection discrete earlier edge engineering existing expanding experimented fact facts females fifth frames frequency funding gating generated gradient grant grants harmonic heart hough identify improve inspected intensity interruption intervention jolly largely leading length load localization localizing males maps medical modified national negative novel occupying outperforms partially pattern phys pixel plot portion position positive precise procedure proof proposed public reason recently recognition reduction scanner science segmentation separated series slice space spacing spectrum star still subjects superiority support suppressed surface technology theory thought transactions transform typically useful utilized ventricle visually votes whole years