16 March 2006 Toward real-time endocardial boundary detection from intracardiac echocardiographic images: a comparative study
Author Affiliations +
Abstract
In this paper a novel method is proposed for live endocardial boundary identification. The goal is to achieve an optimal solution to the problem of real-time automatic detection and tracking of endocardial border in ultrasonic image sequences acquired through Intracardiac Echocardiography (ICE). Border identification of 2D ultrasonic images, which normally consists of a number of stages namely preprocessing, segmentation, detection and visualization of the border, is a cumbersome task. ICE's potential in guidance of minimally invasive interventions requires online boundary detection of its inherently less speckled images. Numerous studies have addressed this issue in echographic images by proposing various methods applicable at each stage. With this repository of methods available, a comparative study is performed on single-image segmentation approaches. An algorithm based on order-statistics operators is proposed to achieve fast border delineation in a sequence of images. This method can outperform other approaches in terms of time and robustness, and does not require user interaction.
© (2006) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Yusof Ganji, Farrokh Janabi-Sharifi, Safieddin Safavi-Naeini, "Toward real-time endocardial boundary detection from intracardiac echocardiographic images: a comparative study", Proc. SPIE 6147, Medical Imaging 2006: Ultrasonic Imaging and Signal Processing, 61470P (16 March 2006); doi: 10.1117/12.653329; https://doi.org/10.1117/12.653329
PROCEEDINGS
10 PAGES


SHARE
Back to Top