No Cover Image

Conference Paper/Proceeding/Abstract 385 views

Segmentation and Tracking of Coronary Arterial Wall / Xianghua, Xie

Computational and Mathematical Biomedical Engineering

Swansea University Author: Xianghua, Xie

Full text not available from this repository: check for access using links below.

Abstract

In this paper, we present a segmentation and tracking method based on hidden Markov model (HMM) to detect the outer coronary arterial wall in intravascular images. The proposed method tracks a set of hidden states representing the border location on a set of normal lines obtained from the previ- ous...

Full description

Published in: Computational and Mathematical Biomedical Engineering
ISSN: 2227-3085 2227-9385
Published: 2017
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa32107
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract: In this paper, we present a segmentation and tracking method based on hidden Markov model (HMM) to detect the outer coronary arterial wall in intravascular images. The proposed method tracks a set of hidden states representing the border location on a set of normal lines obtained from the previ- ous frame. The border observation is derived from a classification-based cost function and a shape prior model. The emission probability is defined based on two Gaussian probability distributions for the vessel border and background. The transition probability is learned by using the Baum-Welch algorithm. The optimal sequence of the hidden states is obtained by using Viterbi algorithm. The proposed method shows promising results on tracking and segmenting the arterial wall.
Item Description: The paper is available freely online as part of CMBE 2017 Proceedings Vol. 1: http://www.compbiomed.net/2017/cmbe-proceedings.htm
Keywords: Image analysis, machine learning
College: College of Science