Computing synthetic echocardiography volumes for automatic validation of 3D segmentation results

B. Stender, B. Wang, A. Schlaefer

Abstract

Within the last years some segmentation algorithms working on 3D echocardiography data have been published. Usually such algorithms have been validated against contours manually drawn within the acquired imaging data. This praxis requires considerable expert knowledge and is time-consuming. We propose a different workflow based on synthetic ultrasound volumes. These volumes can be generated from computed tomography (CT) volumes or a surface models. The echocardiography simulation includes reflection, attenuation, speckle and interference between adjacent receiver elements. Based the synthetic echocardiography volumes we present a tool to validate segmentation algorithms. We illustrate the use of our tool in quantifying the segmentation result for a 3D region-based active contour approach.

Original languageEnglish
JournalBiomedizinische Technik
Volume57
Issue numberSUPPL. 1 TRACK-B
Pages (from-to)123-127
Number of pages5
ISSN0013-5585
DOIs
Publication statusPublished - 24.08.2012

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

  1. SDG 3 - Good Health and Well-being
    SDG 3 Good Health and Well-being
  2. SDG 9 - Industry, Innovation, and Infrastructure
    SDG 9 Industry, Innovation, and Infrastructure

Fingerprint

Dive into the research topics of 'Computing synthetic echocardiography volumes for automatic validation of 3D segmentation results'. Together they form a unique fingerprint.

Cite this