• JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
 
  Bookmark and Share
 
 
Master's Dissertation
DOI
https://doi.org/10.11606/D.3.2016.tde-29082016-145113
Document
Author
Full name
Veronica Meyer Gaiarsa
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
São Paulo, 2016
Supervisor
Committee
Furuie, Sérgio Shiguemi (President)
Lemos Neto, Pedro Alves
Rittner, Letícia
Title in Portuguese
Abordagens para segmentação de imagens de tomografia intravascular por coerência ótica.
Keywords in Portuguese
Bioengenharia
Contenedores
FD-IOCT
Neoíntima
Processamento de imagens
Tomografia de coerência óptica
Abstract in Portuguese
Devido à sua alta taxa de mortalidade as doenças cardiovasculares vem sendo foco de pesquisas nos últimos anos. Estima-se que neste grupo, 42% das mortes foram consequência de doença coronariana (CHD) em 2012. Para auxiliar o acompanhamento de pacientes com CHD, este estudo teve como objetivo investigar um método simples e robusto que segmenta de maneira semi-automática a área da neoíntima em Tomografias Intravasculares por Coerência Óptica no Domínio da Frequência (FD-IOCT), a mais recente técnica de imagear vasos internamente. O método foi dividido em duas etapas. A primeira segmenta a área contida pelo lúmen através de operações morfológicas com valores de intensidade e aplicação da limiarização Otsu. Na segunda, o foco foi segmentar a área contida por partes de stent onde duas estratégias foram desenvolvidas e comparadas. Ambas as etapas (segmentação de área de lúmen e de stent) obtiveram resultados acurados com aproximadamente 98% de Verdadeiro Positivo, enquanto o Falso Positivo foi próximo de 3% para o lúmen e 5% para stent, onde uma das estratégias (para delimitar a área contida por stent) apresentou um tempo de execução 50 vezes maior que a outra. O método foi utilizado em 443 imagens com diferentes características e os resultados são encorajadores.
Title in English
Approaches to segmentation of intravascular optical coherence tomography.
Keywords in English
FD-IOCT
Neointimal
Stent strut
Abstract in English
Due to its high mortality rate cardiovascular diseases have been the focus of research in recent years. It is estimated that among these, 42% of deaths were due to coronary heart disease (CHD) in 2012. To monitor patients with CHD, the goal of this study was to investigate a simple and robust method that segments semi-automatically the neointimal area on Frequency Domain Intravascular Optical Coherence Tomography (FD-IOCT), the latest technology to view vessels internally. The method was divided into two steps. The first one segments the area contained by the lumen through morphological operations on intensity values and the Otsu threshold. In the second one, the focus was to segment the area contained by stent struts where two strategies were developed and compared. Both steps (lumen and stent strut area segmentation) obtained accurate results with true positives approximately 98%, while the false positives were close to 3% for the lumen and 5% for the stent, where one strategy (to delimit the area contained by stent struts) showed run time of 50 times the other. The method was executed on 443 images with different characteristics and the results are encouraging.
 
WARNING - Viewing this document is conditioned on your acceptance of the following terms of use:
This document is only for private use for research and teaching activities. Reproduction for commercial use is forbidden. This rights cover the whole data about this document as well as its contents. Any uses or copies of this document in whole or in part must include the author's name.
Publishing Date
2016-08-30
 
WARNING: Learn what derived works are clicking here.
All rights of the thesis/dissertation are from the authors
CeTI-SC/STI
Digital Library of Theses and Dissertations of USP. Copyright © 2001-2024. All rights reserved.