Multi-Resolution Level Sets with Shape Priors: A Validation Report for 2D Segmentation of Prostate Gland in T2W MR Images

Abstract

The level set approach to segmentation of medical images has received considerable attention in recent years. Evolving an initial contour to converge to anatomical boundaries of an organ or tumor is a very appealing method, especially when it is based on a well-defined mathematical foundation. However, one drawback of such evolving method is its high computation time. It is desirable to design and implement algorithms that are not only accurate and robust but also fast in execution. Bresson et al. have proposed a variational model using both boundary and region information as well as shape priors. The latter can be a significant factor in medical image analysis. In this work, we combine the variational model of level set with a multi-resolution approach to accelerate the processing. The question is whether a multi-resolution context can make the segmentation faster without affecting the accuracy. As well, we investigate the question whether a premature convergence, which happens in a much shorter time, would reduce accuracy. We examine multiple semiautomated configurations to segment the prostate gland in T2W MR images. Comprehensive experimentation is conducted using a data set of a 100 patients (1,235 images) to verify the effectiveness of the multi-resolution level set with shape priors. The results show that the convergence speed can be increased by a factor of ≈ 2.5 without affecting the segmentation accuracy. Furthermore, a premature convergence approach drastically increases the segmentation speed by a factor of ≈17.9.

Segmentação Automática da Próstata em Imagens de Ressonância Magnética utilizando Redes Neurais Convolucionais e Mapa Probabilístico

“…No trabalho de [Al-Qunaieer et al 2014]é proposto a utilização de Level Sets em multiresolução. Esta estratégia de multiresoluçãoé aplicada para acelerar as Level Sets.…”

“…São apresentadas a base utilizada e as métricas usadas para validação. Este trabalho apresenta um melhor resultado em relação ao trabalho [Li et al 2013] que utiliza a mesma base, mostrando resultado superior aos trabalhos de [Maan and van der Heijden 2012], [Al-Qunaieer et al 2014] e [Tian et al 2018] que utilizam outras bases. Nosso método apresenta um resultado inferior ao trabalho de [Yang et al 2014], que por sua vez utiliza uma base privada com um método semiautomático, que depende de uma segmentação inicial feita por especialista.…”

Sequential Registration-Based Segmentation of the Prostate Gland in MR Image Volumes

“…The image dataset in DICOM format and manual markings were provided by Segasist Technologies. The same dataset has been a part of multiple algorithm and software validation processes as reported in the literature [21,22].…”