A survey of prostate segmentation methodologies in ultrasound, magnetic resonance and computed tomography images

Abstract

Prostate segmentation is a challenging task, and the challenges significantly differ from one imaging modality to another. Low contrast, speckle, micro-calcifications and imaging artifacts like shadow poses serious challenges to accurate prostate segmentation in transrectal ultrasound (TRUS) images. However in magnetic resonance (MR) images, superior soft tissue contrast highlights large variability in shape, size and texture information inside the prostate. In contrast poor soft tissue contrast between prostate and surrounding tissues in computed tomography (CT) images pose a challenge in accurate prostate segmentation. This article reviews the methods developed for prostate gland segmentation TRUS, MR and CT images, the three primary imaging modalities that aids prostate cancer diagnosis and treatment. The objective of this work is to study the key similarities and differences among the different methods, highlighting their strengths and weaknesses in order to assist in the choice of an appropriate segmentation methodology. We define a new taxonomy for prostate segmentation strategies that allows first to group the algorithms and then to point out the main advantages and drawbacks of each strategy. We provide a comprehensive description of the existing methods in all TRUS, MR and CT modalities, highlighting their key-points and features. Finally, a discussion on choosing the most appropriate segmentation strategy for a given imaging modality is provided. A quantitative comparison of the results as reported in literature is also presented.

Multimodal image-guided prostate fusion biopsy based on automatic deformable registration

“…In this context, Ghose et al [8] review the extensive research on semi-automatic and automatic segmentation of the prostate from TRUS and MRI images. The main approaches in delineating prostate boundaries are contour-, shape-or region-based and can be distinguished in supervised or unsupervised classification, as well as combinations of these.…”

Clinical value of prostate segmentation and volume determination on MRI in benign prostatic hyperplasia

“…Contourand shape-based models use prostate boundary information for segmentation, whereas region based models use local intensity or statistics for segmentation and supervised and unsupervised models use features such as signal intensity on images or additional features, like filters, to separately classify the prostate and background regions (34). Technical details of these methods have been described extensively elsewhere and are beyond the scope of this review (34).…”

“…Fully automated prostate segmentation algorithms have been developed recently and they can be summarized as contour shape-based, region-based, and supervised and unsupervised classification techniques (34). Contourand shape-based models use prostate boundary information for segmentation, whereas region based models use local intensity or statistics for segmentation and supervised and unsupervised models use features such as signal intensity on images or additional features, like filters, to separately classify the prostate and background regions (34).…”

Topological Prostate Segmentation Method in MRI

“…Prostate segmentation methods based on images acquired using ultrasound, magnetic resonance and computed tomography could be generally divided into four major categories: contour and shape based methods, region based methods, supervised and un-supervised classification methods, hybrid methods [2].…”

“…In order to provide accurate information various imaging techniques are used in the clinical practice. Nowadays, trans rectal ultrasound (TRUS) is probably the most common and widespread medical imaging technique employed for cancer detection [2], [3], [4] as well as for guided needle biopsy [5]. This is mainly due to its low cost, portability and real-time acquisition.…”