By Topic

TRUS image segmentation using morphological operators and DBSCAN clustering

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

2 Author(s)
Manavalan, R. ; Dept. of Comput. Sci. & Applic., K.S.R. Coll. of Arts & Sci., Tiruchengode, India ; Thangavel, K.

Ultrasound imaging is a widely used technology for prostate cancer diagnosis and prognosis among the different medical image modalities. The ultrasound images are very difficult to segment because of poor image contrast, speckle noise, and missing or diffuse boundaries in the transrectal ultrasound (TRUS). So the significant application is the segmentation of the prostate in transrectal ultrasound image. Generally there is no common approach for prostate image segmentation. The extraction of the prostate region from the original TRUS medical image is still a challenging research. This paper proposes a novel segmentation procedure for the TRUS medical image of prostate. It consists of four main stages. In the first stage, aM3-Filter is used to generate a despeckled image, since the speckle noise is commonly found in the ultrasound medical images. And the despeckled image is enhanced by top-hot filter. In the second stage, this enhanced image is used to compute thresholded image by local adaptive threshold method and Morphological operators are applied to extract an area containing the prostate (or large portions of it). In the third stage, The DBSCAN algorithm is applied to identify the core pixels, border pixels and noise pixels. The Clusters are formed by considering the density relations of the points. The clusters of core pixels and border pixels are used to automatically characterize the prostate region. The performance of the proposed algorithm is compared with manual segmentation using statistical parameters such as Rand Index (RI), Global Consistency Error (GCE), Variations of Information (VOI) and Boundary Displacement Error (BDE).

Published in:

Information and Communication Technologies (WICT), 2011 World Congress on

Date of Conference:

11-14 Dec. 2011