Auto threshold region-growing method for edge detection of nuclear medicine images

Wei Gao; Yongjie Jin

doi:10.1117/12.440258

18 September 2001 Auto threshold region-growing method for edge detection of nuclear medicine images

Wei Gao, Yongjie Jin

Proceedings Volume 4549, Medical Image Acquisition and Processing; (2001) https://doi.org/10.1117/12.440258
Event: Multispectral Image Processing and Pattern Recognition, 2001, Wuhan, China

Abstract

In the analysis of nuclear medicine images, it is generally necessary to recognize organ's boundary and determine some special regions of interest (ROl). This paper will introduce a method of auto-edge detection of the ROl. Due to some spatial properties of nuclear medicine images, traditional method, such as using edge detection operators, sometimes can not give us the correct ROl. The region-growing method is frequently used in image segmentation. Though this method needs a large calculation, it can get exact edge of ROl by utilize several properties ofthe image directly. So it is adaptive to nuclear medicine image that has small size and a lot of local spatial properties. The procedure of region-growing method is: At first, the image is segmented into several cells. Secondly pick up one typical pixel, examine the value of the pixel with the clinical statistical criterion to decide whether the pixel is a part of the target organ. If it meets the criterion, merge it into the target organ. If not, reject this pixel. Then go on checking all neighbors of the pixel until all image have been examined. By means of the spatial relationship of the pixels, we can easily find the edge ofthe target organ. Using this method in edge detection of some clinical images such as kidney, heart, liver, we find that the result is better than some other methods.

Citation Download Citation

Wei Gao and Yongjie Jin "Auto threshold region-growing method for edge detection of nuclear medicine images", Proc. SPIE 4549, Medical Image Acquisition and Processing, (18 September 2001); https://doi.org/10.1117/12.440258

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available