|
KMID : 1147120090150010039
|
|
Journal of the Korean Society of Imaging Informatics in Medicine
2009 Volume.15 No. 1 p.39 ~ p.44
|
|
|
A Study on the Detection of Rib Fractures in Chest Radiography Using Texture Analysis
|
|
Bae Jang-Pyo
Kim Kwang-Gi
Jeong Chang-Bu
Yeom Chol-Hee
Kim Young-Jae
Kim Sung-Jun
|
|
|
Abstract
|
|
|
|
This research proposed the method to detect the broken rib from chest radiography using texture analysis. The used features in texture analysis are kurtosis, invariant moment, homogeneity, entropy and max value of region of interest. 5 features were tested through 16 images which consists of 8 normal and 8 broken images. It was showed that 5 features could be the basises to distinguish the broken rib from the normal image. B-spline preprocessing is proposed and the reason that texture analysis is used instead of the approach through the segmentation of rib in chest radiographs is explained.
|
|
|
KEYWORD
|
|
|
rib fracture, detection, texture analysis, chest radiograph
|
|
|
FullTexts / Linksout information
|
|
|
|
|
|
Listed journal information
|
|
|
|