Publication detail

X-ray Image Processing in Studying Jawbone Tissues

MIKULKA, J. KABRDA, M.

Original Title

X-ray Image Processing in Studying Jawbone Tissues

English Title

X-ray Image Processing in Studying Jawbone Tissues

Type

conference paper

Language

en

Original Abstract

Image processing in biomedical applications is strongly developing issue. Many methods and approaches for image preprocessing, segmentation and visualization were described. This paper describes X-ray image processing. The aim of processing is to segment regions of jawbone cysts and evaluate their local descriptors. It is necessary to choose suitable segmentation method because of adverse parameters of regions. The regions of the cysts are of low contrast and the pixel intensity distribution is not homogenous. The semiautomatic live-wire method was chosen. This approach to image segmentation is faster and more accurate than manual segmentation. It is very good compromise between simple manual edge tracing and automatic methods such as thresholding, watershed segmentation or other methods whose results must be post-processed. The second step of processing is to evaluate local descriptors of segmented regions which correspond to cysts. Several parameters were chosen to describe these regions: region area, mean gray value of intensities, modal gray value of intensities, standard deviation of intensities, minimal and maximal gray value of intensities, integrated intensity, median of intensities and shape descriptors of region (perimeter, circularity, aspect ratio, roundness and solidity). Values of these parameters will be used in following development of semiautomatic processing method with regard to current assessment of cysts by doctors.

English abstract

Image processing in biomedical applications is strongly developing issue. Many methods and approaches for image preprocessing, segmentation and visualization were described. This paper describes X-ray image processing. The aim of processing is to segment regions of jawbone cysts and evaluate their local descriptors. It is necessary to choose suitable segmentation method because of adverse parameters of regions. The regions of the cysts are of low contrast and the pixel intensity distribution is not homogenous. The semiautomatic live-wire method was chosen. This approach to image segmentation is faster and more accurate than manual segmentation. It is very good compromise between simple manual edge tracing and automatic methods such as thresholding, watershed segmentation or other methods whose results must be post-processed. The second step of processing is to evaluate local descriptors of segmented regions which correspond to cysts. Several parameters were chosen to describe these regions: region area, mean gray value of intensities, modal gray value of intensities, standard deviation of intensities, minimal and maximal gray value of intensities, integrated intensity, median of intensities and shape descriptors of region (perimeter, circularity, aspect ratio, roundness and solidity). Values of these parameters will be used in following development of semiautomatic processing method with regard to current assessment of cysts by doctors.

Keywords

X-ray, jawbone, cysts, necrosis, image segmentation, image classification

RIV year

2012

Released

19.08.2012

Location

Moscow

ISBN

978-1-934142-22-6

Book

PIERS 2012 in Moscow Proceedings

Pages from

145

Pages to

148

Pages count

4

BibTex


@inproceedings{BUT93421,
  author="Jan {Mikulka} and Miroslav {Kabrda}",
  title="X-ray Image Processing in Studying Jawbone Tissues",
  annote="Image processing in biomedical applications is strongly developing issue. Many methods and approaches for image preprocessing, segmentation and visualization were described. This paper describes X-ray image processing. The aim of processing is to segment regions of jawbone cysts and evaluate their local descriptors. It is necessary to choose suitable segmentation method because of adverse parameters of regions. The regions of the cysts are of low contrast and the pixel intensity distribution is not homogenous. The semiautomatic live-wire method was chosen. This approach to image segmentation is faster and more accurate than manual segmentation. It is very good compromise between simple manual edge tracing and automatic methods such as thresholding, watershed segmentation or other methods whose results must be post-processed. The second step of processing is to evaluate local descriptors of segmented regions which correspond to cysts. Several parameters were chosen to describe these regions: region area, mean gray value of intensities, modal gray value of intensities, standard deviation of intensities, minimal and maximal gray value of intensities, integrated intensity, median of intensities and shape descriptors of region (perimeter, circularity, aspect ratio, roundness and solidity). Values of these parameters will be used in following development of semiautomatic processing method with regard to current assessment of cysts by doctors.",
  booktitle="PIERS 2012 in Moscow Proceedings",
  chapter="93421",
  howpublished="electronic, physical medium",
  year="2012",
  month="august",
  pages="145--148",
  type="conference paper"
}