%0 Conference Proceedings
%T Skull Stripping for MRI Images Using Morphological Operators
%+ AGH University of Science and Technology [Krakow, PL] (AGH UST)
%A Swiebocka-Wiek, Joanna
%Z Part 3: Images, Visualization, Classification
%< avec comité de lecture
%( Lecture Notes in Computer Science
%B 15th IFIP International Conference on Computer Information Systems and Industrial Management (CISIM)
%C Vilnius, Lithuania
%Y Khalid Saeed
%Y Władysław Homenda
%I Springer International Publishing
%3 Computer Information Systems and Industrial Management
%V LNCS-9842
%P 172-182
%8 2016-09-14
%D 2016
%R 10.1007/978-3-319-45378-1_16
%K Skull stripping
%K Brain extraction
%K Morphological operators
%K Image segmentation
%K MRI
%Z Computer Science [cs]
%Z Humanities and Social Sciences/Library and information sciencesConference papers
%X One of the most common MRI (Magnetic Resonance Imaging) use is a brain visualisation. Brain anatomy is highly complicated therefore it might be difficult to extract only these structures which have diagnostic value. In a consequence it is so necessary to develop and apply most efficient brain’s segmentation algorithms. One of the first steps in case of neurological MRI analysis is skull stripping. It involves removing extra-meningeal tissue from the head image, therefore it is essential to find the best method to determine the brain and skull boundaries. In T1-weighted images, cerebrospinal fluid (CSF) space and skull are dark, that is why the edges between the brain and the skull are well-marked but even strong edges might be unsettled because of finite resolution during MRI acquisition or the presence of other anatomical partial structures within the brain (connections between the brain and optic nerves or brainstem). There are many ways to perform this operation, none of them is not so great as to constitute a standard proceedings. In many cases, there are limitations associated with the development environment, license and images input that hinder skull stripping without specialised software. Proposed method is free of these constraints. It is based on application of morphological operations and image filtration to enhance the result of the edge detection and to provide better tissues separation. The efficiency was compared with other methods, common in commercial use, and the results of this comparison was presented in this paper.
%G English
%Z TC 8
%2 https://inria.hal.science/hal-01637463/document
%2 https://inria.hal.science/hal-01637463/file/419526_1_En_16_Chapter.pdf
%L hal-01637463
%U https://inria.hal.science/hal-01637463
%~ SHS
%~ IFIP-LNCS
%~ IFIP
%~ IFIP-TC
%~ IFIP-TC8
%~ IFIP-CISIM
%~ IFIP-LNCS-9842