International Journal of Innovation and Applied Studies
ISSN 2028-9324 Vol. 3 No. 2 June 2013, pp. 560-569
© 2013 Innovative Space of Scientific Research Journals
http://www.issr-journals.org/ijias/
Corresponding Author: Kalpana Saini (kal_2312@rediffmail.com) 560
Level set based on new Signed Pressure Force Function for Echocardiographic image
segmentation
Kalpana Saini
1
, M.L. Dewal
1
, and Manojkumar Rohit
2
1
Deptt. of Electrical Engg., IIT Roorkee,
India
2
Deptt. of Cardiology, PGIMER Chandigarh,
India
Copyright © 2013 ISSR Journals. This is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
ABSTRACT: In the present paper a novel region based active contour method is developed by formulating a new signed
pressure force (SPF) function. The method has been applied to the echocardiographic images for getting the desired
boundary. The method is useful for finding the automatic boundary detection of other images (Microbiological, MRI, CT,
Natural and welding joint etc.) as well. Level set method in combination with original SPF has not been able to give
satisfactory results during the segmentation of echocardiographic images. There are lots of noises present in the
echocardiographic images those create difficulties in the segmentation process. The proposed method resolves all these
difficulties in such a manner that the output image is having the proper boundary detection without any disturbances and
noises. The very important advantage of this method is that it gives a very fast response in terms of time taken by CPU and
the number of iterations. Fast response is very important in the clinical area especially in diagnosis purpose. The presented
model is an advancement of Selective Binary and Gaussian Filtering Regularized Level Set (SBGFRLS) method. Proposed
model is more robust against images with weak edge and intensity inhomogeneity when compared with the performance of
earlier methods.
KEYWORDS: Echocardiographic, Active contour, Signed pressure force function, Level set, Segmentation, Gaussian filtering,
Regularized level Set.
1 INTRODUCTION
Segmentation is a well-known step which is carried out for the clinical diagnosis from medical images. But the major
problem encountered during the segmentation of echcardiographic images. The literature shows that the active contour
based segmentation techniques are being extensively used in medical imaging [1-5]. Active contour models can be
categorized as edge based active contour models [1-4, 6, 9, 10, 11] and region based active contour models [5, 7, 8, 12-14].
Some of the edge based active contour models are used as the edge-detector. The operation of edge detector depends
on the gradient of the image to stop the initial contour on the boundary of the interested objects. This technique has
advantage when the objects and background of segmented image are heterogeneous. Drawbacks of these active contour
models are that the satisfactory results cannot be achieved in case of objects with discrete or with the presence of blur
boundaries or noise. Some active contour models as in [6] introduce the balloon force to shrink and enlarge the capture
range of the force. However some undesired effects occur during balloon method. If weak balloon force is there then contour
is not able to pass through the narrow part of the object, and if the balloon force is large, the contour will pass through the
weak edges of the object. The geodesic active contour (GAC) [4] is the most popular methods in this category. This method
uses an edge stopping function (ESF) which stops the contour on the object boundaries.