Foetus Ultrasound Medical Image Segmentation via Variational Level Set Algorithm M.Y. Choong M.C. Seng S.S. Yang A. Kiring K.T.K. Teo Modelling, Simulation and Computing Laboratory, School of Engineering and Information Technology, Universiti Malaysia Sabah, Kota Kinabalu, Malaysia. msclab@ums.edu.my ktkteo@ieee.org Abstract—There is a challenge to segment the medical image which is often blurred and consists of noise. The objects to be segmented are always changing shape. Thus, there is a need to apply a method to automated segment well the objects for future analysis without any assumptions about the object’s topology are made. In general, when performing pregnancy ultrasound scanning, obstetrician needs to find out the best position or angle of the foetus and freeze the scene. The obstetrician will click on the crown and the rump of the foetus to get the foetus length. The segmentation technique applied is level set method. A variational level set algorithm has been successfully implemented in medical image segmentation (X- ray image, MRI image and ultrasound image). The results showed the level set contour evolved well on the low contrast and noise consisting medical image, especially the ultrasound image. Keywords-image segmentation; level set algorithms; foetus ultrasound medical image I. INTRODUCTION Image segmentation plays an important role in image analysis, appearing in many applications including pattern recognition, object detection, and medical imaging. Image segmentation means to partition an image into meaningful region with respect to a particular application and corresponding to individual surfaces, objects, or natural parts of objects. In general, image segmentation is the first stage in image analysis which seeks to simplify the data into its basic component elements or objects within the scene. The purpose of image segmentation is to simplify the representation of an image into something that is more meaningful and easier to analyze. There are some applications of image segmentation which include identify an object in a screen for object based measurements such as size and shape. For example, by segment the body of the foetus from the ultrasound image, future calculation of the foetus weight can be obtained through the region of segmentation. This can help the obstetrician to check whether the foetus is growing healthily. Image segmentation can be generally categorized into two categories which are parametric and non-parametric image segmentation. Parametric approaches are more generative. Non-parametric method does not require the image regions to have a particular type of probability distribution and does not require the extraction and use of a particular statistic. The image segmentation method that used in the project is a parametric method. This paper is organized as follows. In section II, image segmentation methods described. Explanation of the level set algorithm is presented in section III. In section IV, the experimental results on foetus ultrasound medical image segmentation and discussions are provided. Lastly, this paper is concluded in section V. II. IMAGE SEGMENTATION METHODS Some previous approaches to image segmentation, which provide the basis for a variety of more recent methods, include boundary-based segmentation such as Canny edge detection [1], region-based segmentation such as region growing and global optimization approaches. Andreetto propose a simple probabilistic generative model for image segmentation. The model is principled, provides both hard and probabilistic cluster assignments, as well as the ability to naturally incorporate prior knowledge [2]. In many vision problems, the performance of the segmentation step is highly dependent on the algorithm selection and its parametrization. These tasks are tricky and time-consumming. Martin and Thonnat present an approach to perform task-oriented segmentation based on segmentation algorithm parameter tuning and learning techniques [3]. W.Tao, H.Jin and Y.Zhang developed a novel approach that provides effective and robust segmentation of colour images [4]. Colour image segmentation methods can be seen as an extension of the gray image segmentation method in the colour images, however, many of the original gray image segmentation methods cannot be directly implemented to colour images. Jun Tang proposes a colour image segmentation method of automatic seed region growing on basis of the region with the combination of the watershed algorithm with seed region growing algorithm which based on the traditional seed region growing algorithm [5]. Michael and Anil propose a method of segmentation by using the object classification subsystem as an integral part of the segmentation. This will provide contextual information regarding the objects to be segmented, and the probability of correct classification as a metric to determine the quality of the segmentation [6]. Clustering is a vital element of model identification field means distinguishing and classifying things that are provided with similar properties. Based on image segmentation and model identification technologies and considering application characteristics of clustering method into image segmentation, Z.Wang and M.Yang 2012 Third International Conference on Intelligent Systems Modelling and Simulation 978-0-7695-4668-1/12 $26.00 © 2012 IEEE DOI 10.1109/ISMS.2012.102 225