Nasrul Humaimi Mahmood, Noraishikin Zulkarnain, Nor Saradatul Akmar Zulkifli / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 www.ijera.com Vol. 2, Issue 3, May-Jun 2012, pp.691-694 691 | P a g e Ultrasound Liver Image Enhancement Using Watershed Segmentation Method Nasrul Humaimi Mahmood*, Noraishikin Zulkarnain** and Nor Saradatul Akmar Zulkifli** *Faculty of Health Science and Biomedical Engineering **UTM Razak School of Engineering and Advanced Technology Universiti Teknologi Malaysia, MALAYSIA ABSTRACT This paper proposed the ultrasound liver image enhancement based on watershed segmentation method. Image segmentation is an important problem in medical image processing fields. The focus of this study is to enhance the region of liver based on watershed algorithm of segmentation and visualization technique. The MATLAB is used as a tool for this study. The watershed segmentation entirely relay presented the good result base on the contrast of the image. In this study, an ultrasound image is transformed into a binary image using the threshold method, which means that the color of the output image appears only black and white. After the image is converted into binary, the image is modified using Watershed technique together with the visualization process. The result is really helpful in medical diagnostics. Keywords - Ultrasound liver image, image processing, watershed algorithm, image segmentation and visualization. I. INTRODUCTION Nowadays, ultrasound imaging is an important diagnosis method in a medical analysis. It is important to segment out the cavities, different types of tissues and organs in the ultrasound image for effective and correct diagnosis. In the medical field, the human experts are very good in segmenting out the required region of the medical image. But humans lack efficiency when size of data set increases. The need of high reproducibility and need of increasing efficiency motivates the development of computer-assisted and automated segmentation. These automated procedures segment out different regions in medical images by applying different types of image segmentation methods. The main disadvantage of ultrasound images is the poor quality of images, which are also affected by speckle noise. Therefore, in general, many of the image segmentation methods may not be suitable in case of ultrasound images [2]. Image segmentation is a process to partition an image into non-overlap regions, which is an important step in the image processing area and is fundamental to the analysis and identification in image processing. Image segmentation is an important process for most of the medical image analysis tasks, which is basic for higher-level image comprehension and analysis. A good segmentation will benefit clinicians and patients as it provides important information for surgical planning, early disease detection and 3D visualization [1, 4]. In order to solve the problems of medical image segmentation, many practical methods have been advanced in this field. These include watershed segmentation, thresholding method, region-growing method, fuzzy cluster method and so on. The watershed algorithm is a classical and an effective segmentation method by which one-pixel- wide continuous edge can be extracted. The most effective methods in complex segmentation problems are watershed segmentation. The segmented region is obtained when the algorithm uses watershed transform applied to the image. However, segmentation of noisy ultrasound image using watershed transform always leads to over-segmentation [3]. There are many applications whether on synthesis of the objects or computer graphic images require precise segmentation. In general, image noise should be eliminated through image pre-processing [5, 6]. The overall objective of this study is to understand the concept of image processing in medical image analysis. A liver image is scanned by using a medical ultrasound. By using this image, this study involves improving the contrast of image by using histogram equalization techniques, converting the gray scale image to binary image by using thresholding, segmenting the image by using watershed algorithm and transforming the liver region to the color image in visualization technique. II. LITERATURE REVIEW An effective and correct diagnosis of ultrasound image is very important to avoid faulty in segmenting out cavities, tissues and organs which can lead to other problems in treating the patient. Thus, automated segmentation is a must to help clinicians and doctors make the diagnosis as the ultrasound images come out with poor quality of images due to the relatively low resolution and reduced contrast of the images. Some segmentation methods have been proven to be effective in handling the mentioned problem, such as Active Contour Model, Fuzzy C Means (FCM) and Graph-based method [3]. Fei Mao et. al. [5], describe a segmentation algorithm based on a discrete dynamic model approach with only one seed point to guide the initialization of the deformable model for each lumen cross-section. The initial contour of the deformable model is generated by using the entropy map of the original image and mathematical morphology operations with one seed. Meanwhile, the definition of the deformable model is the driven to fit the lumen contour by an internal force and an external force that are calculated respectively