Abstract—The main objective of medical image segmentation is to extract and characterize anatomical structures with respect to some input features or expert knowledge. This paper describes a way of medical image segmentation based on level set method to extract the region of interest of various medical images such as magnetic resonance imaging (MRI), computer tomography (CT), and positron emission tomography (PET) image. The overall technique of the paper is divided into three steps. The first step is to do threshold of the input image to make the entire pixel under threshold value to 0 and others to take the value as original image. This helps to keep original properties of original image and keep all value fixed as original image except the pixel that are not necessary for our analysis. Next, a morphological technique is used to remove some small ignorable parts. Finally we apply variational level set method for final segmentation. The experimental results show the efficacy of the proposed method.

Index Terms—Active contour, Medical image segmentation, thresholdng, Variational level set method.

The authors are with the Department of Computer Science and Engineering, Mawlana Bhashani Science and Technology University, Tangail, Bangladesh. (e-mail: m_nurnobi@yahoo.com, faruk_ce06003@ yahoo.com, faruk.mbstu@gmail.com, faysal_mbstu06@yahoo.com, usuf672@yahoo.com).

Cite: Mohammad N. Nobi, Faruk Chowdhury, Faysal Mahmood, and Mohammad A. Yousuf, "A New Medical Image Segmentation Technique Based on Variational Level Set Method," International Journal of Computer and Electrical Engineering vol. 3, no. 5, pp. 690-694, 2011.