Abstract—Internal structure of the human body obtained by different modalities such as Computed Tomography (CT), Magnetic Resonance Imaging (MRI), etc. Images from these modalities are generally used for the diagnosis purpose. In the Brain MRI image, the tumor part can be seen clearly but the physician need the accurate size and the correct measurement for the treatment. This requires to segment the tumor part from the given Brain MRI image and quantify it. We have used Brain tumor (Metastatic bronchogenic carcinoma) MRI scans of 42-year-old woman from the Whole Brain Atlas (WBA) for the experiment; in this paper, we presented the basic level set method to catch the boundaries of the tumor area in 2D MRI images. After catching the accurate boundaries, we applied the inverse threshold and binary mask to segment the exact tumor region from the given different MRI brain scans. At last, we applied some of the statistical function on the region of interest (ROI) to quantify the tumor region.

Index Terms—CT, MRI, WBA, level set, ROI.

Nitish Zulpe is with the College of Computer Science and Information Technology, Latur, Maharashtra (e-mail: nitish.zulpe@gmail.com)
Vrushsen Pawar is with department of computational science, SRTM University Nanded, Maharashtra (e-mail: vrushvijay@yahoo.co.in)

Cite: N. S. Zulpe and V. P. Pawar, "Level set and Thresholding for Brain Tumor Segmentation," International Journal of Computer and Electrical Engineering vol. 4, no. 1, pp. 82-86, 2012.