Segmentation of Lung Tumor in Multimodality PET-CT Image using a New Modified Variational Level Set Method

  Guruprasad S,  Dr M Z Kurian,  Dr H N Suma

Medical image segmentation is a vital process in medical diagnosis and evaluation of tumor response to therapy. Current segmentation methods works only on single modality image like Positron Emission Tomography (PET) has low resolution and gives only functional information; Computed Tomography (CT) has low contrast and provides structural information. This paper focus on segmentation of multimodality PET-CT image. In recent days PET-CT is advanced multimodal imaging equipment, which gives both functional and ana-tomical information in a single image. This multimodal image is very essential in detection of cancer, such as rectal cancer, lung cancer, head and neck cancer so on. The manual delineation of gross tumor volume (GTV) is done by radiologist on CT with the help PET image is inaccurate hence this paper presents a method to segment and identify the other information about size, shape and location of the tumor. Basic deformation models for image segmentation such as snakes, gradient vector flow, active contour and level sets have some limitations that is convergence rate is slow hence these methods are not more suitable for multimodal image segmentation. Modified Variational level set method is introduced in this paper to increase the convergence rate and computational efficiency. Both qualitative and quantitave results shows that efficiency of the algorithm compared to existing methods. This method is robust in delineation of tumor in therapy treatment.

Keywords- CT, Deformation model, GTV, Level set, PET

Unique Identification Number - IJEDR1901011
Page Number(s) - 52-59
Pubished in - Volume 7 | Issue 1 | January 2019
DOI (Digital Object Identifier) -   
Publisher - IJEDR (ISSN - 2321-9939)

  Guruprasad S,  Dr M Z Kurian,  Dr H N Suma,   "Segmentation of Lung Tumor in Multimodality PET-CT Image using a New Modified Variational Level Set Method", International Journal of Engineering Development and Research (IJEDR), ISSN:2321-9939, Volume.7, Issue 1, pp.52-59, January 2019, Available at :http://www.ijedr.org/papers/IJEDR1901011.pdf