Segmentation of MRI Prostate Images

In this work, we investigate the performance of two segmentation methods; level set,
and texture-based, in segmentation of prostate region. Both segmentation methods are
applied onto transverse view of T2-W-MRI slice of prostate acquired using a 3T
scanner. Level set method is one of the popular partial differential equations (PDEs)
based in image processing especially in image segmentation as it relies on an initial
value PDEs for a propagating level set function. “It also has been introduced in many
disciplines, such as computer graphics, computational geometry, and optimization
because this method acts as a tool for numerical analysis of surfaces and shapes.
Besides, level set method can perform numerical computations involving curves and
surfaces on a fixed Cartesian grid without having to parameterize the object. Prostate
gland in MRI images is categorized as a texture image because the structures are not
homogeneous and its surface has grey level values close to the neighbouring organs
around the prostate which making it more difficult to detect the damaged tissues.