A study on Segmentation of renal scintigraphy using Level Sets algorithm

Abstract

Image Segmentation is one of the most vital step leading to the analysis of image data. It plays an important role in diagnosis of various disorders. This paper presents a novel approach to segment the image using Level Sets algorithm. Level sets based active contour technique uses Partial Differential Equations to evaluate the implicit curve equations. In order to achieve proposed Level Sets algorithm to renal data, an open source image processing tool named Fiji is used. The Level Sets method uses an Eikonal equation to segment the data. The tool has integrated program with an Eikonal equation in Level Sets algorithm to estimate the implicit curve equation for every iteration. An Eikonal equation can be defined as a nonlinear partial differential equation encountered in problems of wave propagation. These segmentation algorithm uses a standard rectangular implicit function which is later evolved into the desired segmentation mask as the algorithm proceeds. This paper presents a detailed view of the Level Sets model and its application on Renal Images. The Renal uptake images are obtained through nuclear medicine imaging techniques which involves injection of radiotracer into the patient intervenously and study the uptake of the desired organ. A radiotracer is a radioactive element that relatively has very low activity. This radiotracer when injected undergoes gamma decay inside the subject and emits gamma photons, which are obtained using special purpose camera called Gamma Camera. The advantage of using this Nuclear Medicine imaging is that it provides both the structural and functional information of the kidneys. This renal study helps in computing Glomerualr Filtration Rate, a parameter to determine the kidney functionality. Level Sets based algorithm was tested on 20 real time thyroid cases and the uptake values obtained are compared with those obtained from the existing software tool.