Liver segmentation using marker controlled watershed transform

<span lang="EN-US">The largest organ in the body is liver and primarily helps metabolism detoxification. Liver segmentation a crucial step cancer detection computer vision-based biomedical image analysis. critical task results under-segmentation over-segmentation due to complex structure of abdominal computed tomography (CT) images, noise, textural variations over image. This paper presents CT images using marker-based watershed transforms. In pre-processing stage, modified double stage gaussian filter (MDSGF) used enhance contrast, preserve edge texture information images. Further, marker controlled transform utilized for from suggested MDSGF help diminish object. The performance proposed system evaluated on LiTS dataset based Dice score (DS), relative volume difference (RVD), volumetric overlapping error (VOE), Jaccard index (JI). method gives (Dice 0.959, RVD 0.09, VOE 0.089, JI 0.921).</span>