Liver cancer is one of the most common internalmalignancies worldwide and also one of the most leading deathcauses disease. Early detection and accurate staging of livercancer is considered an important issue in practical radiology. Inthis paper, a hybrid segmentation approach based on the modifiedWatershed algorithm and Neutrosophic logics is proposed forliver segmentation from abdominal CT images. The proposedapproach consists of three fundamental phases: (1) preprocessing,(2) CT image transformation to Neutrosophic domain and (3)post-processing phase. At preprocessing phase, histogram equal-ization and median filter are applied to enhance the contrastand intensity values of the liver CT image as well as removingthe noise. The enhanced CT liver image is transformed andrepresented in the Neutrosophic set domain via three membershipsets. Finally, at post-processing phase, mathematical morphologyand modified watershed algorithm are used to enhance theobtained truth image produced from the previous phase and toextract liver from CT image. Several measurements are used toevaluate the performance of the proposed segmentation approach.It obtains overall accuracy almost 95%. Moreover, it comparedwith other approaches and achieves better results.Index Terms—Watershed, Segmentation, Neutrosophic Set,Morphological Operators, CT, Parenchyma.
A hybrid segmentation approach based on Neutrosophic sets and modified watershed: A case of abdominal CT Liver parenchyma. Available from: https://www.researchgate.net/publication/296706420_A_hybrid_segmentation_approach_based_on_Neutrosophic_sets_and_modified_watershed_A_case_of_abdominal_CT_Liver_parenchyma [accessed Jun 2, 2016]. |
