CT Liver Segmentation Using Artificial Bee Colony Optimisation

• DOI: 10.1016/j.procs.2015.08.272
• Publication Date: 2015-08-31T22:34:04Z
• Authors: Abdalla Mostafa; Ahmed Fouad; Mohamed Abd Elfattah; Aboul Ella Hassanien; Hesham Hefny; Shao Ying Zhu; Gerald Schaefer
• Keywords: Centroid; Similarity (geometry); Region growing

The automated segmentation of the liver area is an essential phase in diagnosis from medical images. In this paper, we propose artificial bee colony (ABC) optimisation algorithm that used as a clustering technique to segment CT our algorithm, ABC calculates centroids clusters image together with region corresponding each cluster. Using mathematical morphological operations, then remove small and thin regions, which may represents flesh regions around area, sharp edges organs or lesions inside liver. extracted are integrated give initial estimate area. final step, further enhanced using growing approach. experiments, employed set 38 images, taken pre-contrast phase, similarity index calculated judge performance proposed This experimental evaluation confirmed approach afford very good accuracy 93.73% on test dataset.