Metaheuristic models have proven very flexible in tackling optimization problems and, in particular, in image processing . The aim of this work is the analysis and comparison of possible metaheuristics that allow oneto evolve a generic segmentation algorithm for ultrasound images guided by the examples provided by a human expert. During this process the algorithm adapts itself to perform the automatic identification of the structures related to a specificclinical procedure during both diagnostic and therapeutic tasks-Metaheuristic models have proven very flexible in tackling optimization problems　and, in particular, in image processing . The aim of this work is the analysis and comparison of possible metaheuristics that allow oneto evolve a generic segmentation algorithm for ultrasound images guided by the examples provided by a human expert. During this process the algorithm adapts itself to perform the automatic identification of the structures related to a specificclinical procedure during both diagnostic and therapeutic tasks