Segmentation of uterine fibroid using morphology: An automatic approach

Image segmentation is an important task in medical image analysis. Automatic segmentation of ultrasound image is a difficult task as it suffers from speckle noise. This paper presents a fully automatic approach in which there is no need for the user to provide a seed point to segment the image. It proposes a new method for segmenting the fibroid in uterus. The method used in this paper uses concepts in mathematical morphology. It completely avoids over segmentation which is a major problem in morphological segmentation. The performance of this method is also commendable.