Segmentation of mitral regurgitant jet using the combination of wavelet and watershed transformation

This paper concentrates on automatic segmentation of the regurgitation jet from the color Doppler echocardiographic image during the mitral regurgitation. Segmentation of color jet area is not an easy task especially when the background image is also color image. In this case it is not possible to do pixel classification for the separation of the color area. Jet area is directly related to the severity of mitral regurgitation. Here the main objective is to segment the color mosaic pattern from the color Doppler image. For this a multistep algorithm has been applied. The first stage of the proposed algorithm is dual-tree complex wavelet transform. Next stage is the watershed transform and then similarity measure has been done to get the final segmented image. Jet area obtained with the proposed method satisfies the clinical results.