The invention discloses a sequential medical image quick segmentation method based on interactive dictionary migration, for mainly solving the problems of the prior art, such as low speed and large work load in segmenting single images. The sequential medical image quick segmentation method comprises inputting a CT (computed tomography) abdominal sequential image; primarily segmenting the sequential image by a mean shift method, and extracting 5*5 blocks based on each pixel as the center as the feature; subjecting one image to interactive segmentation to obtain a target and the background; respectively training the target and the background regions to produce a target and a background dictionary, approaching the non-segmented image in the sequence, marking all of the pixels to target or background according to the minimum error principle, and carrying out Gaussian smoothing; calculating the density of the target pixel points in each region in the primarily segmented image, and removing the region with small density to obtain the segmentation result; subjecting the image that is not segmented correctly to interactive segmentation again; and updating the target dictionary by dictionary migration. The method provided by the invention has the advantages of good effect and high speed in segmenting the medical images, and can be used for segmenting medical CT sequential images.