ABDOMINAL IMAGE SEGMENTATION METHOD, COMPUTER DEVICE, AND STORAGE MEDIUM

摘要

An abdominal image segmentation method, a computer device, and a storage medium. The method comprises: acquiring an abdominal image to be tested; classifying pixels in said abdominal image on the basis of a trained full convolutional neural network to determine a segmented image corresponding to said abdominal image, wherein the trained full convolutional neural network is determined on basis of the training of a first training set and a second training set, the first training set comprises respective first sample abdominal images and pixel classification tag images corresponding to the first sample abdominal images, and the second training set comprises respective second sample abdominal images and the number of pixels of respective categories corresponding to the second sample abdominal images. By means of the present method, the accuracy of segmentation can be improved.