ABDOMINAL ORGAN STATUS DIAGNOSIS BASED ON ABNOMINAL ORGAN VOLUME CHANGE ANALYSIS USING ABNOMINAL ORGAN AUTOMATIC SEGMENTATION

摘要

A method for judging the state of abdominal organs based on an analysis of changes in abdominal organ volume using automatic abdominal organ segmentation is provided. The abdominal organ status determination method based on the abdominal organ volume change analysis using the automatic segmentation of the abdominal organs includes a plurality of axial images, a plurality of coronal images, and a plurality of sagittal images of medical images including abdominal organs. ) by inputting the image into a two-dimensional deep neural network. generating a plurality of prediction maps into which regions of interest are divided through the two-dimensional deep neural network; generating two-dimensional segmentation result data in which the region of interest is segmented by using the plurality of generated prediction maps using majority voting; generating a 3D image by cropping the ROI from the 2D segmentation result data and resizing the cropped 2D segmentation result data; and inputting the three-dimensional image into a three-dimensional deep neural network. It may include; obtaining 3D segmentation result data in which the region of interest is segmented through the 3D deep neural network.