A CT image data processing method, device and storage medium

Abstract

The present invention provides a CT image data processing method, device, and storage medium. The method includes: each acquired CT image includes a group of layered images of human body parts, and selects a preset number of layered images from the layered images of the CT image in a position interval manner. Layer image, using the selected layer image as the training image of the CT image, and synthesizing the image; inputting the composite image into a plurality of pre-established network models for training; merging the trained network models to obtain the fusion network model; The fusion network model is trained; the target CT image is input into the trained fusion network model, and the pathological prediction result of the target CT image is output. Select multiple training images from different positions in each CT image, and synthesize them to obtain a composite image, train each basic network model through the composite image, improve the accuracy of network training, and integrate the trained network models into one, get Fusion of network models to improve robustness and prediction accuracy.

Description

technical field [0001] The present invention mainly relates to the technical field of image processing, in particular to a CT image data processing method, device and storage medium. Background technique [0002] The original format of the image data obtained by CT tomography is a DCM file. A DCM file is a file that follows the DICOM (DICOM: Digital Imaging and Communications in Medicine, medical digital image transmission protocol) standard, and is widely used in the medical industry. CT images are layered images, commonly used as cross-sections, and often contain multiple continuous layered images in order to display the entire organ. At present, the image processing network model is mainly used to analyze and process CT images, and the image is preprocessed before the analysis and processing. During the preprocessing process, image features are likely to be lost, and the output analysis data is not accurate enough. Contents of the invention [0003] The technical prob...

AI Technical Summary

Problems solved by technology

At present, the image processing network model is mainly used to analyze and process CT images, and the images are preprocessed before analysis and processing. During the preprocessing process, image features are likely to be lost, and the output analysis data is not accurate enough.

Images