Three-dimensional image blood vessel enhancement method based on deep learning network

The embodiment of the invention provides a three-dimensional image blood vessel enhancement method based on a deep learning network, and the method comprises the following steps: S1, collecting medical image three-dimensional image data, and carrying out the artificial blood vessel boundary delineation to obtain a corresponding blood vessel marking graph; S2, preprocessing and resampling the acquired three-dimensional image data and the blood vessel labeling graph thereof; S3, performing data enhancement operation on the three-dimensional data of all the candidate sensing areas; S4, training the generator deep learning network and the discriminator deep learning network; S5, collecting new medical three-dimensional image data, and performing preprocessing and resampling to obtain three-dimensional data of a plurality of candidate regions of interest; and S6, inputting the three-dimensional data of all the candidate regions of interest into a generator deep learning network to obtain ablood vessel enhancement