Three-dimensional image arteriovenous segmentation method and system based on deep learning network

The embodiment of the invention provides a three-dimensional image arteriovenous segmentation method and system based on a deep learning network, and the method comprises the following steps: S1, collecting three-dimensional image data, and carrying out the preprocessing; s2, resampling the preprocessed three-dimensional image data, and generating three-dimensional data of a plurality of candidateregions of interest; s3, inputting the three-dimensional data of all candidate regions of interest into a blood vessel extraction deep learning network to obtain a blood vessel segmentation map in each region of interest; and S4, inputting the blood vessel segmentation map into an arteriovenous recognition deep learning network to obtain an arteriovenous segmentation map of each region of interest, and splicing and combining all the segmentation maps together to obtain a complete arteriovenous segmentation result. According to the three-dimensional image arteriovenous segmentation method andsystem, the arteriovenous