Processing method and system for reconstructing blood vessel three-dimensional model based on 2D-DSA images

The invention discloses a processing method for reconstructing a blood vessel three-dimensional model based on 2D-DSA images. The method comprises the following steps of: S1, collecting 2D-DSA imagesbased on a normal position angle and a side position angle, and constructing a sparse blood vessel point cloud based on the collected 2D-DSA images; S2, performing preprocessing based on the constructed sparse blood vessel point cloud so as to obtain a point cloud sheet and a standard result, taking the obtained point cloud sheet, the standard result and a known intracranial blood vessel data setas a training set, inputting the training set into a PU-GCN deep learning network for training, and obtaining a trained PU-GCN deep learning network; and S3, obtaining a to-be-reconstructed sparse point cloud of a to-be-reconstructed 2D-DSA image based on the step S1, inputting the to-be-reconstructed sparse point cloud into the trained PU-GCN deep learning network, outputting the to-be-reconstructed sparse point cloud to