Lung artery and vein automatic segmentation method based on neural network

• Main Authors: AO QIANG, WANG XIAO, WANG YUNBING, TIAN GUANGYE, CHENG YUSHENG
• Format: Patent
• Language: Chinese; English
• Published: 10.11.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS

The invention discloses a pulmonary artery and vein automatic segmentation method based on a neural network, and the method comprises the steps: (S100), enabling a bronchial part in CT to be enhanced through an image enhancement algorithm; (S200) constructing a neural network, combining the enhanced CT data with the original CT image to serve as input data of the neural network, and performing neural network training to complete arteriovenous segmentation or classification training; and (S300) adopting an image enhancement algorithm to enhance the lung bronchus for CT scanning data to be analyzed, merging the enhanced CT data and an original CT image, inputting the merged data into a neural network for prediction, calculating a probability maximum value of each voxel point of an output result, taking the probability maximum value as a prediction target, namely a background, an artery or a vein, generating a corresponding organ mask, and performing three-dimensional reconstruction. According to the method, the