Low-dose CT image quality enhancement method based on Bayesian neural network

The invention provides a low-dose CT image quality enhancement method based on a Bayesian neural network. The low-dose CT image quality enhancement method based on the Bayesian neural network comprises the following steps: S1, using a Mayo Clinic data set as a training set; and S2, constructing a Bayesian neural network for enhancing the quality of the low-dose CT image based on a Bayesian convolutional layer. And S3, performing data enhancement on the training set, inputting the data into the Bayesian neural network, and training the Bayesian neural network until the network converges. And S4, calculating the low-dose CT image by using the trained Bayesian neural network to generate a quality-enhanced low-dose CT image. 本发明提供一种基于贝叶斯神经网络的低剂量CT图像质量增强方法，包括如下步骤：步骤S1，使用Mayo Clinic数据集作为训练集。步骤S2，基于贝叶斯卷积层构建用于增强低剂量CT图像质量的贝叶斯神经网络。步骤S3，对训练集进行数据增强后，输入到贝叶斯神经网络中，训练贝叶斯神经网络，直到网络收敛。步骤S4，使用训练完成的贝叶斯神经网络对低剂量CT图像进行计算，生成质量增强的低剂量CT图像。