Disclosed is a three-dimensional low-dose PET reconstruction method based on deep learning. The method comprises the following steps: back projecting low-dose PET raw data to the image domain to maintain enough information from the raw data; selecting an appropriate three-dimensional deep neural network structure to fit the mapping between the back projection of the low-dose PET and a standard-dose PET image; after learning from the training samples the network parameters are fixed, realizing three-dimensional PET image reconstruction starting from low-dose PET raw data, thereby obtaining a low-dose PET reconstructed image which has a lower noise and a higher resolution compared with the traditional reconstruction algorithm and image domain noise reduction processing.

Proposed are a method and device for regularizing rapid three-dimensional tomographic imaging using a machine-learning algorithm. A method for regularizing three-dimensional tomographic imaging using a machine-learning algorithm according to an embodiment comprises the steps of: measuring a three-dimensional tomogram of a cell to acquire a raw tomogram of the cell; acquiring a regularized tomogram by using a regularization algorithm; and learning the relationship between the raw tomogram and the regularized tomogram through machine-learning.

A processor acquires a plurality of first projection images acquired by imaging an object at a plurality of radiation source positions and acquires a lesion image indicating a lesion. The processor combines the lesion image with the plurality of first projection images on the basis of a geometrical relationship between the plurality of radiation source positions and a position of the lesion virtually disposed in the object to derive a plurality of second projection images. The processor reconstructs the plurality of second projection images to generate a tomographic image including the lesion.

The disclosure provides a low-dose image reconstruction method and system based on prior anatomical structure difference. The method includes: determining the weights of different parts in the low-dose image based on prior information of anatomical structure differences; constructing a generative network being taking the low-dose image as input extract features, and integrating the weights of the different parts in the feature extraction process, outputting a predicted image; constructing a determining network being taking the predicted image and standard-dose image as input, to distinguish the authenticity of the predicted image and standard-dose image as the first optimization goal, and identifying different parts of the predicted image as the second optimization goal, collaboratively training the generative network and the determining network to obtain the mapping relationship between the low-dose image and the standard-dose image; and reconstructing the low-dose image by using the obtained mapping relationship. The disclosure can obtain more accurate high-definition images.

A system for few-view computed tomography (CT) image reconstruction is described. The system includes a preprocessing module, a first generator network, and a discriminator network. The preprocessing module is configured to apply a ramp filter to an input sinogram to yield a filtered sinogram. The first generator network is configured to receive the filtered sinogram, to learn a filtered back-projection operation and to provide a first reconstructed image as output. The first reconstructed image corresponds to the input sinogram. The discriminator network is configured to determine whether a received image corresponds to the first reconstructed image or a corresponding ground truth image. The generator network and the discriminator network correspond to a Wasserstein generative adversarial network (WGAN). The WGAN is optimized using an objective function based, at least in part, on a Wasserstein distance and based, at least in part, on a gradient penalty.

This present invention discloses a dynamic dual-tracer PET reconstruction method based on a hybrid-loss 3D CNN, which selects a corresponding 3D convolution kernel for a 3D format of dual-tracer PET data, and performs feature extraction in a stereoscopic receptive field (down-sampling) and the reconstruction (up-sampling) process, which accurately reconstructs the three-dimensional concentration distributions of two different tracers from the dynamic sinogram. The method of the invention can better reconstruct the simultaneous-injection single-acquisition dual-tracer sinogram without any model constraints. The scanning time required for dual-tracer PET can be minimized based on the method of the present invention. Using this method, the raw sinogram data of dual tracers can be reconstructed into two volumetric individual images in a short time.

The present disclosure is related to systems and methods for image processing. The method may include obtaining an image including at least one of a first type of artifact or a second type of artifact. The method may include determining, based on a trained machine learning model, at least one of first information associated with the first type of artifact or second information associated with the second type of artifact in the image. The trained machine learning model may include a first trained model and a second trained model. The first trained model may be configured to determine the first information. The second trained model may be configured to determine the second information. The method may include generating a target image based on at least part of the first information and the second information.

A method for calibrating defective channels of a CT device involves in a step S10, acquiring original data collected by the CT device; in a step S20, capturing to-be-recovered areas from the original data, wherein the to-be-recovered areas contain the defective channels of the CT device; in a step S30, inputting data of the to-be-recovered areas to a neural network for training so as to generate training results; and in a step S40, using the training results to repair the to-be-recovered areas. The method eliminates effects of artifacts caused by defective channels on image reconstruction.

A computer-implemented method is for providing a difference image data record. In an embodiment, the method includes a determination of a first real image data record of an examination volume in respect of a first X-ray energy, and a determination of a multi-energetic real image data record of the examination volume in respect of a first X-ray energy and a second X-ray energy, the second X-ray energy differing from the first X-ray energy. The method further includes the determination of the difference image data record of the examination volume by applying a trained function to input data, wherein the input data is based upon the first real image data record and the multi-energetic real image data record, as well as the provision of the difference image data record.

Disclosed is an approach for generating interactive visualizations for multi-nodal transfers that may involve terminal nodes and multiple transitional nodes by using various protocols to acquire data from computing systems or devices associated with each node. A first visualization layer comprising a set of geographic or physical indicators in a multi-nodal transfer route (which comprises a set of three or more nodes) may be generated. API protocols (and/or non-API protocols) corresponding to each node in the transfer route may be identified. The protocols may be executed to obtain, from computing systems and devices associated with the nodes, data packets used to generate a second visualization layer, which may comprise graphics that visually depict details of a transfer along the transfer route. An overlay of visualization layers may be displayed such that the graphics are displayed in association with multiple nodes.