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.

Methods and systems are provided for adjusting medical imaging parameters based on imaging parameters used during a previous imaging session. In one embodiment, a method for a computed tomography (CT) system includes reducing a radiation level of at least one CT scan of one or more successive CT scans performed on a patient based on CT scan information obtained from a previous CT scan performed on the patient, the radiation level reduced relative to a radiation level of the previous CT scan.

The present disclosure relates to image reconstruction with favorable properties in terms of noise reduction, spatial resolution, detail preservation and computational complexity. The disclosed techniques may include some or all of: a first-pass reconstruction, a simplified datafit term, and/or a deep learning denoiser. In various implementations, the disclosed technique is portable to different CT platforms, such as by incorporating a first-pass reconstruction step.

Disclosed herein is a medical image reconstruction apparatus for reconstructing a medical image to assist the reading of a medical image. The medical image reconstruction apparatus includes a computing system, which includes: a receiver interface configured to receive a first medical image to which a first reconstruction parameter adapted to diagnose or analyze a first type of lesion is applied; and at least one processor configured to generate a second reconstruction parameter to be applied to the first medical image in response to a diagnosis order for the diagnosis or analysis of a second type of lesion. The at least one processor provides the second reconfiguration parameter to a user via a user interface, or generates a second medical image for the diagnose or analysis of the second type of lesion by executing the second reconstruction parameter on the first medical image and provides the second medical image to the user.

A method of statistical image restoration for a low-dose CT image using a deep learning, the method includes increasing a number of channels of the low-dose CT image, which is an input image, and decreasing a size of an activation map of the low-dose CT image using an encoder, passing the activation map generated by the encoder to a plurality of residual blocks, and increasing the size of the activation map passed through the residual blocks and generating a denoised result image using a decoder.

An image reconstruction system generates de-noised, attenuation corrected, and scatter corrected images using AI processing. The system receives a low-dose PET image and applies a machine learning algorithm via a convolutional neural network to the low-dose PET image to generate an output image. The output image includes correction for scatter and attenuation associated with the image being low-dose. The system provides the output image to a computing device comprising a user interface.

Disclosed herein are a method and system for processing medical image data. The method can comprise querying, using one or more monitor processors of a Picture Archiving and Communication System (PACS) monitor, a storage unit on a PACS server for available image data; determining, using the one or more monitor processors, if the available image data is new image data; retrieving, using the one or more monitor processors, the new image data from the storage unit on the PACS server if the available image data is new image data; processing, using one or more model processors, the new image data using a machine learning model to obtain a model result; generating, using the one or more model processors, at least one of an enhanced image data and a model result report based on the model result; and storing the at least one of the enhanced image data and the model result report for retrieval by a computing device.

A low-dose imaging method includes receiving a sparse image set of a portion of a patient's anatomy; up-sampling the sparse image set, in the sinogram domain and using a first neural network, to yield an up-sampled sinogram; generating, from the up-sampled sinogram, an initial reconstruction; and removing, from the initial reconstruction and using a second neural network, one or more artifacts in the initial reconstruction to yield a final output volume.

A method for establishing a 3D medical imaging model of a subject is to be implemented by an X-ray computed tomography (CT) scanner and a processor. The method includes: emitting X-rays on the subject sequentially from plural angles with respect to the subject to obtain M number of X-ray images of the subject in sequence; obtaining, for each pair of consecutive X-ray images, K number of intermediate image(s) by using the pair of consecutive X-ray images as inputs to a convolutional neural network (CNN) model that has been trained for frame interpolation; and establishing the 3D medical imaging model by using a 3D reconstruction technique based on the M number of X-ray images and the intermediate images obtained for the M number of X-ray images.

Provided is a low-dose imaging method, including continuously acquiring projection data; generating a first image by processing the acquired projection data before a data volume of the acquired projection data reaches a preset volume, and displaying the first image; and generating a second image by processing the preset volume of projection data when the data volume of the acquired projection data reaches the preset volume, and displaying the second image.