Abstract:
An imaging method includes calculating a derivative back projection (DBP) result value using a DBP method with respect to a projection image of a field of view (FOV) inside an object, and reconstructing an image of the FOV by applying a regulation function to the FOV while reconstructing the image of the FOV using the DBP result value.
Abstract:
A method for processing an X-ray computed tomography (CT) image using a neural network and an apparatus therefor are provided. An image reconstruction method includes receiving low-dose X-ray CT data, obtaining an initial reconstruction image for the received low-dose X-ray CT data using a predetermined analytic algorithm, and reconstructing a denoised final image using the obtained initial reconstruction image and a previously trained neural network.
Abstract:
Disclosed is a method of reconstructing an image. The method of reconstructing an image includes receiving low dose X-ray computed tomography (CT) data, applying an analytic principle to an optimization approach for low dose imaging to transform the low dose X-ray CT data, and removing a noise included in the low dose X-ray CT data to reconstruct a high-quality image.
Abstract:
A method for identifying plaque erosion in a vessel. The method includes: obtaining, using a processor, a sequence of images of the vessel; extracting, using the processor, one or more image features from the sequence of images using a convolutional neural network model; contextually classifying, using the processor, the one or more extracted image features using a cascaded self-attention trained model; and generating, using the processor, one or more diagnostic labels associated with the sequence of images based on contextually classifying the one or more extracted image features, where the one or more diagnostic labels may include an indication of a presence of plaque erosion or an absence of plaque erosion.
Abstract:
A method for processing an X-ray computed tomography (CT) image using a neural network and an apparatus therefor are provided. An image reconstruction method includes receiving low-dose X-ray CT data, obtaining an initial reconstruction image for the received low-dose X-ray CT data using a predetermined analytic algorithm, and reconstructing a denoised final image using the obtained initial reconstruction image and a previously trained neural network.
Abstract:
Disclosed is a method and apparatus for restoring an image. The method and apparatus may detect boundary information associated with a boundary in an image, generate a reproducing kernel used to restore a hole in the image based on the detected boundary information, estimate hole information using the generated reproducing kernel, and restore the hole based on the estimated hole information.
Abstract:
Disclosed is a method and apparatus for restoring an image. The method and apparatus may detect boundary information associated with a boundary in an image, generate a reproducing kernel used to restore a hole in the image based on the detected boundary information, estimate hole information using the generated reproducing kernel, and restore the hole based on the estimated hole information.
Abstract:
A method and apparatus for reconstructing an image using a low rank Fourier interpolation scheme are provided. The method may reconstruct information of a k-space domain based on the information of the k-space domain using an imaging apparatus, for example, a magnetic resonance imaging (MRI) apparatus, a computed tomography (CT) apparatus, and a diffraction tomography apparatus. The method may generate a block Hankel matrix based on the information of the k-space domain, and may complete the block Hankel matrix using a low rank matrix completion algorithm, to reconstruct corresponding information of the k-space domain.
Abstract:
An X-ray imaging apparatus includes a back projected image generator configured to generate a back projected image with respect to a projected image of a field of view (FOV), and an image restorer configured to obtain frequency components of the back projected image, generate restored images for frequencies based on the frequency components of the back projected image, and generate a restored image with respect to the back projected image by synthesizing the restored images for the frequencies.