A display control unit displays a tomographic image of a specific tomographic plane in a first medical image on a display unit. A finding classification unit classifies each pixel of a partial region of the first medical image into at least one finding. A feature amount calculation unit calculates a first feature amount for each finding in the partial region. A weighting coefficient setting unit sets a weighting coefficient indicating a degree of weighting, which varies depending on a size of each finding, for each finding. A similarity derivation unit performs a weighting operation for the first feature amount for each finding calculated in the partial region and a second feature amount for each finding calculated in a second medical image on the basis of the weighting coefficient to derive a similarity between the first medical image and the second medical image.

The present technology relates to an imaging system. The imaging system can comprise at least one processor configured to apply a projection precomputation algorithm and an x-ray tomography image reconstruction system. The projection precomputation algorithm can be configured to: generate a projection operator matrix that can be used to calculate a plurality of voxels from a plurality of projection measurements before the plurality of projection measurements is acquired and store the projection operator matrix in memory. The projection operator matrix can be at least one of: a compressed matrix, a multi-iteration projection operator matrix, and a combination thereof. The x-ray tomography image reconstruction system can be configured to apply the projection operator matrix to generate a reconstructed three-dimensional image of at least an internal portion of a selected object under a surface of the selected object when the plurality of projection measurements is acquired.

A method for machine learning includes learning, during a training stage, network parameter values of a neural network to obtain a trained neural network configured to perform reconstruction of medical images; refining, during a subsequent refinement stage, the learned network parameter values to generate refined network parameter values defining a refined neural network; and applying input medical image data to the refined neural network to generate a reconstructed medical image. The method retains benefits of machine learning image reconstruction to obtain a desired reconstructed image.

A superimposing images analysis system for AVM includes an image capturing unit, an image superimposing and aligning unit, and a transparency adjustment unit. The image superimposing and aligning unit is connected to the image capturing unit, and the transparency adjustment unit is connected to the image superimposing and aligning unit. The image capturing unit captures a brain MRA image and a brain MRI image. The image superimposing and aligning unit receives the brain MRA image and brain MRI image from the image capturing unit, and performs superimposing and aligning with the received images to generate a superimposing brain image. The transparency adjustment unit adjusts the transparency of the brain MRA image or the brain MRI image in the superimposing brain image according to a transparency adjustment instruction so as to generate a compared superimposing brain image.

The present disclosure provides a system and method for image processing. The method may include obtaining multiple projection images of a subject acquired by an imaging device from multiple view angles; generating an initial slice image of the subject by image reconstruction based on the multiple projection images; determining, based on the multiple projection images, a target out-of-plane artifact of the initial slice image; and generating a corrected slice image by correcting the initial slice image with respect to the target out-of-plane artifact.

Provided are a method and device for deep learning-based patch-wise three-dimensional (3D) bone microstructure reconstruction from clinical CT scan data. A computer device may be configured to segment a low-resolution skeletal image into a plurality of low-resolution image patches, to acquire a plurality of high-resolution image patches from the low-resolution image patches, respectively, using a pretrained artificial neural network, and to reconstruct a high-resolution bone microstructure by assembling and postprocessing the high-resolution image patches.

Disclosed are a three-dimensional image generation method and an electronic device for performing same, according to various embodiments. The electronic device according to one embodiment of the present invention comprises: an image capture device for acquiring a plurality of radiological images for a sample moving on a transport device; and a processor, wherein the processor can: determine feature points of the plurality of radiological images, for reconstructing a three-dimensional image of the sample; use the location of the feature points to calculate the location information of the feature points; generate a feature point image on the basis of the location information; and generate the three-dimensional image by using the feature point image and the location information.

Described embodiments provide systems and methods for generating a corrected planar scintigraphy image (CPSI) corrected for image artifacts. A computing system can obtain a plurality of planar scintigraphy images of a subject. The plurality of planar scintigraphy images may contain image artifacts caused by one or more physical processes. The computing system may generate a corrected CPSI by applying a planar scintigraphy image reconstruction model to the plurality of planar scintigraphy images. The planar scintigraphy image reconstruction model may comprise a first non-negativity constraint and a second non-negativity constraint, and be based on a first regularization term, a second regularization term, a coupling term and a fidelity term. The computing system may present the CPSI for evaluation of a condition of the subject. Presenting the CPSI may comprise at least one of transmitting the CPSI to a computing device or displaying the CPSI on a display screen.

The image processing apparatus derives an inclination amount of a reference line on the basis of an optical image obtained by imaging a part used for deriving the reference line of a subject or information indicating an irradiation direction of a light beam emitted onto the reference line, and generates a tilt image which is a tomographic image of an inclined cross section inclined according to the inclination amount by tilting and rotating an axial cross section using a slice image.

Technology is described for handling truncated data in iterative reconstruction. A method comprises iterating on a volume of an object including a non-truncated part based on image data and at least one truncated part representing deficiently imaged data. The volume is represented by voxels. The iterating includes regularizing the non-truncated part of the volume using a first regularizer, and regularizing the truncated part of the volume using a second regularizer different from the first regularizer.