Abstract: An apparatus for and a method of correcting an image for an image artifact. An initial image is corrected by an image artifact corrector (190). The so corrected sample correction image is compared with the initial image to obtain information on the corrective action. The corrective action is then adaptively reapplied by a controller (140) to obtain an improved corrected image thereby ensuring previously present artifacts are removed and creation of new artifacts are avoided.

Abstract: A method includes obtaining contrast-enhanced image data having a plurality of voxels, each voxel having an intensity value. The method further includes determining a vesselness value for each voxel. The method further includes determining a hypo-density value for each voxel. The method further includes weighting each of the intensity values by a corresponding vesselness value. The method further includes weighting each of the hypo-density values by the corresponding vesselness value. The method further includes combining the weighted intensity values and the weighted hypo-density values, thereby generating composite image data. The method further includes visually displaying the composite image data.

Abstract: The invention relates to a method for OCR detection of valuable documents in a cash dispenser in the case of which an image of the valuable document is detected by means of a digital video or matrix camera. A Hough transformation is used to calculate edge lines of the valuable document and a rotation angle is calculated therefrom such that the edges of the valuable document are aligned with the image edges. The detected image is homogenized to compensate an inhomogeneous image background. This is followed by OCR detection of alphanumeric information on the valuable document.

Abstract: For image quality scoring of an image from a medical scanner, a generative model of an expected good quality image may be created using deep machine-learning. The deviation of an input image from the generative model is used as an input feature vector for a discriminative model. The discriminative model may also operate on another input feature vector derived from the input image. Based on these input feature vectors, the discriminative model outputs an image quality score.

Abstract: The disclosure describes a grading method and device for digital image quality. The method comprises: obtaining a digital image and extracting n main body region blocks from the image; calculating a first ratio value of the area of a plurality of main body region blocks to the total area of the image, a second ratio value of the area of background region blocks to the total area of the image, and a normalized distance value from a plurality of the pixel points in the plurality of main body region blocks to a center pixel point of the image; calculating to obtain a quality score value of the image through a preset digital image quality score conversion relation based on the first ratio value, the second ratio value, and the normalized distance value; and grading a quality of image according to a quality score value of the image and a preset digital image quality score threshold value.

Abstract: A computed tomography (CT) imaging apparatus that reduces metal artifacts for digital subtraction angiography by, in circuitry, obtaining mask frames and contrast frames generated during scans of an object; performing a first reconstruction of the mask frames to generate metal volume data; identifying metal voxels in the metal volume data; re-projecting the metal voxels into the mask frames to generate metal-mask frames; defining a region of interest in each of the mask frames, each region of interest including metal regions; re-projecting the contrast frames; blending the mask frames with the re-projected contrast frames; registering the contrast frames with respect to the blended mask frames performing a cross-correlation analysis of the mask frames and the contrast frames, frame-pair-by-frame-pair; subtracting the registered contrast image from the corresponding mask image, frame-pair-by-frame-pair, to generate subtracted frames; and performing a second reconstruction on the subtracted frames.

Abstract: A method of quantifying a medical image is disclosed herein. The method of quantifying a medical image includes acquiring regions of interest based on a medical image; filtering the sizes of the acquired regions of interest using length scale analysis; classifying the regions of interest whose sizes have been filtered, according to the sizes of the regions of interest; and visualizing the regions of interest whose sizes have been filtered, so that the regions of interest whose sizes have been filtered are distinguished from each other in the medical image according to the sizes of the regions of interest.

Abstract: The invention relates to a system and method for estimating a quantity of interest from perfusion data resulting from the acquisition of a plurality of a patient volumes or stations. Such a method may include a step for triggering the output of said quantity of interest in the form of a consolidated map by means of an adapted man-machine interface.

Abstract: An apparatus for processing volumetric image data to identify vessel regions having a predetermined condition, comprises a measurement unit for measuring at least one vessel parameter, a reference identification unit for obtaining data representing a branch-free vessel and identifying at least one reference section of the branch-free vessel, a calculation unit for calculating an expected value of the parameter in a further section of the branch-free vessel based on at least one measured value of the parameter in the at least one reference section, and a region identification unit for identifying a region of the vessel having the predetermined condition in dependence on both the expected value of the parameter in the further section and a measured value of the parameter in the further section.

Abstract: Provided is a magnetic resonance imaging (MRI) apparatus.

Abstract: A system and method are presented for neural network based feature extraction for acoustic model development. A neural network may be used to extract acoustic features from raw MFCCs or the spectrum, which are then used for training acoustic models for speech recognition systems. Feature extraction may be performed by optimizing a cost function used in linear discriminant analysis. General non-linear functions generated by the neural network are used for feature extraction. The transformation may be performed using a cost function from linear discriminant analysis methods which perform linear operations on the MFCCs and generate lower dimensional features for speech recognition. The extracted acoustic features may then be used for training acoustic models for speech recognition systems.

Abstract: Systems and methods for implementing a voxel 3D modeling technique in client server systems, e.g., using a web browser as the main user interface.

Abstract: Apparatus and methods are described including designating within an image of blood vessels of a subject, a target portion of the blood vessels. In response to a first input from a user that designates a single point on the image, a first location within the blood vessels in a vicinity of the designated point is designated. In response to second and third inputs from the user, proximal and distal locations within the target portion are designated. The target portion is designated such that the target portion passes from the proximal location of the target portion to the distal location of the target portion, and such that the target portion includes the first location. Quantitative vessel analysis is performed on the target portion of the blood vessels, and an output is generated based upon the quantitative vessel analysis. Other applications are also described.

Abstract: An image processing device includes an acquiring unit configured to acquire a medical video image obtained by imaging of lungs, a holding unit configured to hold a lung field motion model, the lung field motion model simulating lung field motion, and a processing unit configured to process the medical video image by using the lung field motion model.

Abstract: A method for obtaining object surface topology in which image frames of a scene (e.g., video frames from a user passing a smartphone camera over an object) are transformed into dense feature vectors, and feature vectors are correlated to obtain high precision depth maps. Six dimensional pose is determined from the video sequence, and then used to register patches of pixels from the frames. Registered patches are aligned and then correlated to local shifts. These local shifts are converted to precision depth maps, which are used to characterize surface detail of an object. Feature vector transforms are leveraged in a signal processing method comprising several levels of interacting loops. At a first loop level, a structure from motion loop process extracts anchor features from image frames. At another level, an interacting loop process extracts surface texture, as noted. At additional levels, object forms are segmented from the images, and objects are counted and/or measured.

Abstract: Magnetic resonance imaging (MRI) systems and methods to effect improved MR image reconstruction for undersampled data acquisitions are described. The improved MR image reconstruction is performed by iteratively using compressed sensing to reconstruct an MR image based upon at least one sensitivity map by minimizing a predetermined function which is based upon the MR image and coefficients of the at least one sensitivity map, and updating the at least one sensitivity map by minimizing the predetermined function.

Abstract: According to one embodiment, an image processing device includes processing circuitry. The processing circuitry acquires a plurality of first image data indicating a plurality of time-sequential bloodstream images obtained after administering contrast agent to an object, and sets a target region on second image data. The second image data are generated based on the plurality of the first image data so as to indicate information on temporal change of pixel values of each pixel. Further, the processing circuitry selects a reproduction section by selecting at least reproduction-start image data and reproduction-end image data from the plurality of first image data based on bloodstream information of the target region.

Abstract: A system and method for obtaining a composite image by combining multiple sub-images are provided. In some embodiments, the method may include retrieving overlapping images corresponding to sub-images including 3D volume data, generating two-dimensional (2D) projection images and pixel maps based on the overlapping images, performing one or more registrations based on the 2D projection images and the pixel maps, calibrating the sub-images based on the results of the registration(s), and fusing the sub-images to produce a composite image. In some embodiments, the method may include setting a plurality of parameters relating to an X-radiation source or a radiation detector based on a preliminary number of exposures and a preliminary exposure region, controlling, based on at least one of the plurality of parameters, a motion of the X-radiation source or a motion of the radiation detector to capture a plurality of sub-images, and combining the plurality of sub-images.

Abstract: The present invention, in one form, is a method for deriving respiratory gated PET image reconstruction from raw PET data. In reconstructing the respiratory gated images in accordance with the present invention, respiratory motion information derived from individual voxel signal fluctuations, is used in combination to create usable respiratory phase information. Employing this method allows the respiratory gated PET images to be reconstructed from PET data without the use of external hardware, and in a fully automated manner.

Abstract: A medical image processing apparatus according to an embodiment includes an estimation circuitry and a tracking circuitry. The estimation circuitry is configured to estimate the activity of the myocardium across a plurality of images at different time phases from a group of images where a plurality of images containing a myocardium are chronologically arranged. The tracking circuitry is configured to set a search range for tracking the myocardium in the group according to the activity of the myocardium and perform the tracking.

Abstract: Systems and methods are disclosed for predicting coronary plaque vulnerability, using a computer system. One method includes acquiring anatomical image data of at least part of the patient's vascular system; performing, using a processor, one or more image characteristics analysis, geometrical analysis, computational fluid dynamics analysis, and structural mechanics analysis on the anatomical image data; predicting, using the processor, a coronary plaque vulnerability present in the patient's vascular system, wherein predicting the coronary plaque vulnerability includes calculating an adverse plaque characteristic based on results of the one or more of image characteristics analysis, geometrical analysis, computational fluid dynamics analysis, and structural mechanics analysis of the anatomical image data; and reporting, using the processor, the calculated adverse plaque characteristic.

Abstract: The invention relates to a method for OCR detection of valuable documents in a cash dispenser in the case of which an image of the valuable document is detected by means of a digital video or matrix camera. A Hough transformation is used to calculate edge lines of the valuable document and a rotation angle is calculated therefrom such that the edges of the valuable document are aligned with the image edges. The detected image is homogenized to compensate an inhomogeneous image background. This is followed by OCR detection of alphanumeric information on the valuable document.

Abstract: Systems and methods are provided for data reconstruction. In accordance with one aspect, data interpolation is performed on a time-varying three-dimensional (3D) image dataset of one or more vessel-like structures to generate at least one interpolated voxel value. The interpolated voxel value is used to correct at least one value of a vessel voxel representing the one or more vessel-like structures in the time-varying 3D dataset.

Abstract: A method, medical imaging workstation (1000) and hybrid medical imaging scanner (1100) are provided for defining a region of interest (RoI) for display on at least two medical scan images. When displaying a first medical scan image (740), input data defining a RoI on the image is captured, and stored as at least a first region representation (760). The RoI is displayed on a second medical scan image (750), based on the first region representation (760). Changes to the RoI on the second medical scan image (750) are used to update the first region representation (760). There may be separate region representations (760, 770) associated with each of several medical scan images. The invention may improve the definition of a region of interest, by allowing editing on each of multiple image displays (820, 830, 880) to feed through to all medical scan images.

Abstract: A nondestructive inspection device 1 comprises an X-ray indicator 20, a low-energy detector 32, a high-energy detector 42, a low-energy transmittance calculation unit 72, a high-energy transmittance calculation unit 74, a detection unit 76, and a correction unit 78. The calculation unit 72 calculates a value indicating the transmittance of transmission X-rays in a low energy range. The calculation unit 74 calculates a value indicating the transmittance of transmission X-rays in a high energy range. The detection unit 76 detects a positional deviation detail of the X-ray indicator 20 according to a ratio between the transmittances calculated by both of the calculation units 72, 74. When the positional deviation detail of the X-ray indicator 20 is detected by the detection unit 76, according to the positional deviation detail, the correction unit 78 corrects X-ray luminance data detected by the detectors 32, 42.

Abstract: A method of analyzing structure of a network of vessels in a medical image, comprising: receiving the medical image depicting the network of vessels; obtaining a mask of the network of vessels in the image; and generating a non-forest graph mapping a plurality of paths of vessels in the network to directed paths in the graph, with each edge in the graph either directed to indicate a known direction of flow in the corresponding vessel, or undirected to indicate a lack of knowledge of direction of flow in the corresponding vessel, and with all directed edges in a path directed in a same direction as the path.

Abstract: An image processing device classifies a region of interest included in an image into multiple classification items. The image processing device has: an initial region detector that detects at least part of the region of interest and sets the part as an initial region; an expansion region detector that detects an expansion region by expanding the initial region; and a region determining unit that calculates feature data of the initial region and the expansion region, and determines, based on the feature data, to which of the multiple classification items the region of interest belongs.

Abstract: Aspects provide for notifications of image recognition analysis matches to streamed image data on augmented reality device displays. Data identifies one or more visual appearance attributes of a target object that are discernible within image data acquired by a camera. Image data captured by a camera is analyzed to determine an occurrence of the visual appearance attribute(s) within a stream of images of captured image data. In response to the analyzing determining an occurrence of the visual appearance attribute(s) within the stream of captured images, the method drives a display screen of the augmented reality device to distinguish a possible location of the target object within surroundings of a user of the augmented reality display device that are displayed to the user by the display screen.

Abstract: A system and a method are disclosed that allow for generation of a model or reconstruction of a model of a subject based upon acquired image data. The image data can be acquired in a substantially mobile system that can be moved relative to a subject to allow for image acquisition from a plurality of orientations relative to the subject. The plurality of orientations can include a first and final orientation and a predetermined path along which an image data collector or detector can move to acquire an appropriate image data set to allow for the model of construction.

Abstract: An image forming apparatus includes: a first correction unit configured to obtain a measurement value of a correction amount based on a detection result by a detection unit, and correct an image forming condition based on the measurement value; a second correction unit configured to obtain a variation value by a prediction calculation at a predetermined timing, obtain a prediction value by accumulating the variation value since the first correction unit performed the correction, and correct the image forming condition based on the prediction value, the variation value being a estimated value of the correction amount; and a determination unit configured to obtain an evaluation value related to a prediction error based on the variation value since the first correction unit performed the correction, and determine, based on the evaluation value, the correction timing by the first correction unit.

Abstract: An apparatus for visualizing medical image data comprises a data processing unit for obtaining first image data representing a first medical image, second image data representing a second medical image, and subtraction image data representing a subtraction image, wherein the subtraction image data is generated by subtracting one of the first medical image data and the second medical image data from the other of the first medical image data and the second medical image data, and the apparatus further comprises a feature identification unit for processing the first image data representing the first medical image to identify a feature, and a rendering unit for rendering a combined image comprising an overlay of the identified feature from the first medical image data and one of the subtraction image and the second medical image.

Abstract: A system and method for obtaining a composite image by combining multiple sub-images are provided. In some embodiments, the method may include retrieving overlapping images corresponding to sub-images including 3D volume data, generating two-dimensional (2D) projection images and pixel maps based on the overlapping images, performing one or more registrations based on the 2D projection images and the pixel maps, calibrating the sub-images based on the results of the registration(s), and fusing the sub-images to produce a composite image. In some embodiments, the method may include setting a plurality of parameters relating to an X-radiation source or a radiation detector based on a preliminary number of exposures and a preliminary exposure region, controlling, based on at least one of the plurality of parameters, a motion of the X-radiation source or a motion of the radiation detector to capture a plurality of sub-images, and combining the plurality of sub-images.

Abstract: An image processing apparatus according to an embodiment includes a lung field region extracting unit, a lung field bottom region extracting unit, and a detecting unit. The lung field region extracting unit is configured to extract, based on pixel values of pixels constituting a three-dimensional medical image capturing a chest of a subject, a lung field region from the three-dimensional medical image. The lung field bottom region extracting unit is configured to extract a lung field bottom region from the lung field region. The detecting unit is configured to detect a vertex position of the lung field bottom region on the head side of the subject.

Abstract: A method and system for extracting coronary vessels fluoroscopic image sequences using coronary digital subtraction angiography (DSA) are disclosed. A set of mask images of a coronary region is received, and a sequence of contrast images for the coronary region is received. For each contrast image, a motion estimate is calculated between each of the mask images and a background region of the contrast image and a covariance is calculated for each motion estimate. Multiple background layer predictions are generated by generating a background layer prediction for each mask image based on the calculated motion estimate and covariance. The multiple background layer estimates are combined using statistical fusion to generate a final estimated background layer. The final estimated background layer is subtracted from the contrast image to extract a coronary vessel layer for the contrast image.

Abstract: A method for automatically detecting the presence of a contrast agent in an x-ray image includes acquiring a preliminary x-ray image. A background image is estimated. The contrast agent is administered. A plurality of image frames is acquired. The background image is subtracted from each image frame. An image having a highest image intensity is selected. A predefined shape model is fitted to the selected image using a semi-global optimization strategy. The fitting of the shape model is used to fit the shape model to each of the subtracted images. A feature value is calculated for each image frame based on pixel intensities of each pixel fitted to the shape model for the corresponding subtracted image. An image frame of peak contrast is determined by selecting the image frame with the greatest feature value.

Abstract: An image display method and an image display apparatus to solve the three-dimensional cueing confliction problem of the maximum intensity projection in the stereoscopic display method, while enabling users to select and render with emphasis the maximum intensity projection of the objects of interest, so that to realize the stereoscopic display. A three-dimensional surface which has equal distance to a sight-point is utilized as a reference surface, and the distances from all of the local maximum intensity points to this reference surface are calculated, then a weighting factor of each local maximum intensity point is calculated according to the distances and predetermined weighting function. After, the values of local maximum intensity points are adjusted according to the obtained weighting factors, and finally, the maximum intensity projection value is produced by synthesizing all of the adjusted values of the local maximum intensity points.

Abstract: A method is described for determining brain shift, such as the brain shift that occurs following a neurosurgical intervention. The method includes taking a first image of the brain of a subject, the first image showing the position of blood vessels in the brain relative to a reference position. A second image of the brain that has been acquired following an intervention on the subject is also taken. This second image has been acquired using computed tomography imaging apparatus. The second image shows the position of blood vessels in the brain relative to the reference position. Brain shift is then determined from the shift in position of the blood vessels in at least one region of interest of the brain, with respect to the reference position, between the first and second images. Corresponding apparatus is also described.

Abstract: A structure extraction element obtains three-dimensional structure extraction data by extracting a tubular structure from volume data obtained by imaging a predetermined subject. A relating element relates positions on paths of a two-dimensional template representing a schematic diagram of a three-dimensionally branching tubular structure obtained by imaging a predetermined subject and positions on paths of the three-dimensional structure extraction data to each other. A projection image generation element generates a two-dimensional projection image by projecting voxel values of the tubular structure that is present on the paths of the three-dimensional structure extraction data onto corresponding positions on the paths of the two-dimensional template.

Abstract: Systems and methods are provided for data reconstruction. In accordance with one aspect, data interpolation is performed on a time-varying three-dimensional (3D) image dataset of one or more vessel-like structures to generate at least one interpolated voxel value. The interpolated voxel value is used to correct at least one value of a vessel voxel representing the one or more vessel-like structures in the time-varying 3D dataset.

Abstract: In a Workflow Management System with a workflow having a sequence of activities of an underlying business process, logical wait situations occur when a given activity depends upon completion of a prerequisite preparation activity. A method for decreasing the resulting run-time delays includes: identifying a primary activity and an associated preparation activity required to be done before executing the primary activity; navigating through the process template of the underlying workflow and calculating the probability at a node of the process template that the primary activity will be reached for a current process instance; and pre-executing the preparation activity in parallel to the activity sequence if the probability exceeds a predetermined threshold.

Abstract: Different candidate windows in an image are identified, such as by sliding a rectangular or other geometric shape of different sizes over an image to identify portions of the image (groups of pixels in the image). The candidate windows are analyzed by a set of convolutional neural networks, which are cascaded so that the input of one convolutional neural network layer is based on the input of another convolutional neural network layer. Each convolutional neural network layer drops or rejects one or more candidate windows that the convolutional neural network layer determines does not include an object (e.g., a face). The candidate windows that are identified as including an object (e.g., a face) are analyzed by another one of the convolutional neural network layers. The candidate windows identified by the last of the convolutional neural network layers are the indications of the objects (e.g., faces) in the image.

Abstract: A device is used to precisely deliver the treatment plan created by an automatic planning system by positioning a single low energy radiation source, or a plurality of low energy sources connected to each other, in a predetermined parallel, planar, or similar geometry, each source equipped with blocking and attenuation mechanisms, thereby delivering a plurality of parallel overlapping beams indexed on a millimeter or submillimeter grid such that a concentration of dose is achieved at a variable depth in tissue relative to the dose where the radiation first enters the tissue. A plurality of overlapping beams indexed on a millimeter or submillimeter grid can converge on a target volume loaded with gold nanoparticles to deliver a tumorcidal dose of radiation in as little as a single session to tumor cells but not to normal cells or to deliver serial radiosurgical treatments.

Abstract: An image processing apparatus includes: a subtraction unit that generates subtraction images constituted by a plurality of frames from X-ray images of an object acquired before and after injecting a radiopaque dye; an evaluation unit that calculates evaluation values indicating states of the injection of the radiopaque dye for the frames constituting the subtraction images; and a selection unit that selects at least one of the frames constituting the subtraction images using the evaluation values and outputs the selected frame as a reference image.

Abstract: A method of providing a UI and a display apparatus applying the same include the operations of selecting a main thumbnail, searching thumbnails related to the main thumbnail on a basis of the main thumbnail, and displaying the selected main thumbnail and the searched thumbnails on radial axes according to a preset arrangement criterion.

Abstract: An apparatus for providing a motion-corrected time series of image data sets, comprises an image selection unit for obtaining a series of sets of image data, each set of image data representative of a region at a different time, and for automatically determining a sub-region of interest, and a registration unit that is configured to, for each of a plurality of sets of image data in the series, perform a registration, for the sub-region of interest, to register one of a reference data set of the series and the set of image data to the other of that reference data set and the set of image data.

Abstract: An angiographic examination method for the representation of flow properties of vessels of an object under examination is presented. To determine blood-flow parameters in 3-D with high temporal resolution, at least one 4-D DSA sequence is acquired for the generation of measurement-based 4-D DSA data sets. A model-based method determines time-dependent volume data sets, which, in terms of time, lie between the time-dependent volume data sets of the measurement-based 4-D DSA method.

Abstract: A system (100) for segmenting an object in an image adapts a first model for segmenting the object to the image. A feature is extracted from the image based on the adapted first model. A second model is selected for segmenting the object from a plurality of models for segmenting the object, based on the feature extracted from the image. The second model includes additional detail of the object. The second model is utilized based on the adapted first model and/or the feature extracted from the image; the initialized second model is adapted to the image. The features extracted from the image based on the adapted first model help the system (100) to select the second model for segmenting the object from a plurality of models for segmenting the object. The adapted first model and/or the extracted features are also used for initializing the second model.

Abstract: A method and apparatus for utilizing mammogram images for verification, including obtaining one or more mammogram image of a patient; determining a first set of breast characteristics from the one or more mammogram image, wherein the first set of breast characteristics include breast landmark, breast tissue density, tissue pattern and breast finding; comparing the first set of breast characteristics with a second set of breast characteristics to yield a comparison, wherein the second set of breast characteristics is from one or more previous mammogram image identified as potentially belonging to the patient using an identification not related to content of a mammogram image; and determining based on the comparison whether the one or more previous mammogram image belongs to the patient.

Abstract: An imaging apparatus includes a reconstruction unit configured to reconstruct a tomographic image of a predetermined range of a measurement object based on interference light produced by interference between returning light of measurement light from the measurement object and reference light, and a generation unit configured to, for each pixel value row in a depth direction of the tomographic image, generate a two-dimensional image based on a pixel value selected in order of magnitude of the pixel values.

Abstract: A method and apparatus of image compensation are provided. The method may include: calculating an overlapping position between two adjacent scannings; obtaining two images at the overlapping position of the two adjacent scannings, and calculating mutual information of the two images by using a three dimensional non-rigid registration method; and acquiring a corresponding transformational matrix when the mutual information reaches a threshold, and compensating one of the two scans which needs to be compensated by using the transformational matrix. Images at the overlapping position of two neighboring scannings can be used to find a motion law at an identical z position of the two adjacent scannings and to obtain a corresponding match factor. Thus inconsistence of images caused by a patient's movement can be compensated.