Abstract: Techniques are described for generating reformatted views of a three-dimensional (3D) anatomy scan using deep-learning estimated scan prescription masks. According to an embodiment, a system is provided that comprises a memory that stores computer executable components, and a processor that executes the computer executable components stored in the memory. The computer executable components comprise a mask generation component that employs a pre-trained neural network model to generate masks for different anatomical landmarks depicted in one or more calibration images captured of an anatomical region of a patient. The computer executable components further comprise a reformatting component that reformats 3D image data captured of the anatomical region of the patient using the masks to generate different representations of the 3D image data that correspond to the different anatomical landmarks.

Abstract: An embodiment of the present disclosure provides a method of generating a tomography image and an X-ray imaging apparatus operating according to the method, whereby when a material with a high X-ray attenuation rate is inserted into an object, an image quality may be improved by reducing ripple artifacts and/or undershoot artifacts that may occur in a tomography image that is a captured X-ray image of the object.

Abstract: Improved (e.g., high-throughput, low-noise, and/or low-artifact) X-ray Microscopy images are achieved using a deep neural network trained via an accessible workflow. The workflow involves selection of a desired improvement factor (x), which is used to automatically partition supplied data into two or more subsets for neural network training. The neural network is trained by generating reconstructed volumes for each of the subsets. The neural network can be trained to take projection images or reconstructed volumes as input and output improved projection images or improved reconstructed volumes as output, respectively. Once trained, the neural network can be applied to the training data and/or subsequent data—optionally collected at a higher throughput—to ultimately achieve improved de-noising and/or other artifact reduction in the reconstructed volume.

Abstract: Brain tumor or other tissue classification and/or segmentation is provided based on from multi-parametric MRI. MRI spectroscopy, such as in combination with structural and/or diffusion MRI measurements, are used to classify. A machine-learned model or classifier distinguishes between the types of tissue in response to input of the multi-parametric MRI. To deal with limited training data for tumors, a patch-based system may be used. To better assist physicians in interpreting results, a confidence map may be generated using the machine-learned classifier.

Abstract: A system includes a reconstructor (314) configured to reconstruct cone beam projection data to generate cone beam artifact corrected short scan cone beam volumetric image data. A method includes reconstructing, with a reconstructor, cone beam projection data to generate cone beam artifact corrected short scan cone beam volumetric image data. A computer-readable storage medium storing computer executable instructions which when executed by a processor of a computer cause the processor to: reconstruct cone beam projection data to generate cone beam artifact corrected short scan cone beam volumetric image data.

Abstract: An object of the technique of the present disclosure is to obtain a demosaic network whose robustness for noise is high. In the technique of the present disclosure, first, a data set including a set of a teacher image with noise and a pupil image with noise is generated. Then, learning of a demosaic network is performed by using an image pair of the teacher image with noise and the pupil image with noise, to both of which noise is added.

Abstract: Disclosed herein are visualization systems, methods, devices and database configurations related to the real-time depiction, in 2 D and 3 D on monitor panels as well as via 3 D holographic visualization, of the internal workings of patient surgery, such as patient intervention site posture as well as the positioning, in some cases real time positioning, of an object foreign to the patient.

Abstract: An information processing apparatus including: a classifier that is applied for medical treatment data; and a hardware processor that sets a property of the classifier, evaluates a sufficiency level of the property by using evaluation data for the classifier, and selects and sets one classifier as the classifier to be used in a facility by referring to an evaluation result.

Abstract: The present subject matter discloses a system and method for detecting Large Vessel Occlusion (LVO) on a Computational Tomography Angiogram (CTA) automatically. the system comprises a vascular-territory-segmentation module, an ICV segmentation module, MCA-LVO classifier and ICA-LVO classifier. The vascular territory segmentation module is configured to receive a set of CTA images and to mark a territory of vascular segments in the ICV region for each slice of the ROI. The ICV segmentation module is configured to process each slice of the ROI. The processed slices of the ROI are combined to develop a CTA image after application of MIP and the developed CTA image is segmented into a Middle Cerebral Artery (MCA) region and an Internal Cerebral Artery (ICA) region. The MCA-LVO and ICA-LVO classifiers determine presence of the LVO on the received MCA and ICA region using Deep Learning techniques and accordingly the presence of the LVO is reported.

Abstract: An image analysis system including at least one processor and at least one memory is provided. The image analysis system is configured to receive image data associated with a brain of a patient, the image data including a first three-dimensional (3D) diffusion weighted imaging (DWI) image acquired using a magnetic resonance imaging (MRI) system and a second 3D DWI image, concurrently provide the first 3D DWI image to a first channel of a trained model and the second 3D DWI image to a second channel of the trained model, receive an indicator associated with the first 3D DWI image and the second 3D DWI image from the model, generate a report based on the indicator, and cause the report to be output to at least one of a memory or a display.

Abstract: A system and method for training a neural network to denoise images. One noise realization is paired to an ensemble of training-ready noise realizations, and fed into a neural network for training. Training datasets can also be retrospectively generated based on existing patient studies to increase the number of training datasets.

Abstract: The invention provides systems, apparatus and methods for digital image processing providing enhanced display of and radiomics generated from images generated from x-ray and other medical imaging data. Such systems, apparatus and methods capture and use that data to identify distinct gradations, e.g., of grayscale within and beyond the spectrum of human vision, then delineate borders based on ranges of gradation, forming irregular multi-layer visual objects with delineated internal contouring and an outer boundary, and then enhance the delineated layers and superimpose the enhancing display over and/or determine radiomic measures of the corresponding areas of the original image, thereby revealing underlying morphology of masses previously obscured, hidden or “masked” from human vision.

Abstract: An organ evaluation device, system, or method is configured to receive electrophysiological data from a patient or model organism and integrates the data in a computational backend environment with anatomical data input from an external source, spanning a plurality of file formats, where the input parameters are combined to visualize and output current density and/or current flow activity having ampere-based units displayed in the spatial context of heart or other organ anatomy.

Abstract: Devices, systems, methods, and computer-readable media for registering an electromagnetic registration of a luminal network to a 3D model of the luminal network include accessing a 3D model of a luminal network based on computed tomographic (CT) images of the luminal network, selecting a plurality of reference points within the 3D model of the luminal network, obtaining a plurality of survey points within the luminal network, dividing the 3D model of the luminal network and the luminal network into a plurality of regions, assigning a plurality of weights to the plurality of regions, determining an alignment of the plurality of reference points with the plurality of survey points based on the plurality of weights, and generating a registration based on the alignment, the registration enabling conversion of the plurality of survey points within the luminal network to points within the 3D model of the luminal network.

Abstract: Method and system for determining the angle of anteversion of for example bone fragments of a fractured bone are disclosed. Supported by artificial intelligence, a first object is classified in a first X-ray projection image. A second object is classified in a second X-ray projection image. A spatial arrangement of the objects relative to each other can be determined based on a respective determination of a representation and a localization of both objects.

Abstract: An image segmentation method is provided for a computing device. The method includes obtaining a general tumor image, performing tumor localization on the tumor image to obtain a candidate image indicating a position of a tumor region in the general tumor image, inputting the candidate image to a cascaded segmentation network constructed based on a machine learning model, and performing image segmentation on the general tumor region in the candidate image using a first-level segmentation network and a second-level segmentation network in the cascaded segmentation network to obtain a segmented image.

Abstract: Systems and methods are provided for generating a pseudo-CT prediction model that can be used to generate pseudo-CT images. An exemplary system may include a processor configured to retrieve training data including at least one MR image and at least one CT image for each of a plurality of training subjects. For each training subject, the processor may extract a plurality of features from each image point of the at least one MR image, create a feature vector for each image point based on the extracted features, and extract a CT value from each image point of the at least one CT image. The processor may also generate the pseudo-CT prediction model based on the feature vectors and the CT values of the plurality of training subjects.

Abstract: A material decomposition apparatus for performing decomposition of a material in an object. The apparatus includes a data storage section for storing correction data preliminarily generated by decomposing one of three or more materials into the other two materials, a data input section to which radiation data of the object is inputted, the radiation data being divided into a plurality of energy levels, and a decomposition processing section for repeatedly performing two-material decomposition for decomposition of the other two materials of the three or more materials using the radiation data at different energy levels and the correction data to perform decomposition of the inside of the object into the three or more materials.

Abstract: System and method of more efficiently identifying and segmenting anatomical structures from 2-D cone beam CT images, rather than from reconstructed 3-D volume data, is disclosed. An image processing system receives, from a cone beam CT device, at least one 2-D x-ray image, which is part of a set of x-ray images taken from a 360 degree scan of a patient with a cone beam CT imaging device. The x-ray image contains at least one anatomical structure such as vertebral bodies to be segmented. The received x-ray is then analyzed in order to identify and segment the anatomical structure contained in the x-ray image based on a stored model of anatomical structures. Once the 360 degree spin is completed, a 3-D image volume from the x-ray image set is created. The identification and segmentation information derived from the x-ray image is then added to the created 3-D image volume.

Abstract: In one embodiment, a diagnosis support apparatus includes: an input circuit configured to acquire a first medical image; and processing circuitry configured to generate a second medical image from the first medical image in such a manner that information included in the second medical image is reduced from information included in the first medical image, extract auxiliary information from the first medical image, and perform inference of a disease by using the second medical image and the auxiliary information.

Abstract: A magnetic particle imaging (MPI) and fluorescence molecular tomography (FMT)-fused multimodal imaging system and method for a small animal are provided. The multimodal imaging system includes an image processing module, a display module, a control module, an object table, a gradient coil, a driving coil, a reception coil, a fluorescence camera, and a light source module, where the gradient coil includes a first rounded rectangular coil and a second rounded rectangular coil; the driving coil includes a third rounded rectangular coil, a fourth rounded rectangular coil, and a fifth rounded rectangular coil; and the reception coil is a circular coil.

Abstract: A method for calculating and reporting image quality properties of an image acquisition device after a subject or object has been scanned consists of a scan quality monitoring system with automated software for receiving scans and radiation dose data from scanners and automated algorithms for analyzing image quality metrics and radiation dose tradeoffs. Image quality assessment methods include algorithms for measuring fundamental imaging characteristics, level and type of image artifacts, and comparisons against large databases of historical data for the scanner and protocols. Image quality reports are further customized to report on expected clinical performance of image detection or measurement tasks.

Abstract: Provided is an infrared imaging device capable of reducing streak noise by calculating correction coefficients for correcting streak noise from a single image. The infrared imaging device includes: a thermal image generator to generate a single thermal image on the basis of a signal from a thermal image sensor; a correction coefficient calculator to calculate multiple correction coefficients included in a correction equation for converting pixel values of pixels included in the single thermal image into target values in which streak noise is reduced; and a thermal image corrector to correct a thermal image generated by the thermal image generator by using the multiple correction coefficients, wherein the correction coefficient calculator calculates the multiple correction coefficients on the basis of difference between pixel values of pixels arranged in a direction in which the streak noise occurs in the single thermal image.

Abstract: Disclosed are a cranial CT-based grading method and a corresponding system, which relate to the field of medical imaging. The cranial CT-based grading method as disclosed solves the problems of relatively great subjective disparities and poor operability in eye-balling ASPECTS assessment. The grading method includes: determining frames where target image slices are located from to-be-processed multi-frame cranial CT data; extracting target areas; performing infarct judgment on each target area included in the target areas to output an infarct judgment outcome regarding the target area; and outputting a grading outcome based on infarct judgment outcomes regarding all target areas.

Abstract: Tomographic images acquired by iterative reconstruction of low quality projection images, are enhanced by the steps of correcting at an iteration step the result of the previous iteration step by means of a back-projection of the result of a comparison of a projection image and the forward projection of the result of the previous iteration step whereby this result is enhanced by subjecting it to a trained neural network.

Abstract: Systems and methods for image processing are provided in the present disclosure. The systems and methods may obtain an image; determine a current resolution level of the image; determine, based on the current resolution level of the image, from a group of resolution level ranges, a reference resolution level range corresponding to the image; determine a target processing model corresponding to the reference resolution level range; and/or determine a processed image with a target resolution level by processing the image using the target processing model, the target resolution level of the processed image being higher than the current resolution level of the image.

Abstract: A computer obtains a chest X-ray image, detects boundary lines in the chest X-ray image using a model constructed through machine learning, sets a third lung area including at least one of a first lung area or a second lung area, extracts a vascular index indicating at least one of thickness or density of at least one pulmonary blood vessel present in an area included in the third lung area, determines whether the area included in the third lung area is in an abnormal state on a basis of the vascular index and a reference index based on indices extracted in advance from an area in chest X-ray images in a normal state corresponding to the area included in the third lung area, and outputs, if determining that the area included in the third lung area is in an abnormal state, information indicating a result of the determination.

Abstract: The present disclosure relates to an image processing method. The method may include: obtaining image data; reconstructing an image based on the image data, the image including one or more first edges; obtaining a model, the model including one or more second edges corresponding to the one or more first edges; matching the model and the image; and adjusting the one or more second edges of the model based on the one or more first edges.

Abstract: Techniques for compensating a TFM delay computation live (e.g., during acquisition) as a function of the measured thickness along the scan axis of a probe of an acoustic inspection system. At various scan positions, the acoustic inspection system can measure the thickness of the object under test. With the measured thickness, the acoustic inspection system can compute the delays used for the TFM computation to reflect the actual thickness at that particular scan position of the probe.

Abstract: Provided herein are methods and systems for high-resolution, cerebrospinal fluid-suppressed T2*-weighted magnetic resonance imaging of cortical lesions.

Abstract: Disclosed are devices and techniques based on optical coherence tomography (OCT) technology in combination with optical actuation. A system for providing optical actuation and optical sensing can include an optical coherence tomography (OCT) device that performs optical imaging of a sample based on optical interferometry from an optical sampling beam interacting with an optical sample and an optical reference beam; an OCT light source to provide an OCT imaging beam into the OCT device which splits the OCT imaging beam into the optical sampling beam and the optical reference beam; and a light source that produces an optical actuation beam that is coupled along with the optical sampling beam to be directed to the sample to actuate particles or structures in the sample so that the optical imaging captures information of the sample under the optical actuation.

Abstract: Systems and methods of facilitating determination of risk of coronary artery disease (CAD) based at least in part on one or more measurements derived from non-invasive medical image analysis. The methods can include accessing a non-invasive generated medical image, identifying one or more arteries, identifying, regions of plaque within an artery, analyzing the regions of plaque to identify low density non-calcified plaque, non-calcified plaque, or calcified plaque based at least in part on density, determining a distance from identified regions of low density non-calcified plaque to one or more of a lumen wall or vessel wall, determining embeddedness of the regions of low density non-calcified plaque by one or more of non-calcified plaque or calcified plaque, determining a shape of the more regions of low density non-calcified plaque, and generating a display of the analysis to facilitate determination of one or more of a risk of CAD of the subject.

Abstract: An image processing apparatus comprises processing circuitry configured to: obtain a respective identity and location for each of a first plurality of landmarks in a reference volume; obtain a respective weighting for each landmark of the first plurality of landmarks; obtain a respective identity and location for each of a second plurality of landmarks in a further volume; and determine a transform based on a fit between the first plurality of landmarks and the second plurality of landmarks, wherein the fit is determined in dependence on the weightings.

Abstract: An apparatus for providing brain lesion information based on an image includes a magnetic resonance angiography (MRA) provider configured to provide an environment capable of displaying 3D time-of-flight magnetic resonance angiography (3D TOF MRA) using user input, a brain lesion input unit configured to generate and manage a brain lesion image, a maximum intensity projection (MIP) converter configured to configure MIP image data including at least one image frame corresponding to a projection position of the brain lesion image, a noise remover configured to remove noise of brain lesion information and to configure corrected MIP image data, from which the noise is removed, and an MRA reconfiguration unit configured to reconfigure a corrected brain lesion image by back-projecting the corrected MIP image data.

Abstract: A system for perihematomal edema (PHE) analysis. The system includes a computing device receiving computerized tomography (CT) images from CT imaging devices. The CT images are associated with patients exhibiting perihematomal edema surrounding cerebral hematomas. CT images may be converted into feature vectors and passed as input to a convolution neural network model for identification and diagnosis of perihematomal edema volume changes. Detected changes may be thresholded to determine if the change represents an increase or shrinkage in the volumetry of the perihematomal edema.

Abstract: Disclosed herein is a medical image reading assistant apparatus. The medical image reading assistant apparatus includes a computing system. The computing system includes: a receiver interface configured to receive a first medical image and a second medical image following the first medical image; and at least one processor configured to: generate follow-up information between first findings on the first medical image and second findings on the second medical image by comparing the locations of the first findings and the locations of the second findings; select at least some of the first findings as first valid findings based on a first threshold of diagnostic assistant information for each of the first findings; set a second threshold of diagnostic assistant information for each of the second findings based on the follow-up information; and select at least some of the second findings as second valid findings based on the second threshold.

Abstract: Provided is a magnetic resonance imaging apparatus with no occurrence of artifacts, even after noise removal by applying Wavelet transform to a zero-fill reconstructed image. Nuclear magnetic resonance signals acquired by the magnetic resonance imaging apparatus are processed to perform reconstruction with a reconstruction matrix extended by zero-filling an acquisition matrix, and then a zero-fill reconstructed image is produced. This reconstructed image is subjected to an iterative operation combining the Wavelet transform and L1 norm minimization to remove noise. Before the noise removal, a pre-processing is performed to change the reconstruction matrix size so that an artifact does not occur in the image after noise removal, an artifact portion appears outside the reconstruction matrix after the noise removal, or cutting out is performed so that no artifact appears after the noise removal. The matrix size is restored to its original size in the post-processing after the noise removal.

Abstract: A method for reconstructing target cardiac images is provided. The method may include: obtaining a plurality of projection data corresponding to a plurality of cardiac motion phases; determining a plurality of cardiac motion parameters corresponding to at least a portion of the plurality of cardiac motion phases based on the plurality of projection data; determining a phase of interest based on the plurality of cardiac motion parameters; and/or reconstructing the one or more target cardiac images of the phase of interest.

Abstract: A method for the computing and memory resource-conserving semantic segmentation of image data of an imaging sensor with the aid of an artificial neural network, in particular, a convolutional neural network, the artificial neural network including an encoder path and a decoder path. The method includes: dividing an input tensor into at least one first slice tensor and at least one second slice tensor as a function of a division function, the input tensor being dependent on the image data; outputting the at least one first slice tensor to the decoder path of the neural network; connecting the at least one first slice tensor to the at least one second slice tensor as a function of a connecting function to obtain an output tensor; and outputting the output tensor to the encoder path of the artificial neural network.

Abstract: The present disclosure relates to a method for patient-specific optimization of imaging protocols. According to an embodiment, the present disclosure relates to a method for generating a patient-specific imaging protocol, comprising acquiring scout scan data, the scout scan data including scout scan information and scout scan parameters, generating a simulated image based on the acquired scout scan data, deriving a simulated dose map from the generated simulated image, determining image quality of the generated simulated image by applying machine learning to the generated simulated image, the neural network being trained to generate at least one probabilistic quality representation corresponding to at least one region of the generated simulated image, evaluating the determined image quality relative to a image quality threshold and the derived simulated dose map relative to a dosage threshold, optimizing.

Abstract: Methods and systems are provided for increasing a quality of computed tomography (CT) images reconstructed from high helical pitch scans. In one embodiment, the current disclosure provides for a method comprising generating a first computed tomography (CT) image from projection data acquired at a high helical pitch; using a trained multidimensional statistical regression model to generate a second CT image from the first CT image, the multidimensional statistical regression model trained with a plurality of target CT images reconstructed from projection data acquired at a lower helical pitch; and performing an iterative correction of the second CT image to generate a final CT image.

Abstract: Described herein is a system for 3D proton imaging encompassing both proton radiography (pRad) and proton CT (pCT). The disclosed system can reduce range uncertainties, while providing a fast and efficient check of patient set up and integrated range along a beam's eye view just before treatment. The disclosed system provides a complete solution to the range inaccuracy problem in proton therapy, substantially reducing the uncertainties of treatment planning by directly measuring relative stopping power without being affected by image artifacts and with much lower dose to the patient than comparable x-ray images. Also described herein is a proton imaging algorithm for prompt iterative 3D pCT image reconstruction, where each iteration is fast and efficient, and the number of iterations is minimized. The method offers a unique solution that optimally fits different protons and does not depend on the starting point for the first iteration.

Abstract: Embodiments discussed herein facilitate determination of a response to treatment and/or a prognosis for a tumor based at least in part on features of tumor-associated vasculature (TAV). One example embodiment is a method, comprising: accessing a medical imaging scan of a tumor, wherein the tumor is segmented on the medical imaging scan; segmenting tumor-associated vasculature (TAV) associated with the tumor based on the medical imaging scan; extracting one or more features from the TAV; providing the one or more features extracted from the TAV to a trained machine learning model; and receiving, from the machine learning model, one of a predicted response to a treatment for the tumor or a prognosis for the tumor.

Abstract: An ophthalmologic apparatus includes a refractive power measurement optical system, a fixation projection system, an inspection optical system, and a controller. The refractive power measurement optical system includes a first focusing element capable of changing a focal position, and is configured to project first light onto a subject's eye and to detect returning light of the first light from the subject's eye via the first focusing element. The fixation projection system is configured to project fixation light target onto the subject's eye. The inspection optical system includes a second focusing element capable of changing a focal position in conjunction with the first focusing element, and is used for a predetermined inspection in which second light is projected onto at least the subject's eye via the second focusing element.

Abstract: A control apparatus that controls, using automatic exposure control of controlling a radiation generation apparatus by comparing a radiation dose from the radiation generation apparatus with a target dose, radiation imaging using radiation from the radiation generation apparatus, comprises a processing unit that executes image processing on a radiation image obtained by the radiation imaging; and a setting unit that sets, as the target dose used in the automatic exposure control, a dose which is changed in accordance with an image processing parameter for executing the image processing.

Abstract: A medical image display apparatus according to an embodiment includes processing circuitry. The processing circuitry is configured to: identify imaged ranges respectively corresponding to a plurality of examinations, on the basis of anatomical information from a plurality of medical images taken of an examined subject in the plurality of examinations; map the imaged ranges respectively corresponding to the plurality of examinations onto a single human body model; and cause a display unit to display a list of information about the plurality of examinations and the human body model on which the imaged ranges respectively corresponding to the plurality of examinations are mapped, so as to be kept in correspondence with each other.

Abstract: An x-ray apparatus includes an x-ray source capable of producing an x-ray beam with a focal spot having a spatial shape that is selected to pre-amplify predetermined spatial frequencies exceeding half a cutoff frequency as compared to spatial frequencies below half the cutoff frequency; an x-ray detector capable of detecting the x-ray beam; and a processor adapted to reconstruct an image from the detected x-ray beam using a filter that compensates the pre-amplified predetermined spatial frequencies. The spatial shape comprises two or more disconnected regions that preferably have widths less than that of a nominal focal spot, combined widths greater than or equal to that of a nominal focal spot, and are separated by less than three times a width of a nominal focal spot.

Abstract: A method to reconstruct images from raw MRI data, and recover an image by solving an optimization problem. Hence, the method is an image reconstruction system for image recovery contrary to similar methods, the method reconstructs all images using both joint and individual objective functions. The method includes: acquiring with the MRI system, multiple images under an influence of different contrast mechanisms, wherein the different contrast mechanisms belong to a same anatomy, solving an optimization problem with an optimization algorithm using both joint and individual objective functions simultaneously for each of multiple image contrasts or a subset of the multiple image contrasts, to reconstruct the multiple images from acquired data.

Abstract: The present disclosure discloses a method and an apparatus for classifying a brain anomaly and apparatus based on a 3D MRI image, wherein the classifying method comprises: receiving a to-be-processed 3D MRI image, performing a convolution operation on an imaging sequence corresponding to the 3D MRI image based on a first neural network algorithm to obtain segment masks; and performing a classification operation on the imaging sequence corresponding to the 3D MRI image based on a second neural network algorithm and the segment masks to obtain a classification result of the brain anomaly. Supported by the technologies of artificial intelligence and big data processing, embodiments of the present disclosure enable classification of a brain anomaly shown in the received MRI image through training an MRI recognition model, thereby effectively improving classification accuracy of the brain anomaly and further enhancing diagnosis accuracy of the brain anomaly based on the MRI image.

Abstract: An image processing apparatus includes a generation unit configured to generate a first distance image having a pixel value based on a distance from an outline of a region that indicates a predetermined part of the subject and is extracted from a first image, and having a resolution lower than a resolution of the first image, and generate a second distance image having a pixel value based on a distance from an outline of a region that indicates the predetermined part and is extracted from a second image, and having a resolution lower than a resolution of the second image, a first calculation unit configured to calculate first deformation information by registering the first distance image and the second distance image, and a second calculation unit configured to calculate second deformation information by registering the first image and the second image based on the first deformation information.