Abstract: The disclosure relates to a method and an imaging device for generating a new roadmap image of a target object. For the method, a first subtraction image is generated from a first mask image and a contrast agent image of the target object. A parameter value of a parameter of the imaging device that affects the acquisition of images by the device is then changed. A third mask image of the target object is acquired automatically by the imaging device using the changed parameter value. From a second examination image subsequently acquired using the changed parameter value, a third subtraction image is generated by subtraction of the third mask image. The new roadmap image is generated using the first subtraction image and the third subtraction image.

Abstract: Aspects of the current disclosure include systems and methods for identifying an entity in a query image by comparing the query image with digital images in a database. In one or more embodiments, a query feature may be extracted from the query image and a set of candidate features may be extracted from a set of images in the database. In one or more embodiments, the distances between the query feature and the candidate features are calculated. A feature, which includes a set of shortest distances among the calculated distances and a distribution of the set of shortest distances, may be generated. In one or more embodiments, the feature is input to a trained model to determine whether the entity in the query image is the same entity associated with one of the set of shortest distances.

Abstract: Methods are herein provided for decision support in diagnosis of a disease in a subject, and for extracting features from a multi-slice data set. Systems for computer-aided diagnosis are provided. The systems take as input a plurality of medical data and produces as output a diagnosis based upon this data. The inputs may consist of a combination of image data and clinical data. Diagnosis is performed through feature selection and the use of one or more classifier algorithms.

Abstract: An imaging system, such as a DBT system, capable of providing an operator of the system with information concerning the location, magnitude and direction of motion detected by the system during performance of the scan to enhance image processing. The imaging system provides the motion information to the operator directly in conjunction with the images processed by the imaging system thereby providing the operator with sufficient information for decisions regarding the need for additional images for completing the scan with the imaging system before the patient is discharged, or even before the breast is decompressed.

Abstract: In hemodynamic determination in medical imaging, the classifier is trained from synthetic data rather than relying on training data from other patients. A computer model (in silico) may be perturbed in many different ways to generate many different examples. The flow is calculated for each resulting example. A bench model (in vitro) may similarly be altered in many different ways. The flow is measured for each resulting example. The machine-learnt classifier uses features from medical scan data for a particular patient to estimate the blood flow based on mapping of features to flow learned from the synthetic data. Perturbations or alterations may account for therapy so that the machine-trained classifier may estimate the results of therapeutically altering a patient-specific input feature. Uncertainty may be handled by training the classifier to predict a distribution of possibilities given uncertain input distribution.

Abstract: An automated analysis instrument operates to detect inadequate dispensation of a fluidic substance on a tray. The instrument includes an image capturing device to capture an image of at least a portion of the tray including a receptacle portion and a surrounding portion around the receptacle portion. The instrument then identifies the surrounding portion of the at least the portion of the tray in the image, evaluates color components of the image corresponding to the surrounding portion of the at least the portion of the tray, and determines whether the fluidic substance is present on the surrounding portion of the at least the portion of the tray based on at least one of the color components.

Abstract: A medical image classification method such as CT (or CAT) scans includes receiving the CT scan or medical image, inputting the medical image into an image classification model, which provides a cross entropy (CE) loss function and an aggregated cross entropy (ACE) loss function. According to the ACE loss function, image samples with generic label are used as input data during model training. The medical image can be classified by using the image classification model, and a classification of the medical image is thereby obtained. The present disclosure can classify indeterminate or general medical images and even unlabeled images and thus realize supervision of medical data. A device for applying the method is also provided.

Abstract: A method of machine learning model development includes receiving a plurality of images of a scene, and performing an unsupervised image selection. This includes applying the images to a pre-trained model to extract and embed the images with respective feature vectors, and performing a cluster analysis to group the images in a clusters based on correlations among the respective feature vectors. The unsupervised image selection also includes selecting at least some but not all images in each of the clusters, and any images considered outliers that belong to none of the clusters, for a subset of the images that includes fewer than all of the images. And the method includes receiving user input to label or labeling objects depicted in the subset of the images to produce a training set of images, and building a machine learning model for object detection using the training set of images.

Abstract: Methods and systems for controlling aneurysm initiation or formation in an individual are presented; the technique comprises receiving morphological data of an artery being indicative of at least first and second geometrical parameters of the artery along its trajectory; analyzing the data to identify at least one flow-diverting location along the artery satisfying first and second predetermined conditions of the geometrical parameters; classifying the individual as having or not having disposition for future formation of an aneurysm, depending respectively on whether the at least one flow-diverting location is identified or not and generating classification data; and generating prediction data for the individual with regard to future aneurysm formation.

Abstract: Knowledge of the anatomical, post-operative position and orientation influences not only the selection of the IOL to be implanted but also the result of the refractive operation on the eye. In the method for selecting an IOL to be implanted into an eye on the basis of the prediction of the anatomical, post-operative position (ALP) and orientation thereof, based on pre-operative measuring values such as, for example, anterior chamber depth (VKT), lens thickness (LD) and axial eye length (AL), the invention additionally or exclusively uses the curvature(s) of the eye lens or measuring values derived therefrom. The proposed method is used to predict the anatomical, post-operative position (ALP) and orientation of an intraocular lens (IOL) to be implanted into an eye.

Abstract: An image simulation method for orthodontics is provided. First, receive a face video and capturing a face image from the face video and find a plurality of face feature points on the face image to determine a mouth area. Then, acquire a tooth image in the mouth area. Define a plurality of tooth image feature points according to the tooth image. Subsequently, compare a plurality of preset tooth model feature points of the stereo tooth model and the tooth image feature points, to adjust the preset tooth model feature points to conform to the tooth image feature points and further to form an adjusted stereo tooth model. Then, project the adjusted stereo tooth model to the face video. An image simulation device for orthodontics is also provided.

Abstract: Disclosed is a method for controlling the pattern-width of a coating liquid dispensed from a nozzle. The method includes the steps of: dispensing the coating liquid downward from the nozzle in a dispensing position on a measurement reference line to photograph with a camera positioned at a predetermined horizontal distance from the measurement reference line arranged perpendicularly to a coating reference surface; calculating the pattern-width of the coating liquid on the coating reference surface from image data obtained by photographing the coating liquid dispensed from the nozzle in the dispensing position with the camera; comparing the calculated pattern-width of the coating liquid with a coating liquid reference pattern-width; and adjusting dispensing pressure by increasing or decreasing pressure of air supplied to a pressure container that provides the coating liquid to the nozzle when the calculated pattern-width of the coating liquid is outside the reference allowed error.

Abstract: An automated or semi-automated process to prepare karyotypes from metaphase cell images with improved accuracy involves the use of deep convoluted neural networks for both chromosome segmentation and chromosome classification.

Abstract: Provided is a method of analyzing a dental image for a correction diagnosis and an apparatus using the same. The method includes the steps of obtaining a dental image of an examinee and detecting at least some of a plurality of landmarks for a correction diagnosis in the dental image using a landmark detection module, wherein the landmark is an anatomical reference point indicative of a relative position of at least one of a facial skeleton, a tooth and a face contour necessary for t correction diagnosis, and the landmark detection module may include a machine learning module based on an artificial neural network.

Abstract: A method and an apparatus of screening samples, and a method and an apparatus of searching for service object data are provided. The method of screening samples includes setting up a plurality of screening layers; selecting training samples needed by a current screening layer in the plurality of screening layers; extracting target sample features suitable for the current screening layer from the training samples; determining a screening quantity suitable for the current screening layer using the target sample features; and screening target samples based on the target sample features and the screening quantity. The embodiments of the present disclosure can adaptively adjust respective screening quantities of various screening layers, and can maximally optimize resources, thus balancing effects and performance.

Abstract: An ophthalmic surgical system includes a first imaging system configured to generate a first image of an eye, a second imaging system configured to generate a second image of the eye, and an image registration system to receive the first image generated by the first imaging system, receive the second image generated by a second imaging system, track a location of a distal tip of a surgical instrument in the first image, define a priority registration region in the first image, register the priority registration region in the first image with a corresponding region in the second image, and update the registration of the priority registration region in the first image with the corresponding region in the second image in real time, without registering portions of the first or second images that are outside the registration regions.

Abstract: A method for determining geometric transformation relation for images is provided. The method includes: obtaining a first image and a second image collection; obtaining coordinates of three non-collinear pixels in the first image to forma coordinate collection; generating a target matrix according to the coordinate collection, and for the at least two second images in the second image collection, obtaining coordinates of pixels in the at least two second images corresponding to coordinate-indicative pixels in the coordinate collection to form a corresponding coordinate collection; according to the corresponding coordinate collection, generating a target column vector matching a column number of the target matrix; pre-multiplying the target matrix by the target column vector to obtain a transformation column vector; determining the geometric transformation relation of the first images and the at least two second images according to the transformation column vector.

Abstract: Described herein are systems and methods for extracting image features from magnetic resonance imaging scans. These methods and systems are useful for both determining cancer lesion severity and disease prognosis. In particular, methods and systems are provided herein for extracting and analyzing imaging features that are correlated with breast cancer subtype and prognosis.

Abstract: An image quality control system is disclosed. The example system includes an artificial intelligence modeler to process image data and metadata from patient data to generate first feature(s) from the image data and to parse the metadata to identify study information. The example system includes a computer vision processor to identify second feature(s) in the image data. The example system includes a results evaluator to compare the first feature(s) and second feature(s) to generate a comparison and to evaluate the comparison, first feature(s), and second feature(s) with respect to the study information to generate an evaluation. The example system includes a quality controller to compare the evaluation to quality criterion(-ia) to produce an approval or rejection of the patient data, the approval to trigger release of the patient data and the rejection to deny release of the patient data.

Abstract: A computer-assisted surgery system comprises a calibrating instrument adapted to be applied to a pelvis in a known manner, and a surgical instrument. A computer-assisted processor unit operating a surgical assistance procedure and comprises at least one portable inertial sensor unit configured to be connected to the at least one calibrating instrument and the at least one surgical instrument, the portable inertial sensor unit outputting readings representative of its orientation. A geometrical relation data module provides a geometrical relation data between the orientation of the portable inertial sensor unit, of the calibrating instrument and of the surgical instrument.

Abstract: An end-member extraction method based on segmented VCA, includes: conducting rough segmentation on a hyperspectral image by using an unsupervised classification method to partition image elements having a similar substance into the same block; conducting end-member extraction on an area in each partitioned block by using VCA, inverting the abundance by using a least square method after the end-member extraction, and determining one main end-member for each block according to the abundance value; and extracting the main end-members in all blocks and forming an end-member matrix of a global image. The VCA end-member extraction method is used in relatively simple partitioned environment blocks, and the main end-members in the blocks are then controlled by using the abundance inversion result feedback in the blocks, so as to prevent missing main end-members.

Abstract: The invention is a method for characterizing an object (20), comprising the following steps: a) placing the object between a radiation source (10) and a radiation detector (30); b) irradiating the object with the radiation source and detecting radiation transmitted by the object (14) using the radiation detector, the radiation detector defining a plurality of pixels; c) for each pixel (30i), forming an energy spectrum (Si) of the detected radiation, each spectrum comprising at least two distinct energy bands; d) from each spectrum formed in c), estimating, in each pixel, at least two equivalent thicknesses ({circumflex over (L)}i,1 . . . {circumflex over (L)}i,M,{circumflex over (L)}?i,1 . . . {circumflex over (L)}?i,M) respectively associated with at least two basic materials (mat1 . . . matM,mat?1 . . .

Abstract: Systems and methods are disclosed for anatomic structure segmentation in image analysis, using a computer system. One method includes: receiving an annotation and a plurality of keypoints for an anatomic structure in one or more images; computing distances from the plurality of keypoints to a boundary of the anatomic structure; training a model, using data in the one or more images and the computed distances, for predicting a boundary in the anatomic structure in an image of a patient's anatomy; receiving the image of the patient's anatomy including the anatomic structure; estimating a segmentation boundary in the anatomic structure in the image of the patient's anatomy; and predicting, using the trained model, a boundary location in the anatomic structure in the image of the patient's anatomy by generating a regression of distances from keypoints in the anatomic structure in the image of the patient's anatomy to the estimated boundary.

Abstract: An obstacle distribution simulation method, device and terminal based on multiple models. The method can include: acquiring a point cloud, the point cloud including a plurality of obstacles labeled with real labeling data; extracting the real labeling data of the obstacles, and training a plurality of neural network models based on the real labeling data of the obstacles; extracting unlabeled data in the point cloud, inputting the unlabeled data into the neural network models, and outputting a plurality of prediction results. The plurality of prediction results can include a plurality of simulated obstacles with attribute data; selecting at least one simulated obstacle based on the plurality of prediction results; and inputting the attribute data of the selected simulated obstacle into the neural network models to obtain position coordinates of the simulated obstacle, and further obtain a position distribution of the simulated obstacle.

Abstract: A method and apparatus for segmenting a two-dimensional image of an anatomical structure includes acquiring (202) a three-dimensional model of the anatomical structure. The three-dimensional model includes a plurality of segments. The acquired three-dimensional model is adapted to align the acquired three-dimensional model with the two-dimensional image (204). The two-dimensional image is segmented by the plurality of segments of the adapted three-dimensional model.

Abstract: An object recognition and data retrieval system interfaces with a target system, such as an automated storage cabinet, to share object data for storage and retrieval.

Abstract: A method and apparatus that control lateral movement of the vehicle during backward motion are provided. The method includes loading a desired backward path of vehicle, the backward path comprising waypoints to be traveled along during a rearward motion of the vehicle, reflecting the waypoints along a reflection axis perpendicular to a longitudinal axis that runs from front to back of the vehicle such that the reflected waypoints define virtual forward path; and controlling lateral movement of the vehicle to follow the waypoints along the forward path while the vehicle is traveling in a backward direction.

Abstract: Disclosed herein is a camera calibration method based on deep learning including acquiring an image captured by a camera and predicting an intrinsic parameter of the camera by applying, to the acquired image, a neural network module trained to predict the intrinsic parameter.

Abstract: A medical image processing apparatus comprises processing circuitry configured to: acquire a series of medical imaging data sets acquired across multiple time phases, wherein the medical imaging data sets are representative of a first body part and a second body part in the vicinity of the first body part, and wherein the first body part and second body part undergo relative motion across the multiple time phases; and automatically determine a viewing surface for rendering a series of images from the medical imaging data sets, wherein the determining of the viewing surface is based on respective positions of the first body part and second body part in each of the series of medical imaging data sets.

Abstract: A post-processing leakage correction system and method that accounts for bidirectional contrast agent transport between the intravascular and interstitial spaces that commonly occurs in angiogenic high-grade gliomas without a substantial increase in post-processing computation time.

Abstract: A method includes retrieving from a memory a stored sensitivity table that associates magnetic position sensor readings with measured magnetic fields. One or more calibration values for the magnetic position sensor are estimated during a catheterization procedure in which a magnetic position sensor, fitted at a distal end of a catheter, is placed in an organ of a patient, based on (i) the stored sensitivity table and (ii) readings acquired by the magnetic position sensor while in the organ. Based on the one or more calibration values, a location of the distal end in the organ is magnetically tracked.

Abstract: Disclosed are method and apparatus for SAR image recognition based on multi-scale features and broad learning. A region of interest of an original SAR image is extracted by centroid localization, the image is rotated and added with noise for enhancing the data volume, the image is downsampled, LBP features and PPQ features are extracted, an LBP feature vector XLBP and an LPQ feature vector XLPQ are cascaded to achieve dimension reduction by principal component analysis to obtain a fusion feature data Xm, the fusion feature data Xm is input to a broad learning network for image recognition and a recognition result is output. By fusing the LBP features and the LPQ features, complementary information is fully utilized and redundant information is reduced. The broad learning network is used to improve the training speed and reduce the time cost. As a result, the recognition effect is more stable, robust and reliable.

Abstract: Systems and methods for performing quantitative histopathology analysis for determining tissue potency are disclosed. According to some embodiments, a method training a tissue classifier is provided. According to the method, training the tissue classifier includes generating feature fingerprints of detected nuclei within slide images in a control library and clustering the slide images based on their corresponding feature fingerprints. According to some embodiments, a method for utilizing the trained tissue classifier is provided. According to the method, the trained tissue classifier determines whether tissue in an unknown slide image corresponds to slide images clustered during the training of the tissue classifier.

Abstract: A method and device for determining a flow situation in a vessel are disclosed. According to an embodiment of the method, a first image data set containing image information relating to the vessel is used and a vascular tree of the vessel is segmented based upon the first image data set. An organ is also segmented based upon the first image data set or of a second image data set and the organ is assigned at least one area of a parenchyma of the organ. Via texture analysis, a texture of the area of the organ is determined and, depending on the texture, the area of the organ is assigned a flow characteristic. Depending on the vascular tree and the flow characteristic assigned to the area of the organ, a value of a measured variable characteristic of the flow situation within the vessel is then determined via a numerical method.

Abstract: A radiographic image display apparatus 3 included in a radiographic imaging system 100 includes a hardware processor that obtains an image data item on each of the frame images generated by the radiographic imaging apparatus 2, detects a situation of the subject at least at a time point in the dynamic imaging, associates the detected situation of the subject with the obtained image data items, and issues a specific output for notification that the subject is in a specific situation when the subject is in a state of a specific frame image fs, in a case where the display is caused to display the specific frame image fs, this specific frame image fs being taken when it is detected that the situation of the subject is the specific situation.

Abstract: An information processing method includes: obtaining noise region estimation information output from a first converter by a first image including a noise region being input to the first converter; obtaining a second image, on which noise region removal processing has been performed, output from a second converter by the noise region estimation information and the first image being input to the second converter; generating a fourth image including the estimated noise region by using the noise region estimation information and a third image including no noise region and a scene corresponding to the first image; training the first converter by using machine learning in which the first image is reference data and the fourth image is conversion data; and training the second converter by using machine learning in which the third image is reference data and the second image is conversion data.

Abstract: A method includes obtaining contrast enhanced spectral image data that includes voxels representing a tubular structure. The method further includes generating at least a contrast map based on the obtained contrast enhanced spectral image data. The method further includes generating an updated contrast map based on a spectral model. The method further includes segmenting the tubular structure based on updated contrast map. A computing system (120) includes a spectral analyzer (202) that receives contrast enhanced spectral image data and generates a spectral analysis data based thereon, wherein the spectral analysis data includes a contrast map, The computing system further includes a spectral analysis data processor (204) that refines the spectral analysis data, generating refined spectral analysis data.

Abstract: A facility providing a medical outcome prediction model data structure is described. The data structure constitutes a trained statistical model that can be applied to image data and electronic health record data for a patient to predict a cancer survival outcome for the patient.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing fundus images using fundus image processing machine learning models. One of the methods includes obtaining a model input comprising one or more fundus images, each fundus image being an image of a fundus of an eye of a patient; processing the model input using a fundus image processing machine learning model, wherein the fundus image processing machine learning model is configured to process the model input comprising the one or more fundus image to generate a model output; and processing the model output to generate health analysis data.

Abstract: A system for embedding or attaching to one or more attributes to a graphical object may is disclosed. The system includes a user device configured to obtain a graphical object. The graphical object includes at least one of an image or a geolocation. The system also includes a server communicatively coupled to the user device. The server includes a marker subsystem configured to create at least one marker on the graphical object. The server also includes an embedding subsystem configured to embed one or more attributes in the graphical object. The one or more attributes includes a video, an augmented reality video, a 3-dimensional content, a hyperlink or a text. The server further includes a queue manager configured to process and upload the embedded graphical object. The server further includes an alert subsystem configured to generate a notification upon uploading of the embedded graphical object.

Abstract: The present disclosure relates to a method and a system for detecting an anomaly within a biological tissue. A first image of the biological tissue is obtained, the first image containing light at a first wavelength. A second image of the biological tissue is obtained, the second image containing light at a second wavelength. A texture analysis of the biological tissue is performed using spatial information of the first and second images. The texture analysis is resolved over the first and second wavelengths.

Abstract: A medical image processing apparatus comprises processing circuitry configured to: obtain volumetric data corresponding to a three dimensional region of a subject; perform a sampling process that comprises sampling the volumetric data at a plurality of sampling points along a plurality of sampling paths; for each of a plurality of intervals between sampling points on sampling paths, determine a position of a shading evaluation point in the interval based on intensity values and/or opacity values for a front sampling point and back sampling point of the interval; and determine a value for a shading parameter at the determined shading evaluation point; and perform a rendering process using the determined values for the shading parameter.

Abstract: An anatomic structure of interest is contoured in 3D source data by selecting a first subset of data in a first image slice, which has a first orientation in the source data. A first set of instructions identifies a first edge (E1) of the anatomic structure of interest in the first image slice. Then, a second subset of data is selected in a second image slice, which has a second orientation in the source data. A second set of instructions identifies a second edge (E2) of the anatomic structure of interest in the second image slice. A three-dimensional shell (3DS) is calculated based on the first and second edges (E1; E2) and the source data (SD). The three-dimensional shell (3DS) represents an approximation of a delimiting surface that separates the anatomic structure of interest from adjoining tissues in the 3D source data.

Abstract: An image processing apparatus 10 includes a first generation unit 11 which generates a plurality of lower-resolution images having different row-direction resolutions by changing a resolution in a row direction of an image to be processed to a plurality of lower resolutions, a second generation unit 12 which generates a differential image by taking a difference between two of the plurality of lower-resolution images, and a computation unit 13 which computes a correction amount for the pixel values of a predetermined column of the image to be processed by use of a statistical quantity of the pixel values of a predetermined column of the differential image.

Abstract: A method for determining a local tissue function of tissue in a body region of interest of an examination object is disclosed. In an embodiment, the method includes segmentation of an outer contour of the tissue using at least one medical image recording representing the body region of interest of the examination object comprising the tissue; subdivision of the segmented tissue into at least two tissue regions; and ascertaining a function parameter relating to the tissue function for each of the at least two tissue regions. Embodiments also relate to a corresponding computing unit for determining a tissue function of tissue, a corresponding medical imaging system, a computer program and a computer-readable data carrier.

Abstract: A system for processing medical image data stores high information content volumetric (HICV) data reconstructed from different medical imaging devices. Images normally requiring differing image imaging parameters can instead be created by processing of the HICV data, importantly without access to the original reconstruction algorithms or machine. The invention also allows different medical images to be harmonized for ready comparison or according to a desired image sequence for particular diagnostic purpose without any need to rescan the patient or to enforce cumbersome universal protocol standards.

Abstract: The image acquisition unit acquires an image captured by an endoscope. The extraction unit extracts a blood vessel of an observation target. The index value calculation unit calculates a plurality of blood vessel index values based on the blood vessel. The determination unit determines whether each of the plurality of blood vessel index values is a normal value or an abnormal value. The color information setting unit sets color information for an abnormal index value, which is the blood vessel index value determined to be the abnormal value by the determination unit, or sets color information for a normal index value, which is the blood vessel index value determined to be the normal value by the determination unit. The image generation unit generates an emphasized image, in which the blood vessel is emphasized, based on the color information.

Abstract: An example method includes obtaining an image of a lead implanted in a patient, the lead including one or more electrodes positioned along a longitudinal axis of the lead and a plurality of orientation markers; determining, in the image, respective locations of the one or more electrodes and respective locations of the plurality of orientation markers; determining, based on the respective locations of the one or more electrodes, an orientation of the longitudinal axis; projecting the respective locations of the orientation markers into a plane orthogonal to the longitudinal axis; and determining, based on a projected location of a first orientation marker of the plurality of orientation markers in the plane and a projected location of a second orientation marker of the plurality of orientation markers in the plane, a rotational orientation of the lead.

Abstract: In a console, a control unit functions as an acquisition unit that acquires a radiographic image of a breast, a derivation unit that derives a mammary gland content rate for each pixel of a breast region in the radiographic image, and a detection unit that detects a mammary gland concentrated region in which mammary glands are concentrated on the basis of a result of specifying whether a specific pixel which is each pixel of the breast region is a pixel included in the mammary gland concentrated region of the breast region on the basis of the mammary gland content rate of the specific pixel and a mammary gland content rate of a pixel in a local region around the specific pixel.

Abstract: A system and method for improving ablation procedures is presented. The method may comprise measuring EGM signals; performing three dimensional mapping of the EGM signals; and detecting a location for ablation based on the mapping. The method may further comprise creating a potential duration map, a local activation time map, and a bipolar voltage map. The method may further comprise creating a PDM by defining a window of interest (WOI) comprising at least a cycle length; calculating previous heart beats based on the cycle length and reference annotation; and assigning the heart beats within the cycle length of the WOI to the WOI. The method may further comprise detecting a location for ablation by finding start and end potential durations for the WOI; calculating a potential duration value using the start and end potential durations; selecting an ablation point; and using the selected ablation point as the detected location.