Abstract: A method for obtaining a medical image at a target plane. The method comprises obtaining a three-dimensional, 3D, volume containing an anatomical structure and identifying the anatomical structure in the 3D volume. The geometry of the anatomical structure in the 3D volume is analyzed and the target plane is registered to the 3D volume based on the analysis of the geometry of the anatomical structure. The medical image is extracted from the 3D volume at the registered target plane.

Abstract: A method for acquiring a set of target planar views through an anatomy of interest through processing of 3D ultrasound image data. The method processes received 3D image frames in real time to extract target planes and to process each extracted plane with a quality check algorithm. A record is kept of all target planes which have been acquired with sufficient quality, for example by removing them from a list of planes still to be captured. Optionally, each new frame need only be processed to extract planes which are yet to be acquired with sufficient quality. A user interface display is dynamically controlled so that a user is always presented with a real-time display of a planar view of at least one of the planes still to be captured with sufficient quality. The user is thus continually provided the optimal guidance imagery for controlling the acquisition apparatus to acquire new 3D frames which include image data from which the still required 2D planes can be successfully extracted with high quality.

Abstract: A medical imaging and segmentation method comprising utilizing segmentation results obtained for 3D frames earlier in a same imaging session to initialize a segmentation model for subsequent 3D frames. This accelerates and stabilizes segmentation results, particularly in the context of a scanning procedure in which segmentation is performed in real time (on-the-fly) for each acquired frame.

Abstract: The present invention relates to a device (3) and a method for generating a visualization of an anatomical volume of interest.

Abstract: It is an object of the invention to provide an apparatus (110) that allows for assisting a user in determining an improved subdivision into arterial feeding regions in an organ of a patient (121). An image providing unit (111) provides an image of an organ of a subject. A geodesic distance map generation unit (113) generates a geodesic distance map for a region of interest based on the anatomical image, wherein a voxel of the geodesic distance map is given by a cumulated path cost from a predetermined seed to the respective voxel and wherein the path cost is determined based on a cost function taking into account a vesselness filter response and/or a contrast agent density determined, respectively, based on the image. A geodesic distance map providing unit (114) provides the geodesic distance map to a user and/or to further processing to determine the arterial feeding regions.

Abstract: A method of spectral clustering includes: obtaining a set of coronary image data comprising voxels of a volume of interest around a heart; processing the voxels to rank the voxels for likelihood of characterizing a vessel; identifying similarities between pairs of voxels to quantify the strength of a link between the voxels in each pair of voxels; spectral clustering the voxels based on the likelihood of characterizing a vessel and the strength of links between the pairs of voxels; selecting at least one subtree of voxels based on the spectral clustering; classifying each subtree based on the set of coronary image data; and reconstructing a representation of at least one vessel in the volume of interest as a first reconstruction to include at least one subtree classified based on the set of coronary image data.

Abstract: A mechanism for determining a score indicative of a success of a segmentation of a 3D image, i.e. a success score. The mechanism proposes to obtain one or more 2D images of different target views of a target object in the 3D image, by processing a segmentation result of the 3D image. (A view of) each 2D image is classified using an automated classifier. The classification results are used to determine a success score, which may indicate, for example, whether, or how closely, the 3D segmentation result represents a ground truth segmentation result with sufficient accuracy, e.g. for clinical decision making.

Abstract: A method and system for providing saliency maps for a deep learning model. The method comprises inputting an input image into the deep learning model trained to output a metric score, from a plurality of metric scores, for the input image. A supportive saliency map is generated for the input image, from the deep learning model, corresponding to a first range of the metric scores for the image, thereby providing one or more supportive regions of the image indicative of the first range of metric scores. A distractive saliency map is also generated for the image, from the deep learning model, corresponding to a second range of the metric scores for the image, thereby providing one or more distractive regions of the image indicative of the second range of metric scores.

Abstract: The present invention relates to a device and method for segmenting the membranous septum. The method comprises segmenting the left ventricular outflow tract, LVOT, of the heart in a 3D image data set; determining wall thickness information indicative of the wall thickness of the septum at different locations of a part of the segmented LVOT that is oriented towards the right heart chambers and right atrium; mapping the determined wall thickness information onto the surface of said part of the segmented LVOT; and segmenting the membranous septum in the 3D image data set based on the mapped wall thickness information.

Abstract: A system and method for achieving more accurate results when applying an image processing task to a series of medical images of a patient, without significantly increasing processing resource. The proposed system and method is based on receiving a plurality of image sequences of a particular anatomical region, each capturing cyclical movement of an anatomical object. Each image sequence is supplied to a classifier module which employs use of one or more machine learning algorithms to derive at least one score for each image sequence indicative of predicted success or quality of a result of the image processing task if applied to the given image series. This permits an assessment to be made in advance of which of the plurality of image series is most likely to result in the best (e.g. highest quality, or greatest amount of information) results from the image processing task.

Abstract: A method is provided for adapting a 3D field of view (FOV) in ultrasound data acquisition so as to minimize the FOV volume in a manner that is controlled and precise. The method comprises defining a volumetric region across which 3D ultrasound data is desired, and then adapting the data acquisition field of view (FOV) in dependence upon the defined volumetric region, to encompass the region. This is achieved based on adapting a scan line length (or scan depth) of each individual scan line based on the defined volumetric region. In some embodiments, the volumetric region may be defined based on anatomical segmentation of a reference ultrasound dataset acquired in an initial step, and setting the volumetric region in dependence upon boundaries of an identified object of interest. The volumetric region may in a subset of embodiments be set as the region occupied by a detected anatomical object of interest.

Abstract: A method and system for automatically detecting within an input 3D ultrasound image one or more suspected disease features and automatically computing an optimum one or more planes through the 3D field of view for acquiring imagery that would best assist in confirming or further analyzing the suspected disease features. The determined optimum one or more planes are used to control acquisition by an ultrasound acquisition system of new 2D images which correspond to said determined optimum planes.

Abstract: The invention provides a method for determining a confidence value for an image segmentation. The method includes obtaining an image, wherein the image comprises a view of an anatomical structure and a model of the anatomical structure is obtained, wherein the model comprises a plurality of nodes. The image is processed to generate a plurality of image segmentation outputs, wherein each image segmentation output comprises a set of values for the view, wherein each value of the set of values is associated with a node of the plurality of nodes of the model. For each node of the model, a confidence value is determined based on the plurality of values corresponding to the node. A confidence map of the anatomical structure is generated based on the confidence value of each node.

Abstract: The present invention relates to multispectral imaging. In order to improve an identification of relevant multispectral material transitions (in particular caused by injected contrast agent), an apparatus is proposed to use the local maxima of the variances and/or covariances of the intensities of the multi-channel images to locate material transitions. In comparison to gradient vectors, the local variance is not directed and not prone to noise. An alternative apparatus is proposed to use the local covariance deficits of the intensities of the multi-channel images to locate material transitions. The proposed alternative approach is independent of spatial drifts across the image volume.

Abstract: A mechanism for creating/synthesizing realistic training data, for training a machine-learning model, using anatomical knowledge. An anatomical model can be obtained. Information from annotated training data entries (i.e. “ground truth” information), can be used to model the anatomical variation, from the obtained model, in the population of the training data. This anatomical model can then be modified, e.g. incorporating some random factors, in order to generate one or more augmented models of realistic anatomies. The augmented anatomy is then transferred from the model domain to the data entry domain to thereby generate a new data entry or data entries for training a machine-learning model. This latter process can be achieved in various ways, e.g. using GANs, such as CycleGANs and label images, or deformation vector fields.

Abstract: The invention relates to an image processing device (3) comprising a data input unit (4) for receiving volumetric medical image data (7) organized in voxels and processing unit (5), wherein the volumetric medical image data (7) os spectral computed tomography data. The processing unit (5) is adapted to perform an automatic anatomical shape model segmentation (8) on the volumetric medical image data (7). It is further adapted to perform a determination of a first layer of interest (9) and of a second layer of interest (11). A first projection (10) of the first layer of interest (9) is performed, yielding perfusion information data, and a second projection (14) of the second layer of interest (11) is performed, yielding vascular information data. Finally. a graphical combination (15) of the perfusion information data and the vascular information data is performed, yielding combined information data (21).

Abstract: The invention provides a method for refining a mapped surface mesh of a cardiac chamber. The method includes obtaining a mapped surface mesh of the cardiac chamber anatomy, wherein the mapped surface mesh comprises a central region representing a cardiac chamber and an outer region representing a peripheral cardiac structure connected to the cardiac chamber, and wherein the mapped surface mesh comprises a first view of an anatomical landmark within the cardiac chamber, and obtaining image data of a cardiac chamber anatomy of a subject. The central region of the mapped surface mesh is deformed based on a first segmentation algorithm configured according to one or more predetermined shape-constraints and the outer region of the mapped surface mesh is deformed based on a second segmentation algorithm configured according to the image data, thereby generating a deformed outer region. The deformed central region and the deformed outer region are then combined, thereby generating a refined mapped surface mesh.

Abstract: A computer implemented method of identifying changes in a subject's heart or an adjacent region over time. The method comprising: receiving a set of imaging data relating to a subject's heart that has been obtained at a plurality of points in time; generating an anatomical model of the subject's heart for each of the images in the set of imaging data so as to provide a set of anatomical models of the subject's heart corresponding to the plurality of points in time; and aligning each of the anatomical models in the set of anatomical models relative to one another so as to provide a set of aligned data of the subject's heart. The aligned data are for identifying changes in at least one region of the subject's heart by comparing the anatomical models in the set of aligned anatomical models using a machine learning model.

Abstract: The invention relates to a system and computer-implemented method for enabling correction of a segmentation of an anatomical structure in 3D image data. The segmentation may be provided by a mesh which is applied to the 3D image data to segment the anatomical structure. The correction may for example involve a user directly or indirectly selecting a mesh part, such as a mesh point, that needs to be corrected. The behaviour of the correction, e.g., in terms of direction, radius/neighbourhood or strength, may then be dependent on the mesh normal direction, and in some embodiments, on a difference between the mesh normal direction and the orientation of the viewing plane.

Abstract: A method is provided for adapting a 3D field of view (FOV) in ultrasound data acquisition so as to minimize the FOV volume in a manner that is controlled and precise. The method comprises defining a volumetric region across which 3D ultrasound data is desired, and then adapting the data acquisition field of view (FOV) in dependence upon the defined volumetric region, to encompass the region. This is achieved based on adapting a scan line length (or scan depth) of each individual scan line based on the defined volumetric region. In some embodiments, the volumetric region may be defined based on anatomical segmentation of a reference ultrasound dataset acquired in an initial step, and setting the volumetric region in dependence upon boundaries of an identified object of interest. The volumetric region may in a subset of embodiments be set as the region occupied by a detected anatomical object of interest.