Abstract: An image segmentation method is disclosed that allows a user to select image component types, for example tissue types and or background, and have the method of the present invention segment the image according to the user's input utilizing the superpixel image feature data and spatial relationships.

Abstract: Disclosed, among other things, is a computer device and computer-implemented method of classifying cells within an image of a tissue sample comprising providing the image of the tissue sample as input; computing nuclear feature metrics from features of nuclei within the image; computing contextual information metrics based on nuclei of interest with the image; classifying the cells within the image using a combination of the nuclear feature metrics and contextual information metrics.

Abstract: The subject disclosure presents systems and methods for improved meso-dissection of biological specimens and tissue slides including importing one or more reference slides with annotations, using inter-marker registration algorithms to automatically map the annotations to an image of a milling slide, and dissecting the annotated tissue from the selected regions in the milling slide for analysis, while concurrently tracking the data and analysis using unique identifiers such as bar codes.

Abstract: Described herein are methods for co-expression analysis of multiple markers in a tissue sample comprising: computing a heat map of marker expression for each of a plurality of single marker channel images, wherein each of the plurality of single marker channel images comprise a single marker; identifying one or more candidate regions of interest in each heat map of marker expression; computing overlay masks comprising the identified one or more candidate regions of interest from each heat map of marker expression; determining one or more co-localized regions of interest from the overlay masks; mapping the one or more co-localized regions of interest to a same coordinate position in each of the plurality of single marker channel images; and estimating a number of cells in at least one of the determined one or more co-localized regions of interest in each of the plurality of single marker channel images.

Abstract: Embodiments disclosed herein are directed, among other things, to imaging systems, methods, and apparatuses for automatically identifying fields of view (FOVs) for regions in an image encompassing tumor are disclosed. In embodiments and in further aspects of the present invention, a computer-implemented method is disclosed for a tumor region based immune score computation. The method, in accordance with the present invention, involves identifying regions, for example, tumor areas or regions around a tumor area, partitioning a whole slide image or portion of a whole slide image into multiple regions related to the tumor, selecting FOVs within each identified region, and computing a number of cells present in each FOV. An immune score and/or immune-related score may be generated based on the cells counted in each FOV.

Abstract: Methods, systems, and apparatuses for detecting and describing heterogeneity in a cell sample are disclosed herein. A plurality of fields of view (FOV) are generated for one or more areas of interest (AOI) within an image of the cell sample are generated. Hyperspectral or multispectral data from each FOV is organized into an image stack containing one or more z-layers, with each z-layer containing intensity data for a single marker at each pixel in the FOV. A cluster analysis is applied to the image stacks, wherein the clustering algorithm groups pixels having a similar ratio of detectable marker intensity across layers of the z-axis, thereby generating a plurality of clusters having similar expression patterns.

Abstract: The subject disclosure presents systems and computer-implemented methods for automatic immune cell detection that is of assistance in clinical immune profile studies. The automatic immune cell detection method involves retrieving a plurality of image channels from a multi-channel image such as an RGB image or biologically meaningful unmixed image. A cell detector is trained to identify the immune cells by a convolutional neural network in one or multiple image channels. Further, the automatic immune cell detection algorithm involves utilizing a non-maximum suppression algorithm to obtain the immune cell coordinates from a probability map of immune cell presence possibility generated from the convolutional neural network classifier.

Abstract: Methods, systems, and apparatuses for automatically identifying glandular regions and tubule regions in a breast tissue sample are provided. An image of breast tissue is analyzed to detect nuclei and lumen candidates, identify tumor nuclei and true lumen from the candidates, and group tumor nuclei with neighboring tumor nuclei and lumina to define tubule glandular regions and non-tubule glandular regions of the image. Learnt supervised classifiers, such as random forest classifiers, can be applied to identify and classify the tumor nuclei and true lumina. Graph-cut methods can be applied to group the tumor nuclei and lumina and to define the tubule glandular regions and non-tubule glandular regions. The analysis can be applied to whole slide images and can resolve tubule areas with multiple layers of nuclei.

Abstract: Systems and methods for generating a locally adaptive threshold image for foreground detection performing operations including creating a saliency edge strength image or layer indicating edge or border pixels of the nuclei by performing tensor voting on pixels neighboring the initial edge pixels within an image region to refine true edges are featured. Further, for each of a plurality of regions or blocks of the image, an adaptive threshold image is determined by sampling a foreground pixel and a background pixel for each initial edge pixel or refined edge pixel, generating histograms for both background and foreground saliency (or gradient magnitude) modulated histograms, determining a threshold range for each block of the image, and interpolating the threshold at each pixel based on the threshold range at each block. Comparing the input image with the resulting locally adaptive threshold image enables extraction of significantly improved foreground.

Abstract: A method for reconstructing a three-dimension image includes receiving a plurality of two-dimensional images and projection information of the two-dimensional images, projecting a plurality of rays onto the plurality of two-dimensional images, determining correspondence information between pixels of different ones of the plurality of two-dimensional images, determining a value of each of the pixels, and reconstructing a three-dimension image by integrating the plurality of rays, wherein a position on each ray can be associated to one pixel of the plurality of two-dimensional images.

Abstract: A method of segmenting images of biological specimens using adaptive classification to segment a biological specimen into different types of tissue regions. The segmentation is performed by, first, extracting features from the neighborhood of a grid of points (GPs) sampled on the whole-slide (WS) image and classifying them into different tissue types. Secondly, an adaptive classification procedure is performed where some or all of the GPs in a WS image are classified using a pre-built training database, and classification confidence scores for the GPs are generated. The classified GPs with high confidence scores are utilized to generate an adaptive training database, which is then used to re-classify the low confidence GPs. The motivation of the method is that the strong variation of tissue appearance makes the classification problem more challenging, while good classification results are obtained when the training and test data origin from the same slide.

Abstract: A method and system for magnetic resonance imaging (MRI) based motion correction in position emission tomography (PET) images is disclosed. A static 3D magnetic resonance (MR) image of a patient is received. PET image data of the patient and a series of 2D MR images of the patient acquired simultaneous to the acquisition of the PET image data are received. A 3D+t motion field is estimated by registering the series of 2D MR images acquired at the plurality of time points to the static 3D MR image. A motion corrected PET image is generated based on the estimated 3D+t motion field using motion corrected PET reconstruction.

Abstract: Multiple users are supported with an ultrasound server. Tiling of images may be used to limit transmission and/or bandwidth. By transmitting parts of images that change and avoiding transmission of other parts, wireless and processing bandwidth may be optimized. On the server side, separate instances are used for scanning each patient or for each of the multiple transducer probes being used. Dynamic assignment of shared resources based on use of the transducer probes may provide further optimization. From an overall perspective, the server may beamform from data received by a transducer probe based on controls routed from a separate tablet used as a display and user input.

Abstract: A method (100) that generates attenuation correction maps for the reconstruction of PET data using MR images, such as, MR ultra-fast TE (UTE) images, Dixon MR images, as well as MR images obtained using other MR imaging methods.

Abstract: A method (100) that registers a 3D heart volume (112, 114) obtained from either a pre-operative MR image or CT image (102) to an intra-operative fluoroscopic image using a mesh of the heart structure (106) as the basis for the registration.

Abstract: A method and system for magnetic resonance imaging (MRI) based motion correction in position emission tomography (PET) images is disclosed. A static 3D magnetic resonance (MR) image of a patient is received. PET image data of the patient and a series of 2D MR images of the patient acquired simultaneous to the acquisition of the PET image data are received. A 3D+t motion field is estimated by registering the series of 2D MR images acquired at the plurality of time points to the static 3D MR image. A motion corrected PET image is generated based on the estimated 3D+t motion field using motion corrected PET reconstruction.

Abstract: A method for automatic initialization of 2D to 3D image registration includes acquiring a 3D model. A plurality of shape descriptor features is calculated from the acquired 3D model representing a plurality of poses of the 3D model. A 2D image is acquired. The plurality of shape descriptors is matched to the acquired 2D model. An optimum pose of the 3D model is determined based on the matching of the plurality of shape descriptors to the acquired 2D model. An initial registration is generated, in an image processing system, between the 3D model and the 2D image based on the determined optimum pose.

Abstract: A method for performing motion compensation in a series of magnetic resonance (MR) images includes acquiring a set of MR image frames spanning different points along an MR recovery curve. A motion-free synthetic image is generated for each of the acquired MR image frames using prior knowledge pertaining to an MR recovery curve. Each of the acquired MR images is registered to its corresponding generated synthetic images. Motion within each of the acquired MR image is corrected based on its corresponding generated synthetic image that has been registered thereto.

Abstract: A method for generating a pseudo-computed tomography (CT) image volume includes acquiring a first magnetic resonance (MR) image volume (UTE1) using an ultra-short echo time and acquiring a second MR image volume (UTE2) using a conventional echo time that is longer than the ultra-short echo time. The acquired UTE1 and UTE2 image volumes are normalized. A mask for an anatomical structure featured in the normalized UTE1 and UTE2 image volumes is created and bone regions are segmented from the normalized UTE1 and UTE2 image volumes using the created mask and one or more trained classifiers. A pseudo-CT image is constructed from the normalized UTE1 and UTE2 image volumes, the created mask, and the segmented bone regions.

Abstract: A method and system for data dependent multi phase image visualization, includes: acquiring a plurality of series of image data acquisitions; registering the plurality of series of image data acquisitions to a same reference series to create a plurality of registered series; combining information from the registered series to create a new series; creating a further new series by a selection decision based on combination rules from information from the plurality of registered series and the new series; and displaying the further new series.