Abstract: An artificial intelligence platform and associated methods of training and use are disclosed. An example apparatus includes a data pipeline to: preprocess data using one or more preprocessing operations applied to features associated with the data; and enable debugging to visualize the preprocessed data. The example apparatus includes a network to: instantiate one or more differentiable operations in a training configuration to train an artificial intelligence model; capture feedback including optimization and loss information to adjust the training configuration; and store one or more metrics to evaluate performance of the artificial intelligence model.

Abstract: Techniques are provided for deep neural network (DNN) identification of realistic synthetic images generated using a generative adversarial network (GAN). According to an embodiment, a system is described that can comprise a memory that stores computer executable components and a processor that executes the computer executable components stored in the memory. The computer executable components can comprise, a first extraction component that extracts a subset of synthetic images classified as non-real like as opposed to real-like, wherein the subset of synthetic images were generated using a GAN model.

Abstract: Systems and techniques for providing concurrent image and corresponding multi-channel auxiliary data generation for a generative model are presented. In one example, a system generates synthetic multi-channel data associated with a synthetic version of imaging data. The system also predicts multi-channel imaging data and the synthetic multi-channel data with a first predicted class set or a second predicted class set. Furthermore, the system employs the first predicted class set or the second predicted class set for the synthetic multi-channel data to train a generative adversarial network model.

Abstract: Techniques are provided for evaluating and defining the scope of data-driven deep learning models. In one embodiment, a machine-readable storage medium is provided comprising executable instructions that, when executed by a processor, facilitate performance of operations comprising employing a machine learning model to extract first training data features included in a training data set and first target data features included in a target data set. The operations further comprise determining whether the target data set is within a defined data scope of the training data set based on analysis of correspondences between the first training data features and the first target data feature, and determining whether application of the target data set to a target neural network model developed using the training data set will generate results with an acceptable level of accuracy based on whether the target data set is within the defined data scope.

Abstract: Systems and computer-implemented methods for facilitating automated compression of artificial neural networks using an iterative hybrid reinforcement learning approach are provided. In various embodiments, a compression architecture can receive as input an original neural network to be compressed. The architecture can perform one or more compression actions to compress the original neural network into a compressed neural network. The architecture can then generate a reward signal quantifying how well the original neural network was compressed. In (?)-proportion of compression iterations/episodes, where ??[0,1], the reward signal can be computed in model-free fashion based on a compression ratio and accuracy ratio of the compressed neural network. In (1??)-proportion of compression iterations/episodes, the reward signal can be predicted in model-based fashion using a compression model learned/trained on the reward signals computed in model-free fashion.

Abstract: Systems, methods and apparatus are provided through which in some implementations changes in an aneurysm in a patient over time are identified by determining temporal differences between segmented aneurysms in a plurality of longitudinal exams and visually presenting the temporal differences.

Abstract: The present invention is a system and method of remote patient monitoring to allow a patient to initiate and activate sensing systems. In the system and method, standard parameters can be sensed, and the information can then be processed and sent to the physician or clinician. The clinician then has the ability to remotely configure or reconfigure the parameters of the sensing system so as to probe for more targeted information based on the initial sensed data.

Abstract: A method includes automatically determining at least one gating signal based on a physiological signal from a subject being imaged by an imaging system, automatically determining, based upon prior analysis and knowledge of the imaging system's capabilities, a timing of each of a plurality of exposures within a single or multiple cycles of the physiologic signal, and performing the multiple acquisitions.

Abstract: Certain embodiments of the present invention provide a system and method for synchronized viewing of a plurality of images of an object. Corresponding landmarks of an object are synchronized between a first image set and a second image set. In an embodiment, the landmarks are folds of a human colon and the first image set and second images sets are computerized tomography scans, at least one image set being a prone scan of a portion of anatomy and at least one image set being a supine scan of a portion of the anatomy. An indicator for at least a first location in a first image set is displayed. The location of a second location in a second image set of an object is determined, wherein the second location corresponds to the first location of the object. The second location in the second image set is displayed.

Abstract: A system, method, and apparatus includes a computer readable storage medium with a computer program stored thereon having instructions that cause a computer to access a first anatomical image data set of an imaging subject acquired via a morphological imaging modality, access a functional image data set of the imaging subject acquired via a functional imaging modality, register the first anatomical image data set to the functional image data set, segment the functional image data set based on the functional image data set, define a binary mask based on the segmented functional image data set, and apply the binary mask to the first anatomical image data set to construct a second anatomical image data set and an image based thereon. The second anatomical image data set is substantially free of image data of the first anatomical image data set correlating to an area outside the region of physiological activity.

Abstract: A data processing technique is provided. In one embodiment, a computer-implemented method includes accessing reference deviation maps for a plurality of disease types. The reference deviation maps may include subsets of maps associated with severity levels of respective disease types, and a disease severity score may be associated with each severity level. The method may include selecting patient severity levels for multiple disease types based on the subsets of reference deviation maps. Also, the method may include automatically calculating a combined patient disease severity score based at least in part on the disease severity scores associated with the selected patient severity levels, and may include outputting a report based at least in part on the combined patient disease severity score. Additional methods, systems, and manufactures are also disclosed.

Abstract: Certain embodiments of the present invention provide a method and system for improved clinical workflow using wireless communication. A system for remote image display includes a data source with image data, wherein the data source is capable of transmitting the image data. The system also includes an identifiable display device capable of displaying image data transferred from the data source and a portable device capable of identifying the display device and requesting image data transfer from the data source to the display device without the transfer of the image data between the portable device and the data source. The system may also include an access point for relaying communication between the portable device and the data source. Communication between the portable device, the data source, and/or the display may include wireless communication, for example.

Abstract: A technique is disclosed for servicing complex systems. Upon detection of a serviceable condition, a system snapshot is made and stored. The snapshot may include a range of data then available on the system, particularly hardware configuration data, such as components then installed, peripherals installed, their states, and so forth. The snapshot may include data available prior to the detection of an indicator, at the time of detection and following detection, facilitating evaluation of the condition and potential responses to it.

Abstract: Systems, methods and apparatus are provided through which coronary plaque is classified in an image and visually displayed using an iterative adaptive process, such as an expectation maximization process.

Abstract: A technique for reconciling two or more reads of an image data set. One or more computer implemented routines is employed to provide computer-assisted reconciliation (CAR) including resolution of discrepancies between the two or more reads. The computer-assisted reconciliation may optimally display the discrepancies, the concurrences and any associated information to a human reconciler, may resolve the discrepancies in a partially automated manner, or may resolve the discrepancies in a fully automated manner. The reconciled data may then be provided to an end user.

Abstract: One or more techniques are provided for determining the overall motion of an organ of interest relative to a viewer or imager. Motion data is acquired for the organ of interest and/or for one or more proximate organs using sensor-based and/or image data-based techniques. The sensor-based techniques may include electrical and non-electrical techniques. The image data-based techniques may include both pre-acquisition and acquisition image data. The motion data for the organ of interest and proximate organs may be used to determine one or more quiescent periods corresponding to intervals of minimal motion for the organ of interest and the proximate organs, which may be used to determine one or more gating points that may be used retrospectively, as well as one or more motion compensation factors that may be used to reduce motion-related artifacts during processing and reconstruction of the acquired image data.

Abstract: Methods and apparatus for detecting aneurysm in vasculatures are provided. The method includes removing bone structures from three-dimensional (3D) computed tomography (CT) angiography image data and extracting spherical shapes from the bone removed 3D CT angiography image data. The method further includes displaying the extracted spherical shapes on an image of the 3D CT angiography image data.

Abstract: Methods and systems for quantification of a selected attribute of an image volume are provided. The system is configured to receive an image dataset for a volume of interest, process the dataset for a selected attribute based at least on one of shape and texture to obtain a plurality of responses, and compute an index of an aggregate of a plurality of obtained responses.

Abstract: Certain embodiments of the present invention provide a system and method for identifying stool particles in virtual dissection data for a colon. A shape classification may be determined for a segmented colon by three-dimensional filtering of a prone data set and a supine data set. The shape classification may be mapped onto a prone virtual dissection image and a supine virtual dissection image. The prone data set and the supine data set may be registered using one-dimensional registration to determine a registration. Shapes may be localized based on the shape classification and the registration for the prone virtual dissection and the supine virtual dissection. A distance metric may be applied to the localized shapes to identify stool particles. The identified stool particles may be suppressed. A prone virtual dissected image and a supine virtual dissected image may be displayed having the stool particles suppressed.

Abstract: A data processing technique is provided. In one embodiment, a computer-implemented method includes accessing patient image and non-image deviation scores derived through respective comparisons of patient image and non-image data to standardized image and non-image data. The method may also include processing the image and non-image deviation scores to generate a visual output indicative of differences between the patient image and non-image data, and the standardized image and non-image data, respectively. Further, the method may include displaying the visual output. Additional methods, systems, and manufactures are also disclosed.