Abstract: The present invention relates to methods for predicting the recurrence risk of hepatocellular carcinoma (HCC). Specifically, it proposes a deep learning model capable of integrating information from different phases of CT images and clinical data to predict the risk of HCC recurrence within 1 to 5 years after treatment. Experimental results demonstrate that these models outperform traditional prediction methods based on histological microvascular invasion (MVI) in predicting HCC recurrence risk.

Abstract: A novel medical image generation approach synthesizes thin-cut computerized tomography (CT) images from thick-cut CT ones as inputs. First thick-cut CT images are obtained by maximizing the pixel-wise intensity of five or more than five continuous thin-cut ones after image registration. Second, the obtained thick-cut CT images are fed into a generator block which adopts an encoder-decoder architecture, where each thick-cut image is encoded into low-dimensional embedding space before decoding into multiple thin-cut ones. Third, a discriminator focuses on distinguishing original real thin-cut images from synthetic thin-cut images. An adversarial mechanism between the generator and discriminator causes the discriminator's output to provide an effective gradient update of the network parameters for the generator to increasingly improve the generator's ability to synthesize higher-quality thin-cut images and in turn promotes the discriminator's discriminating capability.

Abstract: A three dimensional classification system for recognizing cross-sectional images automatically contains a processor that executes: (1) rescaling of a plurality of cross-sectional images; and feeding the rescaled plurality of cross-sectional images into two branches; (2) feeding the rescaled plurality of cross-sectional images into a first branch for performing a plurality of convolutions on the rescaled plurality of cross-sectional images directly to learn features for distinguishing phases; (3) feeding the rescaled plurality of cross-sectional images into a second branch for reducing resolution, and then performing a plurality of convolutions on the reduced resolution plurality of cross-sectional images to learn features for distinguishing phases; and (4) concatenating convolutional output channels from the two branches to fuse global and local features, on which two fully-connected layers are stacked as a classifier to recognize cross-sectional volumetric images accurately and quickly.

Abstract: A computer-implemented system (CIS) is provided for processing and/or analyzing non-contrast-enhance computer tomography medical imaging input data is described. The CIS contains (i) twin U-Net architectures with equal weights, which are built on a Siamese architecture, and (ii) a Dissimilar block operably linked to the two U-Net architectures, and built on top of the Siamese-U-Net architecture to form a Dissimilar-Siamese-U-Net architecture. The computer-implemented system can be used in diagnosing acute ischemic stroke and/or thromoboembolus, by analyzing separate and independent input images of the left and right hemispheres of a brain. The diagnosis is based on a detection of the presence of a hyperdense middle cerebral artery sign.

Abstract: Disclosed are systems and methods using artificial intelligence for the detection and characterization of liver cancers.

Abstract: A rotating cutter apparatus including a housing having a housing interior surface defining a housing bore, a shaft extending through the housing bore such that an annular space is provided between the housing interior surface and the shaft, and at least one cutter carried on the shaft. The shaft is rotatable relative to the housing about a shaft axis. The at least one cutter extends radially toward the housing interior surface and is rotatable relative to the housing about the shaft axis. A method for reducing a size of solid objects contained in a fluid, including providing a rotating cutter apparatus, introducing the fluid containing the solid objects into the rotating cutter apparatus, and rotating a shaft and at least one cutter in order to reduce the size of the solid objects while passing the fluid through the rotating cutter apparatus.

Abstract: Embodiments disclosed herein include devices, methods, and systems for direct, selective, and sensitive detection of single-stranded and double-stranded target nucleic acid sequences from various sources in a solution-based system. When activated by binding a target nucleic acid sequence, the Cas protein cleaves a tether separating a reporter molecule from a capture moiety. The capture moiety can then be used to remove, localize, or sequester uncleaved molecule containing intact tethers. In some embodiments, the systems, methods, and devices may include a filter, a membrane, or other molecules that may help to separate the tethered and untethered reporter molecules and/or capture the reporter molecules. These devices, systems, and techniques allow a user to rapidly process samples that may contain the target nucleic acid, in some cases, without needing to amplify the target sequences, and without the need for sophisticated or expensive laboratory equipment.

Abstract: Software program “help” functionality includes features enhancing its relevance to users. An engine may receive/record context data arising from interactions between a user and the software program. The engine may process that context data to reference a plurality of sources (e.g., official help documentation, user blogs, partner web sites, etc.) and harvest relevant help information. Based upon stored context data and help information available from the plurality of sources, certain embodiments may generate suggested questions allowing further exploration of help issues. Some embodiments may offer help information in the form of an interactive “hotspot” focusing upon a specific UI element (e.g., icon). Processing the context data may allow the engine to assign a particular skill level (e.g., novice/intermediate/expert) to individual users, tailoring the help information to that skill level.

Abstract: Software program “help” functionality includes features enhancing its relevance to users. An engine may receive/record context data arising from interactions between a user and the software program. The engine may process that context data to reference a plurality of sources (e.g., official help documentation, user blogs, partner web sites, etc.) and harvest relevant help information. Based upon stored context data and help information available from the plurality of sources, certain embodiments may generate suggested questions allowing further exploration of help issues. Some embodiments may offer help information in the form of an interactive “hotspot” focusing upon a specific UI element (e.g., icon). Processing the context data may allow the engine to assign a particular skill level (e.g., novice/intermediate/expert) to individual users, tailoring the help information to that skill level.

Abstract: Software program “help” functionality includes features enhancing its relevance to users. An engine may receive/record context data arising from interactions between a user and the software program. The engine may process that context data to reference a plurality of sources (e.g., official help documentation, user blogs, partner web sites, etc.) and harvest relevant help information. Based upon stored context data and help information available from the plurality of sources, certain embodiments may generate suggested questions allowing further exploration of help issues. Some embodiments may offer help information in the form of an interactive “hotspot” focusing upon a specific UI element (e.g., icon). Processing the context data may allow the engine to assign a particular skill level (e.g., novice/intermediate/expert) to individual users, tailoring the help information to that skill level.

Abstract: Software program “help” functionality includes features enhancing its relevance to users. An engine may receive/record context data arising from interactions between a user and the software program. The engine may process that context data to reference a plurality of sources (e.g., official help documentation, user blogs, partner web sites, etc.) and harvest relevant help information. Based upon stored context data and help information available from the plurality of sources, certain embodiments may generate suggested questions allowing further exploration of help issues. Some embodiments may offer help information in the form of an interactive “hotspot” focusing upon a specific UI element (e.g., icon). Processing the context data may allow the engine to assign a particular skill level (e.g., novice/intermediate/expert) to individual users, tailoring the help information to that skill level.