Abstract: Methods, apparatus, systems, and articles of manufacture are disclosed to autonomously detect thermal anomalies. Disclosed examples include an example apparatus to detect engine anomalies comprising: at least one memory; instructions in the apparatus; and processor circuitry to execute the instructions to: control a plurality of infrared cameras to capture a baseline image set, the baseline image set including at least two thermal images; generate emissivity data based on the baseline image set; provide the baseline image set and the emissivity data to an artificial intelligence model, the artificial intelligence model to generate a reconstructed image set; determine a difference between the baseline image set and the reconstructed image set; and in response to the difference exceeding a threshold, generate an alert indicating detection of an engine anomaly.

Abstract: Briefly, embodiments are directed to a system, method, and article for acquiring a symbol comprising a representation of input data. The symbol may be converted into an emergent language expression in an emergent language via processing of a first neural network. Transmission of the emergent language expression may be initiated over a communications network, where the emergent language comprises a language based on and specific to the input data. The emergent language expression may be translated back into the representation of the input data via processing of a second neural network.

Abstract: Methods, apparatus, systems, and articles of manufacture are disclosed to autonomously detect thermal anomalies. Disclosed examples include an example apparatus to detect engine anomalies comprising: at least one memory; instructions in the apparatus; and processor circuitry to execute the instructions to: control a plurality of infrared cameras to capture a baseline image set, the baseline image set including at least two thermal images; generate emissivity data based on the baseline image set; provide the baseline image set and the emissivity data to an artificial intelligence model, the artificial intelligence model to generate a reconstructed image set; determine a difference between the baseline image set and the reconstructed image set; and in response to the difference exceeding a threshold, generate an alert indicating detection of an engine anomaly.

Abstract: Improved gimbal systems, apparatus, articles of manufacture and associated methods are disclosed. Examples include a panel including a window, the window to define an aperture for a sensor; a platform to mount the sensor, the platform including a first pinion; a first stepper motor to move the first pinion about a first arched rack; a gimbal body including the first arched rack and a second pinion; and a second stepper motor to move the second pinion about a second arched rack, the second arched rack positioned orthogonally to the first arched rack.

Abstract: Improved gimbal systems, apparatus, articles of manufacture and associated methods are disclosed. Examples include a panel including a window, the window to define an aperture for a sensor; a platform to mount the sensor, the platform including a first pinion; a first stepper motor to move the first pinion about a first arched rack; a gimbal body including the first arched rack and a second pinion; and a second stepper motor to move the second pinion about a second arched rack, the second arched rack positioned orthogonally to the first arched rack.

Abstract: An image inspection computing device is provided. The device includes a memory device and at least one processor. The at least one processor is configured to receive at least one sample image of a first component, wherein the at least one sample image of the first component does not include defects, store, in the memory, the at least one sample image, and receive an input image of a second component. The at least one processor is also configured to generate an encoded array based on the input image of the second component, perform a stochastic data sampling process on the encoded array, generate a decoded array, and generate a reconstructed image of the second component, derived from the stochastic data sampling process and the decoded array. The at least one processor is further configured to produce a residual image, and identify defects in the second component.

Abstract: An image inspection computing device is provided. The device includes a memory device and at least one processor. The at least one processor is configured to receive at least one sample image of a first component, wherein the at least one sample image of the first component does not include defects, store, in the memory, the at least one sample image, and receive an input image of a second component. The at least one processor is also configured to generate an encoded array based on the input image of the second component, perform a stochastic data sampling process on the encoded array, generate a decoded array, and generate a reconstructed image of the second component, derived from the stochastic data sampling process and the decoded array. The at least one processor is further configured to produce a residual image, and identify defects in the second component.

Abstract: Briefly, embodiments are directed to a system, method, and article for acquiring a symbol comprising a representation of input data. The symbol may be converted into an emergent language expression in an emergent language via processing of a first neural network. Transmission of the emergent language expression may be initiated over a communications network, where the emergent language comprises a language based on and specific to the input data. The emergent language expression may be translated back into the representation of the input data via processing of a second neural network.

Abstract: A system and method including receiving input data by an encoder, the encoder reducing a dimensionality of the received data; receiving, by a sender module, the reduced dimensionality data; generating, by the sender module, a sentence comprising a plurality of symbols representative of the input data, the symbols being defined by a predetermined vocabulary and a predetermined sentence length; receiving, by a receiver module, the sentence comprising the plurality of symbols; generating, based on the received sentence, continuous data by the receiver module; receiving, by a decoder, the continuous data from the receiver module; generating an output, by the decoder based on the continuous data, the output including a recreation of the input data.

Abstract: The example embodiments are directed to refinement process for generating an accurate image segmentation map. A refinement network may enhance an initially generated segmentation map using a model that is trained using synthetic images. In one example, the method may include storing an image of content which includes a plurality of categories of data, receiving an initial segmentation map of the image, the initial segmentation map comprising pixel probability values with respect to the plurality of categories, executing a refinement predictive model on the initial segmentation map and the image to generate a refined segmentation map, wherein the predictive model is trained using synthetic images of the plurality of categories of data, and generating a segmented image based on the refined segmentation map.

Abstract: The example embodiments are directed to refinement process for generating an accurate image segmentation map. A refinement network may enhance an initially generated segmentation map using a model that is trained using synthetic images. In one example, the method may include storing an image of content which includes a plurality of categories of data, receiving an initial segmentation map of the image, the initial segmentation map comprising pixel probability values with respect to the plurality of categories, executing a refinement predictive model on the initial segmentation map and the image to generate a refined segmentation map, wherein the predictive model is trained using synthetic images of the plurality of categories of data, and generating a segmented image based on the refined segmentation map.

Abstract: The present approach relates to the use of trained artificial neural networks, such as convolutional neural networks, to classify vascular structures, such as using a hierarchical classification scheme. In certain approaches, the artificial neural network is trained using training data that is all or partly derived from synthetic vascular representations.

Abstract: The subject matter of the present disclosure generally relates to techniques for image analysis. In certain embodiments, various morphological or intensity-based features as well as different thresholding approaches may be used to segment the subpopulation of interest and classify object in the images.

Abstract: The subject matter of the present disclosure generally relates to techniques for image analysis. In certain embodiments, various morphological or intensity-based features as well as different thresholding approaches may be used to segment the subpopulation of interest and classify object in the images.

Abstract: The disclosed subject matter relates to an automated determination of cell-by-cell segmentation quality of a tissue specimen sample. A training set of cells is examined by an expert to determine which cells that include “good” segmentation and which cells include “poor” segmentation. A training model is build based on the image data of the cells in the training set. Image data from cells in a test specimen is obtained and that image data is compared to the training model to determine on a cell-by-cell basis which cells in the test specimen include “good” segmentation and which cells include “poor” segmentation. The accumulated data on the cells analyzed in the test specimen can be utilized to determine an overall segmentation quality score for the area of the test specimen in which the individual cells are located in the test specimen.

Abstract: A method in one embodiment includes acquiring optical image information with a detection unit configured to be operably coupled to a patient. The optical image information corresponds to microcirculation of the patient. The method also includes generating a microcirculation map of microvasculature of the patient using the optical image information. Further, the method includes generating a quantitative microcirculation index based on the microcirculation map, the quantitative microcirculation index corresponding to a condition of the patient.

Abstract: Disclosed are novel computer-implemented methods for creating a blood vessel map of a biological tissue. The methods comprise the steps of, accessing image data corresponding to multi-channel multiplexed image of a fluorescently stained biological tissue manifesting expression levels of a primary marker and at least one auxiliary marker of blood vasculature, and extracting features of blood vessels using the primary marker as an input to create a single channel segmentation of the blood vessels. The method further comprises the steps of extracting features of blood vessels using the auxiliary marker to create auxiliary channels as a second input and apply multi-channel blood vessel enhancement. Blood vessel maps are created using the features and tracing the blood vasculature by iteratively extending the centerlines of the initial segmentation using statistical models and geometric rules.

Abstract: The disclosed embodiments are directed to a method for accurately counting and characterizing multiple cell phenotypes and sub-phenotypes within cell populations simultaneously by exploiting biomarker co-expression levels within cells of different phenotypes in the same tissue sample. The disclosed embodiments are also directed to a simple intuitive interface enabling medical staff (e.g., pathologists, biologists) to annotate and evaluate different cell phenotypes used in the algorithm and the presented through the interface.

Abstract: The present approach relates to the use of trained artificial neural networks, such as convolutional neural networks, to classify vascular structures, such as using a hierarchical classification scheme. In certain approaches, the artificial neural network is trained using training data that is all or partly derived from synthetic vascular representations.

Abstract: The disclosed subject matter relates to an automated determination the stain quality and segmentation quality of a tissue sample. By way of example, separate image data is acquired of an unstained form of a biological specimen, the biological specimen stained with a nuclei marker and the biological specimen stained with a segmentation marker. A correlation map (Cr) from the separate image data and a ridgeness map (Pr) from the image data of the biological specimen stained with a segmentation marker are each determined. A staining quality score and segmentation quality score are then determined from the correlation map (Cr) and the ridgeness map (Pr).