Abstract: Multiple image capture devices are located in an operating room to capture video of the entirety of the operating room. A surgical tracking server applies one or more models to the captured video to determine states of objects in the operating room from which a phase of the operating room is determined. A surgical time-out where information about a surgical procedure is audibly provided by people in the operating room during the surgical procedure. From audio and video of the operating room, the surgical tracking server determines whether the time-out was performed and identifies audio and video corresponding to the surgical time-out. The identified audio and video are compared to criteria to evaluate completeness of the surgical time-out. A quality metric for the time out based on the captured video data based on characteristics of the video or audio of the operating room may also be generated.

Abstract: A method of processing image data includes: receiving image data from a color filter array including N×N same color pixels; converting first pixel data in an N×N array into second pixel data in an (N?L)×(N?M) array, wherein the first pixel data is output from the N×N same color pixels; and generating third pixel data in the N×N array by performing reconstruction on the second pixel data, wherein each of “L” and “M” is a natural number that is greater than or equal to 1 and less than N, and “N” is a natural number that is greater than or equal to 2.

Abstract: An apparatus, system, method, and computer-readable medium include aspects for using surgical video analysis. The aspects utilize intracorporeal video footage, image analysis, and notifications to optimize correspondences among medical procedure information, medical guidelines, and medical transaction information.

Abstract: A labeling method and a computing device by detecting lesion region in images using a learned network function. The method includes obtaining a medical image, detecting and displaying a region of interest corresponding to at least one subject disease through a learned network function, and labeling a lesion region corresponding to the at least one subject disease on the medical image according to the region of interest and user inputs. The labeling includes receiving a first user input for choosing a selected region within the region of interest, setting a reference pixel value based on the selected region, detecting and displaying a lesion estimation region having pixel values within a predetermined range from the reference pixel value, and correcting the lesion estimation region according to a second user input to label the lesion region.

Abstract: A method, apparatus and system for prediction of a neonatal brain development prognosis, using a pregenerated brain neurodevelopment prediction model. The apparatus includes a computing device, which includes a memory; and at least one processor performing communication with the memory. The processor is configured to extract cerebral blood flow (CBF) and brain tissue relaxation times T1 and T2 from the input magnetic resonance imaging (MRI) of a subject newborn, generate brain neurodevelopment prediction data of the subject newborn using a pregenerated brain neurodevelopment prediction model, and control the output of the generated brain neurodevelopment prediction data of the subject newborn.

Abstract: A method for determining a charge characteristic of electrical charges of particles in a fluid stream includes directing the fluid stream containing particles through a fluid line. Then a spatially-resolved determination of a measuring field-less particle velocity in a measurement area without an electrical measuring field is performed. An electrical measuring field is applied transverse to the flow direction in the measurement area. A spatially-resolved determination of a midfield particle velocity in the measurement area is determined. At least one charge characteristic which denotes an electrostatic charge of the particles is determined from the spatially-resolved particle velocities.

Abstract: Apparatus, system, method, and computer-readable medium aspects for using surgical video analysis for improving compliance with medical guidelines and improving processing of medical bills, medical malpractice claims, and insurance claims are described. The aspects utilize intracorporeal video footage, image analysis, and notifications to optimize correspondences among medical procedure information, medical guidelines, and medical transaction information.

Abstract: Systems and methods are provided for ultrafast light field tomography (LIFT), a transient imaging strategy that offers a temporal sequence of over 1000 and enables highly efficient light field acquisition, allowing snapshot acquisition of the complete two, three or four-dimensional space and time. The apparatus transforms targets in object space into parallel lines in the image plane with a cylindrical lens. Beam projections are optionally directed through a Dove prism and an array of cylindrical lenslets to an imaging device such as a SPAD camera, streak camera and CCD camera. By using an array of cylindrical lenslets oriented at distinct angles, enough projections are obtained simultaneously to recover the image with a single snapshot. The time-resolved system and methods were adapted to LIDAR, hyperspectral, non-line-of-sight, and three-dimensional transient imaging.

Abstract: Provided is a remote sensing image feature discretization method based on rough-fuzzy model, comprising the following steps: listing the digital number in each band and category of a selected sample in remote sensing images, and building an image information decision table based on the digital number and category; initializing the class center of each category and the membership degree of a sample example relative to the class center; updating the class center of each category and the membership degree of the sample example relative to the class center iteratively, and obtaining the final value of the class center of each category and the final value of the membership degree; building a rough-fuzzy set, computing the mean approximation accuracy of the rough-fuzzy set, discretizing the image information decision table, evaluating the discretization results based on the mean approximation accuracy and a genetic algorithm, and selecting an optimal discretization solution.

Abstract: A cell quality evaluation apparatus performs a process including estimation processing of estimating a quality of cells in the entirety of a cell-culture container by determining feature quantities from a plurality of images and calculating an average value of the feature quantities; derivation processing of deriving an estimation error of the quality estimated in the estimation processing, based on a variation of the feature quantities in the plurality of images and imaging information related to an area of a plurality of imaging regions; and imaging control processing of causing a imaging apparatus to perform re-imaging on at least one re-imaging region different from the plurality of imaging regions in a case where the estimation error is out of an allowable range.

Abstract: Class agnostic object mask generation uses a vision transformer-based auto-labeling framework requiring only images and object bounding boxes to generate object (segmentation) masks. The generated object masks, images, and object labels may then be used to train instance segmentation models or other neural networks to localize and segment objects with pixel-level accuracy.

Abstract: Techniques to modify an image from a night vision camera so as to reduce smear. An example method comprises receiving at least one image from a night vision camera; modifying the at least one image, based on motion data from a tracker system and a shutter time of night vision camera, to reduce smear in the image; and outputting the modified at least one image.

Abstract: An apparatus for analyzing a blood vessel includes a processor configured to detect a blood vessel area corresponding to the blood vessel and a stenosis area corresponding to a stenosed portion of the blood vessel, based on a machine learning model, from a blood vessel image, determine a partial area of the blood vessel area to be a comparison area for the stenosis area, based on a length of the stenosis area, and calculate a stenosis score indicating a degree of stenosis of a blood vessel of the stenosis area compared to a blood vessel of the comparison area.

Abstract: A computer-implemented method for processing a seismic image obtained from seismic measurements performed on a geological formation includes determining a relative geological time (RGT) image based on the seismic image; obtaining setpoints for the RGT image; retrieving an RGT surface of the RGT image for each of the setpoints, wherein each point of an RGT surface is defined by a vertical position and at least one horizontal position; and determining a reference RGT surface by combining the RGT surfaces determined for the setpoints. The vertical position of a point of the reference RGT surface is obtained by combining the vertical positions of the points of the RGT surfaces having the same at least one horizontal position as the considered point of the reference RGT surface. The combining of the vertical positions uses weighting coefficients.

Abstract: A product identification apparatus includes an acquisition unit, an image processing unit, and a storage processing unit. The acquisition unit acquires an image generated by an image capturing apparatus 10. The image includes a product shelf and a product. The image processing unit determines a position where continuity of the product shelf is broken by processing the image acquired by the acquisition unit, and divides the product shelf into a unit region by using the position. Further, the image processing unit determines a kind and a product name of the product by processing the image acquired by the acquisition unit. The storage processing unit causes a storage unit to store product identification information of the product located in the unit region, for each unit region of the product shelf.

Abstract: Systems and methods for scalable mapping of brain dynamics are capable of mapping brain dynamics in a computationally efficient way. Shape graphs can be produced which can operate as interactive network representations of dynamic brain activity data. In methods for automatically interpreting the structure of the shape graphs, the mapped brain dynamics can be used to indicate which type of treatment protocol is likely to be most effective for the particular individual. The treatment can include transcranial magnetic stimulation, pharmaceutical treatment, and/or any other mental condition treatment as appropriate to the requirements of specific applications.

Abstract: A warpage amount estimation apparatus that estimates a warpage amount of a substrate includes a processor and a memory. The processor acquires a captured image of one surface of an estimation target substrate. The processor calculates a rate of change in pixel value relating to a substrate radial direction in the captured image of the one surface of the estimation target substrate. The processor estimates a warpage amount of the estimation target substrate based on a correlation obtained in advance between a rate of change in pixel value relating to the substrate radial direction in a captured image of the one surface of a substrate and a warpage amount of the substrate, and on a calculation result of the rate of change which is calculated.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a neural network to process images. One of the methods includes obtaining a training image; processing the training image using a first subnetwork to generate, for each of a plurality of first image patches of the training image, a relevance score; generating, using the relevance scores, one or more second image patches of the training image by performing one or more differentiable operations on the relevance scores; processing the one or more second image patches using a second subnetwork to generate a prediction about the training image; determining an error of the training network output; and generating a parameter update for the first subnetwork, comprising backpropagating gradients determined according to the error of the training network output through i) the second subnetwork, ii) the one or more differentiable operations, and iii) the first subnetwork.

Abstract: A computer system of unsupervised learning with deep similarity for optical flow estimation is provided. The computer system is configured to calculate deep similarity by using deep features extracted from a sequence of a plurality of images, and learning optical flow for the images based on the deep similarity.

Abstract: A system for optimizing DBS parameters for a subject includes automatically performing actions via a processor. The actions include obtaining functional MRI data of a brain of the subject acquired utilizing an MRI system during DBS of the brain utilizing a first set of DBS parameters. The actions include generating functional MRI response maps from the functional MRI data. The actions include extracting, utilizing an unsupervised autoencoder-based neural network, features from the functional MRI response maps. The actions include determining, utilizing a deep learning-based DBS parameter classification model, whether the first set of DBS parameters are optimal DBS parameters for the subject based on the features. The actions include, when the first set of DBS parameters are not the optimal DBS parameters, predicting, utilizing a deep learning-based DBS parameter prediction model, a second set of DBS parameters that are the optimal DBS parameters for the subject based on the features.