Abstract: A novel and useful artificial neural network that incorporates emphasis and focus techniques to extract more information from one or more portions of an input image compared to the rest of the image. The ANN recognizes that valuable information in an input image is typically not distributed throughout the image but rather is concentrated in one or more regions. Rather than implement CNN layers sequentially (i.e. row by row) on the input domain of each layer, the present invention leverages the fact that valuable information is focused in one or more regions of the image where it is desirable to apply more attention and for which it is desired to apply more elaborate evaluation. Precision dilution can be applied to those portions of the input image that are not the center of focus and emphasis. A spatial aware function determines the location(s) of the ears of focus and is applied to the first convolutional layer.

Abstract: Methods and systems are provided for direct monochromatic image generation for spectral computed tomography. In one embodiment, a method comprises acquiring projection data during a scan of a subject, reconstructing a low energy image and a high energy image from the projection data, and generating a monochromatic image from the low energy image and the high energy image. In this way, a monochromatic image may be generated directly from low and high energy images with a substantial reduction in image noise, especially when compared to a monochromatic image generated indirectly from material density images.

Abstract: Embodiments described herein provide various examples of a visual-haptic feedback system for generating a haptic feedback signal based on captured endoscopy images. In one aspect, the process for generating the haptic feedback signal includes the steps of: receiving an endoscopic video captured for a surgical procedure performed on a robotic surgical system; detecting a surgical task in the endoscopic video involving a given type of surgical tool-tissue interaction; selecting a machine learning model constructed for analyzing the given type of surgical tool-tissue interaction; for a video image associated with the detected surgical task depicting the given type of surgical tool-tissue interaction, applying the selected machine learning model to the video image to predict a strength level of the depicted surgical tool-tissue interaction; and then providing the predicted strength level to a surgeon performing the surgical task as a haptic feedback signal for the given type of surgical tool-tissue interaction.

Abstract: A system and method for contactless blood pressure determination. The method includes: receiving a captured image sequence; determining, using a trained hemoglobin concentration (HC) changes machine learning model, bit values from a set of bitplanes in the captured image sequence that represent the HC changes of the subject; determining a blood flow data signal; extracting one or more domain knowledge signals associated with the determination of blood pressure; building a trained blood pressure machine learning model with a blood pressure training set, the blood pressure training set including the blood flow data signal of the one or more predetermined ROIs and the one or more domain knowledge signals; determining, using the blood pressure machine learning model trained with a blood pressure training set, an estimation of blood pressure; and outputting the determination of blood pressure.

Abstract: The invention is a method and an apparatus for generating a displacement map of a first input dataset and a second input dataset of an input dataset pair. The apparatus comprises a neural network based feature extractor (25) for processing the first input dataset and the second input dataset so as to generate a feature map hierarchy (50) comprising a base pair of feature maps (20e, 30e) and a refinement pair of feature maps (20b, 30b, 20c, 30c, 20d, 30d), a displacement unit comprising a first comparator unit for obtaining an initial displacement map using the base pair of feature maps (20e, 30e); and a displacement refinement unit for obtain an updated displacement map for the refinement pair of feature maps (20d, 30d).

Abstract: The apparatus enables biometric authentication without the risk of forgery or the like and enables living-tissue discrimination. The roughness distribution pattern of deep-layer tissue of the skin covered with epidermal tissue is detected, thereby extracting a unique pattern of the living tissue. Biometric authentication is performed based upon the detected pattern. The roughness distribution pattern of the deep-layer tissue of the skin is optically detected using difference in optical properties between the epidermal tissue and the deep-layer tissue of the skin. Long-wavelength light, e.g., near-infrared light, is used as illumination light cast onto the skin tissue. A fork structure of a subcutaneous blood vessel is used as the portion which is to be detected, for example. The portion which is to be detected is determined based upon the structure of the fork structure. The living-tissue discrimination may be made using the subcutaneous blood vessel.

Abstract: A checkout lane management system is described that uses object recognition to detect whether one or more items is under a shopping cart during a customer checkout process. The checkout lane management system may comprise one or more of cameras for collecting a stream of images focused on a checkout lane. The checkout lane management system may also comprise one or more of a display device and a speaker device as part of a checkout lane device for displaying a BOB indicator or sounding the BOB indicator.

Abstract: Kits, diagnostic systems and methods are provided, which measure the distribution of colors of skin features by comparison to calibrated colors which are co-imaged with the skin feature. The colors on the calibration template (calibrator) are selected to represent the expected range of feature colors under various illumination and capturing conditions. The calibrator may also comprise features with different forms and size for calibrating geometric parameters of the skin features in the captured images. Measurements may be enhanced by monitoring over time changes in the distribution of colors, by measuring two and three dimensional geometrical parameters of the skin feature and by associating the data with medical diagnostic parameters. Thus, simple means for skin diagnosis and monitoring are provided which simplify and improve current dermatologic diagnostic procedures.

Abstract: In one embodiment, a light sensor, such as a camera, records an image through the surface with the residue to produce a stained image. A processor associated with the camera identifies object outlines within the image using a machine learning model, and smooth the colors within the object outlines. In another embodiment, the light sensor is placed beneath a dual-mode region of a display containing the residue. The dual-mode region can be opaque and function as part of the display, or can be transparent and allow environment light to reach the light sensor. Initially, the processor determines the position of the residue by causing the dual-mode region to display a predetermined pattern, while the light sensor records the predetermined pattern. Using the determined position of the residue, the processor corrects the pixels within the residue in the recorded image, by interpolating the values of the pixels outside of the residue.

Abstract: An input chromaticity data calculation unit 113 calculates input chromaticity data rg from device-dependent color data RGB. A chromaticity data transformation unit 114 transforms the input chromaticity data rg into output chromaticity data xy by referring to an LUT stored by an LUT storage unit 13. A total amount calculation unit 115 refers to the LUT of the LUT storage unit 13 to calculate a total amount ??XYZ of the device-independent color data XYZ from the color data RGB. An output color data calculation unit 116 uses the total amount ??XYZ to calculate color data XYZ from the output chromaticity data xy.

Abstract: A method and apparatus for authenticating a fingerprint image captured through an optical sensor. For at least some embodiments, light scattering characteristics associated with a fingerprint are determined and compared to a reference light scattering characteristic. The fingerprint is authenticated when the light scattering characteristics are within a threshold difference of the reference light scattering characteristic. For some embodiments, the light scattering characteristics associated with the fingerprint are compared to light scattering characteristics associated with one or more reference (enrollment) images. For at least some embodiments, the light scattering characteristics may be based on a correlation value based on identified pixels and one or more pixels neighboring the identified pixel.

Abstract: A method for recognizing the face of a user that uses a device in a facility includes receiving a photo of the user and extracting a face image of the user from the photo. The method includes receiving a full candidate face set of all registered users, a first list of registered users that are within the facility, and a second list of registered users that are using other devices in the facility. The method includes generating a reduced candidate face set from the full candidate face set by excluding at least one of (1) candidate faces corresponding to users that are not included in the first list or (2) candidate faces corresponding to users that are included in the second list, from the full candidate face set. The method also includes determining whether the face image corresponds to a candidate face of the reduced candidate face set.

Abstract: A method for making a segmentation in the sorted sequence of data (100), that comprises arranging a sequential stream of sorted input data (10), where the data are arranged in matrices (frames), in this case by way of a preferred example as a sequence of images for image analysis (20), and that delivers a sequential stream of sorted output data (30), where the sorted output data are arranged in sorted matrices.

Abstract: An apparatus obtains a second straight-line by mapping a first straight-line passing a projection center of a target-image taken by a first imaging-device in a three-dimensional space and a point representing an object in a projection-plane of the first imaging-device, onto each of reference-images respectively taken by second imaging-devices, and generates a line-segment representing an existing-range of the object on the second straight-line for each reference-image. The apparatus stores the coordinate-value of the second-endpoint and a difference between coordinate-values of the first-endpoint and a second-endpoint of the line-segment in a memory.

Abstract: Multi-source, multi-type image registration is provided. Images are received from a plurality of image devices, and images are received from a medical imaging device. A pre-existing diagram of a probe of the medical imaging device is received. A four-dimensional model is determined based on the received images from the image devices. A pose of the probe of the medical imaging device is determined based on the pre-existing diagram of the probe and the received images from the image devices. The plurality of images from the medical imaging device are registered with the four-dimensional model based on a common coordinate system and the determined pose of the probe.

Abstract: An optical fingerprint module includes: an optical fingerprint sensor, wherein the optical fingerprint sensor includes at least one pixel region where photosensitive pixels are disposed; each of the photosensitive pixels includes an optical fingerprint sensing device, a non-opaque region and an opaque region, wherein the optical fingerprint sensing device is disposed in the opaque region, and all the optical fingerprint devices are arranged in rows and columns; and at least one point-shaped backlight source, wherein each of the at least one pixel region corresponds to one of the at least one point-shaped backlight source; in the pixel region, a largest square region whose center is closest to the corresponding point-shaped backlight source is selected, and optical fingerprint sensing devices are divided into different device groups, and optical fingerprint sensing devices in different device groups are set differently, which improves performance of the optical fingerprint module.

Abstract: Aspects of the disclosure include sizing catheters and methods for determining the size and other physical parameters of an internal orifice or lumen. Embodiments include a sizing catheter assembly having a handle assembly and a catheter assembly including at least one catheter extending from the handle assembly. The catheter assembly also includes at least two sizers, which can be adjustably spaced from each other. Each of the sizers are configured to conform to respective portions of a lumen of a patient's anatomy. The sizing catheter is configured such that the sizers are configured to specify or determine first and second dimensions of the lumen and also distance between the sizers. The sizers can include valve leaflets or the sizing catheter can include a temporary valve. Additional aspects include software/methods of assessing and determining an appropriate implantable device that can include assessing the biomechanical interaction between the device and the anatomy.

Abstract: An apparatus that includes a computer server having processing circuitry operatively coupled to a memory and a communication interface is provided. A remote image capture device is communicatively coupled to the computer server via the communication interface. The apparatus acquires a captured image of a storage system from the image capture device. The apparatus identifies a pair of physical items in the image, including determining identities of each item of the pair of physical items, and identifying locations in the storage system of each physical item of the pair of physical items. The apparatus processes the image to measure a physical proximity between the pair of physical items, and generates or updates a measure of similarity between the pair of physical items based on the measured physical proximity. The apparatus can also process multiple images to determine similarity between different storage systems or associated persons based on item proximities.

Abstract: A method of modeling anatomic deformation comprises receiving a reference three dimensional model of a branched anatomical formation in a reference state. The method further comprises detecting deflection of the branched anatomical fomation with a shape sensor to create a three dimensional deformation field. The method further comprises applying the three dimensional deformation field to the reference three dimensional model to create a deformed three dimensional model of a deformed state of the branched anatomical formation. The method further comprises dynamically displaying an image of the deformed three dimensional model of the deformed state of the branched anatomical formation. The method further comprises displaying a composite image including an image of an interventional instrument positioned within the branched anatomical formation and the dynamically displayed image of the deformed three dimensional model of the deformed state of the branched anatomical formation.

Abstract: A surgical assistive apparatus and method to provide assistance during surgery, includes an organ boundary localization circuit that selects a test video frame from a captured sequence of video frames and derives a first global region boundary of an internal organ of interest in the selected test video frame by integration of an appearance likelihood result, a global segmentation result, and a global edge detection result associated with the internal organ of interest. A plurality of local region boundaries are determined for a plurality of local sub regions of the internal organ of interest and a second global region boundary is generated for the internal organ of interest based on the determined plurality of local region boundaries and the first global region boundary. The internal organ of interest is localized based on the generated second global region boundary.