Abstract: An object recognition device includes a recognition unit, a recognition reliability calculation unit, and a combining unit. A recognition unit recognizes an object by a plurality of functions, based on information which is obtained by measuring an object by a plurality of measurement devices. A recognition reliability calculation unit calculates recognition reliability of recognition results that are obtained by recognizing an object by a recognition unit, for each function. A combining unit combines recognition reliability of the object and recognition results, detects a specified object, and outputs detection results of the specified object.

Abstract: An apparatus for producing a fundus image includes: a processor and a memory; an illumination component including a light source and operatively coupled to the processor; a camera including a lens and operatively coupled to the processor, wherein the memory stores instructions that, when executed by the processor, cause the apparatus to: execute an automated script for capture of the fundus image; and allow for manual capture of the fundus image.

Abstract: The present disclosure relates to techniques for predicting a genetically based trait using handprints. In one embodiment, a method generally includes receiving a digital image of a person's handprint. A computing device extracts features from the digital image using at least one feature-extraction technique and inputs the features into a machine-learning model trained to predict at least one genetically based trait when such features are received as input. The machine-learning model predicts whether the user has one or more genetically based traits based on the values of the features. The computing device then sends the prediction to an application that presents the prediction to the user.

Abstract: The subject disclosure presents systems and computer-implemented methods for assessing a risk of cancer recurrence in a patient based on a holistic integration of large amounts of prognostic information for said patient into a single comparative prognostic dataset. A risk classification system may be trained using the large amounts of information from a cohort of training slides from several patients, along with survival data for said patients. For example, a machine-learning-based binary classifier in the risk classification system may be trained using a set of granular image features computed from a plurality of slides corresponding to several cancer patients whose survival information is known and input into the system. The trained classifier may be used to classify image features from one or more test patients into a low-risk or high-risk group.

Abstract: A system and method includes operation of a generation network to generate first generated computed tomography data based on a first instance of surface data, determination of a generation loss based on the first generated computed tomography data and on a first instance of computed tomography data which corresponds to the first instance of surface data, operation of a discriminator network to discriminate between the first generated computed tomography data and the first instance of computed tomography data, determination of a discriminator loss based on the discrimination between the first generated computed tomography data and the first instance of computed tomography data, determination of discriminator gradients of the discriminator network based on the discriminator loss, and updating of weights of the generation network based on the generation loss and the discriminator gradients.

Abstract: A human detection system may detect the presence of a human around a shovel by using an image captured by an imaging device attached to the shovel. Also, the human detection system may have an extracting part that extracts a part of the captured image as an identification process target image, and an identifying part that identifies whether an image included in the identification process target image is an image of a human. A region of the identification process target image is preliminarily associated with one of head image positions in the captured image. The extracting part finds out a helmet image in the captured image, and extracts the identification process target image by associating a representative position of the helmet image with one of the head image positions.

Abstract: A method of improving diagnostic and functional imaging is provided by obtaining at least two input images of a subject, using a medical imager, where each input image includes a different contrast, generating a plurality of copies of the input images using non-local mean (NLM) filtering, using an appropriately programmed computer, where each input image copy of the subject includes different spatial characteristics, obtaining at least one reference image of the subject, using the medical imager, where the reference image includes imaging characteristics that are different form the input images of the subject, training a deep network model, using data augmentation on the appropriately programmed computer, to adaptively tune model parameters to approximate the reference image from an initial set of the input and reference images, with the goal of outputting an improved quality image of other sets of low SNR low resolution images, for analysis by a physician.

Abstract: Systems and methods of stent enhancement in medical images includes obtaining a plurality of medical images of a vessel and a stent. A centerline of the vessel is obtained in medical images of the plurality of medical images. A deformation field across the medical images of the plurality of medical images is estimated based at least in part upon the obtained centerlines. The medical images of the plurality of medical images are registered to a common reference image. The images are temporally integrated to obtain an enhanced visual representation of the stent.

Abstract: Embodiments generally relate to cancer treatment with radiation sources. The present technology discloses techniques that can enable an automatic generation of radiotherapy trajectories using anatomical data of a patient. It can improve conformal dose distributions and target volume coverage by considering a radiation risk decided by an organs-at-risk (OAR)'s relative location to the target volume and the radiation source.

Abstract: A system is disclosed using a data-driven approach to objectively measure the diagnostic accuracy and value of diagnostic imaging reports using data captured routinely as part of the electronic health record. The system further utilizes the evaluation of the diagnostic accuracy of individual radiologists (imagers), subspecialty sections, modalities, and entire departments based on a comparison against a “precision diagnosis” rendered by other clinical data sources such as pathology, surgery, laboratory tests, etc.

Abstract: A medical image processing apparatus includes a port, a processor and a display. The port acquires volume data including a subject. The processor calculates long and short diameters of the subject. The display shows the long and short diameters. A line segment presenting the long diameter and a line segment presenting the short diameter is a pair of skew lines.

Abstract: Transient connected regions are tracked in a video sequence, marking them as foreground layer or background layer when the transient regions become stable. A stack of background/foreground Gaussian mixer models is maintained for each pixel. The decision to mark a stable region as background layer or a new or existing foreground layer is done by matching the region with each model in the model stack. If the new region matches an existing model, then the layers above the matched layers are purged, or else the new region is pushed as a new foreground layer with a new model.

Abstract: A client device configured with a neural network includes a processor, a memory, a user interface, a communications interface, a power supply and an input device, wherein the memory includes a trained neural network received from a server system that has trained and configured the neural network for the client device. A server system and a method of training a neural network are disclosed.

Abstract: There is provided a cell image acquisition device, method, and a non-transitory computer readable recording medium recorded with a program to reduce the amount of data to be processed and stored by limiting a target region in a colony region when imaging a cell colony. There are included: a maturity information acquisition unit 22 that acquires information regarding the maturity of cells being cultured; a colony region specifying unit 21 that acquires a cell image by imaging the cells at a first magnification and specifies a colony region of the cells in the cell image; a target region determination unit 23 that determines a target region in the colony region of the cells based on the information regarding the maturity; and a high magnification image acquisition unit 24 that acquires a high magnification image by imaging the target region at a second magnification that is higher than the first magnification.

Abstract: The present disclosure relates to systems and methods for reconstructing a PET image. The systems may execute the methods to acquire PET data of a subject. The PET data may include position information of a plurality of coincident events. The plurality of coincident events may include scattering events and random events. The systems may execute the methods to select a portion of the PET data from the PET data based on the position information. The systems may execute the methods to reconstruct a first preliminary image of the subject based on the selected portion of the PET data, and project the first preliminary image. The systems may execute the methods to may determine, based on the PET data and the projection of the first preliminary image, preliminary correction data relating to the scattering events and the random events.

Abstract: A product monitoring device is provided that presents, to the user, information by which the user can recognize specific work items of a product management work for improving an inappropriate display state. The product monitoring device acquires a captured image of a display area, sets a state monitoring area on the captured image of the display area, acquires product detection information indicating a position of a product in the state monitoring area, based on a captured image of the state monitoring area, evaluates a display state of the product in the state monitoring area, using a plurality of evaluation indices relating to display disorder of the product, based on the product detection information, and presents information on an evaluation result to a user.

Abstract: Systems and methods are provided for a visualization of a multi-modal medical image for diagnostic medical imaging. The systems and methods receive first and second image data sets of an anatomical structure of interest, register the first and second image data sets to a geometrical model of the anatomical structure of interest to form a registered image. The geometrical model includes a location of an anatomical marker. The systems and methods further display the registered image.

Abstract: Among other things, one or more techniques and/or systems are described for processing images yielded from an examination via radiation to reduce visible noise in the images. After an image is reconstructed, a noise contribution to the image (e.g., an amount of noise in the image) is estimated to determine a target noise contribution for the image. The target noise contribution for the image may vary based upon, among other things, dose of radiation, aspects or properties of an object being imaged, etc. The image is subsequently filtered using one or more filtering techniques to generate a filtered image, and a noise contribution to the filtered image is determined. When the noise contribution to the filtered image satisfies the target noise contribution (e.g., a sufficient amount of noise has been filtered out of the image), the filtered image is combined with the reconstructed image to generate a blended image.

Abstract: A method of visualizing anatomical elements in a medical includes receiving a medical image; detecting a plurality of anatomical elements from the medical image; verifying a location of each of the plurality of anatomical elements based on anatomical context information including location relationships between the plurality of anatomical elements; adjusting the location relationships between the plurality of anatomical elements; and combining the verified and adjusted information of the plurality of anatomical elements with the medical image.

Abstract: In a wind turbine of the type including a tower and a nacelle with the rotor being rotatably connected to the nacelle for rotating about a rotor axis and having a plurality of equally spaced blades, there is provided a method of detecting damage to a rotor requiring replacement. The method includes positioning video cameras on each of the blades at a root of a respective one of the blades so as to provide a line of sight of the camera along the respective one of the blades to the tip to obtain a video image of the rotor and tip as they rotate. From the videos an analysis is carried out of the images of the tip at a common location spaced away from the tower to determine a position of the tip and hence the deflection of the tip which is indicative of damage.