Abstract: The present disclosure describes ultrasound imaging systems and methods configured to identify ovarian follicles by applying a neural network to pre-processed ultrasound image frames. Systems may include an ultrasound transducer configured to acquire echo signals responsive to ultrasound pulses transmitted toward a target region. One or more processors communicatively coupled with the ultrasound transducer may be configured to generate at least one image from the ultrasound echoes. A threshold may be applied to the image frame that differentiates pixels representative of an ovarian follicle present in the target region from other subject matter. The processors may apply a neural network to the thresholded image frame, in which the neural network determines the presence of the ovarian follicle in the thresholded image frame. The processors also may generate an indicator based on the presence of the ovarian follicle and display the indicator on a user interface.

Abstract: A system for imaging reconstruction is provided. The system may obtain a first set of image data of a subject acquired by a scanner and a second set of image data of the subject acquired by the scanner. The first set of image data may correspond to a first angle range of the scanner. The second set of image data may correspond to a second angle range of the scanner. The first angle range may be different from the second angle range. The system may also generate a first image based on the first set of image data and generate a second image based on the second set of image data. The system may further generate a target image based on the first image and the second image.

Abstract: An information acquisition unit acquires at least one information item of subject information, imaging condition information, or positioning information. Next, a body movement feature derivation unit derives body movement feature information indicating a feature of body movement of a breast which occurs at the time of the tomosynthesis imaging. A storage stores correspondence information in which the derived body movement feature information corresponds to at least one information item of the subject information, the imaging condition information, or the positioning information.

Abstract: Provided are a cell image evaluation device, method, and program which are capable of more accurate and high reliable evaluation even though a captured image of each part to be observed within a container deteriorates. The cell image evaluation device includes an image evaluation unit that evaluates a state of a cell included in a captured image obtained by capturing an inside of the container that contains the cell based on the captured image, and a deterioration determination unit that determines whether or not the captured image deteriorates. The image evaluation unit changes an evaluation method of the captured image according to a determination result of the deterioration determination unit.

Abstract: Novel tools and techniques are provided for implementing digital microscopy imaging using deep learning-based segmentation and/or implementing instance segmentation based on partial annotations. In various embodiments, a computing system might receive first and second images, the first image comprising a field of view of a biological sample, while the second image comprises labeling of objects of interest in the biological sample. The computing system might encode, using an encoder, the second image to generate third and fourth encoded images (different from each other) that comprise proximity scores or maps. The computing system might train an AI system to predict objects of interest based at least in part on the third and fourth encoded images. The computing system might generate (using regression) and decode (using a decoder) two or more images based on a new image of a biological sample to predict labeling of objects in the new image.

Abstract: A method for characterising motion of one or more objects in a time ordered image dataset comprising a plurality of time ordered data frames, the method comprising: selecting a reference data frame from the plurality of time ordered data frames (210); extracting a plurality of image patches from at least a part of the reference data frame (220); identifying a location of each image patch of at least a subset of the plurality of image patches in each data frame (230); defining, based on the identified locations, a mesh for each data frame, wherein vertices of each mesh correspond to respective identified locations of image patches in the corresponding data frame (240); and deriving, from the meshes, a motion signature for the time ordered image dataset, the motion signature characteristic of the motion of the one or more objects in the plurality of time ordered data frames (250).

Abstract: A system and method for performing intersection scenario retrieval that includes receiving a video stream of a surrounding environment of an ego vehicle. The system and method also include analyzing the video stream to trim the video stream into video clips of an intersection scene associated with the travel of the ego vehicle. The system and method additionally include annotating the ego vehicle, dynamic objects, and their motion paths that are included within the intersection scene with action units that describe an intersection scenario. The system and method further include retrieving at least one intersection scenario based on a query of an electronic dataset that stores a combination of action units to operably control a presentation of at least one intersection scenario video clip that includes the at least one intersection scenario.

Abstract: The invention relates generally to both a method and apparatus for the creation of full resolution color digital images of diagnostic cassettes or objects of interest using a gray-scale digital camera or sensor combined with time sequential illumination using additive primary colors followed by post exposure digital processing. Such procedures and equipment is of significant economic value when employed in situations such as diagnostic clinical analyzers where space is limited and image quality requirements are high.

Abstract: Disclosed are an apparatus and method for assisting the diagnosis of a medical image by an automated system. A computing system includes a memory or database that stores a plurality of medical image diagnosis algorithms each having a medical image diagnosis function. A processor inside the computing system extracts diagnosis requirements for a medical image by analyzing the medical image, selects a plurality of diagnosis application algorithms to be applied to the diagnose of the medical image from among a plurality of medical image diagnosis algorithms based on the diagnosis requirements, and generates display information including diagnosis results for the image frame by applying the plurality of selected diagnosis application algorithms to the image frame.

Abstract: Provided are apparatuses, a non-transitory computer-readable medium or media, and methods for supporting reading of a fundus image of a subject. In certain aspects, disclosed a method including the steps of: extracting a first feature information from a first fundus image of the subject based on a machine learning model; generating a second fundus image having a second feature information by mapping an adversarial factor to the first fundus image so that the first feature information is changed; and displaying the first fundus image having the first feature information and the second fundus image having the second feature information on a display device.

Abstract: Systems and methods are provided for generating sets of candidates comprising images and places within a threshold geographic proximity based on geographic information associated with each of the plurality of images and geographic information associated with each place. For each set of candidates, the systems and methods generate a similarity score based on a similarity between text extracted from each image and a place name, and the geographic information associated with each image and each place. For each place with an associated image as a potential match, the systems and methods generate a name similarity score based on matching the extracted text of the image to the place name, and store an image as place data associated with a place based on determining that the name similarity score for the extracted text associated with the image is higher than a second predetermined threshold.

Abstract: A processor-implemented method with crosstalk correction includes: determining a region in which a crosstalk is to occur based on a three-dimensional (3D) position relationship between a position of eyes of a user and a position of a virtual image of a virtual content object; generating a concealer image for correcting the region in which the crosstalk is to occur based on the determined region and the virtual content object; and correcting the crosstalk by combining the virtual content object and the generated concealer image.

Abstract: A system including a motorized base unit with a smart device mount for automatically orienting the smart device camera toward a moving target to track the moving target and take pictures or video. The target (e.g., a child playing soccer) wears a tracking tag comprising a GPS chip and transmitter packaged inside an athletic pad. The base unit includes a motorized mast for mounting a smart device. The base unit receives the transmitted GPS data, calculates updated pointing angle and angular velocity for the smart phone based on update location information from the remote tag sensor, calculates the correct angle that the smart phone should be pointed at, translates the new pointing directions to a control signal that turns the mast, which in turn causes the smart device camera to “follow” or track the target.

Abstract: The present disclosure discloses a three-dimensional data alignment apparatus, a three-dimensional data alignment method, and a recording medium, which may align a location between volumetric data and surface data even without a segmentation process of extracting a surface from the volumetric data. A three-dimensional data alignment apparatus according to an exemplary embodiment of the present disclosure includes a three-dimensional data alignment unit for aligning a location between first three-dimensional data and second three-dimensional data expressed in different data forms with regard to a target to be measured. The first three-dimensional data are three-dimensional data acquired in a voxel form with regard to the target to be measured, and the second three-dimensional data are three-dimensional data acquired in a surface form with regard to the target to be measured.

Abstract: An image sensor, electronic device and method thereof that performs on-sensor single object class detection using an on-sensor single object class detection deep neural network (DNN), such as a face detection DNN. The single object class detection DNN includes a pixel array layer configured to capture an image and transfer image data of the captured image, and a logic and single object class detection deep neural network (DNN) layer that receives the image data directly from the pixel array layer and outputs the image data with the single object class detection data to a communication bus of an electronic device.

Abstract: It is an object to realize a correcting process for correcting a defective image in a region of interest (ROI) that is a partial region segmented from a captured image. A transmitting apparatus includes a controlling section that controls the holding of defect correcting information for use in correcting a defect in an image included in a ROI and a transmitting section that sends out image data of the image included in the ROI as payload data and sends out ROI information as embedded data.

Abstract: A video may be captured by an image capture device in motion. A stabilization trajectory for the video may reflect stabilization rotational positions to compensate for at least some of the motion of the image capture device. The stabilization trajectory may have a stabilization trajectory length. The stabilization of the visual content may be assessed based on the stabilization trajectory length.

Abstract: An imaging apparatus includes a storage portion that stores image data obtained by imaging, and a processing portion that processes the image data, in which the processing portion performs processing of reducing the image data, performs first detection processing of detecting a specific subject image showing a specific subject from a reduced image indicated by reduced image data obtained by reducing the image data, and in a case where the specific subject image is not detected by the first detection processing, performs second detection processing on pre-reduction image data that is the image data before reduction stored in the storage portion.

Abstract: A method for automatically enhancing image data comprising receiving image data. Utilizing a neural network to analyze the image data to predict how each of a plurality of image enhancement processes will affect the image data. Utilizing the neural network to calculate a reward value for each of the plurality of enhancement processes that can be applied to the image data. Determining if the image data should be enhanced or not, wherein the determination is based on the predictions how each of the plurality of image enhancement processes will affect the image data. When it is determined that the image data should be enhanced, then determining which of the plurality of image enhancing process should be applied to the image data, wherein determination is based on the reward value for each of the plurality of enhancement processes. Applying the determined enhanced processes to the image data.

Abstract: A fluorescence image registration method includes obtaining at least one fluorescence image of a biochip. An interior local area. Sums of pixel values in the interior local area along a first direction and a second direction are obtained. A plurality of first template lines is selected to find a minimum total value of the sums of pixel values corresponding to the first template lines. Pixel-level correction is performed on a local area of the track line to obtain a pixel-level track cross. Other track crosses on the biochip is obtained, and the pixel-level correction is performed on the other track crosses. The position of the pixel-level track line is corrected by a center-of-gravity method to obtain the subpixel-level position of the track line. The subpixel-level positions of all sites uniformly distributed on the biochip is obtained.