Abstract: Example methods and systems for training deep learning engines for radiotherapy treatment planning are provided. One example method may comprise: obtaining a set of training data that includes unlabeled training data and labeled training data; and configuring a deep learning engine to include (a) a primary network and (b) a deep supervision network that branches off from the primary network. The method may further comprise: training the deep learning engine to perform the radiotherapy treatment planning task by processing training data instance to generate (a) primary output data and (b) deep supervision output data; and updating weight data associated with at least some of the multiple processing layers based on the primary output data and/or the deep supervision output data. The deep supervision network may be pruned prior to applying the primary network to perform the radiotherapy treatment planning task for a patient.

Abstract: Disclosed herein are methods and system for training artificial intelligence models configured to execute image segmentation techniques. The methods and system describe a server that receives a first image including a set of pixels depicting multiple objects. The server also receives a second image having a second set of pixels depicting the same set of objects. The server then analyzes the pixels from the first and second images. When a difference between at least one visual attribute of a pixel within the second image and a corresponding pixel within the first image satisfies a predetermined threshold, it will be encoded as spikes to send to the model, the model will be trained using supervised STDP rule by revising weights associated with the nodes within the AI model where the node corresponds to the pixels within the first and/or the second image.

Abstract: Dental images are processed according to a first machine learning model to determine teeth labels. The teeth labels and image are processed using a second machine learning model to label anatomy. The anatomy labels, teeth labels, and image are processed using a third machine learning model to obtain feature measurements, such as pocket depth and clinical attachment level. The feature measurements, labels, and image may be input to a fourth machine learning model to obtain a diagnosis for a periodontal condition. Machine learning models may further be used to reorient, decontaminate, and restore the image prior to processing. A machine learning model may be trained with images and randomly generated masks in order to perform inpainting of dental images with missing information.

Abstract: The invention relates to a medical image data processing system (101) for image segmentation. The medical image data processing system (101) comprises a machine learning framework trained to receive an anatomical position of a voxel and to provide a tissue type classification. An execution of machine executable instructions by a processor (130) of the medical image data processing system (101) causes the processor (130) to control the medical image data processing system (101) to: —receive medical image data (140) comprising an anatomical structure of interest, —fit an anatomical frame of reference (302, 402) to the medical image data (140) using model-based segmentation, —classify tissue types represented by voxels of the medical image data (140) using the machine learning framework, wherein anatomical positions of the voxels with respect to the anatomical frame of reference (302, 402) are used as the input to the machine learning framework.

Abstract: A gate apparatus includes: an exit gate door; a first biometrics information acquisition unit that acquires, from a user who moves toward the exit gate door in a closed state, first target biometrics information to be compared with registered biometrics information registered in advance; a second biometrics information acquisition unit that acquires second target biometrics information to be compared with the registered biometrics information from the use who stops in front of the exit gate door when there is no matching in a comparison between the first target biometrics information and the registered biometrics information or the comparison is unable to be performed; and a door control unit that opens the closed exit gate door in accordance with a result of a comparison between the first target biometrics information or the second target biometrics information and the registered biometrics information.

Abstract: Technology disclosed herein may involve a computing system that (i) generates (a) a first feature map based on a first visual input from a first perspective of a scene utilizing at least one first neural network and (b) a second feature map based on a second visual input from a second, different perspective of the scene utilizing at least one second neural network, where the first perspective and the second perspective share a common dimension, (ii) based on the first feature map and a portion of the second feature map corresponding to the common dimension, generates cross-referenced data for the first visual input, (iii) based on the second feature map and a portion of the first feature map corresponding to the common dimension, generates cross-referenced data for the second visual input, and (iv) based on the cross-referenced data, performs object detection on the scene.

Abstract: In a computer implemented method of determining a boundary of a tumor region or other diseased tissue, hyper- or multispectral image data of a tissue sample including a tumor region or other diseased tissue is taken. The analysis includes a morphological analysis and a spectral analysis of the hyper- or multispectral image data resulting in a morphological tumor boundary and a spectral tumor boundary or a morphological diseased tissue boundary and a spectral diseased tissue boundary. These two boundaries are combined resulting in a combined tumor boundary or combined diseased tissue boundary, wherein an indication of reliability of the combined tumor boundary or combined diseased tissue boundary is given.

Abstract: An in-vehicle device comprises an image recognizer configured to acquire an image using a camera for photographing the exterior of a vehicle and execute processing for recognizing license plate information included in the image at intervals of a predetermined period; and a period determiner configured to determine the period to perform the recognition processing on the basis of a length of time in which a license plate of another vehicle stays within a field of view of the camera.

Abstract: Provided are an image processing apparatus, an image processing method, and a program that can collect high-quality correct answer data used for machine learning with a simple method. The image processing apparatus includes: a first extractor that extracts a measurement target region from a medical image, using a result of learning performed using correct answer data of the measurement target region; a measurement object determination unit that determines a measurement object used to measure the measurement target region; a measurement object correction unit that corrects the measurement object in response to a command from a user; and a measurement target region correction unit that corrects the measurement target region extracted by the first extractor, using a correction result of the measurement object. The first extractor performs learning using the measurement target region corrected by the measurement target region correction unit as correct answer data.

Abstract: As the capabilities of digital histopathology machines grows, there is an increasing need to ease the burden on pathology professionals of finding interesting structures in such images. Digital histopathology images can be at least several Gigabytes in size, and they may contain millions of cell structures of interest. Automated algorithms for finding structures in such images have been proposed, such as the Active Contour Model (ACM). The ACM algorithm can have difficulty detecting regions in images having variable colour or texture distributions. Such regions are often found in images containing cell nuclei, because nuclei do not always have a homogeneous appearance. The present application describes a technique to identify inhomogeneous structures, for example, cell nuclei, in digital histopathology information. It is proposed to search pre-computed super-pixel information using a morphological variable, such as a shape-compactness metric, to identify candidate objects.

Abstract: A person watching a video shares the realism of the spot. There is provided an information processing apparatus including a video acquirer, an object information collector, a status recognizer, and a display unit. The video acquirer acquires a video from an image capturer that captures a predetermined area. The object information collector collects object information of an object included in the captured video. The status recognizer recognizes a status of the object based on the video and the object information of the object. The display unit displays the recognized status to be identifiable.

Abstract: Provided is a method of controlling an image and sound pickup device, which is includes obtaining a plurality of audio signals and a participant image, which shows a plurality of participants, and generating location information about a sound source location by using comparison information about a comparison among the plurality of audio signals and face recognition that is performed on the participant image; and generating an estimated utterer image, which displays an estimated utterer, by using the location information.

Abstract: Methods and systems described in this disclosure allow customers to personalize their phone experience when calling into an organization. In some embodiments, customers who may benefit from this service are identified based on the content of the customer's previous or current phone calls to the organization. The identified customers may be invited to enroll and to provide preferences for a customized Interactive Voice Response experience. In some embodiments, the customer can elect to hear the balances of one or more of his accounts without going through a phone menu or asking a representative to look up the relevant amounts. Once enrolled, when the customer dials into the organization and upon successful authentication, the organization proactively states the customer's account balances with no further customer request.

Abstract: Aspects of the subject disclosure may include, for example, analyzing media content to recognize an object therein, wherein the media content is provided as a video stream displayed to a user of a communication device, the media content comprising a plurality of digital frames transported to the communication device over a private network; associating product information of a product with the object; receiving a user input selecting the object; determining providers of goods or services of the product, wherein the providers of goods or services are selected during the determining based on a proximity of the providers to the user of the communication device; and providing the product information to the communication device. Other embodiments are disclosed.

Abstract: Systems and methods are provided for automatic detection and quantification for traumatic bleeding. Image data is acquired using a full body dual energy CT scanner. A machine-learned network detects one or more bleeding areas on a bleeding map from the dual energy CT scan image data. A visualization is generated from the bleeding map. The predicted bleeding areas are quantified, and a risk value is generated. The visualization and risk value are presented to an operator.

Abstract: To provide a technology of more accurately detecting spoofing in face authentication, without increasing a scale of a device configuration and a burden on a user. A spoofing detection device includes a facial image sequence acquisition unit, a line-of-sight change detection unit, a presentation information display unit, and a spoofing determination unit. The facial image sequence acquisition unit acquires a facial image sequence indicating the face of a user. The line-of-sight change detection unit detects information about a temporal change in the line-of-sight from the facial image sequence. The presentation information display unit displays presentation information presented to the user as part of an authentication process. The spoofing determination unit determines the likelihood of the face indicated by the facial image sequence being spoofing on the basis of the information about the temporal change in the line-of-sight with respect to the presentation information.

Abstract: A system and method for acquiring magnetic resonance (MR) images, with an MR system, of tissue proximal to a target treatment region in a patient, the tissue including a non-spatially uniform subcutaneous fat layer; in a computer comprising a hardware-based processor, automatically determining a thickness of the subcutaneous fat layer using a trained neural network, the neural network trained using manually-segmented MR images from previous patients; in the computer, automatically adjusting a treatment parameter based on the thickness of the subcutaneous fat layer; and delivering thermal therapy to the target treatment region with a high-intensity focused ultrasound system based on the adjusted treatment parameter.

Abstract: The present disclosure provides methods and devices for determining liver segments in a medical image. The methods may be implemented on the devices. The method may include: obtaining a scan image; obtaining a segmentation protocol; obtaining segmentation information associated with the scan image; determining one or more marked points based on the segmentation information and the segmentation protocol; determining one or more segmentation surfaces based on the one or more marked points; and determining a segmentation result of at least part of a liver in the scan image based on the one or more segmentation surfaces.

Abstract: Described herein is a method for detecting keypoints in three-dimensional images in which a three-dimensional image of a scene captured by a depth sensing imaging system is processed using a distance-independent keypoint filter. Keypoints are derived from the three-dimensional image by determining a mean shift field and using x- and y-components of the mean shift field to derive intersections of 0-isolines thereof. Positive and negative keypoints or nodes are connected to one another, positive to positive and negative to negative, to form a keygraph structure.

Abstract: A discriminator includes a common learning unit and a plurality of learning units that are connected to an output unit of the common learning unit. The discriminator is trained, using a plurality of data sets of a first image obtained by capturing an image of a subject that has developed a disease and an image data of a disease region in the first image, such that information indicating the disease region is output from a first learning unit in a case in which the first image is input to the common learning unit. In addition, the discriminator is trained, using a plurality of data sets of an image set obtained by registration between the first image and a second image whose type is different from the type of the first image, such that an estimated image of the second image is output from an output unit of a second learning unit.