Abstract: Techniques are disclosed for measuring the corpus callosum volume of a fetus using magnetic resonance imaging. A scanogram of a fetus is acquired, and a detection area is determined using the corpus callosum position of the fetus in the scanogram. Magnetic resonance scanning is performed on the detection area to obtain a diffusion weighted image, with a gradient direction that is orthogonal or normal to an extending direction of fiber bundles of the corpus callosum. A fetal head image is cropped in the diffusion weighted image, and a predetermined threshold is applied to obtain an image including pixels having a brightness value that is greater than the threshold. Image processing is performed on the binarized image, with the largest region therein being identified as the corpus callosum, and the sum of voxel dimensions associated with the signal of the largest region being calculated as the corpus callosum volume.

Abstract: A method may include applying, to a corpus of data, a first machine learning technique to identify candidate domains of an ontology mapping brain structure to mental function. The corpus of data may include textual data describing a plurality of mental functions and spatial data corresponding to a plurality of brain structures. A second machine technique may be applied to optimize a quantity of domains included in the ontology and/or a quantity of mental function terms included in each domain. The ontology may be applied to phenotype an electronic medical record and predict a clinical outcome for a patient associated with the electronic medical record. Related systems and articles of manufacture, including computer program products, are also provided.

Abstract: A computer-implemented method for autonomous segmentation of contrast-filled coronary artery vessels, the method comprising the following steps: receiving (101) an x-ray angiography scan representing a maximum intensity projection of a region of anatomy that includes the coronary vessels on the imaging plane; preprocessing (102) the scan to output a preprocessed scan; and performing autonomous coronary vessel segmentation (103) by means of a trained convolutional neural network (CNN) that is trained to process the preprocessed scan data to output a mask denoting the coronary vessels.

Abstract: Disclosed are an intelligence device and a method of selecting a user of the intelligence device. According to an embodiment of the disclosure, the intelligence device may analyze the eye blinks and pupil shapes of persons and select the person gazing at the intelligence device as a user. According to an embodiment, the intelligence device may be related to artificial intelligence (AI) modules, robots, augmented reality (AR) devices, virtual reality (VR) devices, and 5G service-related devices.

Abstract: An authentication method comprising creating electrocardiogram data of users; calculating a similarity between electrocardiogram data of each user and template data created by averaging electrocardiogram data of each user; creating and training a first NNmodel for every user by using similarities between electrocardiogram data of a user and template data of the same user and similarities between electrocardiogram data of a user and template data of another user, and creating and training second NNmodels for users by using similarities between electrocardiogram data of a user and template data of the user and similarities between electrocardiogram data of the user and template data of another user; and executing a first step in which the similarities calculated using electrocardiogram data for authentication of a user to be authenticated and template data are input to the first NNmodel, and executing a second step in which the similarities are input to the second NNmodels.

Abstract: The present invention discloses a method, apparatus and system for detecting a fundus image on the basis of machine learning. The method comprises: acquiring a fundus image to be detected; classifying the entire region of the fundus image by using a first classification model to determine whether the fundus image contains a first feature; and if the fundus image does not contain any first feature, classifying a specific region in the fundus image by using at least one second classification model to determine whether the fundus image contains any second feature, wherein the saliency of the first features are greater than that of the second features.

Abstract: A computer-implemented method for correction of a voxel representation of metal affected x-ray data. The method comprises a first 3D deep neural network receiving an initial voxel representation of x-ray data at its input and generating a voxel map at its output, the map identifying voxels of the initial voxel representation that belong to a region of voxels that are affected by metal. A second 3D deep neural network receives the initial voxel representation and the map generated by the first 3D deep neural network at its input and generating a corrected voxel representation, the corrected voxel representation including voxel estimations for voxels that are identified by the voxel map as being part of a metal affected region, the first 3D deep neural being trained on the basis of training data and reference data that include voxel representations of clinical x-ray data of a predetermined body part of a patient.

Abstract: The present disclosure relates generally to the operation of autonomous machinery for performing various tasks at various industrial work sites, and more particularly to the volumetric estimation and dimensional estimation of a pile of material or other object, and the use of multiple sensors for the volumetric estimation and dimensional estimation of a pile of material or other object at such work sites. An application and a framework is disclosed for volumetric estimation and dimensional estimation of a pile of material or other object using at least one sensor, preferably a plurality of sensors, on an autonomous machine (e.g., robotic machines or autonomous vehicles) in various work-site environments applicable to various industries such as, construction, mining, manufacturing, warehousing, logistics, sorting, packaging, agriculture, etc.

Abstract: Disclosed are a method and an apparatus for processing a blurred image. The method for processing a blurred image includes the steps of generating a first input feature map and a second input feature map with a feature distribution for blur removal from the blurred image, generating a prediction feature map from the first input feature map by using a self-spatial feature transform (SSFT) module which transforms the feature distribution for blur removal into a feature distribution for face recognition without external information, and generating a deblurred image based on the second input feature map and the prediction feature map.

Abstract: A learning model provided in a segmentation device is a learning model which is generated using training data such that segmentation data of a biologically important region is output when data of a constituent maxillofacial region is input.

Abstract: A medical information processing apparatus according to an embodiment includes: a memory storing therein a trained model provided with a function to specify, on the basis of input information including a medical image and medical examination information related to the medical image, at least one selected from between a relevant image relevant to the medical image and an image processing process performed on the basis of the medical image; and processing circuitry configured to give an evaluation to at least one selected from between the relevant image and the image processing process specified by the trained model.

Abstract: Methods and systems are provided for improving image quality of ultrasound images by automatically determining one or more image quality parameters via a plurality of separate image quality models. In one example, a method for an ultrasound system includes determining a plurality of image quality parameters of an ultrasound image acquired with the ultrasound system, each image quality parameter determined based on output from a separate image quality model, and outputting feedback to a user of the ultrasound system based on the plurality of image quality parameters.

Abstract: System and method for simultaneous object detection and semantic segmentation. The system includes a computing device. The computing device has a processor and a non-volatile memory storing computer executable code. The computer executable code, when executed at the processor, is configured to: receive an image of a scene; process the image using a neural network backbone to obtain a feature map; process the feature map using an object detection module to obtain object detection result of the image; and process the feature map using a semantic segmentation module to obtain semantic segmentation result of the image. The object detection module and the semantic segmentation module are trained using a same loss function comprising an object detection component and a semantic segmentation component.

Abstract: A benign tumor development trend assessment system includes an image outputting device and a server computing device. The image outputting device outputs first/second images captured from the same position in a benign tumor. The server computing device includes an image receiving module, an image pre-processing module, a target extracting module, a feature extracting module and a trend analyzing module. The image receiving module receives the first/second images. The image pre-processing module pre-processes the first/second images to obtain first/second local images. The target extracting module automatically detects and delineates tumor regions from the first/second local images to obtain first/second region of interest (ROI) images. The feature extracting module automatically identifies the first/second ROI images to obtain at least one first/second features. The trend analyzing module analyzes the first/second features to obtain a tumor development trend result.

Abstract: A device for spatially normalizing a medical image includes: an adaptive template generation unit configured such that when a plurality of functional medical images are input to a deep learning architecture, the adaptive template generation unit generates, based on prestored learning data, an adaptive template for spatially normalizing the plurality of functional medical images; a learning unit configured to learn by repeating a process of generating an image from an input functional medical image of a user based on the adaptive template through a generative adversarial network (GAN) and determine authenticity of the generated image; and a spatial normalization unit configured to provide the functional medical image of the user which is spatially normalized based on results of the learning.

Abstract: Embodiments of the present systems and methods may provide fissure detection in CT images, with improved performance, accuracy, and specificity. For example, in an embodiment, a method may comprise imaging, using a computed tomography system, at least one lung, to generate, at a computer system comprising a processor, memory accessible by the processor, and computer program instructions stored in the memory and executable by the processor, at least one computed tomography image of the at least one lung, determining, at the computer system, at least one approximate fissure region of interest in the at least one lung image, determining, at the computer system, a more precise fissure location within the at least one region of interest, and generating an image of the lung including indication of the determined fissure location.

Abstract: Systems, methods, and computer-readable media for context-aware synthesis for video frame interpolation are provided. Bidirectional flow may be used in combination with flexible frame synthesis neural network to handle occlusions and the like, and to accommodate inaccuracies in motion estimation. Contextual information may be used to enable frame synthesis neural network to perform informative interpolation. Optical flow may be used to provide initialization for interpolation. Other embodiments may be described and/or claimed.

Abstract: CT and PET are registered, providing a spatial alignment to be used in attenuation correction for PET reconstruction. A model for machine learning is defined to generate a deformation field. The model is trained with loss based, in part, on the attenuation corrected PET data rather than or in addition to loss based on the uncorrected PET or the generated deformation field. Due to the nature of the mapping from CT to attenuation, a separate, pre-trained network is used to form the attenuation corrected PET data in training the model.

Abstract: Disclosed is a system and method for assembling instrument sets that include correct instruments with one or more verified states for different procedures. The system may receive a request for a particular instrument, and may determine instrument states defined for the particular instrument or a procedure involving the particular instrument. The system may scan a first instrument using one or more sensors, may verify that the first instrument matches a make, model, or type of the particular instrument based on the scanning data, and may classify the first instrument states with at least a threshold probability based on the scanning data matching characteristics from a probabilistic model. The system may control the distribution of the first instrument to a first destination or a second destination based on whether or not the first instrument states satisfy the instrument states defined for the particular instrument or the procedure.

Abstract: A system and method for automated localization, enumeration, and diagnoses of a tooth/condition. The system detects a condition for at least one defined localized and enumerated tooth structure within a cropped image from a full mouth series based on any one of a pixel-level prediction, wherein said condition is detected by at least one of detecting or segmenting a condition on at least one of the enumerated tooth structures within the cropped image by a 2-D R-CNN.