Abstract: A computer-implemented method and system of CT reconstruction can include receiving one or more CT projection images; performing smooth filtering on the one or more CT projection images to generate one or more smooth boundary filtered images; and back-projecting the one or more smooth boundary filtered images.

Abstract: A weakly supervised intracranial hemorrhage (ICH) detection workflow includes training a deep learning (DL) model including a coupled convolutional neural network and recurrent neural network on a large dataset of CT scans with expert-labeled slices indicating presence or absence of ICH. Transfer learning (TL) is used to further train the DL model using a second large dataset of CT scans with only scan labels extracted from radiology reports using natural language processing (NLP). The DL model weights each slice of the scan against the final ICH diagnosis using an attention-based bi-directional long-short term memory network, where the attention weights represent slice-level ICH predictions. Model-generated heatmaps highlight significant regions of the CT scans that lead to the provided ICH predictions.

Abstract: An information processing device, an information processing method, and a non-transitory computer-readable medium are disclosed. The information processing device includes a processor configured to acquire transverse tomographic images of a plurality of frames obtained by imaging a blood vessel of a patient, and calculate plaque burden in each of the frames by inputting the acquired transverse tomographic image of each of the frames to a model that has learned to calculate plaque burden in response to the transverse tomographic image inputted, and a display unit configured to display a longitudinal tomographic image based on the transverse tomographic images of the plurality of frames and a first object that is an object displayed in correlation with the longitudinal tomographic image and represents magnitude of the plaque burden at each position of the longitudinal tomographic images along an axial direction of the blood vessel.

Abstract: A system comprises an authentication sub-system configured to authenticate a user and a biochemical analysis sub-system configured to perform a biochemical analysis on a biological sample provided to the biochemical analysis sub-system. The system is configured to authenticate a user using the authentication sub-system, instruct an authenticated user to provide a biological sample to the biochemical analysis sub-system, execute a biochemical analysis on a provided biological sample.

Abstract: A method for training a 3D space consistent medical volume data generative model is provided, which is performed by one or more processors and which includes generating a synthetic volume data using a generator network, receiving real volume data, generating a plurality of 2D images based on the synthetic volume data and the real volume data, discriminating whether each of the plurality of generated 2D images is real or fake by using a plurality of discriminator networks, and updating the generator network based on discrimination results of the plurality of discriminator networks.

Abstract: A system and method for navigation of a luminal network including receiving a computed tomography (CT) image data set, generating a three-dimensional (3D) model, displaying the 3D model in a user-interface on a display operatively connected to the computing device, and receiving an indication of a location of a tumor in the CT image data set. The system and method further including displaying the location of the tumor in the 3D model, receiving an indication of a margin around the tumor to achieve a desired therapy, generating a pathway to the tumor for navigation of a catheter, receiving a location of a sensor associated with a navigation catheter and registering the CT image data set with a luminal network of a patient, displaying the location of the sensor within the 3D model that substantially corresponds to the location of the sensor within the luminal network of the patient.

Abstract: A system and method automatically detect, in MR images, WM lesions exhibiting a central vein sign. A set of MR images of a brain lesion is acquired, using images of the set as input to different ML algorithms. A first ML algorithm classifies inputted image(s) into first or second classes. The first class includes CVS+/? and the second class CVSe lesions. A second ML algorithm classifies inputted image(s) into third or fourth classes. The third class includes CVS+ lesions and the fourth central vein sign? lesions. For each set, probability values are used that the set belongs to classes as inputs to a final classifier performing a final classification of the set into second, third, or fourth classes. For each class, the final classifier outputs final probability that the set belongs to the class. The second, third or fourth class with highest probability value is provided through an interface.

Abstract: An image processing method and apparatus are provided. The method includes: acquiring a sample image, the sample image comprising an image obtained by image-capturing a sample within a preset optical waveband; processing a first image corresponding to at least one preset wavelength within the preset optical waveband in the sample image, to obtain a pseudo-color image; performing region division on the sample image according to a difference in sample element types in the sample image to obtain a region division result, the sample element types comprising a recognition element type; and determining an image region comprising the recognition element type from the sample image based on the pseudo-color image and the region division result.

Abstract: Systems and methods for generating attenuation maps for reconstructing medical images are disclosed. In some examples, measurement data, such as positron emission tomography (PET) data or single-photon emission computed tomography (SPECT) data, is received for a subject. A machine learning process is applied to the measurement data to generate initial synthetic images for multiple values of an imaging parameter. Further, patient data is received and classified to determine an object imaged with the subject. A second medical image is selected that includes the object, and a region-of-interest (ROI) of the initial synthetic images is determined. Further, based on the ROI an anatomical mask is generated for each initial synthetic image. A second image-to-image network process is applied to the patient data, the second medical image, a portion of each initial synthetic image that includes the ROI, and the corresponding anatomical mask to generate a final synthetic image.

Abstract: Detecting motions associated with a body part of a patient may include using an image sensor installed inside a medical scanner to capture first and second images of the patient inside the medical scanner, wherein the first image may depict the patient in a first state and the second image may depict the patient in a second state. A first area, in the first image, that corresponds to the body part of the patient may be identified and a second area, in the second image, that corresponds to the body part may also be identified so that a first plurality of features may be extracted from the first area of the first image and a second plurality of features may be extracted from the second area of the second image. A motion associated with the body part of the patient may be determined based on the first and second pluralities of features.

Abstract: A computer-implemented method of evaluating a user image of a patient to enable identification of a historical twin of the patient. The method includes organizing a plurality of medical images in an archive and receiving from a medical professional each of: (i) a region of interest; (ii) a textual description; (iii) selections for binary criteria; and (iv) weights of weighable criteria. The method comprises using a natural language search to create a relevant set of medical images and creating an optimal set from the relevant set of medical images by discarding medical images from the relevant set based at least on the selections for binary criteria. The method includes image processing medical images in the optimal set using the weight of the features of the region of interest to create medical image results. The relevant set comprises less than ten percent of the medical images in the archive.

Abstract: Systems and methods for intraocular lens selection include receiving, by one or more computing devices implementing a prediction engine, pre-operative multi-dimensional images of an eye; extracting, by the prediction engine, pre-operative measurements of the eye based on the pre-operative images; estimating, by the prediction engine using a prediction model based on a machine learning strategy, a post-operative position of an intraocular lens based on the extracted pre-operative measurements; selecting a power of the intraocular lens based on the estimated post-operative position of the intraocular lens; and selecting the intraocular lens based on the selected power. In some embodiments, the systems and methods further include receiving post-operative multi-dimensional images of the eye after implantation of the selected intraocular lens, extracting post-operative measurements of the eye, and updating the prediction model based on the pre-operative measurements and the post-operative measurements.

Abstract: A method for analyzing and displaying functional lumen imaging probe (FLIP) measurement data includes receiving FLIP measurement data. A heatmap or topograph of the measurement data is generated and displayed to thereby create an image of a diameter along the lumen. The image is segmented to display only diameters of the lumen within a range of diameter values specified by a user.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating an output video conditioned on an input. In one aspect, a method comprises receiving the input; initializing a current intermediate representation; generating an output video by updating the current intermediate representation at each of a plurality of iterations, wherein the updating comprises, at each iteration: processing an intermediate input for the iteration comprising the current intermediate representation using a diffusion model that is configured to process the intermediate input to generate a noise output; and updating the current intermediate representation using the noise output for the iteration.

Abstract: An inspection system includes: an identification model learning means that performs machine-learning of a model identifying the type of a target object from time-series data representing the movement trajectory of the target object obtained by observation; a confidence level prediction model learning means that performs machine-learning of a confidence level prediction model estimating the confidence level of an estimation result by the identification model from the observation specification of time-series data representing the movement trajectory of a target object; and a determining means that uses the learned identification model to estimate the type of a target object from the movement trajectory of the target object obtained by observation, and uses the learned confidence level prediction model to predict the confidence level of an estimation result by the identification model from the observation specification of the time-series data.

Abstract: A cardiac catheterization image recognition and evaluation method is disclosed. The first deep learning algorithm is used to conduct an object recognition process on the cardiac catheterization image to obtain the vessel object image. The image processing process is conducted to the cardiac catheterization image to obtain the vessel location image. The vessel object image and the vessel location image are combined to obtain the vessel contour image. The vessel type judging process is conducted to the vessel contour image to determine the type of vessel in the cardiac catheterization image. The second deep learning algorithm is used on the vessel contour image to detect the vessel occlusion location and to judge the vessel occlusion rate. Based on the type of vessel and the vessel occlusion rate at the vessel occlusion location, the cardiac catheterization image is evaluated to obtain the SYNTAX Score.

Abstract: Systems and methods for detecting the cracks in illuminated electronic device screens are disclosed. In one embodiment, the method includes receiving an image of an electronic device screen and retrieving a plurality of kernels, each having values corresponding to a line region and a non-line region, with the orientation of the line region and the non-line region differing for each kernel. At least some of the kernels are applied to the image to obtain, at various locations of the image, values corresponding to the line regions and the non-line regions. Based on the values corresponding to the line regions and the non-line regions, cracks are automatically identified in the electronic device screen.

Abstract: The present invention provides an image analysis method for generating data indicating a region of a cell nucleus in an image of a tissue or a cell. The image analysis method is a method for analyzing an image of a tissue or a cell using a deep learning algorithm of a neural network structure, and generates data indicating a region of a cell nucleus in an analysis target image by the deep learning algorithm by generating analysis data from the analysis target image including an analysis target tissue or cell, and inputting the analysis data in the deep learning algorithm.

Abstract: A computer-implemented technique is described herein for interpolating input data that includes image and/or audio content. The technique identifies plural sizes associated with different respective phenomena exhibited by the input data. The technique then interpolates the input data in a pipeline that includes plural passes. The plural passes are controlled using plural respective parameter values. The plural respective parameter values, in turn, are selected based on the plural respective sizes, arranged from largest to smallest. In other implementations, the technique chooses pass-specific algorithmic changes to be applied by the interpolation algorithms used by the different passes. In other implementations, the technique chooses its configurations without regard to sizes of phenomena that may be exhibited in the input data. The technique is advantageous because it reduces the presence of artifacts in output data produced by the computer-implemented technique.

Abstract: Systems and corresponding methods are provided for moving object predictive locating, reporting, and alerting. This method includes receiving moving object data corresponding to a moving object; receiving sensor data from a sensor and merging the received moving object data and the received sensor data into a set of merged data. The example method further includes based on the set of merged data, automatically determining one or more of a predicted location or range of locations for the moving object, a potential path of travel for the moving object, an alert concerning the moving object, and providing the alert. The automatically determining may be further based on one or more historical traits concerning the object, and the geographic medium the object is moving through. The geographic medium may include one or more of terrain, air, water, and space. The object may be a soldier, vehicle, drone, or ballistic.