Abstract: Embodiments include accessing an image of a region of tissue demonstrating cancerous pathology; detecting a plurality of cells represented in the image; segmenting a cellular nucleus of a first member of the plurality of cells and a cellular nucleus of at least one second, different member of the plurality of cells; extracting a set of nuclear morphology features from the plurality of cells; constructing a feature driven local cell graph (FeDeG) based on the set of nuclear morphology features and a spatial relationship between the cellular nuclei using a mean-shift clustering approach; computing a set of FeDeG features based on the FeDeG; providing the FeDeG features to a machine learning classifier; receiving, from the machine learning classifier, a classification of the region of tissue as a long-term or a short-term survivor, based, at least in part, on the set of FeDeG features; and displaying the classification.

Abstract: An action selection learning device includes a memory; and a processor coupled to the memory and configured to generate a reference model that is a set of model parameter vectors that indicate an influence level of each factor that influences selection of an action alternative, calculate a selection probability for each action alternative, for each of the model parameter vectors, calculate a model parameter vector for each user using a subset of model parameter vectors extracted from the reference model, based on the selection probability for each action alternative and a selection history of the action alternative by each user, generate the action alternatives based on the model parameter vector for each user, and transmit the generated action alternatives to a terminal device.

Abstract: Samples each including a feature amount and a label indicating one of a first value and a second value are classified into three or more clusters according to their feature amounts. Some of the clusters are combined to form a combined cluster on the basis of information on the number of samples with labels of the first value. The accuracy level of each sample with respect to the first value or the second value is calculated on the basis of the result of classifying into the combined cluster and the remaining clusters. Then, machine learning is performed using samples whose accuracy levels do not meet a prescribed condition more preferentially than samples whose accuracy levels meet the prescribed condition.

Abstract: Systems and methods for detecting defects are disclosed. According to certain embodiments, a method of performing image processing includes acquiring one or more images of a sample, performing first image analysis on the one or more images, identifying a plurality of first features in the one or more images, determining pattern data corresponding to the plurality of first features, selecting at least one of the plurality of first features based on the pattern data, and performing second image analysis of the at least one of the plurality of first features. Methods may also include determining defect probability of the plurality of first features based on the pattern data. Selecting the at least one of the plurality of first features may be based on the defect probability.

Abstract: A method comprising: receiving, as input, a video sequence comprising a plurality of frames and depicting a scene; detecting, in each of said frames, one or more regions of interest (ROI) associated with an object in said scene; extracting, from each of said ROIs with respect to all of said frames, a feature set representing spectral reflectance; at a training stage, train a machine learning classifier on a training set comprising: (i) all of said feature sets, and (ii) labels indicating whether each of said ROIs depicts living skin tissue; and at an inference stage, applying said trained machine learning classifier to a target feature set extracted from a target ROI detected in a target video sequence, to determine whether said ROI depicts living skin tissue.

Abstract: A piece of medical information, e.g., a medical image of tissue, may be received for processing and analysis on a computing device or system. A region of the medical image may be analyzed to determine a presence of one or more contours in the region. One or more properties of the one or more contours may be extracted, where the one or more properties are inputted into a first algorithm to determine an indication of cancer for the region. The indication of cancer may be inputted into a second algorithm to generate a cancer score for the region.

Abstract: Methods and systems are provided for automatically selecting a target area to project content thereon. For example, a projection device receives content to be projected and content attributes of the content. The projection device also captures images of candidate areas and determines candidate area characteristics based on the captured images. The projection device generates a respective quality-of-projection indicator based on the content attributes and the candidate area characteristics. The projection device selects the candidate area with the highest quality-of-projection indicator as the target area on which the content is to be projected.

Abstract: Systems and methods may utilize a predictive analysis model to analyze a contract or other document. A system may parse a document and/or a repository of information associated with the document. The system may identify one or more terms in the document and corresponding terms in the repository. The system may determine a difference parameter between a first term extracted from the document and a second term extracted from the repository. The system may determine whether the difference between the first term and the second term, represented by the difference parameter, is likely to be acceptable to the user using a predictive analysis model. The system may report a validation parameter indicating a level of acceptability associated with the difference. User feedback on the accuracy of the predictive analysis model is used to train, modify, and improve the predictive analysis model.

Abstract: Embodiments of the present invention provide a computer system, a computer program product, and a method that comprises determining potential risks associated with a building by analyzing images of the building; generating an interactive diagram that includes a visual representation of structures associated with the building that depicts the determined potential risks associated with the building; and creating a geo-fence surrounding the building upon the program determining potential risks.

Abstract: Certain aspects of the present disclosure provide a method for assessing the performance of a physical activity, including: recording motion capture data while a training subject demonstrates a physical activity sequence; identifying one or more primary body elements based on the motion capture data; deriving one or more path characteristic metrics based on a state variable set and the one or more primary body elements, wherein the state variable set defines the state of a body at any given time; and defining an ideal activity path of the physical activity sequence based on the one or more path characteristic metrics.

Abstract: Methods, systems, and computer program products for machine learning model decision boundary enhancement are provided. Aspects include determining a decision boundary associated with a first machine learning model, wherein the first machine learning model is trained with an initial training set of features and associated classifications from a verification model, obtaining a plurality of new feature sets, analyzing, by the first machine learning model, the plurality of new feature sets to determine that a subset of feature sets have classification predictions within a threshold range of the decision boundary associated with the first machine learning model, inputting the subset of feature sets into the verification model to determine enhanced training data based on outputs of the verification model, and creating an enhanced machine learning model by further training the first machine learning model with the initial training data and the enhanced training data.

Abstract: The present disclosure generally relates to the measurement of objects. Using unique processes, the present systems and methods can determine the size of an object using a computing device. For example, in certain embodiments, the present systems and methods receive a physical object such as a foot determine the size of the foot in millimeters, and convert the size of the foot in millimeters into a shoe size.

Abstract: Provided are a training data creation method and device, and a defect inspection method and device capable of securing the accuracy of defect inspection even though the number of samples of a defect to be used in creating training data is small.

Abstract: According to one embodiment, a learning method for causing a second statistical model to learn using a first statistical model is provided. The method includes obtaining a first learning image, cutting out each local area of the obtained first learning image, and obtaining a first prediction value output from the first statistical model by inputting each local area to the first statistical model and obtaining a second prediction value output from the second statistical model by inputting the entire area of the first learning image to the second statistical model, and causing the second statistical model to learn based on a difference between the first prediction value and the second prediction value.

Abstract: A system and method for constructing fluoroscopic-based three-dimensional volumetric data of a target area within a patient from two-dimensional fluoroscopic images acquired via a fluoroscopic imaging device.

Abstract: Systems and methods for classification and mutation prediction from histopathology images using deep learning. In particular, the examples described herein utilize lung cancer as an example, in particular adenocarcinoma (LUAD) and squamous cell carcinoma (LUSC) as two subtypes of lung cancer to identify and distinguish between.

Abstract: In various embodiments, an experiment analysis application detects executional artifacts in experiments involving microwell plates. The experiment analysis application computes one or more sets of spatial features based on one or more heat maps associated with a microwell plate. The experiment analysis application then aggregates the set(s) of spatial features to generate a feature vector. The experiment analysis application inputs the feature vector into a trained classifier. In response, the trained classifier generates a label indicating that the microwell plate is associated with a first executional artifact.

Abstract: This disclosure relates to system and method for enabling a robot to perceive and detect socially interacting groups. Various known systems have limited accuracy due to prevalent rule-driven methods. In case of few data-driven learning methods, they lack datasets with varied conditions of light, occlusion, and backgrounds. The disclosed method and system detect the formation of a social group of people, or, f-formation in real-time in a given scene. The system also detects outliers in the process, i.e., people who are visible but not part of the interacting group. This plays a key role in correct f-formation detection in a real-life crowded environment. Additionally, when a collocated robot plans to join the group it has to detect a pose for itself along with detecting the formation. Thus, the system provides the approach angle for the robot, which can help it to determine the final pose in a socially acceptable manner.

Abstract: A method of generating a combined image of a body part from a sequence of partially overlapping source images of the body part, each of the partially overlapping source images showing the body part at one of a plurality of different times, the source images being ordered in the sequence according to the different times, the method including defining a temporally coherent sequence of transformations, for registering the partially overlapping source images in the sequence with each other, registering the source images to each other using the defined temporally coherent sequence of transformations, to obtain co-registered images, and combining at least some of the co-registered images into a combined image. Related apparatus and methods are also described.

Abstract: Systems and methods for detecting an aneurysm are disclosed. The method includes forming a virtual skeleton model. The virtual skeleton model has a plurality of edges with each edge having a plurality of skeleton points. Each skeleton point is associated with a subset of the plurality of blood vessel surface points. The method includes virtually fitting elliptically shaped tubules for each edge of the virtual skeleton model and identifying a potential aneurysm based on the fitted elliptically shaped tubules.