A method of manufacturing a porous part includes controlled freeze casting of a slurry. After freezing, a solvent in the slurry is removed by sublimation and the remaining material is sintered to form the porous part. Spatial and temporal control of thermal conditions at the boundary and inside of the mold can be controlled to create parts with controlled porosity, including size, distribution, and directionality of the pores. Porous parts with near-net-shape from ceramics, metals, polymers and other materials and their combinations can be created.
B22F 1/107 - Metallic powder containing lubricating or binding agents; Metallic powder containing organic material containing organic material comprising solvents, e.g. for slip casting
B22F 3/22 - Manufacture of workpieces or articles from metallic powder characterised by the manner of compacting or sintering; Apparatus specially adapted therefor for producing castings from a slip
B28B 1/00 - Producing shaped articles from the material
B29C 39/02 - Shaping by casting, i.e. introducing the moulding material into a mould or between confining surfaces without significant moulding pressure; Apparatus therefor for making articles of definite length, i.e. discrete articles
Disclosed herein is a system and method for the automatic detection of persons engaged in the open carry of firearms at a venue. The system and method comprise strategically placed cameras at the venue which are connected to edge devices which extract frames from video generated by the cameras. The video frames are sent to a server for analysis by an AI/ML model trained to detect firearms and, specifically, to detect persons carrying firearms. If a person wielding a firearm is detected in any image, an alert is raised and local authorities are automatically contacted. The system is designed to run continuously such as to be able to quickly detect a person in a venue carrying a firearm.
Disclosed herein is a system and method for data augmentation for general object recognition which preserves the class identity of the augmented data. The system comprises an image recognition network an image generation network that take as input ground truth images and classes respectively and which generates a predicted class and an augmented image. A discriminator evaluates the predicted class and augmented image and provides feedback to the image recognition network and the image generation network.
Disclosed herein is an Overmoded Bulk Acoustic Resonator (OBAR) and a solidly-mounted OBAR (SBAR), which operate in a partially transduced 2nd overtone split between piezoelectric and electrode layers using dual all metal Bragg mirrors. The devices may be deployed in a series configuration. The devices have arbitrarily thick electrodes to minimize ohmic loss and bandwidths high enough to meet filtering requirements of 5G networks. The devices provide sharp filtering which can be performed directly at each antenna element in a form factor much smaller than the half-wavelength separation between adjacent antenna elements required when using electromagnetic resonators.
Disclosed herein is a framework for the modeling and design of materials based on differentiable programming, where the models can be trained by gradient-based optimization. Within this framework, all the components are differentiable and can be seamlessly integrated and unified with deep learning. The framework can design and optimize materials for a variety of applications.
G06F 30/27 - Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
6.
DEVICE AND METHOD FOR IMPROVED POLICY LEARNING FOR ROBOTS
A computer-implemented method of learning a policy for an agent. The method includes: receiving an initialized first neural network, in particular a Q-functionor value-function, an initialized second neural network, auxiliary parameters, and the initialized policy; repeating the following steps until a termination condition is fulfilled: sampling a plurality of pairs of states, actions, rewards and new states from a storage. Sampling actions for the current states, and actions for the new sampled states; computing features from a penultimate layer of the first neural network based on the sampled states and actions and updating the second neural network and the auxiliary parameters as well as updating parameters the first neural network using a re-weighted loss.
The present disclosure relates to structures and methods of use of textiles for protection, and more particularly. relates to single layer knitted structures that can be worn as garments to block mosquito bites. The knit structures of the present embodiments can include single layer structures having a unique geometric combination of yarn properties and textiles to block mosquito bites in a single. comfortable layer of fabric. The knit structures can include one or more controllable parameters that can be adjusted to increase and/or decrease the bite blocking ability of the knit structure.
Disclosed herein is a system and method for reducing false positives in object detection frameworks. A human form view of objects detected by the object detection framework and indicates the object is a false positive. When an indication of a false positive been received, a feature representation of displayed object stored in the gallery. During an inference or testing phase, the gallery is searched for a feature representation matching the feature representation of the detected objects, and, if a match is found, the detected object is deemed to be a false positive and is not displayed to the user.
A system includes an inertial sensing device having an inertial sensor and plurality of stress sensors configured to measure stress applied to the inertial sensing device, and at least one computing device configured to: receive sensor data from the inertial sensor and the plurality of stress sensors; and determine a drift compensation of the inertial sensor based on the sensor data. Other systems, methods, and devices are disclosed.
G01D 3/036 - Measuring arrangements with provision for the special purposes referred to in the subgroups of this group mitigating undesired influences, e.g. temperature, pressure on measuring arrangements themselves
B81B 7/02 - Microstructural systems containing distinct electrical or optical devices of particular relevance for their function, e.g. microelectro-mechanical systems (MEMS)
G01P 15/02 - Measuring acceleration; Measuring deceleration; Measuring shock, i.e. sudden change of acceleration by making use of inertia forces
G01P 21/00 - Testing or calibrating of apparatus or devices covered by the other groups of this subclass
Disclosed herein is a system and method using an equivariant neural network for predicting quantum mechanical charge density. The equivariant neural network serves as a surrogate for the density-functional theory used to calculate a selfconsistent field and predicts the central observable charge density, which, in addition to enabling force calculations, can also accelerate DFT itself and compute a full range of chemical properties.
UNIVERSITY OF PITTSBURGH - OF THE COMMONWEALTH SYSTEM OF HIGHER EDUCATION (USA)
CARNEGIE MELLON UNIVERSITY (USA)
Inventor
Ohodnicki, Paul Richard
Shen, Sheng
Abstract
A near-field probe (and associated method) compatible with near-infrared electromagnetic radiation and high temperature applications above 300°° C. (or 500° C. in some applications) includes an optical waveguide and a photonic thermal emitting structure comprising a near-field thermally emissive material coupled to or part of the optical waveguide. The photonic thermal emitting structure is structured and configured to emit near-field energy responsive to at least one environmental parameter of interest, and the near-field probe is structured and configured to enable extraction of the near-field energy to a far-field by coupling the near-field energy into one or more guided modes of the optical waveguide.
THE BOARD OF TRUSTEES OF THE UNIVERSITY OF ILLINOIS (USA)
Inventor
Bao, Zhipeng
Tokmakov, Pavel
Wang, Yuxiong
Gaidon, Adrien David
Hebert, Martial
Abstract
A method for learning a representation of a sequence of frames includes encoding, via an encoder network, the sequence of frames to obtain a set of feature maps and extracting, a motion-guided slot learning mechanism, mid-level features from the set of feature maps. The method further includes quantizing the mid-level features via a vector quantization process to obtain a set of tokens, and decoding, via a decoder network, the tokens to obtain a reconstructed sequence of frames. The method still further includes optimizing a combination of a reconstruction loss and a motion loss to train the encoder and decoder networks.
G06T 7/246 - Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
G06V 20/58 - Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
13.
SYSTEM AND METHOD FOR PROVIDING PROVABLE END-TO-END GUARANTEES ON COMMODITY HETEROGENEOUS INTERCONNECTED COMPUTING PLATFORMS
Disclosed herein is a system architecture that structures commodity heterogeneous interconnected computing platforms around universal object abstractions, which are a fundamental system abstraction and building block that provides practical and provable end-to-end guarantees of security, correctness, and timeliness for the platform.
A method for sequential point cloud forecasting is described. The method includes training a vector-quantized conditional variational autoencoder (VQ-CVAE) framework to map an output to a closest vector in a discrete latent space to obtain a future latent space. The method also includes outputting, by a trained VQ-CVAE, a categorical distribution of a probability of V vectors in a discrete latent space in response to an input previously sampled latent space and past point cloud sequences. The method further includes sampling an inferred future latent space from the categorical distribution of the probability of the V vectors in the discrete latent space. The method also includes predicting a future point cloud sequence according to the inferred future latent space and the past point cloud sequences. The method further includes denoising, by a denoising diffusion probabilistic model (DDPM), the predicted future point cloud sequences according to an added noise.
In an example, a method may include deforming a first ray associated with a dynamic object at a first time using a first neural network and a latent code to obtain a deformed ray. The method may also include obtaining a hyperspace code associated with the first ray by inputting the first ray, the first time, and the latent code into a second neural network. The method may further include sampling one or more points from the deformed ray. The method may also include combining the sampled points and the hyperspace code into a network input. The method may further include inputting the network input into a third neural network to obtain RGB values for rendering images of a three-dimensional scene representative of the dynamic object at a second time.
Computer-based systems and methods for discovering neighborhood clusters in a geographic region, where the clusters have a mix of venues and are determined based on venue check-in data. The mix of venues for the clusters may be based on the social similarity between pairs of venues; or emblematic of certain neighborhood typologies; or emblematic of temporal check-in pattern types; or combinations thereof. The neighborhood clusters that are so discovered through venue-check in data could be used for many commercial and civic purposes.
Disclosed herein is a method for optimizing electrode placement that directly exploits the thresholding phenomenon of neurons. The method employs a loss function which only becomes non-zero when the electric field is above a user-specified threshold in the cancel region, thereby allowing for fields which can have significant non-zero current in the cancel region, but still provide more focused neural activation.
Lipid-containing particles and formulations containing the lipid-containing particles are provided. Methods of delivery of therapeutic agents to the pancreas using the lipid-containing particles are provided.
A61K 31/711 - Natural deoxyribonucleic acids, i.e. containing only 2'-deoxyriboses attached to adenine, guanine, cytosine or thymine and having 3'-5' phosphodiester links
19.
METHOD FOR DETECTING AN APPLICATION PROGRESS AND HANDLING AN APPLICATION FAILURE IN A DISTRIBUTED SYSTEM
A method for detecting an application progress and handling an application failure in a distributed system. The method includes: monitoring an interaction between modules of at least one application, the at least one application being deployed across different physical nodes, the interaction being carried out by exchanging messages between the modules using a message broker, the monitoring being carried out at least partially using the message broker; detecting the application progress based on the monitoring; initiating a failure handling based on the detecting.
A method for providing a secondary backup application as a backup for a primary application, particularly for a predictive standby in distributed systems. The method includes the following steps are carried out by a predictive standby manager: receiving application-specific state data, the application-specific state data being obtained from monitoring a state of the primary application; receiving platform-specific state data, the platform-specific state data being obtained from monitoring a state of at least one platform that executes the primary application; initiating a backup process for using the secondary backup application based on the received application-specific state data and the platform-specific state data.
G06F 11/14 - Error detection or correction of the data by redundancy in operation, e.g. by using different operation sequences leading to the same result
21.
ELECTROCHEMICAL MATERIALS INCLUDING SOLID AND LIQUID PHASES
Electrochemical devices, and associated materials and methods, are generally described. In some embodiments, an electrochemical device comprises an electroactive material. The electroactive material may comprise an alloy having a solid phase and a liquid phase that co-exist with each other. As a result, such a composite electrode may have, in some cases, the mechanical softness to permit both high energy densities and an improved current density as compared to, for example, a substantially pure metal electrode.
Disclosed herein is a system and method implementing a trusted IoT security gateway architecture, based on a microhypervisor, that provides a guarantee that the correct security protections are applied to each IoT device's network traffic at all times, including when under attack. The disclosed architecture provides robust trust properties to a broad range of legacy hardware platforms utilizing existing software with a reasonable performance overhead.
A novel method for using the widely-used electroencephalography (EEG) systems to detect and localize silences in the brain is disclosed. The method detects the absence of electrophysiological signals, or neural silences, using noninvasive scalp electroencephalography (EEG) signals. This method can also be used for reduced activity localization, activity level mapping throughout the brain, as well as mapping activity levels in different frequency bands. By accounting for the contributions of different sources to the power of the recorded signals and using a hemispheric baseline approach and a convex spectral clustering framework, the method permits rapid detection and localization of regions of silence in the brain using a relatively small amount of EEG data.
A61B 5/00 - Measuring for diagnostic purposes ; Identification of persons
A61B 5/055 - Detecting, measuring or recording for diagnosis by means of electric currents or magnetic fields; Measuring using microwaves or radio waves involving electronic [EMR] or nuclear [NMR] magnetic resonance, e.g. magnetic resonance imaging
25.
System, Method, and Computer Program Product for Extracting Features From Imaging Biomarkers With Machine-Learning Models
Provided are systems, methods, and computer program products for extracting features from imaging biomarkers with machine-learning models. The method includes training a first artificial intelligence (AI) model based on first training data including images labeled with imaging biomarkers, the first AI model trained to identify a plurality of imaging biomarker features in at least one image, training a second AI model based on second training data including sets of imaging biomarker features associated with task-specific labels, the second AI model trained to identify at least one task-specific feature based at least partially on a set of imaging biomarker features, processing at least one input image with the first AI model to generate a first AI model output, and processing the first AI model output with the second AI model to generate a second AI model output.
G06V 10/44 - Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
University of Pittsburgh - Of the Commonwealth System of Higher Education (USA)
Carnegie Mellon University (USA)
Inventor
Falo, Jr., Louis D.
Erdos, Geza
Ozdoganlar, O. Burak
Abstract
A method of forming a microneedle array can include forming a sheet of material having a plurality of layers and micromilling the sheet of material to form a microneedle array. At least one of the plurality of layers can include a bioactive component, and the microneedle array can include a base portion and plurality of microneedles extending from the base portion.
Disclosed here are various techniques for improving the testing and training of datasets comprising sequences of skeletal representations performing various actions. The dataset can be denoised by applying various techniques to determine noisy frames within each sequence and eliminating the sequences from the dataset when the number of noisy frames in the sequence is too large. In addition, the dataset may be augmented by various data augmentation techniques to manipulate the skeletal representations, after denoising.
Disclosed herein is a method of reducing the complexity of a neural network using PRC-NPTN layers by applying a pruning technique to remove a subset of filters in the network based on the importance of individual filters to the accuracy of the network, which is determined by the frequency with which the response of the filter is activated.
Disclosed herein is a system and method implementing a battery avionics system for integrating battery monitoring, control, and management functions with an avionics system of an aircraft. The system uses a model implementing a battery pack digital twin, which is a continuous simulation of the operation of the battery pack within the aircraft, receives data regarding the battery pack generated by the digital twin model and provides optimized parameters to the battery avionics system. The system enables high precision, cell-level resolution control of the battery pack. The system estimates the state of charge, state of health, state of safety, and state of function of the cells and the battery pack as a whole and uses this information to manage the battery pack, given a particular flight profile of the aircraft.
B60L 58/18 - Methods or circuit arrangements for monitoring or controlling batteries or fuel cells, specially adapted for electric vehicles for monitoring or controlling batteries of two or more battery modules
A method for carrying out a decision for upgrading and/or deploying software on multiple heterogenous devices. The method includes: receiving a request to upgrade and/or deploy software on at least one of the devices; initiating a connection to the at least one of the devices; initiating a process for determining at least one capability of the at least one connected device for executing the software, the initiating being carried out via the connection, the process being initiated for being executed by the at least one connected device; receiving a result of the initiated process; and carrying out the decision for the upgrade and/or deployment of the software based on the received result.
A photonic crystal for detection of an analyte includes: a first layer including a first material with a first refractive index; a second layer over the first layer and including a second material with a second refractive index that is higher than the first refractive index; where the second layer includes a hole, the hole including: a first diameter from an outer surface of the second layer to a first hole depth; a second diameter from the first hole depth to a second hole depth; where the first diameter is larger than the second diameter; and a member of a binding pair with the analyte linked to a surface of the hole.
G01N 21/77 - Systems in which material is subjected to a chemical reaction, the progress or the result of the reaction being investigated by observing the effect on a chemical indicator
G01N 21/35 - Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light
G01N 21/3577 - Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light for analysing liquids, e.g. polluted water
Provided is a system, method, and computer program product for training a machine-learning model. The method includes labeling each object of a plurality of objects with a probabilistic confidence label including a probability classification score for each class of at least two classes, resulting in a plurality of probabilistic confidence labels associated with the plurality of objects, and training, with at least one computing device, the machine-learning model based on the plurality of objects and the plurality of probabilistic confidence labels.
Provided are systems, methods, and computer program products for segmenting an image. A method includes segmenting each image in a sequence of images including a needle into a needle and at least one needle artifact based on processing each image with a first machine-learning model trained with a plurality of hard labels for a plurality of images, resulting in a plurality of hard-labeled images, transforming each hard-labeled image of the plurality of hard-labeled images into a soft-labeled image including pixel values corresponding to an effect of the at least one needle artifact, resulting in a plurality of soft-labeled images, and segmenting at least one image of the sequence of images based on processing the at least one image with a second machine-learning model trained at least partially with the plurality of soft-labeled images.
G06V 10/762 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using clustering, e.g. of similar faces in social networks
Disclosed herein is a system providing a mixed reality combination system that pairs augmented reality technology and an inertial measurement unit sensor with 3D printed objects such that user motions tracked by the inertial measurement unit as the user interacts with the 3D printed object is reflected in a virtual environment display of dynamic 3D imagery and augmented reality imagery.
Provided are methods including the steps of receiving, with at least one computing device, an image of a portion of a subject, assigning: with the at least one computing device and based on a machine-learning model, a label to one or more pixels of the image to generate a diagnostically segmented image: and classifying, with the at least one computing device and based on a machine-learning model, the diagnostically segmented image and the one or more pixels into at least one class to generate a classified image, wherein the classified image includes a classification label indicating a clinical assessment of the portion of the subject and wherein the one or more pixels include a clinical label indicating a diagnosis of a portion of a subject contained within each pixel, based on the diagnostically segmented image having labels assigned to each pixel of the segmented image.
G06V 10/26 - Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
G06V 10/764 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
G06V 10/82 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
G06V 20/70 - Labelling scene content, e.g. deriving syntactic or semantic representations
37.
Modular DNA Nanoshells for Cell Encapsulation and Ruggedization
Ruggedized particles or live cells are provided. The particles or cells comprise a cross-linked matrix of nucleic acid and/or nucleic acid analog nanostructures forming a shell about the particle or cell. Methods of making and using the ruggedized particles or live cells are provided. The ruggedized particles or cells may be decorated with environmental sensors, for example, which are prepared from nucleic acid and/or nucleic acid analog nanostructures and may include a FRET pair.
Provided is a system, method, and computer program product for determining a needle injection site. The method includes segmenting, with at least one computing device, an image of a sequence of images into at least one object based on a machine-learning model configured to estimate its uncertainty for each segmentation, generating, with the at least one computing device, a 3D model of the at least one object, and determining, with the at least one computing device, an insertion location of the at least one object based at least partially on an output of the machine-learning model.
A hierarchical deep-learning object detection framework provides a method for identifying objects of interest in high-resolution, high pixel count images, wherein the objects of interest comprise a relatively a small pixel count when compared to the overall image. The method uses first deep-learning model to analyze the high pixel count images, in whole or as a patchwork, at a lower resolution to identify objects, and a second deep-learning model to analyze the objects at a higher resolution to classify the objects.
G06F 18/2135 - Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
G06F 18/214 - Generating training patterns; Bootstrap methods, e.g. bagging or boosting
G06F 18/241 - Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
G06F 18/2415 - Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
Provided is a system, method, and computer program product for tracking a needle. The method includes determining a visibility of the needle being inserted into a subject in an image of a sequence of images, in response to determining that the visibility satisfies a visibility threshold, detecting a location of the needle based on at least one first algorithm and a detected curvature of the needle, in response to determining that the visibility does not satisfy the visibility threshold, detecting the location of the needle being inserted based on at least one second algorithm, and tracking the location of the needle in the sequence of images based on locations detected with the at least one first algorithm and the at least one second algorithm.
University of Pittsburgh - Of the Commonwealth System of Higher Education (USA)
Carnegie Mellon University (USA)
Mississippi State University (USA)
Inventor
Madhani, Shalv
Frankowski, Brian Joseph
Federspiel, William J.
Burgreen, Gregory
Antaki, James F.
Abstract
An extracorporeal system for lung assist includes a housing having a blood flow inlet in fluid connection with a pressurizing stator compartment within the housing. A fiber bundle compartment within the housing is above and in fluid connection with the pressurizing stator compartment via a flow channel formed within the housing and extending from the pressurizing stator compartment to an inlet manifold of the fiber bundle compartment. A blood flow outlet in is fluid connection with an outlet man fold of the fiber bundle compartment. The blood flow inlet extends through the housing parallel to a plane of rotation of an impeller in the pressurizing stator compartment. The blood flow inlet turns to deliver blood into a central portion of the impeller. A has inlet is in fluid connection with the housing and in fluid connection with inlets of the plurality of hollow gas permeable fibers of a cylindrical fiber bundle positioned within the fiber bundle compartment. A gas outlet is in fluid connection with the housing and in fluid connection with outlets of the plurality of hollow gas permeable fibers.
A61M 60/113 - Extracorporeal pumps, i.e. the blood being pumped outside the patient’s body incorporated within extracorporeal blood circuits or systems in other functional devices, e.g. dialysers or heart-lung machines
A61M 60/419 - Blood pumps; Devices for mechanical circulatory actuation; Balloon pumps for circulatory assistance - Details relating to driving for non-positive displacement blood pumps the force acting on the blood contacting member being permanent magnetic, e.g. from a rotating magnetic coupling between driving and driven magnets
A61M 60/422 - Blood pumps; Devices for mechanical circulatory actuation; Balloon pumps for circulatory assistance - Details relating to driving for non-positive displacement blood pumps the force acting on the blood contacting member being electromagnetic, e.g. using canned motor pumps
Disclosed herein are training strategies for query-based object detectors, referred to herein as Query Recollection (QR). In one variation or QR, dense query recollection, every intermediate query is collected and independently forwarded to every downstream stage. In a second variation or QR, selective query recollection, intermediate queries are collected from the two nearest previous stages and forwarded to the next downstream stage. This eliminates the phenomena wherein intermediate stages of the decoder produce more accurate results than later stages of the decoder.
In an example, a method may include obtaining, from a data source, first data including multiple frames each including a human face. The method may include automatically detecting, in each of the multiple frames, one or more facial landmarks and one or more action units (AUs) associated with the human face. The method may also include automatically generating one or more semantic masks based at least on the one or more facial landmarks, the one or more semantic masks individually corresponding to the human face. The method may further include obtaining a facial hyperspace using at least the first data, the one or more AUs, and the semantic masks. The method may also include generating a synthetic image of the human face using a first frame of the multiple frames and one or more AU intensities individually associated with the one or more AUs.
A wearable device has a plurality of sensors surrounding a user's arm or wrist and provides depth information about the user's environment. Each sensor in the plurality of sensors has a field-of-view that may include the user's arm, torso, and surrounding environment. A controller receives data from the plurality of sensors and merges the data to create a composite image or depth point cloud. The device utilizes low-resolution sensors, with the composite image having a greater resolution and field-of-view than any individual sensor. The device is worn on the user's arm or wrist and can be used for static or continuous hand pose estimation, whole-arm pose estimation, and object detection, among other applications.
Provided is a method for locating an epilepsy seizure onset zone and prediction of seizure outcome including receiving interictal electroencephalographs from two or more points in a patient's cerebral cortex. The interictal electroencephalographs are used to determine directional information flow values which indicate dominant information flow from a non-seizure zone to a seizure onset zone. The directional informational flow values may be input into a classification model trained to predict whether the two or more points in the patient's cerebral cortex are a seizure onset zone and/or classify the patient's predicted post-treatment seizure outcome after epilepsy treatment based on the directional information flow values. An output from the classification model may indicate a location of seizure onset zone in the patient's cerebral cortex and/or the patient's predicted post-treatment seizure outcome after epilepsy treatment. Systems and computer program products are also provided.
A61B 5/374 - Detecting the frequency distribution of signals, e.g. detecting delta, theta, alpha, beta or gamma waves
A61B 5/00 - Measuring for diagnostic purposes ; Identification of persons
A61B 5/291 - Bioelectric electrodes therefor specially adapted for particular uses for electroencephalography [EEG]
G16H 20/40 - ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to mechanical, radiation or invasive therapies, e.g. surgery, laser therapy, dialysis or acupuncture
A bipedal walking robot uses a quasi-passive control scheme and a simplified mechanical design. The walking robot has a pair of legs connected to a body through a passive hip joint, which is offset from a center of gravity of the walking robot. A nonconcentric, curved foot is attached at to each leg by a prismatic joint. Extension and retraction of the prismatic joint initiates the walking sequence of the robot, with each foot retracted during the swing phase and extended during the stance phase. Directional changes are controlled by changing a phase offset in the actuation of each foot.
B62D 57/032 - Vehicles characterised by having other propulsion or other ground-engaging means than wheels or endless track, alone or in addition to wheels or endless track with ground-engaging propulsion means, e.g. walking members with alternately or sequentially lifted feet or skid
Disclosed herein are devices comprising stretchable 3D circuits and methods for fabricating the circuits. The fabrication process includes providing in the elastomeric polymer as a substrate and providing conductive interconnects within the substrate encased in an insulating polymer, such as polyimide, to provide a stiffness gradient between the conductive interconnects and the flexible elastomeric substrate. The circuit may be fabricated as a multilayer construction using three-dimensional pillars as vias and as external interconnects to the circuit.
Systems and methods for generating new images for training a machine-learning model are disclosed. Image data is produced regarding an image captured by an image sensor. The image data is altered such that the style of the image (e.g., color, shading, orientation, etc.) is altered. The altered image data is encoded into a first latent space. An image from a database is selected based on its similarity to the altered image and a decoding of the first latent space. Style encodings of the first latent space are extracted to classify a style of the altered image data in a second latent space. New images are then generated utilizing a reconstructor model that combines the two latent spaces. These new images can be used to train an image-recognition model.
G06V 10/764 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
G06V 10/82 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
G06V 20/58 - Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
This document describes a process of producing gel microparticles, which are consistent in size and morphology. Through the process of coacervation, large volumes of gel microparticle slurry can be produced by scaling up reactor vessel size. Particles can be repeatedly dehydrated and rehydrated in accordance to their environment, allowing for the storage of particles in a non-solvent such as ethanol. Gel slurries exhibit a Bingham plastic behavior in which the slurry behaves as a solid at shear stresses that are below a critical value. Upon reaching the critical shear stress, the slurry undergoes a rapid decrease in viscosity and behaves as a liquid. The rheological behavior of these slurries can be adjusted by changing the compaction processes such as centrifugation force to alter the yield-stress. The narrower distribution and reduced size of these particles allows for an increase in FRESH printing fidelity.
Methods for assisting one or more users in organizing content-items, such as location information (e.g., URLs) for online information resources (e.g., webpages) and clips taken from information resources, accessed via content-access software, such as one or more web browsers, using a content item-bundle primitive that allows users to create, build, manipulate, and/or populate their own content-item bundles according to their information investigation and collection desires/needs. In some embodiments, the methods include automatically bundling content items into suggested content-item bundles based on learned relationships among various content items. In some embodiments, the methods can be implemented to provide bundle-based task managers that allow users to not only organize their content items but also define tasks and/or projects rooted in the content-item-bundle primitive. Further embodiments are disclosed, as is software for executing disclosed methods.
In one aspect, a method includes providing support material within which the structure is fabricated, depositing, into the support material, structure material to form the fabricated structure, and removing the support material to release the fabricated structure from the support material. The provided support material is stationary at an applied stress level below a threshold stress level and flows at an applied stress level at or above the threshold stress level during fabrication of the structure. The provided support material is configured to mechanically support at least a portion of the structure and to prevent deformation of the structure during the fabrication of the structure. The deposited structure material is suspended in the support material at a location where the structure material is deposited. The structure material comprises a fluid that transitions to a solid or semi-solid state after deposition of the structure material.
B29C 64/112 - Processes of additive manufacturing using only liquids or viscous materials, e.g. depositing a continuous bead of viscous material using individual droplets, e.g. from jetting heads
B29C 64/118 - Processes of additive manufacturing using only liquids or viscous materials, e.g. depositing a continuous bead of viscous material using filamentary material being melted, e.g. fused deposition modelling [FDM]
B29C 64/40 - Structures for supporting 3D objects during manufacture and intended to be sacrificed after completion thereof
A system for classifying structured medical data, with each item of structured medical data, the system comprising a processing module that parses items of structured medical data to retrieve values of respective fields of the one or more items of structured medical data, the one or more retrieved values representing a set of medical attributes; a classification module that selects a classifier based at least one of the attributes in the set and applies the classifier to the set of attributes to classify one or more items of structured medical data into a particular risk profile; a user interface that renders one or more controls for input data that confirms one or more of the risk factors of the risk profile; and a transmitter to transmit to a remote medical device, an alert that specifies confirmation of the one or more of the risk factors.
G16H 50/70 - ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
A61B 5/00 - Measuring for diagnostic purposes ; Identification of persons
G06F 9/451 - Execution arrangements for user interfaces
G06F 16/901 - Indexing; Data structures therefor; Storage structures
G16H 10/20 - ICT specially adapted for the handling or processing of patient-related medical or healthcare data for electronic clinical trials or questionnaires
G16H 40/67 - ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices for remote operation
G16H 50/30 - ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for individual health risk assessment
53.
High-Density Automated Storage and Retrieval System
A high-density storage system for goods is described in which totes carrying the goods are storage in a storage structure and stored and retrieved by robotic carriers. The carriers move laterally and/or longitudinally along the exterior of the support structure and retrieve totes from the interior of the structure by manipulating rows of coupled totes. Totes at the ends of rows are quickly removed and stored in another row until the desired tote appears at the end of the row, at which point the carrier proceeds with the tote to the exit point of the storage system. Storing totes is also a quick action by pushing them into any row. As a tote is pushed into the row, it will automatically couple with a tote inside the row that it comes into contact with.
A high-density storage system for goods is described in which totes carrying the goods are stored in a storage structure and stored and retrieved via stationary or mobile conveyors running along opposite ends of each layer of the storage structure. The totes may be moved to or from the conveyors as the rows move at a constant velocity toward or away from the conveyors. Totes at the ends of rows are quickly moved and stored in another row until the desired tote appears at the end of the row, at which point the desired tote is carried to an exit point of the storage structure by one of the conveyors.
Disclosed herein is a system and method for increasing the confidence of a match between the test image and an image stored in a library database. Features are extracted from the test image and compared to features stored in the image database and, if a match is determined, one or more transformations are performed on the test image to generate pose-altered images. Features are then extracted from the pose-altered images and matched with pose-altered images in the database. The scores for the subsequent matchings can be aggregated to determine an overall probability of a match between the test image in an image in the library database.
G06V 10/74 - Image or video pattern matching; Proximity measures in feature spaces
G06T 3/60 - Rotation of a whole image or part thereof
G06T 19/20 - Editing of 3D images, e.g. changing shapes or colours, aligning objects or positioning parts
G06V 10/44 - Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
G06V 10/56 - Extraction of image or video features relating to colour
G06V 10/82 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
Disclosed herein is a system and method for generating quadrilateral bonding boxes which tightly cover the most representative faces of retail products having arbitrary poses. The quadrilateral boxes do not include unnecessary background information or miss parts of the objects, as would the axis-aligned bounding boxes produced by prior art detectors. A simple projection transformation can correct the pose of products for downstream tasks.
G06T 7/70 - Determining position or orientation of objects or cameras
G06V 10/25 - Determination of region of interest [ROI] or a volume of interest [VOI]
G06V 10/44 - Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
G06V 10/764 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
57.
METHOD FOR COMPRESSING AN AI-BASED OBJECT DETECTION MODEL FOR DEPLOYMENT ON RESOURCE-LIMITED DEVICES
Disclosed herein is a method for efficiently reducing the computational footprint of any AI-based object detection model, so as to enable its real-time deployment on computing resource-limited (i.e., low-power, embedded) devices. The disclosed method provides a step-by-step framework using an optimized combination of compression techniques to effectively compress any given AI-based object detection model.
Disclosed herein is a system and method for novel neural architecture search using a random graph network backbone to facilitate the creation of an efficient network structure. The method utilizes reinforcement learning algorithms to build a complex relationship between intra-connections (i.e., links between blocks in a random graph network) and extra-connections (i.e., links among blocks across the random graphs network) for discovering an efficient random neural architecture.
A system comprises a IoT resource and a computing device of a user. The computing device comprises a processor that executes a personal privacy app that receives data about the IoT resource and communicates a preference setting for the user with respect to the IoT device. The preference setting is based on the data received about the IoT resource.
Disclosed herein is a system and method for evolving a deep neural network model by searching for hidden sub-networks within the model. The model is evolved by adding convolutional layers to the model, then pruning the model to remove redundant filters. The model is exposed to training samples of increasing complexity each time the model is evolved, until a desired level of performance is achieved, at which time, the model is exposed to all available training data.
A process for micro-tissue encapsulation of cells includes coating a tissue scaffold stamp with an extracellular matrix compound. The process includes depositing the tissue scaffold stamp onto a thermoresponsive substrate and seeding the tissue scaffold stamp with a cell culture. A cell culture forms a cell patch that is attached to the extracellular matrix compound. A monolayer on the tissue scaffold stamp for which borders of the monolayer maintain expressions for cell-cell junctions, wherein the cell-cell junctions of the monolayer are configured to express tension forces. The process includes removing the thermoresponsive substrate. The process includes folding the micro-tissue structure by suspending the micro-tissue in the solvent. The folded micro-tissue structure is collected from the solvent and administered to an organism.
A61F 9/00 - Methods or devices for treatment of the eyes; Devices for putting in contact-lenses; Devices to correct squinting; Apparatus to guide the blind; Protective devices for the eyes, carried on the body or in the hand
Provided is a system, method, and device for automated energy remediation. The system includes at least one processor programmed or configured to: store energy usage data for a plurality of households, store environmental data associated with the plurality of households, the environmental data including outdoor temperature measurements, determine an inflection temperature for each household of the plurality of households based on a nonlinear regression model, determine a gap metric value based on a maximum median inflection temperature and a minimum inflection temperature from the plurality of households, form a plurality of groups based on the plurality of households and household data associated with each household, each group including a subset of households of the plurality of households, determine at least one group of the plurality of groups, and automatically initiate at least one energy protocol for households in the at least one group.
H02J 13/00 - Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network
H02J 3/00 - Circuit arrangements for ac mains or ac distribution networks
63.
SYSTEM AND METHOD FOR DOMAIN-AGNOSTIC BIAS REDUCTION WITH SELECTED SAMPLING FOR FEW-SHOT LEARNING
Disclosed herein is a methodology for refining novel-class features in a few-shot learning scenario by fine-tuning the feature extractor by reducing both class-agnostic biases and class-specific biases. A distribution calibration module is used to reduce the class-agnostic bias by normalizing the overall feature distribution for novel classes and further reshaping the feature manifold for fast adaptation during fine-tuning. Selected sampling is used to reduce class-specific bias by augmenting more data for better estimation.
G06V 10/77 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
G06V 10/764 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
G06V 10/771 - Feature selection, e.g. selecting representative features from a multi-dimensional feature space
64.
System and Method for Tracking an Object Based on Skin Images
University of Pittsburgh - Of the Commonwealth System of Higher Education (USA)
Inventor
Galeotti, John Michael
Stetten, George Dewitt
Huang, Chun-Yin
Abstract
Provided is a system, method, and computer program product for tracking an object based on skin images. A method includes capturing, with at least one computing device, a sequence of images with a stationary or movable camera unit arranged in a room, the sequence of images including the subject and an object moving relative to the subject, and determining, with at least one computing device, the pose of the object with respect to the subject in at least one image of the sequence of images based on computing or using a prior surface model of the subject, a surface model of the object, and an optical model of the stationary or movable camera unit.
Disclosed herein is a system and method for selecting a battery for a particular application, for example, batteries used in portable electronics, electric vehicles, satellites, etc. The method uses an end-to-end differentiable modeling approach that allows the selection of batteries directly from the parameters of the battery and a specification of the particular application for which the batteries are being selected.
G01R 31/36 - Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
G01R 31/367 - Software therefor, e.g. for battery testing using modelling or look-up tables
Disclosed herein is a method of soft anchor-point detection (SAPD), which implements a concise, single-stage anchor-point detector with both faster speed and higher accuracy. Also disclosed is a novel training strategy with two softened optimization techniques: soft-weighted anchor points and soft-selected pyramid levels.
G06V 10/25 - Determination of region of interest [ROI] or a volume of interest [VOI]
G06V 10/44 - Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
G06V 10/74 - Image or video pattern matching; Proximity measures in feature spaces
G06V 10/764 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
67.
SYSTEM AND METHOD FOR SCENE RECTIFICATION VIA HOMOGRAPHY ESTIMATION
Disclosed herein is a system and method for performing pose-correction on images containing objects within a scene, or the entire scene, to compensate for off-centered camera views. The system and method generates a more frontal view of the object or scene by applying planar homography by identifying corner endpoints of the object or the scene and repositioning the corner endpoints to provide a more frontal view. The pose-corrected scene may then be input to an object detector to determine a location of a bounding box of an object-of-interest which would be more accurate than a bounding box from the original off-centered image.
G06V 10/24 - Aligning, centring, orientation detection or correction of the image
G06V 10/25 - Determination of region of interest [ROI] or a volume of interest [VOI]
G06V 10/44 - Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
G06V 10/764 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
Disclosed herein is a system and method for pooling local features for fine-grained image classification. The deep features learned by the deep network are augmented with low level local landmark features by learning a pooling strategy that pools landmark features from earlier layers of the deep network. These low level landmark features are combined with the deep features and sent to the classifier.
G06V 10/80 - Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level
G06V 10/82 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
G06V 10/77 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
G06V 10/46 - Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
69.
SYSTEM AND METHOD FOR PHOTOREALISTIC IMAGE SYNTHESIS USING UNSUPERVISED SEMANTIC FEATURE DISENTANGLEMENT
Disclosed herein is a method to disentangle linear-encoded facial semantics from facial images without external supervision. The method uses linear regression and sparse representation learning concepts to make the disentangled latent representations easily interpreted and manipulated. Generated facial images are decomposed into multiple semantic features and latent representations are extracted to capture interpretable facial semantics. The semantic features may be manipulated to synthesize photorealistic facial images by sampling along vectors representing the semantic features, thereby changing the associate semantics.
G06V 10/44 - Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
G06V 40/16 - Human faces, e.g. facial parts, sketches or expressions
A MEMS/NEMS actuator based on a phase change material is described in which the volumetric change observed when the phase change material changes from a crystalline phase to an amorphous phase is used to effectuate motion in the device. The phase change material may be changed from crystalline phase to amorphous phase by heating with a heater or by passing current directly through the phase change material, and thereafter quenched quickly by dissipating heat into a substrate. The phase change material may be changed from the amorphous phase to a crystalline phase by heating at a lower temperature. An application of the actuator is described to fabricate a phase change nano relay in which the volumetric expansion of the actuator is used to push a contact across an airgap to bring it into contact with a source/drain.
Provided is a method for classification of diseases including receiving image data associated with an image at a first resolution. The image may be processed, for example by removing a background from the image, deconstructing the image into separate layers, and segmenting the image to define a plurality of single-cell images. A single-cell image may be processed, for example, by applying a filter to the single-cell image to decrease a resolution of the single-cell image as compared to the first resolution, to a second resolution. A label may be assigned to the single-cell image. A machine learning model is trained to predict a classification of the single-cell image based on inputting a plurality of single-cell images into the model. The trained machine learning model may be used to predict the outcome of a treatment. Systems and computer program products are also provided.
G06V 10/82 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
G06T 3/40 - Scaling of a whole image or part thereof
G16H 20/40 - ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to mechanical, radiation or invasive therapies, e.g. surgery, laser therapy, dialysis or acupuncture
72.
SYSTEM AND METHOD FOR DOMAIN ADAPTIVE OBJECT DETECTION VIA GRADIENT DETACH BASED STACKED COMPLEMENTARY LOSSES
Disclosed herein an effective detach strategy which suppresses the flow of gradients from context sub-networks through the detection backbone path to obtain a more discriminative context by forcing the representation of context sub-network to be dissimilar from the detection network. A sub-network is defined to generate the context information from early layers of the detection backbone. Because instance and context focus on perceptually different parts of an image, the representations from either of them should also be discrepant. In addition, a stacked complementary loss is generated to and backpropagated to the detection network.
G06V 10/82 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
G06V 10/77 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
73.
SYSTEM AND METHOD FOR UNSUPERVISED OBJECT DEFORMATION USING FEATURE MAP-LEVEL DATA AUGMENTATION
Disclosed herein is a methodology implementing feature map-level data augmentation in a feature map. Two or more units in the feature map are selected and the values of locations in the two or more units are swapped among the two or more units. Value perturbations applied around local units in the feature map implicitly lead to an unused data augmentation at the image level.
Disclosed herein is a system and method implementing an automated, generalizable model for tracking cortical spreading depressions using EEG. The model comprises convolutional neural networks and graph neural networks to leverage both the spatial and the temporal properties of CSDs in the detection. The trained model is generalizable to different head models such that it can be applied to new patients without re-training. Further, the model is scalable to different densities of EEG electrodes, even when trained on a specific electride density.
University of Pittsburgh - Of the Commonwealth System of Higher Education (USA)
Inventor
Li, Lu
Schwerin, Michael B.
Choset, Howie
Cook, Keith E.
Rose, Jason
Abstract
Provided is a system for operating a ventilator. The system includes a motorized proportional valve actuator including a stepper motor and an actuator. The actuator is connected to the stepper motor and configured to output pressurized air by controlling a pressure on a valve diaphragm. A conduit provides for fluid communication of the pressurized air to a breathing apparatus. A sensor arrangement is in fluid communication with the conduit between the at least one motorized proportional valve actuator and the breathing apparatus. The sensor arrangement includes: (i) an intake manifold configured to output a restricted flow of air from the pressurized air transported in the conduit, and (ii) a sensor device in fluid communication with an outlet of the intake manifold, the sensor device configured to measure an air pressure of the conduit based on the restricted flow of air.
Disclosed herein is an improved method for identifying images containing objects-of-interest from a large set of images. The method comprises mixing two or more of the images to create a grouped image and exposing the grouped image to an object detector trained on grouped images to make an initial determination that the grouped image was formed from at least one image containing an object-of-interest. The images which formed the grouped image are then exposed to regular object detectors to determine a classification of the object-of-interest.
Disclosed herein is an improved few-shot detector which utilizes a dynamic semantic network which takes as input a language feature and generates trainable parameters for a visual network. The visual network takes a visual feature as input and generates a classification and localization of an object.
G06F 18/2136 - Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on sparsity criteria, e.g. with an overcomplete basis
G06N 3/04 - Architecture, e.g. interconnection topology
Disclosed herein are designs for two baselines to detect products in a retail setting. A novel detector, referred to herein as RetailDet, detects quadrilateral products. To match products using visual texts on 2D space, text features are encoded with spatial positional encoding and the Hungarian Algorithm that calculates optimal assignment plans between varying text sequences is used.
G06V 10/774 - Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
G06V 10/77 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
G06V 10/82 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
G06V 10/766 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using regression, e.g. by projecting features on hyperplanes
G06V 10/764 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
A computer-implemented system and method relate to test-time adaptation of a machine learning system from a source domain to a target domain. Sensor data is obtained from a target domain. The machine learning system generates prediction data based on the sensor data. Pseudo-reference data is generated based on a gradient of a predetermined function evaluated with the prediction data. Loss data is generated based on the pseudo-reference data and the prediction data. One or more parameters of the machine learning system is updated based on the loss data. The machine learning system is configured to perform a task in the target domain after the one or more parameters has been updated.
A method includes receiving a trajectory dataset including a plurality of geospatial points forming a point cloud and acquired along a trajectory wherein for each of the plurality of geospatial points there is a defined an x-coordinate, a y-coordinate and a z-coordinate and at least one mapping device orientation attribute, segmenting the trajectory dataset into a plurality of segments, determining at least one relative constraint for each of the plurality of segments and utilizing, for each of the plurality of segments, at least one of the determined relative constraints to determine a relative position of at least two of the plurality of segments.
The Trustees of the University of Pennsylvania (USA)
Carnegie Mellon University (USA)
Inventor
Hsu, David Hwei-Yu
Tambe, Prasanna
Lee, Dokyun
Abstract
Methods, systems, and computer readable media for using machine learning models to determine predicted values of patent documents. In some examples, a method includes training, by at least one processor, a machine learning model to predict patent value based on unstructured text from training patents and, for each training patent, a measure of patent value. The method includes supplying, by the at least one processor, unstructured text from a patent document to the machine learning model. The method includes outputting, by the at least one processor, a predicted measure of value of the patent document.
G06N 3/0442 - Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
82.
METHOD AND SYSTEM OF CROWN BASED FOR ADVERSARIAL ATTACKS
A method of identifying an attack comprising receiving an input of one or more images, wherein the one or more images includes a patch size and size, divide the image into a first sub-image and a second sub-image, classify the first sub-image and the second sub-image, wherein classifying is accomplished via introducing a variable in a pixel location associated with the first and second sub-image, and in response to classifying the first and second sub-image and identifying an adversarial patch, output a notification indicating that the input is not certified.
G06F 21/64 - Protecting data integrity, e.g. using checksums, certificates or signatures
G06V 10/764 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
G06V 10/44 - Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
83.
LIQUID METAL CIRCUITS AND METHODS OF MAKING THE SAME
Manufacturing technology to fabricate liquid metal-based soft and flexible electronics (sensors, antennas, etc.) in a high-throughput fashion, with fabrication rates that may approach that of the traditional integrated circuit components and circuits, are described. The technique allows creation of liquid-metal-only circuits, as well as seamless integration of solid IC chips into the circuits, in which liquid metal traces are used as flexible interconnects and/or as other circuit elements. The process may be applied at the wafer scale and may be integrated into the traditional microelectronics fabrication processes. Many sensors, antennas, and other circuit elements may be directly created using liquid metal, and when combined with the IC chips, a broad range of electronic functionality may be provided in a flexible, soft circuit that can be conformable, wearable.
H01Q 1/36 - Structural form of radiating elements, e.g. cone, spiral, umbrella
H05K 3/38 - Improvement of the adhesion between the insulating substrate and the metal
H05K 3/12 - Apparatus or processes for manufacturing printed circuits in which conductive material is applied to the insulating support in such a manner as to form the desired conductive pattern using printing techniques to apply the conductive material
H01L 21/48 - Manufacture or treatment of parts, e.g. containers, prior to assembly of the devices, using processes not provided for in a single one of the groups
84.
System and Method for Domain Generalization Across Variations in Medical Images
Provided is a method of training a machine-learning-based artificial intelligence (AI) model to handle diverse types of motions occurring during image acquisition, including capturing image data including motion between an imaging device and tissue, modifying the captured image data, resulting in modified image data, by at least one of: altering an amount of time between any two frames; removing a subsequence of frames from the captured image data; and adding a subsequence of one or more new frames to the captured image data, and training a machine-learning-based AI model based on the modified image data. Other systems and methods are also described.
Disclosed herein is a design for a 1-bit CNN that closes the performance gap between binary neural networks and real-valued networks on challenging large-scale datasets. The design starts with a high-performance baseline network. Blocks with identity shortcuts which bypass 1-bit generic convolutions are adopted to replace the convolutions in the baseline network. Reshaping and shifting of activation functions is introduced. Finally, a distributional loss to further is adopted enforce the binary network to learn similar output distributions as those of a real-valued network.
A method to form a coated cathode material may generally include forming, via chemical vapor deposition, an interfacial layer coating on an exterior surface of a cathode active material, wherein the interfacial layer comprises an organic polymer; and wherein the interfacial layer is substantially uniform on and conformal to the exterior surface of the cathode active material. The polymer may include poly(3,4-ethylenedioxythiophene) (PEDOT). Methods of making and using the same are also described.
H01M 4/525 - Selection of substances as active materials, active masses, active liquids of inorganic oxides or hydroxides of nickel, cobalt or iron of mixed oxides or hydroxides containing iron, cobalt or nickel for inserting or intercalating light metals, e.g. LiNiO2, LiCoO2 or LiCoOxFy
H01M 10/0525 - Rocking-chair batteries, i.e. batteries with lithium insertion or intercalation in both electrodes; Lithium-ion batteries
H01M 4/505 - Selection of substances as active materials, active masses, active liquids of inorganic oxides or hydroxides of manganese of mixed oxides or hydroxides containing manganese for inserting or intercalating light metals, e.g. LiMn2O4 or LiMn2OxFy
C23C 16/44 - Chemical coating by decomposition of gaseous compounds, without leaving reaction products of surface material in the coating, i.e. chemical vapour deposition (CVD) processes characterised by the method of coating
C23C 16/00 - Chemical coating by decomposition of gaseous compounds, without leaving reaction products of surface material in the coating, i.e. chemical vapour deposition (CVD) processes
H01M 4/36 - Selection of substances as active materials, active masses, active liquids
H01M 4/62 - Selection of inactive substances as ingredients for active masses, e.g. binders, fillers
H01M 4/1391 - Processes of manufacture of electrodes based on mixed oxides or hydroxides, or on mixtures of oxides or hydroxides, e.g. LiCoOx
87.
System and method for pose tolerant feature extraction using generated pose-altered images
Disclosed herein is a system and method for augmenting data by generating a plurality of pose-altered images of an item from one or more 2D images of the item and using the augmented data to train a train a feature extractor. In other aspects of the invention, the trained feature extractor is used to enroll features extracted from images of new products in a library database of known products or to identify images of unknown products by matching features of an image of the unknown product with features stored in the library database.
G06V 10/56 - Extraction of image or video features relating to colour
G06V 10/74 - Image or video pattern matching; Proximity measures in feature spaces
G06V 10/44 - Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
88.
USER-SPACE EMULATION FRAMEWORK FOR HETEROGENEOUS SOC DESIGN
Arizona Board of Regents on Behalf of Arizona State University (USA)
Arizona Board of Regents on Behalf of the University of Arizona (USA)
Board of Regents, The University of Texas System (USA)
Carnegie Mellon University (USA)
Inventor
Ogras, Umit
Marculescu, Radu
Akoglu, Ali
Chakrabarti, Chaitali
Bliss, Daniel
Arda, Samet Egemen
Sartor, Anderson
Kumbhare, Nirmal
Krishnakumar, Anish
Mack, Joshua
Goksoy, Ahmet
Mandal, Sumit
Abstract
A user-space emulation framework for heterogeneous system-on-chip (SoC) design is provided. Embodiments described herein propose a portable, Linux-based emulation framework to provide an ecosystem for hardware-software co-design of heterogenous SoCs (e.g., domain-specific SoCs (DSSoCs)) and enable their rapid evaluation during the pre-silicon design phase. This framework holistically targets three key challenges of heterogeneous SoC design: accelerator integration, resource management, and application development. These challenges are addressed via a flexible and lightweight user-space runtime environment that enables easy integration of new accelerators, scheduling heuristics, and user applications, and the utility of each is illustrated through various case studies. A prototype compilation toolchain is introduced that enables automatic mapping of unlabeled C code to heterogeneous SoC platforms. Taken together, this environment offers a unique ecosystem to rapidly perform functional verification and obtain performance and utilization estimates that help accelerate convergence towards a final heterogeneous SoC design.
A therapeutic delivery system uses an engineered extracellular vesicle-albumin hybrid carrier for curcumin, which is embedded in dissolvable microneedle arrays. The co-encapsulation of curcumin with albumin in extracellular vesicles extends curcumin's stability. The incorporation of therapeutic loaded carrier into microneedle arrays does not alter its cell uptake properties or bioactivity. Moreover, the bioactivity of therapeutic loaded carrier can be preserved for at least one year when encapsulated in microneedle arrays and stored under room temperature storage conditions. The microneedle arrays of the delivery system are fabricated using molding and casting processes. The extracellular vesicle carrier can be loaded using sonication.
A method for a passive single-viewpoint 3D imaging system comprises capturing an image from a camera having one or more phase masks. The method further includes using a reconstruction algorithm, for estimation of a 3D or depth image.
H04N 13/229 - Image signal generators using stereoscopic image cameras using a single 2D image sensor using lenticular lenses, e.g. arrangements of cylindrical lenses
91.
PERFORMANCE OF NEURAL NETWORKS UNDER DISTRIBUTION SHIFT
Methods and systems of estimating an accuracy of a neural network on out-of-distribution data. In-distribution accuracies of a plurality of machine learning models trained with in-distribution data are determined. The plurality of machine learning models includes a first model, and a remainder of models. In-distribution agreement is determined between (i) an output of the first machine learning model executed with an in-distribution dataset and (ii) outputs of a remainder of the plurality of machine learning models executed with the in-distribution dataset. The machine learning models are also executed with an unlabeled out-of-distribution dataset, and an out-of-distribution agreement is determined. The in-distribution agreement is compared with the out-of-distribution agreement. Based on a result of the comparison being within a threshold, an accuracy of the first machine learning model on the unlabeled out-of-distribution dataset is estimated based on (i) the in-distribution accuracies, (ii) the in-distribution agreement, and (iii) the out-of-distribution agreement.
Arizona Board of Regents on Behalf of Arizona State University (USA)
Arizona Board of Regents on Behalf of the University of Arizona (USA)
Board of Regents, The University of Texas System (USA)
Carnegie Mellon University (USA)
Inventor
Ogras, Umit
Marculescu, Radu
Akoglu, Ali
Chakrabarti, Chaitali
Bliss, Daniel
Arda, Samet Egemen
Sartor, Anderson
Kumbhare, Nirmal
Krishnakumar, Anish
Mack, Joshua
Goksoy, Ahmet
Mandal, Sumit
Abstract
Runtime task scheduling using imitation learning (IL) for heterogenous many-core systems is provided. Domain-specific systems-on-chip (DSSoCs) are recognized as a key approach to narrow down the performance and energy-efficiency gap between custom hardware accelerators and programmable processors. Reaching the full potential of these architectures depends critically on optimally scheduling the applications to available resources at runtime. Existing optimization-based techniques cannot achieve this objective at runtime due to the combinatorial nature of the task scheduling problem. In an exemplary aspect described herein, scheduling is posed as a classification problem, and embodiments propose a hierarchical IL-based scheduler that learns from an Oracle to maximize the performance of multiple domain-specific applications. Extensive evaluations show that the proposed IL-based scheduler approximates an offline Oracle policy with more than 99% accuracy for performance- and energy-based optimization objectives. Furthermore, it achieves almost identical performance to the Oracle with a low runtime overhead and high adaptivity.
G06F 9/48 - Program initiating; Program switching, e.g. by interrupt
G06F 15/80 - Architectures of general purpose stored program computers comprising an array of processing units with common control, e.g. single instruction multiple data processors
Provided is a sensing device including an elastomer, a magnetic device positioned within the elastomer and associated with a magnetic field, and a magnetometer configured to sense a change in the magnetic field of the magnetic device. A method and computer program product are also provided.
G01B 7/24 - Measuring arrangements characterised by the use of electric or magnetic techniques for measuring the deformation in a solid, e.g. by resistance strain gauge using change in magnetic properties
A device with image acquisition includes: a first phase mask disposed at a front end of a display layer and configured to modulate external light; the display layer comprising pixel areas between hole areas through which the modulated light that has passed through the first phase mask passes; a second phase mask disposed at a rear end of the display layer and configured to modulate the modulated light that has passed through the first phase mask; an image sensor disposed at a rear end of the second phase mask and configured to generate a raw image by sensing the modulated light that has passed through the second phase mask; and a processor configured to perform image processing on the raw image, based on blur information corresponding to the raw image.
Arizona Board of Regents on Behalf of Arizona State University (USA)
Arizona Board of Regents on Behalf of the University of Arizona (USA)
Board of Regents, The University of Texas System (USA)
Carnegie Mellon University (USA)
Inventor
Ogras, Umit
Marculescu, Radu
Akoglu, Ali
Chakrabarti, Chaitali
Bliss, Daniel
Arda, Samet Egemen
Sartor, Anderson
Kumbhare, Nirmal
Krishnakumar, Anish
Mack, Joshua
Goksoy, Ahmet
Mandal, Sumit
Abstract
Hierarchical and lightweight imitation learning (IL) for power management of embedded systems-on-chip (SoCs), also referred to herein as HiLITE, is provided. Modern SoCs use dynamic power management (DPM) techniques to improve energy efficiency. However, existing techniques are unable to efficiently adapt the mntime decisions considering multiple objectives (e.g., energy and real-time requirements) simultaneously on heterogeneous platforms. To address this need, embodiments described herein propose HiLITE, a hierarchical IL framework that maximizes energy efficiency while satisfying soft real-time constraints on embedded SoCs. This approach first trains DPM policies using IL; then, it applies a regression policy at runtime to minimize deadline misses. HiLITE improves the energy-delay product by 40% on average, and reduces deadline misses by up to 76%, compared to state-of-the-art approaches. In addition, the trained policies not only achieve high accuracy, but also have negligible prediction time overhead and small memory footprint.
Disclosed herein is a method for training a network to detect and classify abnormal pathologies in images if cells. Specifically, the network uses a deep framework optimally trained to detect and classify abnormal cervical cells in pap smear images.
UNIVERSITY OF PITTSBURGH - OF THE COMMONWEALTH SYSTEM OF HIGHER EDUCATION (USA)
Inventor
Yoshida, Masahiro
Bernstein, C. Douglas
Dur, Onur
Pekkan, Kerem
Abstract
Artificial heart valve structures and methods of their fabrication are disclosed. The heart valve structures may be fabricated from a biocompatible polymer and include one or more heart valve leaflet structures incorporated within a conduit. The valve structures may incorporate one or more conduit sinuses, as well as a gap between the lower margin of the valve leaflets and the interior of the conduit. In addition, the valve structures may include one or more valve sinuses created in a space between the valve leaflets and the conduit inner surface. Computational fluid dynamics and mechanical modeling may be used to design the valve leaflets with optimal characteristics. A heart valve structure may also incorporate a biodegradable component to which cells may adhere. The incorporated cells may arise from patient cells migrating to the biodegradable component, or the component may be pre-seeded with cells prior to implantation in a patient.
A61L 27/16 - Macromolecular materials obtained by reactions only involving carbon-to-carbon unsaturated bonds
A61L 33/00 - Antithrombogenic treatment of surgical articles, e.g. sutures, catheters, prostheses, or of articles for the manipulation or conditioning of blood; Materials for such treatment
A system and method utilize capacitance sensor data to identify cell events with single-cell resolution. The method identifies patterns in the sensor data related to events such as mitosis, migration-in to the sensor field, and migration-out. The system may include a processor co-located with the sensor to perform the pattern recognition. Further, microfluidic channels can be provided to direct cells to the sensors.
Disclosed herein is a method providing a flexible way to transfer knowledge from base to novel classes in a few shot learning scenario. The invention introduces a partial transfer paradigm for the few-shot classification task in which a model is first trained on the base classes. Then, instead of transferring the learned representation by freezing the whole backbone network, an efficient evolutionary search method is used to automatically determine which layer or layers need to be frozen and which will be fine-tuned on the support set of the novel class.
An additive manufacturing method, an additive manufacturing system (1200), a support material for additive manufacturing, an assembly of the support material and a structure material, and a product thereof are provided. The method comprises depositing, by a nozzle (1210a), a structure material into a support material based on a computer model of an object, thereby forming a portion of the object. Image data of at least the portion of the object can be obtained in-process by a detector (1240). The image data is compared to the computer model. Based on the comparison, the method can comprise modifying the computer model, modifying a print parameter, modifying machine path instructions for an additive manufacturing machine that comprises the nozzle, aborting the additive formation, indicating a discrepancy, indicating validation of the shape, or a combination thereof. The depositing of the structure material is repeated by the nozzle (1210a) as necessary to additively form the object.
patents
