Abstract: Mid-infrared spectroscopic analysis of tissue specimens can include training a learning model and using the trained learning model to classify tissue into at least two (or more than two) categories such as a tumor category, a non-tumor category, and a histology category. Resulting richer diagnostic indication can be provided and can optionally be mapped to a schema for RGB or other display. Efficient and accurate classification can employ an efficient linear SVM model representation. The model representation can include a ? spectrum including the fuller wavelength set, a central tendency indicator (?) of the training set, a spread indicator (?) of the training set, and a scaling factor. Various types of illuminators and response detectors can include one or more Quantum Cascade Lasers, broadband light source configured to permit spectral separation, a thermal or other imaging array, among others.

Abstract: Systems and methods for digital media-based user authentication are disclosed herein. In some aspects, the system may receive an authentication request. The system may transmit a user verification request based on the authentication request. The system may receive media recordings in response to the user verification request. The system may detect one or more features in the media recordings. The system may generate a plurality of risk metrics for the user based on generating risk metrics for the one or more features. The system may determine an authentication probability based on the plurality of risk metrics.

Abstract: Embodiments of the invention provide a camera array imaging architecture that computes depth maps for objects within a scene captured by the cameras, and use a near-field sub-array of cameras to compute depth to near-field objects and a far-field sub-array of cameras to compute depth to far-field objects. In particular, a baseline distance between cameras in the near-field subarray is less than a baseline distance between cameras in the far-field sub-array in order to increase the accuracy of the depth map. Some embodiments provide an illumination near-IR light source for use in computing depth maps.

Abstract: A method of decoding an encoded video bitstream using at least one processor includes obtaining a video coding layer (VCL) network abstraction layer (NAL) unit; determining whether the VCL NAL unit is a first VCL NAL unit of a picture unit (PU) containing the VCL NAL unit; based on determining that the VCL NAL unit is the first VCL NAL unit of the PU, determining whether the VCL NAL unit is a first VCL NAL unit of an access unit (AU) containing the PU; and based on determining that the VCL NAL unit is the first VCL NAL unit of the AU, decoding the AU based on the VCL NAL unit.

Abstract: A vehicle comprises a Wi-Fi network including at least one Wi-Fi antenna module including a Wi-Fi antenna in communication with a Wi-Fi Integrated Circuit (IC) and a Bluetooth Low Energy (BLE) network including at least one BLE antenna module including a BLE antenna and a BLE IC, the BLE IC configured to detect in-band noise, characterize the in-band noise, and transmit the characterization of the in-band noise to the Wi-Fi IC.

Abstract: A pump system for selectively extracting nonaqueous phase liquids (NAPLs) from groundwater includes a fuel pump with current monitoring and a control assembly. The system utilizes a conductive probe within a probe body to detect a fluid interface, distinguishing NAPLs from water by measuring conductivity. The control assembly monitors the pump's current draw and adjusts operation to avoid water intake, thereby protecting the pump from damage. The probe body features an insulated shell to ensure accurate readings, and the control assembly uses real-time data to activate or deactivate the pump based on fluid type. The system is further configurable to calibrate for various contaminant conditions, optimizing efficient NAPL extraction from contaminated environments.

Abstract: Accurate prompting improves the operation, efficiency, and output of computer models, such as language models. The disclosed systems and methods improve interaction with computer models by facilitating the generation of accurate prompts for targeted interactions with computer models. For example, disclosed systems can be configured to store prompts and responses generated with computer models in session memories. The system can use information in session memories to generate combined prompts when receiving prompts that refer to previous prompts or answers. The system improves prompt accuracy by generating combined prompts—formed by combining the context from the stored information (e.g., in context sub-prompts) and instructions in the new prompt (e.g., in instructions sub-prompts). The system can generate responses based on the combined prompts allowing the computer models to leverage the context in previous interactions, without burdensome or complicated prompts, for more meaningful or accurate responses.

Abstract: A fuel cell system, e.g., of a motor vehicle, includes a fuel cell stack having fuel cells and non-repeating hardware components, the latter including wet and dry end units. The cells are positioned between the end units. The dry end unit includes an end plate and a compression plate assembly that uniformly compresses the cells against the end plate, and a terminal plate surrounded by a seal plate. An insulator frame disposed adjacent to the compression plate assembly is engaged with the insulator frame via a frictional interface provided by o-ring compression seals having a respective post connected to the terminal plate and the seal plate, and a pocket connected to the insulator frame. The post and pocket of at least one of the compression seals together define a fluid passage. One or more shim plates may be disposed between the end plate and insulator frame.

Abstract: An exhaust liquid treatment assembly for a fuel cell system (FCS). The FCS includes a fuel cell stack and an FCS exhaust pipe fluidly connected to the fuel cell stack and configured to expel an exhaust stream from the FCS. The exhaust liquid treatment assembly includes a liquid tank having a liquid inlet in fluid communication with the FCS exhaust pipe and a liquid outlet. A liquid treatment filter separates the liquid inlet from the liquid outlet and includes a pH controlling material configured to mix with a liquid passing through the liquid treatment filter. A liquid level sensor configured to determine a level of liquid in the liquid tank. A controller is in communication with the liquid level sensor and configured to regulate a level of the liquid within the liquid tank by selectively opening and closing an outlet valve in fluid communication with the liquid outlet.

Abstract: A system and method for controlling current distribution to electrical brushes used in a dynamoelectric machine is disclosed. The approach includes using an arrangement of resistor devices electrically connected to the electrical brushes to control a distribution of a flow of the electrical current between an external source and the electrical brushes. The arrangement of the resistor devices is configured to control an amount of the flow of the electrical current that is distributed to individual brushes and/or one or more groups of brushes while the dynamoelectric machine is operating.

Abstract: Optimization of multimode messaging via a private network having a hybrid platform. The optimization may include identifying performance priority parameters included within a plurality of data sets, fragmenting the data sets into one or more subsets, generating reprioritize priority for the subsets, assessing transmission performance for the plurality of radio access, and assigning each of the subsets for transmission from the hybrid platform via one of the radio access points based on the transmission performance and the reprioritized priority.

Abstract: Prostate treatment using fluid stream to resect prostate tissue, thereby relieving symptoms of conditions such as BPH, prostatitis, and prostatic carcinoma. A device having a fluid delivery element is positioned within a lumen of the urethra within the prostate. A fluid stream is directed outwardly from the fluid delivery element toward a wall of the urethral lumen. The fluid delivery element is moved to scan the fluid stream over the wall to remove a volume of tissue surrounding the lumen. The fluid may be combined with therapeutically active substances or with substances that increase resection efficiency. Fluid force may be adjusted to provide selective tissue resection such that soft tissue is removed while harder tissue is left undamaged. In order to gain a working space within the urethra, another fluid may be introduced to insufflate the urethra in the region of treatment.

Abstract: A frame assembly configured for assembly for holding taco shells includes a first frame member and a second frame member. The first frame member has a first arm, a second arm, a third arm and a fourth arm, the first frame member defining a first taco receiving channel between the first and second arms; a second taco receiving channel between the second and third arms; and a third taco receiving channel between the third and fourth arms. The second frame member has; a fifth taco receiving channel between the second and third arms; and a sixth taco receiving channel between the third and fourth arms. The first and second frame members are movable into a configuration such that the first and third taco receiving channels are aligned for receipt of a first taco and the fourth and sixth taco receiving channels are aligned for receipt of a second taco.

Abstract: Methods of manufacturing a radiopaque localization array are provided. A radiopaque layer and a backing layer may be provided. The radiopaque layer and the backing layer may be combined with an adhesive layer between the radiopaque layer and the backing layer. The radiopaque localization array may be cut in the radiopaque layer by a cutter.

Abstract: A jewelry system comprising a necklace and a clasp. The clasp includes a housing having a bottom including a hole disposed in a side of the bottom and a centered anchor post comprising a groove disposed perpendicular to the hole. The clasp also includes an actuator comprising an actuator housing having a housing anchor, an opening, and a connector interface, and a spring comprising a first spring anchor and a second spring anchor, wherein the spring is received in the actuator housing by coupling the second spring anchor with the housing anchor. The actuator housing is received in the bottom such that the anchor post is inserted into the opening and the groove receives the first spring anchor of the spring. The clasp includes a cord having a first end coupled with the connector interface and a second end to be pulled away from the clasp.

Abstract: A distributed computing system may include a code repository server configured to store code, a code deployment server configured to receive a deployment of the code, and a development and operations (DevOps) server configured to construct a pipeline between the code repository server and the code deployment server. The DevOps server may be configured to execute a source code management program to receive a request for information regarding a target resource that has been deployed using the pipeline, generate a permissions model for the target resource, the permissions model including one or more permissions, each permission authorizing a managed identity to execute an action related to the target resource, determine a permissions usage history of the permissions contained in the permissions model, filter the permissions model based on the permissions usage history, and generate a list of events determined to have occurred as the filtered permissions model.

Abstract: In multi-threaded or multi-processor computing systems, a deadlock may occur when two or more processes or threads are unable to proceed because they are each waiting for a resource that the other holds. As a result, progress is halted because conflicting entities are stuck in a circular dependency, and none can release the resources they hold to let the others continue. Systems and methods are provided wherein a resource reservation is carried out in two steps. The first step causes query nodes to add an identifier to a queue and, upon a request and the identifier being in a first position, a non-sharable resource is reserved. As a result, non-sharable resources are reserved in order and when needed, thereby preventing deadlocks.

Abstract: A method of generating a multi-modal prediction is disclosed herein. A computing system retrieves event data from a data store. The event data includes information for a plurality of events across a plurality of seasons. Computing system generates a predictive model using a mixture density network, by generating an input vector from the event data learning, by the mixture density network, a plurality of values associated with a next play following each play in the event data. The mixture density network is trained to output the plurality of values near simultaneously. Computing system receives a set of event data directed to an event in a match. The set of event data includes information directed to at least playing surface position and current score. Computing system generates, via the predictive model, a plurality of values associated with a next event following the event based on the set of event data.

Abstract: A method and related system for training a machine learning model using a federated learning structure by selectively sharing condensed data between devices includes operations to obtain datasets from client devices comprising a first client device and a second client device and updating a datasets subset to comprise a first dataset. The method further includes updating the datasets subset to comprise a second dataset based on a result indicating that a feature space distance between the first dataset and the second dataset satisfies a set of criteria and sending, to a third client device, the datasets subset comprising the first dataset and the second dataset. The method further includes obtaining, from the third client device, a set of model parameters that is derived from training based on the datasets subset and updating a server version of a machine learning model based on the set of model parameters.

Abstract: Methods and systems are described herein for minimizing development time in artificial intelligence models by automating model selection based on dataset fittings of time-series data prior to hyperparameter optimization.