Abstract: Disclosed are a sorbent produced by recycling waste paper, which is an eco-friendly recycling paper resource, to adsorb and remove a marine spilled oil and other types of oil and a method of producing the same. The oil sorbent is produced by hydrophobizing a biomass material such as waste paper using a gas-grafting method. Compared to the oil sorbent of the prior art formed of synthetic resin, it is possible to provide excellent oil adsorption selectivity, a high oil adsorption rate, and a low water absorption rate. Therefore, an incineration process for processing the waste oil sorbent after use is easy advantageously. The oil sorbent generates less harmful substances during combustion because the eco-friendly material is used. Since a biomass material is hydrophobized using vegetable fatty acids, decomposition is performed fast within a short time in a landfill process, and there is no risk of environmental pollution.

Abstract: Systems and methods for providing an artificial intelligence-based solution in a dynamic environment that requires models with varying degrees of nuance and specialization. One such dynamic environment relates to generating dynamic human-like conversational responses based on complex data. In particular, systems and methods recite generating dynamic human-like conversational responses using a modular architecture featuring layered data models with gated neural networks. The modular architecture compartmentalizes the various components and functions of an application. That is, the architecture may support multiple layers, each featuring models performing specific functions and/or having been trained on using specific data and/or algorithms.

Abstract: Compounds of formula (I) wherein Y, n, R1, R2 and R3 have the meanings indicated herein, or a pharmaceutically acceptable solvate or salt thereof, are disclosed. The compounds are useful as inhibitors of protein or lipid kinases (in particular as inhibitors of a member of the PIM family of kinases, e.g. PIM-1, PIM-2 or PIM-3) and can be used in the treatment of a number of conditions, in particular cancers.

Abstract: A method for automatically generating clinical structured reports based on templates using voice recognition, the method implemented by a computing device and includes applying a natural language processing algorithm to a captured voice input from a client to identify one or more keywords or phrases. At least one of a plurality of clinical structured report templates are identified based on the identified one or more keywords or phrases correlated to medical examination data points associated with each of the templates. A clinical structured report is automatically prepared based on the identified clinical structured report templates without a non-voice input and without an explicit separate voice command directed to manage a report generation operation. The clinical structured report includes modifications to the clinical structured report template based on the identified one or more keywords or phrases. The clinical structured report is provided to the client.

Abstract: A system is for use in accessing and treating a carotid artery. The system includes an arterial access sheath configured to be introduced into a common carotid artery. A shunt fluidly communicates with the arterial access sheath and provides a pathway for blood to flow from the arterial access sheath. A flow control assembly is mechanically attached to the pathway of the shunt.

Abstract: A patch has: a silicone body; a primer on an underside of the body; a silicone adhesive on the primer; and a release strip on the adhesive. The body has a plurality of perforations.

Abstract: The present invention relates to a wind turbine rotor blade spacer, a transportation and storage system for wind turbine rotor blades and a related method that prevent the vertical flexion of each one of the blades assemblies, minimizing the stresses and avoiding the contact between vertically and/or horizontally adjacent blades, and in consequence, blade damage. Also, the overall occupied surface by the system once assembled is minimised.

Abstract: A technique may include receiving, by a management service a plurality of instance configurations from a client device. The technique may then include receiving, by the management service, information identifying a launch request for a compute instance. The technique may include determining, by the management service, one or more candidate shapes for the compute instance based at least in part on the plurality of instance configurations. The technique may include selecting, by the management service and from the one or more candidate shapes, a launch shape for the compute instance and launching the compute instance using the launch shape. The technique may then include providing, the client device access to the compute instance, launched based on the launch shape.

Abstract: Techniques herein disclose a method for initializing a home automation device using near-field communication and a separate computing device. The computing device receives a request to initialize a home automation device. The computing device receives an indication of one or more initialization settings for the home automation device. The computing device installs the one or more initialization settings onto the home automation device by sending the one or more initialization settings to the home automation device using near-field communication.

Abstract: A system is configured to: (a) receive multi-modal data from one or more sources, the multi-modal data including text data; (b) analyze the data to determine features; (c) store the features in a database; (d) receive a search query for searching the stored features; (e) analyze the search query using a large language model to extract search features; (f) generate search results from the search features; and (g) display search results on a standardized graphical user interface including (i) a map and (ii) a legend having at least one or more of the search features displayed.

Abstract: A system is configured to: (a) receive multi-modal data from one or more sources; (b) analyze the data to determine features; (c) store the features in a database; (d) receive a search query for searching the stored features; (e) analyze the search query using a large language model to extract search features; (f) generate search results from the search features; and (g) display search results on a standardized graphical user interface including a legend having at least one or more of the search features displayed.

Abstract: A system is configured to: (a) receive multi-modal data from one or more sources; (b) analyze the data to determine features; (c) store the features in a database; (d) receive a search query for searching the stored features; (e) analyze the search query using a large language model to extract search features; (f) generate search results from the search features; and (g) display search results on a standardized graphical user interface including a legend having at least one or more of the search features displayed.

Abstract: A method of artificial-intelligence-based robotic pipetting for spermatozoa preparation includes positioning a vessel containing a semen sample on a staging mechanism, wherein the staging mechanism includes at least one motor to position the vessel at a specified orientation, using an artificial intelligence/machine learning (AI/ML system) system. The method includes confirming a position of the vessel, using the AI/ML system. The method includes directing a robotic pipettor to place a pipette tip within the semen sample at a position specified by the AI/ML system. The method includes aspirating and mixing the semen sample using the robotic pipettor. The method includes receiving an image object from an imaging system that includes a microscopy system, a camera system, and a lighting system. The method includes determining a semen liquefaction state of the semen sample based at least in part on the image object, using the AI/ML system.

Abstract: There is provided a formatting module configured to format spaces in an electronic character sequence. The formatting module supports at least one language and comprises a language identifier configured to identify whether the electronic character sequence is written in a supported language, and a character identifier configured to identify a particular character or a particular sequence of characters in the electronic character sequence. The formatting module is configured to format spaces in the electronic character sequence on the basis of the language identified and the particular character identified or the particular sequence of characters identified, when a supported language is identified. A system and method for formatting text are also provided.

Abstract: A method of artificial-intelligence-based robotic pipetting for spermatozoa preparation includes positioning a vessel containing a semen sample on a staging mechanism, wherein the staging mechanism includes at least one motor to position the vessel at a specified orientation, using an artificial intelligence/machine learning (AI/ML system) system. The method includes confirming a position of the vessel, using the AI/ML system. The method includes directing a robotic pipettor to place a pipette tip within the semen sample at a position specified by the AI/ML system. The method includes aspirating and mixing the semen sample using the robotic pipettor. The method includes receiving an image object from an imaging system that includes a microscopy system, a camera system, and a lighting system. The method includes determining a semen liquefaction state of the semen sample based at least in part on the image object, using the AI/ML system.

Abstract: A power conversion system includes a high voltage (HV) switchgear; a solid-state transformer; and a low voltage (LV) switchgear. The HV switchgear connects to and disconnects from a HV network. The HV switchgear connects to an input of the solid-state transformer, which comprises first and second modules. The first module converts an HV alternating current (AC) signal into a HV DC signal, and the first module converts an HV DC signal into an HV AC signal. The second module converts an HV DC signal into at least one LV DC signal, and the second module converts at least one LV DC signal into an HV DC signal. The LV switchgear connects to an output of the solid-state transformer. The LV switchgear connects to a plurality of applications or devices and disconnects from the plurality of applications or devices.

Abstract: A power conversion system includes a high voltage (HV) switchgear; a solid-state transformer; and a low voltage (LV) switchgear. The HV switchgear connects to and disconnects from a HV or medium voltage (MV) network. The HV switchgear connects to an input of the solid-state transformer. The solid-state transformer comprises a first module and a second module. The first module converts a HV alternating current (AC) signal into a HV direct current (DC) signal. The second module connects to an output of the first module and converts the HV DC signal from the first module into at least one LV DC signal. The second module comprises at least one sub-module that converts the HV DC signal from the first module into a LV DC signal. The LV switchgear connects to an output of the solid-state transformer.

Abstract: Systems and methods for providing an artificial intelligence-based solution in a dynamic environment that requires models with varying degrees of nuance and specialization. One such dynamic environment relates to generating dynamic human-like conversational responses based on complex data. In particular, systems and methods recite generating dynamic human-like conversational responses using a modular architecture featuring layered data models with gated neural networks. The modular architecture compartmentalizes the various components and functions of an application. That is, the architecture may support multiple layers, each featuring models performing specific functions and/or having been trained on using specific data and/or algorithms.

Abstract: A technique may include receiving, by a management service a plurality of instance configurations from a client device. The technique may then include receiving, by the management service, information identifying a launch request for a compute instance. The technique may include determining, by the management service, one or more candidate shapes for the compute instance based at least in part on the plurality of instance configurations. The technique may include selecting, by the management service and from the one or more candidate shapes, a launch shape for the compute instance and launching the compute instance using the launch shape. The technique may then include providing, the client device access to the compute instance, launched based on the launch shape.

Abstract: The present invention relates to a wind turbine rotor blade spacer, a transportation and storage system for wind turbine rotor blades and a related method that prevent the vertical flexion of each one of the blades assemblies, minimizing the stresses and avoiding the contact between vertically and/or horizontally adjacent blades, and in consequence, blade damage. Also, the overall occupied surface by the system once assembled is minimised.