8922 results about "Neuroscience" patented technology

In order to solve the problem that the bottleneck exists when image-text features are fused in a traditional anti-modal recognition model and deep-level interaction information between modals cannot be fully mined in an existing method, the invention provides a multi-modal anti-modal recognition model (MSCAG) fusing two-way cross attention and a gating mechanism, and a technical scheme is provided for solving the problem that the traditional anti-modal recognition model has the bottleneck when the image-text features are fused in the traditional anti-modal recognition model. Related attention features on a text layer and an image layer are respectively refined through a bidirectional cross attention mechanism, the text attention features, the images and basic features of the texts are integrated through residual connection, and the image attention features, the texts and the basic features of the images are integrated through residual connection; a gating mechanism is applied to enhance information interaction and key area focusing between the two modes. And the context features and the local features are combined to realize more reliable chiffon recognition. The invention provides an innovative method for multi-mode siphonage recognition, has important practical application value, and can be applied to siphonage recognition of netizen comments in social media.

A mechanical end effector for a humanoid robot includes a plurality of identical finger assemblies. Each of the finger assemblies is removably connected to a frame. Each of the finger assemblies is fully self-contained and operable independently of every other one of the finger assemblies and independently of every other component connected to the frame. Each of the finger assemblies includes a single electric motor and is configured to be fully operable using only the single electric motor.

The invention relates to a multi-modal intelligent robot system and an interaction method, and belongs to the technical field of robot digital data processing, and the method comprises the following steps: S1, multi-modal input collection; s2, carrying out multi-modal input preprocessing; s3, performing multi-modal information fusion, inputting the information into a pre-trained multi-modal feature fusion model, and performing feature fusion on each modal feature through a quantitative feature data fusion mechanism to obtain a user interaction intention vector and a user emotional state vector; s4, interaction response generation: matching a preset interaction response strategy library based on the user interaction intention vector to obtain basic response content, and performing emotion adaptation adjustment on the basic response content in combination with the user emotion state vector to generate multi-modal response content including text information, voice information and action information; s5, interactive feedback is executed; the method has the beneficial effects that the robot is fused with multi-modal information through the basic response content and the user emotional state vector, and interaction is dynamically optimized.

Systems, devices and methods are provided for stimulating one or more nerves within a user. A system comprises stimulator having an electrode configured for contacting the outer skin surface at a target location and an energy source coupled to the stimulator, The energy source is configured to generate at least one electrical impulse and to transmit the electrical impulse transcutaneously from the electrode through the outer skin surface of the user to a selected nerve in the user. In certain aspects, the systems and methods are particularly useful for non-medical purposes, such as reducing stress and anxiety, enhancing relaxation, improving sleep, energy, concentration and mood, increasing mental or physical performance, promoting mental health, recovery and wellness, generally improving the health and wellbeing of a user and / or for temporarily or permanently improving intelligence, learning capacity, memory retention, recall, mood, alertness and / or sleep efficiency in human beings.

An in vitro training method for training genioglossus muscle strength includes adhering an electrode patch of an in vitro training device to a bottom of a chin of a user during a non-sleep period. The electrode patch receives an electrical stimulation signal from an electrical stimulation module of the in vitro training device to stimulate the genioglossus muscle of the user through transdermal electrical stimulation. The electrode patch includes a body surface adhering face and an assembling face opposite to the body surface adhering face. The body surface adhering face is adhered to the bottom of the chin of the user to align with the genioglossus muscle. The electrical stimulation module is disposed on the assembling face and in electrical connection with the electrode patch. The electrical stimulation module sends an electrical stimulation signal to stimulate the genioglossus muscle through transdermal electrical stimulation.

The invention discloses a multi-modal cognitive impairment evaluation system based on a multi-dimensional cognitive function hypergraph. The multi-modal cognitive impairment evaluation system comprises a multi-modal signal acquisition system, a cognitive function evaluation normal form design, a universal feature extraction framework of each modal and online hypergraph learning based on a multi-dimensional cognitive function. The system constructs a dimensionality evaluation normal form containing the cognitive function based on multi-modal physiological and behavior signals, so that comprehensive evaluation of the cognitive function of the elderly is realized. Firstly, physiological and behavior responses of a user under different tasks and situations are recorded and analyzed in real time by fusing multi-mode signals such as electroencephalogram, electrocardio, voice and facial expressions. And then, mining fusion representation among the multi-modal signals, generating unified feature representation, introducing an online hypergraph of clinical knowledge, capturing a complex multi-modal data relationship, and providing an analysis result containing cognitive function evaluation. Finally, in combination with the technical means, the system can realize early detection of cognitive impairment.

Catheter apparatuses, systems, and methods for achieving renal neuromodulation by intravascular access are disclosed herein. One aspect of the present technology, for example, is directed to a treatment device having a multi-electrode array configured to be delivered to a renal blood vessel. The array is selectively transformable between a delivery or low-profile state (e.g., a generally straight shape) and a deployed state (e.g., a radially expanded, generally helical shape). The multi-electrode array is sized and shaped so that the electrodes or energy delivery elements contact an interior wall of the renal blood vessel when the array is in the deployed (e.g., helical) state. The electrodes or energy delivery elements are configured for direct and / or indirect application of thermal and / or electrical energy to heat or otherwise electrically modulate neural fibers that contribute to renal function or of vascular structures that feed or perfuse the neural fibers.

The present disclosure relates to systems and methods for enhancing the interaction between users and automated agents, such as digital assistants, by employing Large Language Models (LLMs) to infer the intent of spoken language. The invention involves continuously monitoring ambient audio, converting speech to text, and utilizing LLMs to determine whether spoken language is intended for the automated agent. A structured prompt, including the converted text and specific instructions, is sent to the LLM, which is fine-tuned to process domain-specific prompts. The LLM provides a structured output in a standardized format, indicating the user's intent. The system may involve multiple prompts to perform separate tasks, such as identifying intent and generating additional context-specific data. This approach facilitates a more natural and intuitive user experience by eliminating the need for wake words and allowing seamless conversational interaction with virtual assistants across various platforms and devices.

The present invention relates to the field of voice interaction analysis technology, and in particular to a voice interaction analysis method and system for an intelligent voice robot. The method comprises the following steps: improving the clarity of user interaction voice through multi-scale noise decomposition and filtering enhancement, performing granular segmentation and long-distance semantic analysis on clear voice, generating a semantic structure, performing multi-level semantic recognition based on the structure, extracting specific meanings and making preliminary responses, generating personalized semantic graphs through independent graph structure mapping, predicting user interaction dialogues, building an advance response engine based on dialogue templates, preparing to respond to user needs, embedding the engine into an intelligent voice robot, and realizing efficient voice interaction analysis. The present invention realizes a more accurate and efficient voice interaction analysis method for an intelligent voice robot.

To provide methods for treating a subject having or at risk of developing a neurocognitive disorder, such as frontotemporal lobar degeneration or neuronal ceroid lipofuscinosis.SOLUTION: Provided is a method comprising administering, to a subject diagnosed as having a neurocognitive disorder (NCD), a composition comprising a population of cells comprising a transgene encoding a progranulin (PGRN) or a granulin (GRN).SELECTED DRAWING: Figure 1A

The invention relates to a multi-mode intention recognition method and system, and the method comprises the steps: carrying out the time domain and frequency domain enhancement of the features of text, video and audio modes, carrying out the splicing to obtain non-language mode fusion features, combining the features of an original text, modeling the time synchronization relation of audio-text and video-text, and carrying out the recognition of a multi-mode intention. Standardized audio features, video features and text features are obtained through context alignment processing; fusing the standardized features of the three modalities to obtain a fused feature vector, and mapping the fused feature vector back to the text modal space to be connected with the weighted residual error of the original text feature to obtain a fused semantic vector; extracting global semantic anchor points and mask positions from the fused semantic vector, and splicing the global semantic anchor points and the mask positions with the original text features and the fused semantic vector to obtain input features; and obtaining probability distribution of multiple intention categories by using the input features. Three types of heterogeneous modal input can be supported, and the accuracy and robustness of intention recognition are improved through fine-grained semantic supervision and enhancement strategies.

The invention discloses a brain image analysis method and system based on multi-modal fusion, and relates to the technical field of brain image processing. A brain image analysis system based on multi-modal fusion comprises a data acquisition module, a data preprocessing module, a feature extraction and fusion module, a brain network analysis module, a comprehensive analysis module and a focus detection module. The comprehensive analysis model adopts a double-branch structure, deep processing is performed on multi-modal features and brain network features, and interactive fusion of the two types of features is realized through a cross-modal attention mechanism; the model is further combined with a classification branch and a regression branch to cooperatively complete brain disease classification and focus quantitative analysis, and the adaptive capacity and diagnosis performance of an existing model in a complex task scene are improved.

Techniques for detecting attributes and / or gestures associated with pedestrians in an environment are described herein. The techniques may include receiving sensor data associated with a pedestrian in an environment of a vehicle and inputting the sensor data into a machine-learned model that is configured to determine a gesture and / or an attribute of the pedestrian. Based on the input data, an output may be received from the machine-learned model that indicates the gesture and / or the attribute of the pedestrian and the vehicle may be controlled based at least in part on the gesture and / or the attribute of the pedestrian. The techniques may also include training the machine-learned model to detect the attribute and / or the gesture of the pedestrian.

The invention relates to the technical field of brain-computer interfaces, and discloses a self-adaptive dynamic feedback brain-computer interface training method and system based on virtual reality. The training method comprises the steps that a user performs motor imagery according to a training scene presented by the VR display device; the electroencephalogram acquisition equipment acquires electroencephalogram signals; configuring a data processing module, and operating a motor imagery continuous control model to calculate characteristic parameters by taking the electroencephalogram signals as input; under the condition that the task exists, the characteristic parameters are used as input, and a dynamic self-adaptive characteristic-feedback mapping model is operated to calculate control parameters; sending the control parameters to a functional electrical stimulation device to load electrical stimulation of corresponding intensity to the muscle to obtain tactile feedback; meanwhile, the dynamic adaptive feature-feedback mapping model is dynamically updated according to latest input feature parameters; the control parameters are sent to the VR display device to present visual feedback, and the user adjusts the motor imagery based on the visual feedback and the haptic feedback. The participation sense of the user and the training efficiency can be improved.

The invention discloses an alimentary canal pathological diagnosis visual language large model construction method based on reinforcement learning and application thereof, and belongs to the technical field of medical image processing. The method comprises the following steps: firstly, extracting pathological information through layout analysis and adaptive threshold processing, and recombining the pathological information into a structured data set containing an inference chain; secondly, a visual encoder and a multi-branch classifier are used for extracting features and confidence coefficients, and dynamic structured cue words are generated; and finally, inputting the image and the cue word into a multi-modal large model, carrying out supervised fine-tuning hot start, and carrying out reinforcement learning training by adopting a group relative strategy optimization algorithm in cooperation with a composite reward function containing format, semantics and diagnosis dimensions. The problems that a general model is prone to generating illusion in a pathological scene and lacks reasoning logic are solved, and the accuracy and logicality of pathological report generation are remarkably improved.

A generative mediator engine can perform a requested interaction with a conversational agent of a target entity on behalf of a user. An internal conversational platform can identify intents for the requested interaction. An external artificial intelligence engine can perform intent discovery when an intent is not identified above a confidence threshold. A discovered intent unknown to the generative mediator engine can be received from the external artificial intelligence engine and used, with input requirements determined by the generative mediator for the requested interaction, by a dialog generator to generate a sample dialog for the requested interaction. User feedback can be received after review of action items and expected inputs identified from the sample dialog. The generative mediator engine can perform the requested interaction with the conversational agent on behalf of the user and without receiving user intervention during the requested interaction.

A method for facilitating conversational interaction with users to help the users includes transmitting a conversational interaction interface for conversationally interacting with a user to a user device, receiving a request of the user through the conversational interaction interface from the user device, identifying an information based on the request, generating an input comprising the request and the information for a machine learning model based on the request and the information, processing the input using the machine learning model, generating a response for the request based on the processing of the input, transmitting the response through the conversational interaction interface for conversationally interacting with the user to the user device, and storing the machine learning model, the request, and the response.

Aspects of the disclosure relate to methods of reducing fatigue in a subject having myotonic dystrophy type 1 (DM1). Aspects of the disclosure relate to methods of treating one or more symptoms assessable by the MDHI (e.g., a GI symptom, myotonia, upper extremity function impairment, fatigue, mobility impairment, impairment in the ability to perform activities, pain, vision impairment, communication impairment, sleep impairment, emotional issues, cognitive impairment, social satisfaction impairment, social performance impairment, breathing impairment, swallowing impairment, and / or hearing impairment) in a subject having myotonic dystrophy type 1 (DM1). In some embodiments, the methods comprise administering to the subject a composition comprising complexes (e.g., muscle targeting complexes) comprising an oligonucleotide (e.g., a DMPK- targeting oligonucleotide) covalently linked to an antibody (e.g., anti-TfRl antibody).

The invention discloses a transcranial electrical stimulation intracranial vector electric field detection device and method, and belongs to the transcranial electrical stimulation intracranial electric field detection technology, the device inputs an individual head model in a stimulation system, sets a target spot and calculates transcranial electrical stimulation parameters; the implantation position and angle of the vector detection electrode are planned through the navigation system, electrical stimulation is applied to a brain target area through the stimulation electrode, and intracranial electric signals are collected in real time through the vector detection electrode; performing high-precision amplification on the acquired weak signal, and converting the weak signal into a digital signal; the collected electric field signals are calculated and processed in real time, and electric field distribution generated by transcranial electrical stimulation in the brain is analyzed and visualized; according to the method, time domain and frequency domain analysis is carried out on electric signals before and after stimulation, the influence of stimulation on neural signals in a detection area is evaluated, and high-precision measurement of the intensity and direction of a three-dimensional electric field is achieved through the design of a high-precision and multi-layer vector detection electrode.

In an approach to coordinating a conversational agent with a large language model for conversation repair, one or more computer processors receive a failure indicator from a first conversational agent. One or more computer processors retrieve a descriptive prompt associated with the first conversational agent. One or more computer processors transmit the descriptive prompt to a large language model. One or more computer processors transfer control of the failed conversation from the first conversational agent to the large language model. One or more computer processors determine the intent of the user associated with the failed conversation using the large language model. One or more computer processors determine whether the intent of the user associated with the failed conversation matches a capability of the first conversational agent. One or more computer processors transfer by one or more computer processors, the user back to the first conversational agent.

The invention discloses a virtual-real fusion exhibition display interaction system and a multi-mode perception method, and belongs to the technical field of exhibition display interaction. The system collects audience eyeball fixation points, gesture actions and ambient light data through AR / VR equipment, analyzes coordinates of a region of interest through an eyeball fixation point attention mechanism, a gesture space-time encoder and a multi-modal fusion unit, triggers holographic projection explanation and virtual exhibition stand light and shadow dynamic adjustment (including illumination intensity, color, Gaussian blur and the like) based on a threshold value, and performs real-time display on the virtual exhibition stand. And multi-user collaborative interaction is realized through federal learning. According to the method, reinforcement learning is adopted to optimize an event-driven threshold value, and virtual and real visual splitting is eliminated in combination with ambient light adaptive mapping. The problems of low participation degree, insufficient single-mode interaction information and multi-user cooperation of traditional exhibition are solved, interest analysis accuracy is improved through multi-mode fusion, personalized experience is enhanced through dynamic interaction, the method is suitable for multiple scenes such as museums and science and technology museums, and exhibition intellectualization, immersion and group interaction efficiency are effectively improved.

The delivery of neurostimulation to a subject using a closed-loop neurostimulation device (e.g., using a brain stimulation device to provide neurostimulation to a subject's brain) is controlled based on a global optimization-based temporal prediction framework. As a result, the brain stimulation device (e.g., a transcranial magnetic stimulation (“TMS”) device) is synchronized with the ongoing neural state (e.g., brain state) in real time. For instance, a brain recording is analyzed to extract the brain process of interest (e.g., frequency of brain oscillations) and used train a prediction algorithm. After that, a stimulation stage is implemented, in which the individual brain state is analyzed in real-time, the occurrence of the biomarkers (e.g., brain oscillation phase) is predicted, and the stimulation is triggered at the expected time.

The present invention describes a system and a method for monitoring and optimizing a user interaction session within an interactive voice response system during a human-computer interaction, the user interaction management system (100) for monitoring and optimizing the user interaction session comprising a conversation controller module (109), the conversation controller module (109) receives the audio features and processes outputs of ASR and NLU during the user interaction session to optimize the user interaction session duration. According to an embodiment of the present invention, the conversation controller may suggest the TTS to increase the rate of speech. According to yet another embodiment, the conversation controller may suggest the TTS to decrease the rate of speech.

The invention discloses a real-time evaluation method for a transcranial electrical stimulation effect, and relates to the technical field of brain electrical stimulation effects, and the real-time evaluation method comprises the following steps: fusing MRI modeling, HD-tDCS electrode optimization, electric field simulation and genetic algorithm parameter configuration, combining multi-modal nerve response collection, multi-task brain state prediction based on an attention mechanism, and Granger causal path modeling. And a real-time evaluation mechanism oriented to a full chain of'stimulation-nerve-behavior 'is constructed. Compared with an existing evaluation mode which only depends on electroencephalogram change or a single behavior index, the method has the advantages that the space targeting, the nerve activation specificity, the behavior improvement degree and the causal connection strength of stimulation can be quantified at the same time, and parameter self-feedback optimization is driven through comprehensive scores; and the instantaneity, the accuracy and the individualized regulation and control capability of transcranial electrical stimulation effect evaluation are remarkably improved.

The disclosure provides methods for treating a subject in need thereof comprising administering to the subject a therapeutically-effective dose of psilocybin. The methods described herein may be used to treat a variety of diseases, disorders, and conditions. For example, the methods may be used to treat anxiety disorders, eating disorders, and headache disorders.

The invention discloses a missing perception prompting method and system based on modal specific and general information, and belongs to the field of multi-modal learning. Performing modal missing processing on the initial features of the original video sample; unifying feature dimensions and sizes of modal missing features; wherein after unifying the feature dimensions, splicing to obtain a multi-modal joint feature; performing size unification and then performing fusion to obtain a modal universal prompt; using a gating fusion mechanism to control the fusion of the learnable matrix corresponding to the modal to obtain a modal specific prompt; fusing the modal general prompt and the modal specific prompt to obtain a specific-general prompt as a missing perception prompt; and the specific-general prompt and the multi-modal joint features are spliced to serve as input of a multi-modal model, and a full connection layer is introduced to complete a multi-modal downstream task. According to the method, the learnable specific-universal prompt of mode missing perception is introduced into the multi-mode model, so that cross-mode application in a data mode incomplete scene is realized, and the performance of a multi-mode downstream task is improved.

Described herein is an Autonomous Conversational AI system, which does not require any human configuration or annotation, and is used to have multi-turn dialogs with a user. A typical Conversational AI system consists of three main models: Natural Language Understanding (NLU), Dialog Manager (DM) and Natural Language Generation (NLG), which requires human provided data and configuration. The system proposed herein leverages novel Conversational AI methods which automatically generates conversational AI configuration from any historical conversation logs. The automatically generated configuration contains Auto-Topics, Auto-Subtopics, Auto-Intents, Auto-Responses and Auto-Flows which are used to automatically train NLU, DM and NLG models. Once these models are trained for given conversation logs, the system can be used to have dialog with any user.

The invention relates to an intelligent dialogue system and method based on an AI multi-mode large model. The method comprises the following steps: collecting biological characteristic data of a user in real time; cloning a personalized expression mode of the user by using the generative adversarial network; driving the audio avatar to perform multi-round dialogue interaction with the user, capturing a real-time physiological signal of the user through an expression recognition module, and generating a dynamic response content suggestion in combination with a dialogue context; and analyzing the interaction process based on the reinforcement learning model. The multi-modal deep fusion of the voice rhythm features, the language structure features and the facial dynamic features is realized, so that the limitation of single-modal or simple feature splicing in the prior art is broken through, the emotional state of the user can be accurately and comprehensively captured, the intention is expressed and the fine physiological reaction is expressed, and the user experience is improved. And a solid foundation is laid for constructing high-fidelity user representation.

Disclosed herein are systems, devices, and methods for peripheral nerve stimulation, particularly for treating tremor. The nerve stimulation may be accomplished by a wearable nerve stimulation device, such as a band configured to be worn on the wrist or other body part. The device can accomplish targeted nerve stimulation using circumferentially spaced electrodes. In some embodiments, the device may use only the same number of electrodes as the number of nerves that are to be stimulated. A biphasic charge-balanced waveform may be used to selectively stimulate a nerve near one of the activated electrodes but not to stimulate the nerve near the other activated electrode. The device may use dry electrodes for long-term, repeated use. The dry electrodes may include a conductive base layer and a polymeric, plastic or rubber skin contact layer comprising a conductive filler. The filler may be a powder, fiber, conductive coating, etc.