107 results about "EMG - Electromyography" patented technology

Methods and systems are disclosed for collecting EMG speech signals using a speech signal detection device. The methods and systems collect a combination of signals comprising electromyograph (EMG) data signals and one or more non-EMG data signals. The methods and systems process the combination of signals by a machine learning (ML) model to detect inner speech of the user, the ML model trained to establish a relationship between training signals comprising training EMG data signals and training non-EMG data signals and ground-truth inner speech data and, in response, performing one or more operations associated with the speech signal detection device.

Methods and systems are disclosed for collecting EMG speech signals. The methods and systems present a target word for electromyograph (EMG) data collection on a graphical user interface (GUI) and receive input to initiate recording of EMG data. The methods and systems, in response to receiving the input, collect, by an EMG communication device, a set of EMG signals generated based on an individual user of the EMG communication device over a threshold period of time. The methods and systems determine whether the set of EMG signals collected over the threshold period of time corresponds to the target word and present feedback in the GUI based on whether the set of EMG signals collected over the threshold period of time corresponds to the target word.

Methods and systems are disclosed for collecting electromyograph (EMG) speech signals using a speech signal detection device and calibrating the speech signal detection device using online learning. The system accesses a machine learning (ML) model that has been trained based on a collection of training data to detect presence of inner speech (silent speech or any other form of speech) and collects, by a speech signal detection device, a combination of signals comprising EMG data signals and one or more non-EMG data signals. The system processes the combination of signals by the ML model to predict presence of inner speech and updates the collection of training data based on the combination of signals and prediction made by the ML model. The system retrains the ML model in an online learning approach using the updated collection of training data.

A closed-loop neuromodulation system includes at least one implantable neurostimulator configured to deliver electrical stimulation to a target nerve innervating a muscle; at least one biosignal sensor configured to detect a physiological signal associated with the target muscle, wherein the physiological signal may include an electromyographic (EMG) or electroneurographic (ENG) signal; and a controller operatively coupled to the neurostimulator and the sensor. The controller is configured to receive the physiological signal from the sensor, determine whether the signal satisfies a predefined condition indicative of weakness, strength, fatigue, muscle underactivity, or overstimulation, and, in response to detecting that the condition is satisfied, automatically adjust one or more stimulation parameters selected from amplitude, frequency, pulse width, train duration, or train count.

An electromyography (EMG) measurement device includes a garment configured to be worn on an anatomical region of an associated wearer, a plurality of electrodes arranged on the garment to contact skin of the anatomical region when the garment is worn on the anatomical region of the associated wearer, electronics operatively connected with the plurality of electrodes and configured to measure EMG data emanating from the anatomical region, and an electronic processor programmed to derive a contribution of spatial muscle activity of a target muscle or muscle group to the measured EMG data.

Methods and systems are disclosed for training a user-specific machine learning (ML) model to detect inner speech. The system accesses the ML model trained to detect inner speech based on a general population dataset. The system collects, by an electromyograph (EMG) communication device, a set of EMG signals generated based on an individual user of the EMG communication device. The system updates parameters of the ML model based on the set of EMG signals associated with the individual user. The system detects inner speech of the individual user by applying the ML model with the updated parameters to a new set of EMG signals received from the EMG communication device.

The invention discloses a gesture recognition system and method based on multi-channel mixing of EMG, FMG and IMU. The system comprises a multi-channel signal acquisition module, a signal conditioning and synchronous acquisition circuit, a data preprocessing module, a deep neural network fusion recognition module and an output interface module. By integrating a surface electromyogram, a muscle force diagram and an inertial measurement unit, synchronous acquisition of multi-modal physiological signals of the forearm of a human body is realized; a double-flow convolution-time sequence attention fusion network (DCTANet) is adopted for feature extraction and classification decision making, and neuromuscular activity, muscle mechanical response and limb movement information are effectively fused. The accuracy, robustness and real-time performance of gesture recognition are improved, and the method is suitable for the fields of intelligent prostheses, wearable interaction, rehabilitation training and the like. The system is compact in structure, supports embedded deployment, and has a good practical prospect.

An intelligent interactive neuromodulation system by intravesical and pelvic floor stimulation is provided, including: an electromyography (EMG) acquisition module, an intelligent diagnostic module, a display interaction module, and an electrical stimulation module. The intelligent diagnostic module is configured to receive an original EMG signal acquired by the EMG acquisition module and obtain feature information, generate a first treatment protocol by using the feature information and a preconfigured intelligent fitting model, and generate a classification result of an abnormal bladder activity based on the feature information and a preconfigured intelligent classification model; in response to identifying that a detrusor activity of a patient changes, automatically complete dynamic switching between a first treatment mode and a second treatment mode. The display interaction module is configured to receive a first treatment protocol and a second treatment protocol, and send the first treatment protocol to the electrical stimulation module to apply electrical stimulation pulses.

An electromyography (EMG) measurement system includes a garment configured to be worn on an anatomical region, electrodes arranged on the garment to contact skin of the anatomical region when the garment is worn on the anatomical region, electronics connected with the electrodes to measure EMG data emanating from the anatomical region, and an electronic processor programmed to filter the EMG data to suppress or remove artifacts using filters computed using approximate joint diagonalization of covariance (AJDC) matrices or by transforming the EMG data to source signals using iteratively adjusted forward filters.

Methods and systems are disclosed for collecting EMG speech signals using a speech signal detection device. The methods and systems collect, by a speech signal detection device worn by a first person, a combination of signals comprising electromyograph (EMG) data signals and one or more non-EMG data signals. The methods and systems process the combination of signals by a machine learning (ML) model to detect presence of a second person, the ML model trained to establish a relationship between training signals comprising training EMG data signals and training non-EMG data signals and ground-truth presence of people data. The methods and systems control operation of the speech signal detection device being worn by the first person in response to detecting the presence of the second person.

A user interface system enables communication through inaudible speech by detecting physiological signals associated with inner voice communication. The system includes a wearable device positioned in, on, or around a user's ear, housing one or more sensor modalities such as electromyography (EMG) sensors for detecting muscle movements, functional near-infrared spectroscopy (fNIR) sensors for brain activity, and / or detection-and-ranging systems using sonar or radar for micro-deformations. Pre-processing modules condition the sensor data, which is then analyzed by one or more machine learning models to reconstruct the content of the user's inaudible communication and produce output representations. The system enables high-bandwidth natural language interaction without audible vocalization, maintaining privacy while supporting applications including AI assistant interaction, real-time translation, secure authentication, and inter-party communication.

Systems, apparatuses, and methods described herein facilitate the placement of spinal cord stimulator (SCS) electrode in relationship to neural tissue of the spinal cord based on electromyography (EMG) data. The systems, apparatuses, and methods are designed to assist physicians and surgeons target specific neurophysiological locations through stimulation and subsequent visualization of EMG activity in response to stimulation to more precisely identify the location of the SCS electrodes that will maximize the intended therapeutics effects of SCS.

An electromyography (EMG) measurement system includes a garment configured to be worn on an anatomical region, electrodes arranged on the garment to contact skin of the anatomical region when the garment is worn on the anatomical region, electronics connected with the electrodes to measure EMG data emanating from the anatomical region, and an electronic processor programmed to filter the EMG data to suppress or remove artifacts using filters computed using approximate joint diagonalization of covariance (AJDC) matrices or by transforming the EMG data to source signals using iteratively adjusted forward filters.

The present disclosure relates to an augmented reality input device and method based on an inertial measurement unit and electromyography, the augmented reality input device being configured to be connected through a network to an arm band worn on an arm of a user and an augmented reality module worn on a face of the user and providing virtual augmented reality, and the armlet is provided with an inertial measurement unit (IMU) sensor and an electromyogram (EMG) sensor, and the augmented reality input device is configured to arrange a virtual cursor on augmented reality, the virtual cursor moving corresponding to a change in angle and a change in position of a user's arm detected by the IMU sensor, and providing an input corresponding to the gesture of the user's arm recognized by the EMG sensor.

The present application relates to devices for monitoring pregnancy or labor. In one embodiment, the device includes an electromyography (EMG) sensor having two or more EMG electrodes that monitor fetal or maternal activity during pregnancy or labor, and one or more position sensors that monitor the relative positioning of the two or more EMG electrodes during fetal or maternal activity. In one embodiment, the device includes a monitoring apparatus that is placed on a body and has a plurality of sensors integrated into the monitoring apparatus, the plurality of sensors including at least a first sensor configured to detect a first type of signal from the body indicative of a first type of fetal or maternal activity during pregnancy or labor, and a second sensor configured to detect a second type of signal from the body different from the first type of signal also indicative of the first type of fetal or maternal activity during pregnancy or labor.

A bruxism training device is disclosed that is capable of detecting bruxism events (e.g., jaw clenching and teeth grinding) via external electromyography (EMG). The training device is shaped to be easily placed in a proper location and proper orientation over a masseter muscle of a user. A locator region is provided to facilitate easy locating of a reference electrode over the gonial angle of the mandible, while an alignment edge of the training device is provided to facilitate easy locating of end electrodes adjacent an end of the masseter muscle coupled to the mandible. With the locator region and alignment edge in place, an array of mid electrodes is automatically positioned over a bulk of the masseter muscle, allowing useful EMG measurement to be acquired of masseter activation.

The invention discloses a lumbar surgery AI dynamic simulation system and decision making method based on multi-modal data fusion, and belongs to the field of medical artificial intelligence, the technical scheme is that the system comprises a multi-modal data input interface, a dynamic modeling engine, a surgical decision making matrix and a three-dimensional comparison unit, and specifically, CT / MRI, electromyogram and gait mechanics data are fused; constructing a spinal canal 4D model through improved U-Net < 3 + >, and predicting a nerve root dynamic path by means of GCN-LSTM; and in combination with a lumbar stenosis type, matching an adaptive operation form, and generating a pre-operation and post-operation quantitative comparison report marked by three colors. A spinal canal-nerve-biomechanics dynamic correlation model is innovatively established, 217 L4-S1 segment anatomical variation templates are arranged in the model, and the anatomical variation adaptation rate reaches 92.3%. Clinical test results show that the operation time planned by the system is averagely shortened by 26 minutes, and the postoperative reoperation rate is reduced to 2.4%.

Therapeutic dressings for wounds are provided, as well as methods of using the same and methods of fabricating the same. A therapeutic dressing can include a central non-adherent portion, an adherent portion surrounding the central non-adherent portion, a sensor, at least one stimulating electrode, and a circuit in operable communication with the sensor and the at least one stimulating electrode. The electrode(s) can deliver an electrical stimulus to one or more nerves through the skin of a patient / user. The sensor can be, for example, an electromyography (EMG) sensor, and the therapeutic dressing can be configured such that the sensor and the electrode(s) are in direct physical contact with the skin during use.

The invention is suitable for the technical field of deep learning, and provides a multi-attention-based hand electromyogram continuous motion estimation method, which comprises the following steps: carrying out preprocessing and feature extraction processing on a surface electromyogram signal, and inputting the processed surface electromyogram signal into a motion estimation model; feature extraction and dimension improvement are carried out through a depth separable convolution module, and the weight of each channel is adaptively adjusted through efficient channel attention; further capturing a global dependency relationship by using additive attention and a feed-forward network through a Transform encoder module; feature compression is carried out through a one-dimensional adaptive average pooling layer, each channel obtains respective average value, and global information is extracted; and inputting the global information into a single-layer MLP layer for nonlinear mapping, and outputting a joint angle. According to the method, depth separable convolution is used, so that the calculation amount and the parameter scale are reduced; and an efficient channel attention module is introduced, so that the feature expression capability is improved while the relatively high calculation efficiency is maintained.

A system and method for visualizing a position of a needle inside a patient's body for use in providing regional anesthesia and a system and method for providing electro anesthesia during surgery, post surgery and during rehabilitation. The method and system use EMG / AMG information regarding activity of a muscle associated with a target nerve to provide a visualization of a position of a needle or other device relative to the target nerve and to determine efficacy of the electro anesthesia. EMG / AMG information and feedback information such as EEG information may be used to provide automatic adjustment of a waveform provided for electro anesthesia to maintain a suitable pain level.

Methods and systems are disclosed for triggering operations using inner speech. The system detects presence of inner speech in electromyograph (EMG) data. The system adjusts a progress indicator in response to detecting the presence of inner speech. The system determines that a current value of the progress indicator has transgressed a specified threshold value and, in response, triggers an operation.

A computer-implemented method for estimating sleep stages that comprises receiving an electromyography (EMG) signal representative of electric activity of a muscle of the subject during the sleep session and providing only the EMG signal as an input to a machine learning model. The method comprises estimating a sleep stage during the sleep session based on an output of the machine learning model. The machine learning model is trained by providing, as a first input, a reference EMG signal representative of electric activity of a muscle of a reference subject during a reference sleep session, and providing, as a second input, a reference sleep stage signal representative of sleep stages of the reference subject during the reference sleep session. The machine learning model is trained by using the first input and the second input to estimate sleep stages based on only an EMG signal.

A neuromuscular assessment system and method of operation may include: attaching a first electrode array to an active muscle configured to detect a first electromyogram (EMG) signal; attaching a second electrode array to an antagonist muscle configured to detect a second EMG signal, the active muscle and the antagonist muscle forming an active / antagonist muscle pair; decomposing the first EMG signal into a first motion unit spike sequence; decomposing the second EMG signal into a second motion unit spike sequence; correlating the first motion unit spike sequence and the second motion unit spike sequence to generate a correlation signal; determining synchronism and periodicity within the correlated signals; and generating a tremor score, the tremor score being a percentage of the correlated signal determined to have both synchronism and periodicity.

Methods and systems are disclosed for collecting EMG speech signals using a speech signal detection device. The methods and systems collect a combination of signals comprising electromyograph (EMG) data signals and one or more non-EMG data signals. The methods and systems process the combination of signals by a machine learning (ML) model to detect inner speech of the user, the ML model trained to establish a relationship between training signals comprising training EMG data signals and training non-EMG data signals and ground-truth inner speech data and, in response, performing one or more operations associated with the speech signal detection device.

A system for monitoring autonomic nervous system (ANS) function in a subject includes at least one sensor configured to acquire a first physiological signal related to heart activity of the subject, one or more additional sensors configured to acquire one or more physiological signals, a processing unit configured to receive, process and analyze the first and additional physiological signals to monitor / detect, in real time, at least one of an autonomic nervous system dysfunction and a change in a physiological state of the subject and determine an output condition based on the detection, and an output mechanism configured to perform, based on the output condition, generating an alert, initiating a treatment, and / or storing data. At least one of the additional sensors is a blood glucose sensor, a respiration sensor, a sudomotor activity sensor, a pulse wave sensor, an electroencephalography (EEG) sensor, an electromyography (EMG) sensor or a motion sensor.

A system includes at least one wearable device having a housing, at least one sensor disposed within the housing, at least one output device disposed within the housing, and at least one processor operatively connected to the sensors and output devices, wherein one or more sensors are configured to detect electrical activity from a user's facial muscles and to transmit a data signal concerning the electrical activity of the user's facial muscles to one of more of the processors. A method of controlling a wearable device includes determining facial muscular electrical data of a facial gesture made by a user, interpreting the facial muscular electrical data to determine a user response, and performing an action based on the user response.

The present invention relates to an apparatus for treating neurological or psychiatric disorders using i-wave periodicity-dependent Quadri-Pulse Theta Burst Stimulation (iQPS). The system comprises a transcranial magnetic stimulation (TMS) stimulator configured to deliver test pulses to a target motor cortical area and therapeutic theta bursts to a treatment area with pulse frequency tailored to the patient's individual i-wave periodicity. An electromyography (EMG) device records motor evoked potentials (MEPs) from a target muscle and analyzes the intervals between MEPs to determine the i-wave periodicity. The personalized stimulation enhances cortical plasticity by synchronizing stimulation with the brain's intrinsic neural timing. The invention is applicable to neurological and psychiatric conditions that include depression, schizophrenia, stroke recovery and neuropathic pain. Advantages include improved treatment efficacy, reduced variability in patient response and a non-invasive, adaptive therapy model that aligns with the patient's unique neurophysiological characteristics, offering a precise and scalable approach to brain stimulation.

A system for early detection of movement instability includes a wearable device affixed to a body part of a subject, featuring at least one electromyography (EMG) sensor for detecting electrical muscle activity and at least one motion sensor for detecting movement. A computing device, communicatively connected to the wearable device, receives EMG data from the EMG sensor and motion data from the motion sensor. The computing device identifies movement instability patterns using a machine-learning model trained on datasets of known stable and unstable muscle activity data and movement patterns. It determines impairment indicators based on the identified movement instability patterns and generates a user interface displaying these indicators. This system facilitates early detection and monitoring of movement instability.

A neuromorphic device for imitating and reproducing a person's hand or arm motion according to the present invention may comprise: a communication unit for receiving information on an electromyography (EMG) signal related to a motion of a person's hand or arm measured by an EMG sensor, and receiving information on a movement of the person's hand or arm detected by a dynamic vision sensor (DVS) camera; and a processor for converting the received information on the EMG signal into a spike signal, applying, to a predetermined trained spiking neural network (SNN) model, the converted spike signal and the information on the movement of the person's hand or arm received from the DVS camera, and outputting a class corresponding to the motion of the person's hand or arm among classes defined in advance on the basis of movement and position.

A method and system for monitoring a patient for postpartum hemorrhage (PPH) is configured to receive PPH factor information for the patient, wherein the PPH factor information includes at least two of the following: ultrasound data generated based on one or more abdominal ultrasound images of the patient obtained in a measurement period, electromyography (EMG) data obtained during the measurement period from abdominal electrodes on the patient, cardiac measurement data obtained from the patient during the measurement period, and a uterine health indicator for the patient. The system and method are configured to process the PPH factor information to determine a PPH probability index indicating a probability that the patient will develop PPH and a PPH severity index predicting a severity of PPH, and generate a PPH risk index for the measurement period based on the PPH probability index and the PPH severity index.