KR102588532B1 - System and method for multi muscle stimulation - Google Patents

Abstract

The present invention is a multi-muscle stimulator system and method that measures EMG signals, amplifies the measured EMG signals, passes only the EMG signals using a low-pass filter, and then stimulates muscle locations requiring stimulation based on the passed EMG signals. It's about.
In addition, it relates to an artificial intelligence learning system that measures and learns electromyography signals of repeated movements, predicts movements through the learned data, and stimulates muscles.

Description

Multiple muscle stimulator system and method {SYSTEM AND METHOD FOR MULTI MUSCLE STIMULATION}

The present invention relates to a multi-muscle stimulator system and method, which measures an electromyogram signal, amplifies the measured electromyogram signal, passes only the electromyogram signal using a low-pass filter, and then determines the muscle location requiring stimulation based on the passed electromyogram signal. It relates to a multi-muscle stimulator system and method for stimulating the muscle.

In addition, it relates to an artificial intelligence learning system that measures and learns electromyography signals of repeated movements, predicts movements through the learned data, and stimulates muscles.

Stroke is divided into cerebral infarction caused by a punctured cerebral blood vessel, cerebral hemorrhage caused by a ruptured cerebral blood vessel, or hemorrhagic cerebrovascular disease. Patients who have a stroke experience paralysis of the entire body, half of the body, or parts of the body, such as limbs.

Previously, technology was used to apply electrical stimulation to muscles by attaching electrode pads to paralyzed body parts to help with recovery. However, data was learned through repetition of specific movements, and movements were not predictable, so electrode pads were used. There was a problem that only specific muscles attached were stimulated.

In addition, recovery from paralysis in stroke patients usually takes between 2 months and 6 months.

While a stroke patient recovers from paralysis, the electromyographic signal of the corresponding muscle changes. Therefore, there is a need for an artificial intelligence learning system and prediction system that identifies movement intention through changing electromyography signals and provides muscle stimulation that matches the movement intention.

The present invention was developed to solve the above-mentioned problems, and measures the electromyogram signal, amplifies the measured electromyogram signal, passes only the electromyogram signal using a low-pass filter, and then stimulates the muscle that needs stimulation based on the passed electromyogram signal. It relates to a multi-muscle stimulator system and method for stimulating locations.

In addition, it relates to an artificial intelligence learning system that measures and learns electromyography signals of repeated movements, predicts movements through the learned data, and stimulates muscles.

According to one aspect of the present invention, in a multi-muscle stimulator system for measuring electromyographic signals and stimulating muscles based on the measured electromyographic signals, the electrode suit unit receives the electromyographic signal measurement and muscle stimulation signals, and the measured electromyographic signals. Controls an amplifier unit for amplifying, a stimulation unit for transmitting the muscle stimulation signal, a signal processing unit for processing the EMG signal data, a low-pass filter unit for passing low frequency signals, and a plurality of switches connected to the electrode suit unit. and a processor unit, wherein the low-pass filter unit removes noise from the EMG signal and allows only the EMG signal to pass.

In one embodiment, the electrode suit unit measures a channel electrode for transmitting the EMG signal to the amplifier unit, a ground electrode that serves as a reference potential for at least one of the muscle stimulation signal and the EMG signal, and the EMG signal, It may be characterized by including a reference electrode that serves as a comparison potential.

In one embodiment, the electrode suit unit may have a matrix structure and may include an electrode switch for each electrode to operate by receiving the muscle stimulation signal from the stimulation unit to a specific electrode.

In one embodiment, the stimulation unit turns on the channel switch of the first electrode and turns on the ground switch of the remaining electrodes except the first electrode to transmit the first muscle stimulation signal. can do.

In one embodiment, the electrode suit unit may be configured to turn on a channel switch of the first electrode and a reference switch of the second electrode in order to simultaneously measure the muscle stimulation signal and measure the electromyogram.

In one embodiment, the stimulation unit may transmit the muscle stimulation signal of a higher frequency than the frequency of the EMG signal to the electrode suit unit.

According to another aspect of the present invention, a multi-muscle stimulator system that measures EMG signals and stimulates muscles based on the measured EMG signals, a sensor data processing module unit that processes data, and a sensor data transmission module unit that transmits sensor data. An edge system comprising: a server data processing module unit that transmits and receives data from a processing module unit of the edge system; And an artificial intelligence system server including a sensor data database (DB) that stores data received from the sensor data transmission module unit of the edge system, connected to a mobile communication bridge and wireless communication server to improve the mobility of the edge system. It may be characterized by including the mobile communication repeater.

In one embodiment, the multi-muscle stimulator system may include an electrode suit having a matrix structure, and may be characterized in that it measures the EMG signal in the same operation through the electrode suit.

In one embodiment, the artificial intelligence system server performs artificial intelligence learning based on muscle stimulation signals transmitted during the multiple measurements through multiple measurements of the electromyogram signal, and performs multiple artificial intelligence learning to determine neural network weights. It can be characterized as calculating.

In one embodiment, the artificial intelligence learning system further includes an artificial intelligence prediction system that generates prediction data through training data learned through the artificial intelligence learning system and the multi-layer neural network weights, and the artificial intelligence prediction system Receiving the measured real-time EMG signal and the prediction data, and transmitting a control signal to the multi-muscle stimulator system to transmit the muscle stimulation signal to at least one electrode among a plurality of electrodes through the multi-layer neural network weight. can do.

In another aspect of the present invention, in the multiple muscle stimulation method performed in a multiple muscle stimulator system, the electromyographic signal measurement and muscle stimulation step is performed in the electrode suit unit and receives the electromyographic signal measurement and muscle stimulation signals, and an amplifier. An amplification step for amplifying the measured EMG signal, performed in a stimulation unit, a stimulation signal transmission step for transmitting the muscle stimulation signal, performed in a stimulation unit, a data processing step for processing the EMG signal data, performed in a signal processing unit , is performed in a low-pass filter unit, and includes a low-pass filter step that passes a low-frequency signal, and a control step that is performed in a processor unit and controls a plurality of switches connected to the electrode suit unit, wherein the low-pass filter step includes the It may be characterized by including the step of removing noise from the EMG signal and allowing only the EMG signal to pass.

According to the present invention, the device induces movement by stimulating the muscle area of a patient paralyzed by a stroke, learns the electromyographic signal for repeated movements, predicts it, and performs an action appropriate for the electromyographic signal through a stimulation signal to the muscle. It has an advantage.

Figures 1 and 2 are reference diagrams for explaining the multi-muscle stimulator system 100 according to an embodiment of the present invention.
Figure 3 is a reference diagram showing a protocol for stimulation according to an embodiment of the present invention.
Figure 4 is a reference diagram for explaining the structure of a frame for controlling the movement of closing and extending the elbows by contracting the biceps and triceps muscles according to an embodiment of the present invention.
Figure 5 is a reference diagram for explaining the structure of a super frame capable of stimulating multiple muscles simultaneously according to an embodiment of the present invention.
Figure 6 is a reference diagram for explaining a device, edge system, server, mobile communication bridge, and mobile communication repeater according to an embodiment of the present invention.
Figure 7 is a reference diagram for explaining the structure of an artificial intelligence learning system according to an embodiment of the present invention.
Figure 8 is a reference diagram for explaining the structure of an artificial intelligence prediction system according to an embodiment of the present invention.
Figure 9 is a flowchart explaining the sequence of the multi-muscle stimulation method.

Since the present invention can be modified in various ways and can have various embodiments, specific embodiments will be illustrated in the drawings and described in detail in the detailed description. However, this is not intended to limit the present invention to specific embodiments, and should be understood to include all transformations, equivalents, and substitutes included in the spirit and technical scope of the present invention. In describing the present invention, if it is determined that a detailed description of related known technologies may obscure the gist of the present invention, the detailed description will be omitted.

Terms such as first and second described in the specification may be used to describe various components, but the components should not be limited by the terms. The above terms are used only for the purpose of distinguishing one component from another.

Additionally, the term "... unit" used in the specification refers to a unit that processes one or more functions or operations, and may be implemented as hardware, software, or a combination of hardware and software.

The terms used in this application are only used to describe specific embodiments and are not intended to limit the invention. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this application, terms such as “comprise” or “have” are intended to designate the presence of features, numbers, steps, operations, components, parts, or combinations thereof described in the specification, but are not intended to indicate the presence of one or more other features. It should be understood that this does not exclude in advance the possibility of the existence or addition of elements, numbers, steps, operations, components, parts, or combinations thereof. Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

Figures 1 and 2 are reference diagrams for explaining the multi-muscle stimulator system 100 according to an embodiment of the present invention.

First, in order to stimulate muscles in the paralyzed area of a stroke patient, an electrode for a stimulation signal and an electrode for grounding must be present within the target muscle area in order to stimulate a specific muscle area among the approximately 600 muscles in the human body. In addition, in order to implement various movements based on installed electrodes, various muscles must be stimulated, which requires an electrode suit with a matrix structure and an electrode switch to select electrodes for stimulating specific muscles.

Looking at Figure 1, the multi-muscle stimulator system 100 includes an electrode suit unit 110, an amplifier unit 120, a signal processing unit 130, a low-pass filter unit 140, a processor unit 150, and a stimulation unit 160. Includes.

Here, the electrode suit unit 110 includes an electrode pad, and can measure an electromyogram signal of a specific area through the electrode pad and receive a muscle stimulation signal. After receiving the EMG signal and the muscle stimulation signal through the electrode sub unit 110, the amplifier unit 120 amplifies at least one of the EMG signal and the muscle stimulation signal. At least one of the EMG signal and the muscle stimulation signal amplified through the amplifier unit 120 is subjected to data processing of the EMG signal through the signal processing unit 130. Here, data processing refers to converting the digitized signal into a desired direction. It refers to numerical processing by an algorithm for the purpose of modifying or improving an information signal. It usually means that an analog signal becomes a discrete signal and becomes a numerical value in the process of digitizing it. The EMG signal processed as data from the signal processing unit 130 is passed through a low-pass filter that passes only low-frequency signals, removes noise, and passes only the EMG signal. In addition, the processor unit 150 controls muscle stimulation control and a plurality of switches connected to the electrode suit unit 110, and the stimulation unit 160 transmits muscle stimulation signals that can stimulate each pad.

The electrode suit unit 110 has a matrix structure, and each electrode included in the electrode suit unit 110 includes a channel electrode for transmitting a stimulation signal to the muscle or an electromyography signal from the muscle to the amplifier unit, and a muscle stimulation device. It includes a ground electrode that serves as a reference for at least one of the signal and the EMG signal, and a reference electrode for measuring the reference signal when amplifying the EMG signal of the muscle, and includes a switch for selecting it.

One electrode can serve as one of a channel electrode, a ground electrode, and a reference electrode.

The channel electrode serves to measure the electromyogram or transmit stimulation, the ground electrode serves as a reference potential when measuring the electromyogram or providing stimulation, and the reference electrode can be used as a comparison potential when measuring the electromyogram.

When stimulating a muscle, the channel electrode and the ground electrode must be present at the same time at the location to be stimulated, so whenever the location to be stimulated changes in the chute electrode of the electrode suit part 110 of the matrix structure, the channel electrode and the ground electrode that stimulate simultaneously The ground electrode must be changed.

Additionally, in order to measure electromyography, it can be measured using the ground electrode or reference electrode of the electrode suit unit 110.

Figure 3 is a reference diagram showing a protocol for stimulation according to an embodiment of the present invention.

The EMG signal has a frequency between 100Hz and 500Hz, and when stimulating, the stimulation must be performed at a higher frequency than the EMG signal in order to distinguish it from the EMG signal. Additionally, a stimulation protocol is needed to stimulate multiple muscles.

One superframe consists of several frames, and one frame consists of several slots.

The length of one slot is determined by the wavelength of one stimulation pulse, and the length of the frame is determined by the interval between stimulation pulses. Additionally, the wavelength of the stimulation pulse and stimulation interval vary depending on the muscle.

When the stimulation unit 160 transmits a muscle stimulation signal to stimulate a muscle, one stimulator signal can be assigned to one slot, and therefore only one channel switch among the plurality of electrodes can be turned on. , the ground switch of the other electrode must also be turned on.

More specifically, in order to deliver a muscle stimulation signal to the first muscle location among the first to tenth muscles, the first channel switch must be turned on, and the ground switch must be turned on for the electrodes at the remaining second to tenth muscle locations. It must be set to (ON).

In addition, in order to perform stimulation by transmitting a stimulation signal and at the same time measure electromyography, both the channel switch and the reference switch of the switch to be measured must be turned on.

Here, muscles can be stimulated by simultaneously transmitting multiple muscle stimulation signals equal to the number of slots. Additionally, the degree of muscle activation varies depending on the degree of repetition of frames containing slots of the same stimulator.

Figure 4 is a reference diagram for explaining the structure of a frame for controlling the movement of closing and extending the elbows by contracting the biceps and triceps muscles according to an embodiment of the present invention.

Referring to FIGS. 4a) and 4b), electrodes 1 to 3 are installed on the biceps brachii, electrodes 4 to 5 are installed on the triceps brachii, and electrode 1 turns on the channel 1 switch in slot 1 to contract the biceps brachii. ON), electrode 2 turns on the ground 2 switch only in slot 1 according to the pulse signal from stimulator 1, and electrode 3 turns on the reference 3 switch for electromyography measurement.

Additionally, electrode 4 turns on the channel 4 switch in slot 2 to contract the triceps muscle, and electrode 5 turns on the ground 5 switch only in slot 2 according to the pulse signal from stimulator 4.

Figure 5 is a reference diagram for explaining the structure of a super frame capable of stimulating multiple muscles simultaneously according to an embodiment of the present invention.

Referring to Figure 5, if the length of one stimulation pulse is 400us and the interval between stimulation pulses is 20ms, a total of 50 muscles can be stimulated simultaneously.

In addition, it is possible to transmit muscle stimulation signals that stimulate each muscle simultaneously and measure the electromyogram signal of each muscle at the same time.

Recovery from paralysis in stroke patients usually takes from as short as 2 months to as long as 6 months or more.

In addition, the EMG signal changes during the recovery period, and therefore, an artificial intelligence learning system for identifying the movement intention through the EMG signal of a stroke patient will be described with reference to FIGS. 6 to 8.

Figure 6 is a reference diagram for explaining a device, edge system, server, mobile communication bridge, and mobile communication repeater according to an embodiment of the present invention.

Figure 7 is a reference diagram for explaining the structure of an artificial intelligence learning system according to an embodiment of the present invention.

Figure 8 is a reference diagram for explaining the structure of an artificial intelligence prediction system according to an embodiment of the present invention.

Referring to Figure 6, the artificial intelligence learning system includes a multiple muscle stimulator system, an edge system, an artificial intelligence system server, a mobile communication bridge, and a repeater.

The multi-muscle stimulator system 100 measures EMG signals and stimulates muscles based on the EMG signals, and the edge system 610 includes a sensor data processing module unit that processes data and a sensor data transmission module unit that transmits sensor data. .

The artificial intelligence system server is a server data processing module unit that transmits and receives data from the sensor data processing module unit of the edge system 610 and a sensor data unit that stores data received from the sensor data transmission module unit of the edge system 610. Includes database (DB). In addition, through multiple measurements of electromyography signals, artificial intelligence learning is performed based on muscle stimulation signals transmitted during multiple measurements, and neural network weights can be calculated by performing multiple artificial intelligence learning.

The mobile communication repeater 630 and the mobile communication bridge 640 can be connected to a server via wireless Internet to improve data mobility of the edge system.

Referring to FIG. 7, an artificial intelligence learning system 700 measures the EMG signal of a specific repeated movement of a patient paralyzed by a stroke through a matrix electrode suit, receives the EMG signal, and uses the signal of the specific movement as the correct answer. Through repeated learning, the weights of the multi-layer neural network are calculated, and the cost function is calculated by combining the signals specified as the correct answer in the learning data with the weights.

The multi-muscle stimulator system includes an electrode suit with a mesh structure, and can measure electromyography signals in the same movement through the electrode suit.

Referring to Figure 8, the weights of the multi-layer neural network calculated through the artificial intelligence learning system 700 of Figure 7 are compared to the artificial intelligence prediction system, and real-time electromyography signal data measured from the matrix electrode suit is input to make artificial intelligence prediction. The intended operation can be predicted through the system.

The weights of the multi-layer neural network are updated by inputting real-time electromyography data and predicted values, and the operation of delivering muscle stimulation signals to electrodes attached to designated muscles is controlled according to the predicted movement intention.

That is, the artificial intelligence learning system 700 further includes a prediction system that generates prediction data through learning data learned through the artificial intelligence learning system and the multi-layer neural network weights, and the artificial intelligence prediction system is a measured real-time electromyogram signal. And a control signal may be transmitted to the multi-muscle stimulator system to receive the prediction data and transmit a muscle stimulation signal to at least one of the plurality of electrodes through multi-layer neural network weights.

Figure 9 is a flowchart explaining the sequence of the multi-muscle stimulation method.

Referring to Figure 9, this is performed in the electrode suit unit, and electromyography signal measurement and muscle stimulation signals can be received (S901). It is performed in the amplifier unit and can amplify the measured EMG signal (S902). It is performed in the signal processing unit, and the EMG signal data can be processed (S903). Afterwards, it is performed in the stimulation unit and transmits the muscle stimulation signal, and it is performed in the low-pass filter unit and passes a low-frequency signal (S904). It is performed in the processor unit and controls a plurality of switches connected to the electrode suit unit (S905). A multi-muscle stimulation method can be implemented through the above sequence.

Claims (11)

In a multi-muscle stimulator system that measures electromyographic signals and stimulates muscles based on the measured electromyographic signals,
An electrode suit unit that measures electromyography signals and receives muscle stimulation signals;
An amplifier unit for amplifying the measured EMG signal;
A signal processor for processing the EMG signal data;
A low-pass filter unit that passes low frequency signals;
a processor unit that controls a plurality of switches connected to the electrode suit unit; and
It includes a stimulation unit that transmits the muscle stimulation signal,
The electrode suit unit includes a channel electrode for transmitting the EMG signal to the amplifier unit, a ground electrode that serves as a reference potential for at least one of the muscle stimulation signal and the EMG signal, and a comparison potential when measuring the EMG signal. Includes a reference electrode,
The stimulation unit determines the movement intention through artificial intelligence prediction using electromyography signal data and delivers a muscle stimulation signal that matches the movement intention, and the channel electrode and the ground electrode are changed whenever the location to be stimulated is changed.
A multiple muscle stimulator system featuring:
delete According to paragraph 1,
The electrode suit part,
It has a matrix structure and includes an electrode switch on each electrode to operate by receiving the muscle stimulation signal from the stimulation unit to a specific electrode.
A multiple muscle stimulator system featuring:
According to paragraph 1,
The stimulation unit,
To transmit the first muscle stimulation signal, turn on the channel switch of the first electrode and turn on the ground switch of the remaining electrodes except the first electrode.
A multiple muscle stimulator system featuring:
According to paragraph 1,
The electrode suit part,
Turning on the channel switch of the first electrode and the reference switch of the second electrode
A multiple muscle stimulator system featuring:
According to paragraph 1,
The stimulation unit,
Transmitting the muscle stimulation signal of a higher frequency than the frequency of the electromyography signal to the electrode suit unit.
A multiple muscle stimulator system featuring:
delete delete delete delete delete