Method and system for prediction of fall on basis of EMG signals in stroke patient

The EMG-based fall prediction system quickly and safely assesses stroke patients' fall risk by measuring muscle signals from specific lower limb muscles during exercises, addressing the limitations of conventional methods by providing a quantitative, cost-effective, and risk-free evaluation.

WO2024248484A1PCT designated stage expired Publication Date: 2024-12-05SEOUL NAT UNIV HOSPITAL +1
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
PCT/KR2024/007338
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-05-30
Filing Date
2024-05-29
Publication Date
2024-12-05

AI Technical Summary

Technical Problem

Conventional methods for predicting fall risk in stroke patients are time-consuming, difficult to quantify, and often expose patients to the risk of falling, especially in cases where they have difficulty walking, and require expensive equipment or the patient to be in a standing position.

Method used

A fall prediction method and system using electromyography (EMG) signals from sensors attached to lower limb muscles, such as tensor fascia latae, rectus femoris, tibialis anterior, and gastrocnemius, to measure EMG signals during exercises like maximum voluntary isometric contraction, allowing for quick, quantitative, and safe assessment of fall risk without the patient needing to stand.

Benefits of technology

Enables rapid, cost-effective, and safe determination of fall risk during rehabilitation, reducing the risk of accidents and eliminating the need for expensive equipment, while providing a quantitative assessment regardless of the therapist's proficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure KR2024007338_05122024_PF_FP_ABST
    Figure KR2024007338_05122024_PF_FP_ABST
Patent Text Reader

Abstract

The present invention provides a method and system for prediction of fall on the basis of EMG signals in a stroke patient. The method for prediction of fall on the basis of EMG signals in a stroke patient, according to one embodiment of the present invention, comprises the steps of: attaching an EMG sensor to a first part among lower limb muscles of the patient; measuring the EMG signals of the patient performing an exercise motion for collecting the EMG signals, by means of the EMG sensor attached to the first part; and determining a fall risk of the patient on the basis of the measured EMG signals.
Need to check novelty before this filing date? Find Prior Art

Description

Fall prediction method and system based on electromyography signals of stroke patients

[0001] The present invention relates to a method and system for predicting falls based on electromyography signals of stroke patients.

[0002] Stroke is a leading cause of disability worldwide. The most common symptoms include muscle weakness and contractures, impaired motor control, and decreased functional independence in performing daily activities, quality of life, and activities. Conventional stroke exercise rehabilitation treatments expose patients to the constant risk of falls during exercise rehabilitation, and falls do occur. Therefore, special attention is required for patients at high risk for falls.

[0003] In addition, although it is an important factor in preventing falls for patients to be aware of their own risk of falling, there is a practical difficulty in that therapists cannot check the patient's risk of falling every time during exercise rehabilitation treatment due to limited treatment time.

[0004] Conventional fall risk prediction methods involve therapists directly assessing and scoring a patient's various balance, gait, and activities of daily living. However, this process can be time-consuming, depending on the therapist's skill level, making quantitative assessments difficult, and requiring patients to perform movements separate from the actual treatment, thus shortening the patient's rehabilitation time. Another conventional fall risk prediction method uses expensive balance assessment equipment to assess a patient's fall risk, making it difficult to obtain even in general hospitals. Furthermore, conventional fall risk prediction methods often require patients to stand or walk when assessing their fall risk. Even with safety assistive devices or therapists, measuring fall risk itself exposes patients who have difficulty standing or walking to the risk of falling.

[0005] Therefore, there is a need for an effective fall prediction method and system that can quickly and easily identify a patient's fall risk even during treatment, and that is quantitative, inexpensive, and can be safely measured in bed without requiring the patient to stand.

[0006] The technical task of the present invention is to resolve the problems of the prior art described above. The technical task to be achieved by this embodiment is not limited to the technical task described above, and other technical tasks can be inferred from the following examples.

[0007] As a technical means for achieving the above-described technical task, a method for predicting a fall based on an electromyography signal of a stroke patient according to one embodiment includes the steps of: attaching an electromyography sensor to a first part of lower extremity muscles of the patient; measuring an electromyography signal of the patient performing a movement motion for collecting the electromyography signal through the electromyography sensor attached to the first part; and determining a fall risk of the patient based on the measured electromyography signal.

[0008] The first part may be any one of the tensor fascia lata (TFL), the rectus femoris (RF), the tibialis anterior (TA), and the gastrocnemius (GC), and after the step of determining the fall risk of the patient, the step of providing the determined fall risk of the patient to the patient or a manager may further include, and the step of measuring the electromyographic signal may include the step of measuring the electromyographic signal using a maximal voluntary isometric contraction (MVIC) of the first part.

[0009] In addition, a method for predicting a fall based on an electromyography signal of a stroke patient according to another embodiment may further include the steps of: attaching an inertial sensor to the first part of the patient's lower extremity muscles; measuring an inertial sensor signal of the patient through the inertial sensor attached to the first part; and determining a risk of falling of the patient based on the measured inertial sensor signal.

[0010] In addition, the step of determining the risk of falling of the patient based on the measured electromyography signal may include the step of determining the risk of falling of the patient based on the amplitude of the electromyography signal, the area of ​​the waveform of the electromyography signal, or the ratio of the electromyography signal of the paretic side to the electromyography signal of the non-paretic side.

[0011] A method for predicting a fall based on an electromyography signal of a stroke patient according to another embodiment may include the steps of: attaching electromyography sensors to first and second portions of lower extremity muscles of the patient; measuring the electromyography signal of the patient performing a movement motion for collecting the electromyography signal through the electromyography sensors attached to the first and second portions; and determining the risk of the patient falling based on the measured electromyography signal.

[0012] The first part is the tensor fascia lata (TFL), the second part is the rectus femoris (RF), and after the step of determining the patient's risk of falling, the step of providing the determined risk of falling of the patient to the patient or a manager may be additionally included.

[0013] In addition, the step of attaching the electromyography sensor further includes a step of attaching the electromyography sensor to a third part of the lower extremity muscles of the patient, and the step of measuring the electromyography signal further includes a step of measuring the electromyography signal of the patient performing an exercise motion for collecting the electromyography signal through the electromyography sensor attached to the third part, in which case the third part may be the tibialis anterior (TA) or the gastrocnemius (GC).

[0014] Alternatively, the step of attaching the electromyography sensor may further include a step of attaching the electromyography sensor to a third part and a fourth part among the lower extremity muscles of the patient, and the step of measuring the electromyography signal may further include a step of measuring the electromyography signal of the patient performing an exercise motion for collecting the electromyography signal through the electromyography sensor attached to the third part and the fourth part, and the third part may be the tibialis anterior (TA) muscle and the fourth part may be the gastrocnemius (GC) muscle.

[0015] In addition, the step of measuring the electromyographic signal may include a step of measuring the electromyographic signal using the maximum voluntary isometric contraction (MVIC) of the first part, the second part, the third part, and the fourth part.

[0016] In another embodiment, a fall prediction system based on an electromyography signal of a stroke patient comprises: a memory; and at least one processor connected to the memory, wherein the at least one processor is configured to: measure an electromyography signal of the patient performing a motor movement for collecting the electromyography signal through an electromyography sensor attached to a first portion of lower extremity muscles of the patient; and determine a fall risk of the patient based on an amplitude of the measured electromyography signal.

[0017] The first part is any one of the tensor fascia lata (TFL), the rectus femoris (RF), the tibialis anterior (TA), and the gastrocnemius (GC), and the at least one processor is further configured to provide the determined fall risk of the patient to the patient or a manager, and may be configured to measure an electromyography signal using a maximal voluntary isometric contraction (MVIC) of the first part.

[0018] In addition, the at least one processor may be further configured to: attach an inertial sensor to the first portion of the patient's lower extremity muscles; measure an inertial sensor signal of the patient through the inertial sensor attached to the first portion; and determine a fall risk of the patient based on the measured inertial sensor signal.

[0019] In addition, determining the risk of falling of the patient based on the measured electromyography signal may include determining the risk of falling of the patient based on the amplitude of the electromyography signal, the area of ​​the waveform of the electromyography signal, or the ratio of the electromyography signal of the paretic side to the electromyography signal of the non-paretic side.

[0020] According to another embodiment, a fall prediction system based on an electromyography signal of a stroke patient comprises: a memory; and at least one processor connected to the memory, wherein the at least one processor is configured to: measure the electromyography signal of the patient performing a motor movement for collecting the electromyography signal through electromyography sensors attached to first and second portions of lower extremity muscles of the patient; and determine the risk of falling of the patient based on the amplitude of the measured electromyography signal.

[0021] The first portion is the tensor fascia lata (TFL), the second portion is the rectus femoris (RF), and the at least one processor may be further configured to provide the determined fall risk of the patient to the patient or a manager.

[0022] Additionally, the at least one processor is further configured to measure the electromyography signal of the patient performing an exercise motion to collect the electromyography signal through the electromyography sensor attached to a third part of the lower extremity muscles of the patient, wherein the third part may be the tibialis anterior (TA) or the gastrocnemius (GC).

[0023] Alternatively, the at least one processor is further configured to measure the electromyography signal of the patient performing an exercise motion to collect the electromyography signal through the electromyography sensor attached to a third part and a fourth part of the lower extremity muscles of the patient, wherein the third part may be the tibialis anterior (TA) muscle and the fourth part may be the gastrocnemius (GC) muscle.

[0024] Additionally, the at least one processor may be configured to measure an electromyography signal using a maximal voluntary isometric contraction (MVIC) of the first portion, the second portion, the third portion, and the fourth portion.

[0025] According to another embodiment, a computer-readable storage medium can record a program for performing a method according to the above embodiments.

[0026] The fall prediction method and system of the present invention assesses the risk of falling in a stroke patient while in bed. The present invention can reduce the risk of falling by enabling the assessment of a patient's fall risk without the patient having to leave the bed.

[0027] Specifically, the fall prediction method and system of the present invention utilizes only minimal muscle signals from stroke patients to measure their risk of falling. Therefore, it is low-cost and requires only a simple attachment process, making it easy to operate. Furthermore, the present invention allows for rapid assessment of fall risk, as after attaching the electromyography sensor, the stroke patient only needs to perform a few simple measurement movements while lying in bed.

[0028] In addition, the fall prediction method and system of the present invention provide the effect of being able to quantitatively measure the patient's fall risk regardless of the skill of the manager by determining the patient's fall risk by measuring the patient's electromyography signal.

[0029] That is, according to the present invention, a patient's fall risk can be quickly and easily determined even during treatment, and an effective fall prediction method and system can be provided that is quantitative, low-cost, and can be safely measured in bed without exposing the patient to the risk of falling.

[0030] The effects of the present invention are not limited to the effects described above, and effects not mentioned can be clearly understood by a person having ordinary skill in the art to which the present invention pertains from this specification and the attached drawings.

[0031] FIG. 1 is a diagram illustrating a fall prediction system according to one embodiment.

[0032] Figure 2 is a diagram showing EMG signals acquired from lower extremity muscles during MVIC of a healthy individual and a stroke patient.

[0033] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. The advantages and features of the present invention, and methods for achieving them, will become clear with reference to the embodiments described in detail below together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but can be implemented in various different forms. These embodiments are provided only to ensure that the disclosure of the present invention is complete and to fully inform those skilled in the art of the scope of the invention, and the present invention is defined only by the scope of the claims. Like reference numerals refer to like elements throughout the specification.

[0034] Unless otherwise defined, all terms (including technical and scientific terms) used herein may be used in their common sense to those of ordinary skill in the art to which the present invention pertains. Furthermore, terms defined in commonly used dictionaries are not to be interpreted ideally or excessively unless explicitly and specifically defined otherwise.

[0035] The terminology used herein is for the purpose of describing embodiments and is not intended to limit the present invention. In this specification, singular forms also include plural forms, unless specifically stated otherwise. As used herein, the terms "comprises" and / or "comprising" do not exclude the presence or addition of one or more other components in addition to the components mentioned.

[0036] Unless stated otherwise herein, “connected” or “coupled” may include one element / feature being directly connected or coupled to another element / feature or indirectly connected or coupled through another element / feature, and does not necessarily mean directly mechanically connected or coupled. Accordingly, while the various schematic diagrams depicted in the drawings illustrate exemplary arrangements of elements and components, additional intervening elements, devices, features, or components may be present in actual embodiments (assuming the functionality of the depicted elements is not adversely affected).

[0037] Additionally, in this specification, “transmitting” or “receiving” may include not only directly transmitting or receiving information between a sender and a receiver, but also transmitting or receiving information through another entity, unless otherwise stated.

[0038] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

[0039] FIG. 1 is a block diagram illustrating a fall prediction system according to one embodiment.

[0040] Referring to FIG. 1, a fall prediction system (100) may include a processor (110) and a memory (120), and may receive signals from a patient and provide information to the patient or a manager.

[0041] The fall prediction system (100) illustrated in FIG. 1 only illustrates components related to the present embodiments. Therefore, it will be apparent to those skilled in the art that the fall prediction system (100) may further include other general components in addition to the components illustrated in FIG. 1, and the fall prediction system (100) may be implemented in the form of one or more servers / devices.

[0042] Although not illustrated in FIG. 1, the fall prediction system (100) may operate independently, or may operate in conjunction with a server and / or mobile terminal. In this case, the fall prediction system (100) may provide an interface for triggering each action to the server and / or mobile terminal.

[0043] The processor (110) may be configured to perform each step performed by the fall prediction system (100). Meanwhile, in another embodiment, certain steps may be performed under the intervention of an administrator. Hereinafter, a detailed description will be given of a specific operation method performed by the fall prediction system (100) using the processor (110).

[0044] A fall prediction system (100) can measure an electromyography signal of a stroke patient on a bed. In the fall prediction system (100), since only a simple attachment operation is required when measuring a patient's risk of falling, there is no great difficulty in the administrator operating the system. That is, in order to measure the patient's risk of falling, the administrator can attach a surface electromyography sensor to at least one part of the patient's lower extremity muscles. At this time, the part of the patient's lower extremity muscles to which the electromyography sensor is attached may be at least one of the tensor fascia lata (TFL), the rectus femoris (RF), the tibialis anterior (TA), and the gastrocnemius medialis (GC), but is not limited thereto.

[0045] Of course, attaching as many electromyography sensors as possible to the patient's lower extremity muscles and analyzing the electromyography signals measured from the attached electromyography sensors to measure the patient's fall risk may be the best in terms of accuracy. However, since treatment time is limited, it is important to provide a fall prediction system that accurately, simply, and quickly measures the patient's fall risk. That is, the fall prediction system (100) of the present invention aims to accurately measure the patient's fall risk by using the minimum number of electromyography sensors targeting the minimum number of muscles.

[0046] In order to achieve these technical tasks, it is important to set a combination of lower extremity muscles of a patient to which electromyography sensors are attached, and then to confirm whether the electromyography signals measured from the corresponding combination of lower extremity muscles can serve as an indicator of the patient's risk of falling. In this regard, the fall prediction system (100) of the present invention measures the electromyography signals of the tensor fasciae latae (TFL), rectus femoris (RF), tibialis anterior (TA), and gastrocnemius medialis (GC) among the lower extremity muscles of the patient.

[0047] The reason why the tensor fascia lata (TFL) is selected instead of the gluteal muscles as the target for measuring the EMG signal in the fall prediction system (100) of the present invention is because the tensor fascia lata (TFL) has relatively thin subcutaneous fat tissue and is less likely to be mechanically stimulated while measuring the EMG signal in the supine and side-lying positions. In addition, the tensor fascia lata (TFL) plays a major role in balance control during single-leg standing and walking, and acts as one of the core muscles for pelvic stabilization during single-leg standing. In addition, the reason why the rectus femoris (RF) is selected as the target for measuring the EMG signal in the fall prediction system (100) of the present invention is because the rectus femoris (RF) is one of the quadriceps femoris muscles that plays an essential role in functional movements such as standing, walking, and running, as well as in balance control, as a major extensor of the knee.

[0048] The fall prediction system (100) can attach an electromyography sensor to a first part among the lower extremity muscles of a patient and measure an electromyography signal from the electromyography sensor attached to the first part. At this time, the fall prediction system (100) can measure the electromyography signal using the maximum voluntary isometric contraction (MVIC) of the first part. In this case, the first part can be selected from the tensor fascia lata (TFL), the rectus femoris (RF), the tibialis anterior (TA), and the gastrocnemius medialis (GC). More preferably, the first part can be the tensor fascia lata (TFL).

[0049] As an example, a location where an electromyography sensor is attached to measure an electromyography signal may be, for example, a point two finger widths in front of the greater trochanter in the case of the tensor fascia lata (TFL); a point halfway along a line from the anterior superior iliac spine to the top of the patella in the case of the rectus femoris (RF); a point one-third along a line from the fibular head to the medial malleolus in the case of the tibialis anterior (TA); and a point one-quarter along a line from the medial side of the popliteus cavity to the medial side of the Achilles tendon in the case of the medial side of the gastrocnemius (GC).

[0050] In the fall prediction system (100), the patient performs an exercise motion to collect an electromyography signal. The patient may be educated in advance by an administrator about the exercise motion to be performed by the patient to collect an electromyography signal. The exercise motion to collect an electromyography signal may be at least one of a supine lying position, a side lying position, and a sitting position on a bed, but is not limited thereto, and any exercise motion that the patient can perform on a bed is possible.

[0051] More specifically, during a maximal voluntary isometric contraction (MVIC), a patient can perform specific exercise movements to contract four lower extremity muscles (TFL, RF, TA, GC). As an example, for the tensor fasciae latae (TFL), the patient can lie on his side with the outer leg elevated 30 degrees against resistance toward the floor; for the rectus femoris (RF), the patient can lie supine with the knee elevated 30 degrees against resistance toward the floor; and for the tibialis anterior (TA) and gastrocnemius medialis (GC), the patient can lie on his side with the ankle flexed 90 degrees to the leg against plantar or dorsal resistance.

[0052] As described above, the fall prediction system and method of the present invention can provide the effect of being able to measure the fall risk of stroke patients who have difficulty even sitting up, because the risk of falling can be determined simply by contracting the lower extremity muscles of the stroke patient even when the stroke patient is in a lying position.

[0053] The fall prediction system (100) can store the measured electromyography signal and its amplitude in the memory (120). In addition, the fall prediction system (100) can determine the patient's risk of falling based on the amplitude of the measured electromyography signal. The patient's risk of falling can be classified into levels according to the degree and can be expressed as a score, but is not limited thereto, and can be expressed and / or determined through a numerical value or classification system that can indicate the patient's risk of falling.

[0054] The fall prediction system (100) can provide information on the amplitude of the measured electromyography signal of the patient and / or the risk of the patient falling determined therefrom to the patient or a manager. For example, the fall prediction system (100) can further include an output device such as a display or a speaker, and can provide information to the patient or a manager through these output devices. In this case, the fall prediction system (100) can display information on the amplitude of the measured electromyography signal of the patient and / or the risk of the patient falling determined therefrom on a display or output it audibly through a speaker.

[0055] Another embodiment of a fall prediction system (100) may attach electromyography sensors to two parts of a patient's lower extremity muscles, i.e., a first part and a second part, measure electromyography signals from the electromyography sensors attached to the first part and the second part, and determine the patient's fall risk based on the amplitude of the measured electromyography sensors. In this case, the first part and the second part may be selected from the tensor fascia lata (TFL), the rectus femoris (RF), the tibialis anterior (TA), and the gastrocnemius medialis (GC). More preferably, the first part may be the tensor fascia lata (TFL), and the second part may be the rectus femoris (RF).

[0056] Another embodiment of a fall prediction system (100) may attach electromyographic sensors to three parts of the lower extremity muscles of a patient, i.e., a first part, a second part, and a third part, measure electromyographic signals from the electromyographic sensors attached to the first part, the second part, and the third part, and determine the risk of the patient falling based on the amplitude of the measured electromyographic sensors. In this case, the first part, the second part, and the third part may be selected from the tensor fascia lata (TFL), the rectus femoris (RF), the tibialis anterior (TA), and the gastrocnemius medialis (GC). More preferably, the first part may be the tensor fascia lata (TFL), the second part may be the rectus femoris (RF), and the third part may be the tibialis anterior (TA).

[0057] Another embodiment of a fall prediction system (100) may attach electromyography sensors to four parts of the lower extremity muscles of a patient, namely, a first part, a second part, a third part, and a fourth part, measure electromyography signals from the electromyography sensors attached to the first part, the second part, the third part, and the fourth part, and determine the risk of the patient falling based on the amplitude of the measured electromyography sensors. In this case, the first part, the second part, the third part, and the fourth part may be the tensor fascia lata (TFL), the rectus femoris (RF), the tibialis anterior (TA), and the gastrocnemius medialis (GC), respectively.

[0058] Additionally, the fall prediction system (100) can measure electromyography signals using the maximal voluntary isometric contraction (MVIC) of the patient's lower extremity muscles (TFL, RF, TA, GC).

[0059] According to another embodiment, a fall prediction system based on electromyography (EMG) signals of a stroke patient can additionally utilize inertial sensor signals from an inertial sensor in addition to the EMG signals. Specifically, an inertial sensor is attached to a first portion of the patient's lower extremity muscles, the inertial sensor signal of the patient is measured through the inertial sensor attached to the first portion, and the patient's fall risk can be determined based on the measured inertial sensor signal. By using the inertial sensor signal together with the EMG signal as input data for a fall prediction model, the patient's fall risk can be determined more accurately.

[0060] In another embodiment, a fall prediction system based on an electromyography signal of a stroke patient may determine the patient's risk of falling based on the measured electromyography signal, wherein the patient's risk of falling may be determined based on the amplitude of the electromyography signal, the area of ​​the waveform of the electromyography signal, or the ratio of the electromyography signal on the paretic side to the electromyography signal on the non-paretic side. For example, the patient's risk of falling may be determined based on the maximum value of the amplitude of the electromyography signal, more specifically, the maximum value of the root mean square (RMS) value of the electromyography signal amplitude.

[0061] Furthermore, a fall prediction system based on EMG signals in stroke patients can determine a patient's fall risk based on the area between the EMG signal waveform and the time axis. The EMG signal waveform of the lower extremity muscles tends to have a wider area in healthy individuals than in individuals with paralysis. That is, as shown in Figure 2, as the BBS value decreases, the area between the EMG signal waveform and the time axis decreases.

[0062] In addition, a fall prediction system based on electromyography signals of stroke patients can determine the risk of a patient's fall based on the ratio of the EMG signal of the paretic side to the EMG signal of the non-paretic side. In the case of stroke patients, there may be muscles on the same lower extremity that show symptoms of paralysis and muscles on the non-paretic side. For example, the tensor fasciae latae (TFL), rectus femoris (RF), tibialis anterior (TA), or gastrocnemius (GC) on the right may show symptoms of paralysis, while the tensor fasciae latae (TFL), rectus femoris (RF), tibialis anterior (TA), or gastrocnemius (GC) on the left may not show symptoms of paralysis and be healthy. In this case, the system of the present invention can calculate the ratio of the EMG signal of the paretic side to the EMG signal of the non-paretic side and determine the risk of a patient's fall based on the ratio. For example, the maximum value of the RMS value of the EMG signal amplitude for the right tensor fascia lata (TFL) muscle, which is paralyzed, can be divided by the maximum value of the RMS value of the EMG signal amplitude for the left tensor fascia lata (TFL) muscle, which is non-paralyzed, and the patient's risk of falling can be determined using that value.

[0063] The processor (110) may correspond to a processor (110) provided in various types of computing devices such as a personal computer (PC), a server device, a mobile device, an embedded device, an Internet of Things (IoT) device, etc. For example, the processor (110) may correspond to a processor (110) such as a central processing unit (CPU), a graphics processing unit (GPU), an application processor (AP), a neural processing unit (NPU), etc., but is not limited thereto.

[0064] The processor (110) may be implemented as an array of a plurality of logic gates, or may be implemented as a combination of a general-purpose microprocessor and a memory (120) storing a program that can be executed on the microprocessor. Furthermore, it will be understood by those skilled in the art to which the present embodiment pertains that the processor (110) may be implemented as other types of hardware.

[0065] The memory (120) is hardware that stores various data processed by the processor (110), and can store information about, for example, data received from a patient or user, data transmitted to a patient or user, and data generated by the fall prediction system (100). In addition, the memory (120) can store various programs or applications to be driven by the processor (110).

[0066] The memory (120) may include at least one of volatile memory and nonvolatile memory. Nonvolatile memory includes Read Only Memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable and Programmable ROM (EEPROM), flash memory, Phase-change RAM (PRAM), Magnetic RAM (MRAM), Resistive RAM (RRAM), Ferroelectric RAM (FeRAM), etc. Volatile memory includes Dynamic RAM (DRAM), Static RAM (SRAM), Synchronous DRAM (SDRAM), PRAM, Magnetic RAM (MRAM), Resistive RAM (RRAM), etc. In an embodiment, the memory (120) may be implemented as at least one of a hard disk drive (HDD), a solid state drive (SSD), a compact flash (CF), a secure digital (SD), a micro secure digital (Micro-SD), a mini secure digital (Mini-SD), an extreme digital (xD), or a memory stick.

[0067] Below, a specific embodiment of the fall prediction system of the present invention, which measures a patient's fall risk using a minimum number of electromyography sensors targeting a minimum number of muscles, is described in order to solve the problems of conventional fall prediction systems.

[0068] Conventionally, there are several assessment tools for measuring fall risk, including the Berg Balance Scale (BBS), the Morse Falls Scale, the Chedoke-McMaster Stroke Impairment Inventory, and the Functional Independence Measure (FIM). For example, with the BBS, managers directly assess a patient's fall risk by observing them perform various balance, gait, and activities of daily living. In other words, conventional fall prediction methods expose patients to the risk of falling while their fall risk is being assessed, and the problem with these methods is that it is difficult to quantitatively measure fall risk depending on the manager's skill level.

[0069] In order to solve the problems of the prior art and provide an indicator of a patient's risk of falling that can replace the Berg Balance Scale (BBS), the present invention evaluates a patient's risk of falling based on electromyography signals of the lower extremity muscles of a patient lying on a bed.

[0070] In the present invention, 10 healthy individuals and stroke patients with a Berg Balance Scale (BBS) score of 29 or less, and 10 healthy individuals and stroke patients with a Berg Balance Scale (BBS) score of 29 or greater were recruited, and their demographic, neurological, and functional characteristics were investigated. Table 1 below presents the results.

[0071]

[0072] Figure 2 shows EMG signals acquired from the tensor fascia lata (TFL), rectus femoris (RF), tibialis anterior (TA), and gastrocnemius medialis (GC) during MVIC in 20 healthy individuals and stroke patients with a BBS score of ≥29 or ≤29. Figure 2 shows that the amplitude of the EMG signal in stroke patients with a BBS score of ≥29 is greater than that in stroke patients with a BBS score of ≤29, and that the amplitude of the EMG signal in healthy individuals is greater than that in stroke patients with a BBS score of ≥29. In other words, it was found that the BBS score has a positive correlation with the amplitude of the EMG signal.

[0073] These characteristics can also be seen in Table 2 below.

[0074]

[0075] As shown in Table 2 above, there was a significant difference in the amplitude of EMG signals between stroke patients with a BBS score of 29 or higher and stroke patients with a BBS score of 29 or lower in all TFL, RF, TA, and GC on the paretic side, as well as in each of the TFL, RF, and TA on the non-paretic side. That is, it was found that the amplitude of EMG signals in the lower extremity muscles (TFL, RF, TA, and GC) measured on the paretic side showed a significant positive correlation with the BBS score.

[0076] The present invention proposes a method to replace the Berg Balance Scale (BBS) assessment tool, which exposes patients to fall risk during the fall risk assessment process. Specifically, the present invention measures electromyography (EMG) signals from the lower extremity muscles of a patient lying in bed, enabling a simple, rapid, and quantitative measurement of fall risk at low cost, without the need for exposure to fall risk.

[0077] Furthermore, in order to determine which combination of lower extremity muscles among the four lower extremity muscle groups can accurately represent a patient's fall risk, a machine learning-based prediction model for a total of 15 fall risk factors was developed. The high and low fall risk factors were set as BBS scores greater than or equal to 29 and less than or equal to 29, and the maximum root mean square (RMS) value of the EMG signal amplitude of the lower extremity muscles of 20 healthy individuals and stroke patients was set as an input variable. Table 3 below presents the performance results of the prediction models for the 15 fall risk factors.

[0078]

[0079] Referring to Table 3 above, among the single lower extremity muscle-based fall prediction models, the TFL-based fall prediction model showed the best performance (MCC: 0.626, AUC: 0.917, sensitivity: 0.771, specificity: 0.889), followed by RF, TA, and GC. In addition, among the two lower extremity muscle-based fall prediction models, the TFL and RF-based fall prediction models showed the best performance (MCC: 0.646, AUC: 0.925, sensitivity: 0.800, specificity: 0.891). In other words, it was confirmed that the amplitude of the EMG signal of the TFL and / or RF is an important indicator for assessing the patient's risk of falling.

[0080] Accordingly, the fall prediction system (100) of the present invention can achieve the aforementioned technical task by measuring the electromyography signal of the tensor fascia lata (TFL) among the lower extremity muscles of the patient, and determining the patient's risk of falling based on the amplitude of the measured electromyography signal. In addition, the fall prediction system (100) of the present invention can achieve the aforementioned technical task by measuring the electromyography signals of the tensor fascia lata (TFL) and the rectus femoris (RF) among the lower extremity muscles of the patient, and determining the patient's risk of falling based on the amplitude of the measured electromyography signal. However, the fall prediction system (100) of the present invention may additionally utilize the electromyography signal of the tibialis anterior (TA) and / or the gastrocnemius medialis (GC) in order to determine the patient's risk of falling.

[0081] In this way, the fall prediction system and method of the present invention can measure the electromyography (EMG) signals of the tensor fascia lata (TFL) or the tensor fascia lata (TFL) and the rectus femoris (RF) among numerous lower extremity muscles of a patient, and determine the patient's fall risk based on the amplitude of the measured EMG signals, thereby allowing the patient's fall risk to be quickly and quantitatively measured in a simple manner even during treatment, and can provide the effect of being able to safely measure the patient's fall risk in bed without the patient having to stand up, in addition to being inexpensive. In particular, since the fall prediction system and method of the present invention can determine the fall risk by simply contracting the lower extremity muscles of a stroke patient even when the patient is in a lying position, it can provide the effect of being able to measure the fall risk of a stroke patient who has difficulty even sitting up.

[0082] Meanwhile, the processing sequence described in the method, fall prediction system, and process disclosed herein is provided as an example. Therefore, the order of each step may be adjusted within the scope of the present invention, as needed. Furthermore, the devices and systems disclosed herein may include means capable of performing the functions described herein, and may be implemented as independent devices or systems, or may be linked or integrated with other systems, as needed.

[0083] The operations described with reference to FIGS. 1 and 2 may be performed or implemented in a computer-readable storage medium recording a program. In addition, the commands stored in the computer-readable medium may cause a processor in each device or system to perform methods and processes related to the commands.

[0084] By way of non-limiting example, computer-readable media can include RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other storage medium that can be used to store desired program code means in the form of instructions or data structures and that can be accessed by a general-purpose or special-purpose computer, or a general-purpose or special-purpose processor. Disk and disc, as used herein, include compact discs (CDs), laser discs, optical discs, digital versatile discs (DVDs), floppy disks, and Blu-ray discs, where disks usually reproduce data magnetically, while discs reproduce data optically with a laser. Combinations of the above should also be included within the scope of computer-readable storage media.

[0085] Although embodiments of the present invention have been described with reference to the attached drawings, those skilled in the art will appreciate that the present invention can be implemented in other specific forms without altering the technical concept or essential features thereof. Therefore, the embodiments described above should be understood to be illustrative in all respects and not restrictive.

Claims

1. A fall prediction method based on electromyography signals of stroke patients. A step of attaching an electromyography sensor to a first part of the lower extremity muscles of the patient; A step of measuring the electromyography signal of the patient performing an exercise movement to collect the electromyography signal through the electromyography sensor attached to the first part; and A method for predicting a fall based on electromyography signals of a stroke patient, comprising a step of determining the risk of falling of the patient based on the measured electromyography signals.

2. In paragraph 1, A method for predicting a fall based on an electromyography signal of a stroke patient, wherein the first region is any one of the tensor fascia lata (TFL), rectus femoris (RF), tibialis anterior (TA), and gastrocnemius (GC).

3. In paragraph 1, After the step of determining the risk of falling of the above patient, A method for predicting a fall based on an electromyography signal of a stroke patient, further comprising a step of providing the determined fall risk of the patient to the patient or a manager.

4. In paragraph 1, A method for predicting a fall based on an electromyography signal of a stroke patient, wherein the step of measuring the electromyography signal includes a step of measuring the electromyography signal using a maximum voluntary isometric contraction (MVIC) of the first part.

5. In paragraph 1, A step of attaching an inertial sensor to the first part of the lower extremity muscles of the patient; A step of measuring the patient's inertial sensor signal through the inertial sensor attached to the first part; and A method for predicting a fall based on an electromyography signal of a stroke patient, further comprising a step of determining the risk of the patient falling based on the measured inertial sensor signal.

6. In paragraph 1, A method for predicting a fall based on an electromyography signal of a stroke patient, wherein the step of determining the patient's risk of falling based on the measured electromyography signal includes the step of determining the patient's risk of falling based on the amplitude of the electromyography signal.

7. In paragraph 1, A method for predicting a fall based on an electromyography signal of a stroke patient, wherein the step of determining the risk of falling of the patient based on the measured electromyography signal includes the step of determining the risk of falling of the patient based on the area of ​​the waveform of the electromyography signal.

8. In paragraph 1, A method for predicting a fall based on electromyography signals of a stroke patient, wherein the step of determining the risk of falling of the patient based on the measured electromyography signals includes the step of determining the risk of falling of the patient based on the ratio of the electromyography signal of the paralyzed side to the electromyography signal of the non-paralyzed side.

9. A fall prediction method based on electromyography signals of stroke patients. A step of attaching electromyography sensors to the first and second parts of the lower extremity muscles of the patient; A step of measuring the electromyography signal of the patient performing an exercise motion to collect the electromyography signal through the electromyography sensor attached to the first part and the second part; and A method for predicting a fall based on electromyography signals of a stroke patient, comprising a step of determining the risk of falling of the patient based on the measured electromyography signals.

10. In paragraph 9, A method for predicting a fall based on electromyography signals of a stroke patient, wherein the first part is the tensor fascia lata (TFL) and the second part is the rectus femoris (RF).

11. In paragraph 9, After the step of determining the risk of falling of the above patient, A method for predicting a fall based on an electromyography signal of a stroke patient, further comprising a step of providing the determined fall risk of the patient to the patient or a manager.

12. In paragraph 9, The step of attaching the above electromyography sensor is: It further includes a step of attaching an electromyography sensor to a third part of the lower extremity muscles of the patient. The step of measuring the above electromyography signal is: A method for predicting a fall based on an electromyography signal of a stroke patient, further comprising a step of measuring the electromyography signal of the patient performing an exercise motion for collecting the electromyography signal through the electromyography sensor attached to the third part.

13. In paragraph 12, A method for predicting falls based on electromyography signals of stroke patients, wherein the third part is the tibialis anterior (TA) or gastrocnemius (GC).

14. In paragraph 9, The step of attaching the above electromyography sensor is: It further includes a step of attaching electromyography sensors to the third and fourth parts of the lower extremity muscles of the patient. The step of measuring the above electromyography signal is: A method for predicting a fall based on an electromyography signal of a stroke patient, further comprising a step of measuring the electromyography signal of the patient performing a movement motion for collecting the electromyography signal through the electromyography sensor attached to the third and fourth parts.

15. In paragraph 14, A method for predicting a fall based on electromyography signals of a stroke patient, wherein the third part is the tibialis anterior (TA) muscle and the fourth part is the gastrocnemius (GC) muscle.

16. In paragraph 15, A method for predicting a fall based on an electromyography signal of a stroke patient, wherein the step of measuring the electromyography signal includes a step of measuring the electromyography signal using the maximum voluntary isometric contraction (MVIC) of the first part, the second part, the third part, and the fourth part.

17. In paragraph 9, A step of attaching an inertial sensor to the first part and the second part among the lower extremity muscles of the patient; A step of measuring an inertial sensor signal of the patient through the inertial sensor attached to the first part and the second part; and A method for predicting a fall based on an electromyography signal of a stroke patient, further comprising a step of determining the risk of the patient falling based on the measured inertial sensor signal.

18. In paragraph 9, A method for predicting a fall based on an electromyography signal of a stroke patient, wherein the step of determining the patient's risk of falling based on the measured electromyography signal includes the step of determining the patient's risk of falling based on the amplitude of the electromyography signal.

19. In paragraph 9, A method for predicting a fall based on an electromyography signal of a stroke patient, wherein the step of determining the risk of falling of the patient based on the measured electromyography signal includes the step of determining the risk of falling of the patient based on the area of ​​the waveform of the electromyography signal.

20. In paragraph 9, A method for predicting a fall based on electromyography signals of a stroke patient, wherein the step of determining the risk of falling of the patient based on the measured electromyography signals includes the step of determining the risk of falling of the patient based on the ratio of the electromyography signal of the paralyzed side to the electromyography signal of the non-paralyzed side.

21. A fall prediction system based on electromyography signals of stroke patients. memory; and comprising at least one processor connected to the memory; At least one processor of the above: Measuring the electromyography signal of the patient performing an exercise movement to collect the electromyography signal through an electromyography sensor attached to a first part of the lower extremity muscles of the patient; and A fall prediction system based on electromyography signals of a stroke patient, configured to determine the risk of falling of the patient based on the measured electromyography signals.

22. In paragraph 21, A fall prediction system based on electromyography signals of a stroke patient, wherein the first region is any one of the tensor fascia lata (TFL), rectus femoris (RF), tibialis anterior (TA), and gastrocnemius (GC).

23. In paragraph 21, At least one processor, A fall prediction system based on electromyography signals of a stroke patient, further configured to provide the determined fall risk of the patient to the patient or a manager.

24. In paragraph 21, At least one processor, A fall prediction system based on electromyography signals of a stroke patient, configured to measure electromyography signals using the maximum voluntary isometric contraction (MVIC) of the first part.

25. In paragraph 21, At least one processor of the above: Attaching an inertial sensor to the first part of the lower extremity muscles of the patient; By means of the inertial sensor attached to the first part, the inertial sensor signal of the patient is measured; and A fall prediction system based on electromyography signals of a stroke patient, further configured to determine the risk of falling of the patient based on the measured inertial sensor signals.

26. In paragraph 21, A fall prediction system based on electromyography signals of a stroke patient, wherein determining the patient's risk of falling based on the measured electromyography signals includes determining the patient's risk of falling based on the amplitude of the electromyography signals.

27. In paragraph 21, A fall prediction system based on electromyography signals of a stroke patient, wherein determining the patient's risk of falling based on the measured electromyography signals includes determining the patient's risk of falling based on the area of ​​the waveform of the electromyography signals.

28. In paragraph 21, A fall prediction system based on electromyography signals of a stroke patient, wherein determining the patient's fall risk based on the measured electromyography signals includes determining the patient's fall risk based on a ratio of the electromyography signal of the paretic side to the electromyography signal of the non-paretic side.

29. A fall prediction system based on electromyography signals of stroke patients, memory; and comprising at least one processor connected to the memory; At least one processor of the above: Measuring the electromyography signal of the patient performing an exercise motion to collect the electromyography signal through electromyography sensors attached to the first and second parts of the lower extremity muscles of the patient; and A fall prediction system based on electromyography signals of a stroke patient, configured to determine the risk of falling of the patient based on the measured electromyography signals.

30. In paragraph 29, A fall prediction system based on electromyography signals of a stroke patient, wherein the first part is the tensor fascia lata (TFL) and the second part is the rectus femoris (RF).

31. In paragraph 29, At least one processor, A fall prediction system based on electromyography signals of a stroke patient, further configured to provide the determined fall risk of the patient to the patient or a manager.

32. In paragraph 29, At least one processor, A fall prediction system based on electromyography signals of a stroke patient, further configured to measure the electromyography signals of the patient performing a motor movement for collecting the electromyography signals through the electromyography sensor attached to the third part of the lower extremity muscles of the patient.

33. In paragraph 32, The third part is a fall prediction system based on electromyography signals of stroke patients, which is the tibialis anterior (TA) or gastrocnemius (GC).

34. In paragraph 29, At least one processor, A fall prediction system based on electromyography signals of a stroke patient, further configured to measure the electromyography signals of the patient performing a motor movement for collecting the electromyography signals through the electromyography sensors attached to the third and fourth parts of the lower extremity muscles of the patient.

35. In paragraph 34, A method for predicting a fall based on electromyography signals of a stroke patient, wherein the third part is the tibialis anterior (TA) muscle and the fourth part is the gastrocnemius (GC) muscle.

36. In paragraph 35, At least one processor, A fall prediction system based on electromyography signals of a stroke patient, configured to measure electromyography signals using maximal voluntary isometric contractions (MVICs) of the first, second, third and fourth regions.

37. In paragraph 29, At least one processor of the above: Attaching inertial sensors to the first and second portions of the lower extremity muscles of the patient; Measuring the patient's inertial sensor signal through the inertial sensor attached to the first part and the second part; and A fall prediction system based on electromyography signals of a stroke patient, further configured to determine the risk of falling of the patient based on the measured inertial sensor signals.

38. In paragraph 29, A fall prediction system based on electromyography signals of a stroke patient, wherein determining the patient's risk of falling based on the measured electromyography signals includes determining the patient's risk of falling based on the amplitude of the electromyography signals.

39. In paragraph 29, A fall prediction system based on electromyography signals of a stroke patient, wherein determining the patient's risk of falling based on the measured electromyography signals includes determining the patient's risk of falling based on the area of ​​the waveform of the electromyography signals.

40. In paragraph 29, A fall prediction system based on electromyography signals of a stroke patient, wherein determining the patient's fall risk based on the measured electromyography signals includes determining the patient's fall risk based on a ratio of the electromyography signal of the paretic side to the electromyography signal of the non-paretic side.

41. A computer-readable storage medium recording a program for performing the method of any one of paragraphs 1 and 20.

Citation Information

Patent Citations

  • Method for Assessing Fall Risk and User Terminal therefor

    KR1020180067233A

  • Electronic device including flexible display

    KR1020230006372A

  • Server providing product sales service and operation method thereof

    KR102314730B1

  • On-off grid hybrid portable ess

    KR102652470B1