WO2023106588A1

WO2023106588A1 - Body-wearable pain management system based on bio-signal collection and analysis

Info

Publication number: WO2023106588A1
Application number: PCT/KR2022/015170
Authority: WO
Inventors: 이성원; 장상현; 조재현
Original assignee: (주)웰스케어
Priority date: 2021-12-06
Filing date: 2022-10-07
Publication date: 2023-06-15
Also published as: KR20230085020A

Abstract

A body-wearable pain management system based on bio-signal collection and analysis is disclosed. According to one aspect of the present invention, provided is a body-wearable pain management system based on bio-signal collection and analysis, comprising: a pain treatment unit performing a procedure for pain relief or pain treatment; and a pain measurement unit, which collects and analyzes changes in bio-signals of a user using the pain treatment unit, so as to derive the degree of pain or changes in bio-signals of the user before and after a procedure, thereby providing same to the user, wherein the pain measurement unit includes a bio-signal collection unit, which collects bio-signals, and an analysis unit, which analyzes the bio-signals collected by means of the bio-signal collection unit so as to derive the degree of pain or changes in bio-signals before and after the procedure, and the analysis unit includes: a pre-processing unit for pre-processing the bio-signal; a determination unit for determining whether the bio-signal pre-processed by the pre-processing unit is a normal signal; an index derivation unit for deriving a set index from the bio-signal determined to be a normal signal by means of the determination unit; and a comparative analysis unit, which compares the set index derived by the index derivation unit with reference data and analyzes same so as to derive the degree of pain or changes in bio-signals before and after the procedure.

Description

Body wearable pain management system based on vital signal collection and analysis

The present invention relates to a body-wearable pain management system based on bio-signal collection and analysis.

[National research and development project supporting this invention]

[Assignment identification number] 1425153809

[Assignment number] S2962695

[Name of Department] Ministry of Small and Medium Venture Business

[Task management (professional) organization name] Small and Medium Business Technology Information Promotion Agency

[Research Project Name] Strategic Entrepreneurship Project

[Research Project Title] Development of a wearable device that can indicate pain relief using complex energy (cold laser + ultrasound) and pain improvement using photoplethysmography

[Contribution rate] 1/1

[Name of project performing organization] Wellscare Co., Ltd.

[Research period] 2020.12.31 ~ 2022.12.30

Ultrasonic waves are sound waves with a frequency exceeding the audible limit, and are used for non-invasive treatment of pain areas in orthopedics, pain medicine, and the like.

In addition, low-level laser (LLL) has various effects such as pain relief, blood flow improvement, blood flow enhancement, skin regeneration, cell activation, lipolysis, and cell stimulation depending on the wavelength range through the results of various clinical trials. It is known that there is

On the other hand, the photoplethysmogram (PPG) is a signal that measures the change in blood volume in blood vessels according to the heartbeat using light absorption, reflection, and scattering. Phenomenon and biological response can be grasped.

Prior Art Document: (Patent Document) Republic of Korea Patent Publication No. 10-2021-0133424 (published on November 8, 2021)

An object of the present invention is to provide a body wearable pain management system based on biosignal collection and analysis, which enables personalized pain assessment through collection and analysis of individual user's biosignals and pain treatment accordingly.

According to one aspect of the present invention, a pain treatment unit that performs a procedure for pain relief or pain treatment, and a user's physiological signal change using the pain treatment unit is collected and analyzed to determine the degree of pain of the user or the bio signal before and after the procedure. It includes a pain measurement unit that derives changes and provides them to the user, and the pain measurement unit analyzes the bio-signals collected by the bio-signal collection unit and the bio-signal collection unit to collect bio-signals and measures the degree of pain or bio-signals before and after the procedure. An analysis unit that derives a change is included, and the analysis unit includes a preprocessing unit that preprocesses the biosignal, a determination unit that determines whether the biosignal preprocessed by the preprocessing unit is a good signal, and a biosignal determined as a good signal by the determination unit. Bio-signal collection and analysis, including an indicator derivation unit that derives a set indicator from the indicator derivation unit, and a comparison analysis unit that compares and analyzes the set indicator derived by the indicator derivation unit with reference data to derive the degree of pain or bio-signal change before and after the procedure. A body wearable pain management system based on the present invention is provided.

The determination unit may determine whether the biosignal is a good signal by iteratively learning the biosignal through a predetermined machine learning-based signal quality evaluation model so as to improve the biosignal discrimination accuracy.

The determination unit includes a matrix transformation unit for matrix transformation of the biosignal image, a layer generation unit for generating a layer by performing a convolution operation on the matrix converted by the matrix transformation unit, and a normalization unit for normalizing the layer generated by the layer generation unit. , An activation unit for nonlinearly activating the layer normalized by the first normalizer, a subsampling unit for subsampling the layer nonlinearly activated by the activation unit, and an integrated connection of the layers subsampled by the subsampling unit to generate an integration matrix. Including an integration unit that normalizes and connects the integration matrix generated by the integration unit, and a determination unit that determines whether the biosignal is a good signal according to a probability derived by applying a probability function to the integration matrix normalized and connected by the connection unit. can do.

The pain measurement unit is provided to enable bi-directional communication with the pain treatment unit, and can provide the user with the user's pain level or changes in biosignals before and after the procedure in-line.

The biosignal includes a photoplethysmogram, and the preprocessor includes a first processing unit that removes signal noise by sequentially applying a nonlinear filter and a linear filter to the photoplethysmogram collected by the biosignal collection unit, and the signal by the first processing unit. A second processing unit may be included to remove and interpolate ectopic beats from the noise-removed photoplethysmogram.

The biosignal includes a photoplethysmogram, and the preprocessing unit includes a first processing unit that removes signal noise by sequentially applying a nonlinear filter and a linear filter to the photoplethysmogram collected by the biosignal collection unit, wherein the first processing unit includes: First, a nonlinear filter is applied to the optical pulse wave to remove baseline fluctuation noise, and a second, linear filter is applied to obtain a signal of a set frequency band excluding the low frequency band including the low frequency component, the respiratory component, and the motion component. can pass

The biosignal includes a photoplethysmogram, and the preprocessor includes a first processing unit that removes signal noise by sequentially applying a nonlinear filter and a linear filter to the photoplethysmogram collected by the biosignal collection unit, and the signal by the first processing unit. A second processing unit may be included to extract poles from the noise-removed photoplethysmogram, detect pole intervals, determine intervals of ectopic beats through the pole intervals, remove ectopic beats, and perform interpolation.

According to the body-wearable pain management system based on bio-signal collection and analysis according to the present invention, personalized and accurate pain evaluation based on user's personal bio-signal collection and analysis is possible.

According to the wearable pain management system based on biosignal collection and analysis according to the present invention, the accuracy of pain evaluation can be improved through repetitive learning of biosignals through an artificial intelligence model.

According to the body-wearable pain management system based on bio-signal collection and analysis according to the present invention, it is possible to evaluate, diagnose, and manage self-pain in a non-face-to-face manner in daily life without a user visiting a hospital.

According to the body-wearable pain management system based on bio-signal collection and analysis according to the present invention, it is possible to provide customized pain treatment according to the result of a user's self-diagnosis of pain.

According to the body-wearable pain management system based on bio-signal collection and analysis according to the present invention, pain diagnosis and pain treatment can be performed in-line.

1 is a schematic diagram showing a body-wearable pain management system based on bio-signal collection and analysis according to an embodiment of the present invention.

2 is a diagram showing a body-wearable pain management system based on bio-signal collection and analysis according to an embodiment of the present invention.

3 is a view showing the normalized setting indicators of the body-wearable pain management system based on bio-signal collection and analysis according to an embodiment of the present invention.

4 is a use state diagram showing a pain treatment unit of a body-wearable pain management system based on bio-signal collection and analysis according to an embodiment of the present invention.

5 is a perspective view illustrating a pain treatment unit of a body-wearable pain management system based on bio-signal collection and analysis according to an embodiment of the present invention.

6 is a plan view and a bottom view illustrating a pain treatment unit of a body-wearable pain management system based on bio-signal collection and analysis according to an embodiment of the present invention.

7 is a cross-sectional view taken along lines AA', (b) lines BB', and (c) lines CC' of FIG. 5;

8 is a diagram of a clinical experiment for evaluating the therapeutic effect of a low-power laser on pain in a pain treatment unit of a body-wearable pain management system based on bio-signal collection and analysis according to an embodiment of the present invention.

9 is a flowchart illustrating a body-wearable pain management process based on bio-signal collection and analysis according to an embodiment of the present invention.

10 is a flowchart illustrating a signal evaluation process of a signal quality evaluation model of a body-wearable pain management system based on bio-signal collection and analysis according to an embodiment of the present invention.

11 is a schematic diagram illustrating a process of pole detection and pole correction through adaptive threshold search according to an embodiment of the present invention.

12 is a schematic diagram illustrating a signal quality evaluation process of a signal quality evaluation model according to an embodiment of the present invention.

13 is a diagram showing the results of signal quality evaluation according to an embodiment of the present invention.

14 is a diagram showing a comparison result of the performance of a signal quality evaluation model according to an embodiment of the present invention and an existing study.

15 is a schematic diagram showing a photoplethysmogram-based setting index according to an embodiment of the present invention.

16 is a schematic diagram showing normalization of an optical pulse wave-based setting index according to an embodiment of the present invention.

17 is a schematic diagram showing a differentiated common pulse wave-based setting index according to an embodiment of the present invention.

Since the present invention can apply various transformations and can have various embodiments, specific embodiments will be illustrated in the drawings and described in detail in the detailed description. However, it should be understood that this is not intended to limit the present invention to specific embodiments, and includes all transformations, equivalents, and substitutes included in the spirit and 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 may be used to describe various components, but the components should not be limited by the terms. These terms are only used for the purpose of distinguishing one component from another.

Terms used in this application are only used to describe specific embodiments, and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this application, the terms "include" or "have" are intended to designate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, but one or more other features It should be understood that the presence or addition of numbers, steps, operations, components, parts, or combinations thereof is not precluded.

Hereinafter, the body-wearable pain management system 300 based on bio-signal collection and analysis according to the present invention will be described in detail with reference to the accompanying drawings. In the description with reference to the accompanying drawings, the same or corresponding components are given the same reference numerals, and overlapping descriptions thereof will be omitted.

Hereinafter, a body-wearable pain management system 300 based on bio-signal collection and analysis according to an embodiment of the present invention will be described. 1 to 3, the body-wearable pain management system 300 based on bio-signal collection and analysis according to an embodiment of the present invention includes a pain treatment unit 100 that performs a procedure for pain relief or pain treatment. and a pain measuring unit 200 that collects and analyzes changes in the user's biosignals using the pain treatment unit 100, derives the degree of pain of the user or changes in biosignals before and after the procedure, and provides the result to the user. there is.

The pain treatment unit 100 may perform procedures for pain relief or pain treatment. Specifically, the pain treatment unit 100 may treat musculoskeletal pain diseases such as herniated disc pain, carpal tunnel syndrome, knee arthritis, knee joint pain, A variety of procedures can be performed to treat epicondylitis pain, turtleneck syndrome, and pain in various joint areas. As an example, the pain treatment unit 110 may perform pain treatment based on Low Level Laser Therapy (LLLT). Low-power laser therapy activates various signal pathways by absorbing photons into the mitochondrial chromophore of skin cells and consequently increasing electron transport, adenosine 3-phosphate (ATP), nitric oxide release, blood flow improvement, and active oxygen. Pain can be cured by creating As another example, the pain treatment unit 110 may perform the above-described treatment for pain using technologies such as ultrasound, radiofrequency, ultrashort waves, extracorporeal shock waves, and thermal therapy.

The pain measurer 200 may collect and analyze changes in biosignals of the user using the pain treatment unit 100 to derive the degree of pain of the user or changes in biosignals before and after the procedure, and provide the result to the user. In this case, the biosignal may include a photoplethysmogram wave, but is not limited thereto, and description will be made mainly on the assumption that the biosignal is a photoplethysmogram wave.

The pain measurer 200 analyzes the bio-signal collection unit 210 that collects bio-signals and the bio-signals collected by the bio-signal collector 210 to derive the degree of pain or changes in bio-signals before and after the procedure. An analysis unit 220 may be included.

Here, the pain measurement unit 200 may be understood as a kind of server, and in this case, the biosignal collection unit 210 may be understood as a kind of database included in the server.

The pain measuring unit 200 is provided to enable bi-directional communication with the pain treatment unit 100, and can provide the user with the user's pain level or changes in biosignals before and after the procedure inline.

The analyzer 220 includes a preprocessor 230 that preprocesses the biosignal, a discriminator 240 that determines whether the biosignal preprocessed by the preprocessor 230 is a good signal, and the discriminator 240. An index derivation unit 250 that derives a set index from the biosignal determined as a good signal, and the index derivation unit 250 compares and analyzes the set index derived by the reference data to determine the degree of pain or changes in the biosignal before and after the procedure. It may include a comparison and analysis unit 260 to derive, and will be described in detail below for each configuration.

The pre-processor 230 includes a first processor 232 that removes signal noise by sequentially applying a nonlinear filter and a linear filter to the optical pulse wave collected by the biosignal collector 210, and the first processor 232 It may include a second processing unit 234 for removing and interpolating ectopic beats from the optical pulse wave from which signal noise has been removed by .

The first processing unit 232 firstly applies a nonlinear filter to the photoplethysmogram to remove baseline fluctuation noise, and secondarily applies a linear filter to a low-frequency band including a low-frequency component, a respiratory component, and a motion component. It is possible to pass the signal of the set frequency band except for .

Here, the nonlinear filter as a first-order filter can remove baseline wander including a respiration signal from the optical pulse wave, and is an infinite impulse response filter (IIR filter), specifically y [n]-0.995y[n-1]=0.9975x[n]-0.9975x[n-1].

In addition, a linear filter as a secondary filter can minimize signal delay and phase distortion, and as a finite impulse response filter (FIR filter), a band pass filter for a frequency band of 0.5 to 10 Hz; BPF), and accordingly, a low frequency band including a low frequency component (less than 0.5 Hz), a respiratory component (0.15 to 0.4 Hz), and a motion component (less than 0.1 Hz) can be filtered.

The second processing unit 234 extracts poles from the photoplethysmogram wave from which signal noise has been removed by the first processing unit 232, detects the pole intervals, determines the intervals of ectopic beats through the pole intervals, and removes the ectopic beats. and can interpolate.

Meanwhile, the second processing unit 234 may perform data normalization prior to peak extraction, and specifically, by 0-1 Normalization, (value-min(value))/(max(value)-min( value)) can be followed.

Specifically, as shown in FIG. 11, the second processing unit 234 may detect a peak through an adaptive threshold search process, and specifically, through the slope of an empirically selected waveform, the next The waveform is tracked until a pulsation is found, and when the next pulsation is detected, a pole is extracted by moving to a maximum threshold, and then the pole can be extracted by repeatedly performing the above-described process.

In addition, the second processing unit 234 may perform correction of the extracted pole point, and specifically, the correction of the pole point may be performed through local maxima / minima. At this time, when there is a reflected wave, the corresponding part is excluded ( may have a predetermined refractory period (0.5 to 1 sec, preferably 0.6 sec) after the pole to be ignored.

The second processing unit 234 may detect a peak to peak interval (PPI) or a heart rate interval (R to R interval) according to peak locs[n]-peak locs[n-1], and detect The ectopic beat-to-beat interval can be determined according to 0.675*PPI[n-1]<PPI[n]<1.245*PPI[n-1] through the polar interval.

Thereafter, the second processing unit 234 may remove the ectopic beat and interpolate the position of the removed ectopic beat through linear, nearest neighbor, cubic spline, shape preservation, and the like.

The determining unit 240 compares good signal indicators including ectopic beats, heart rate, continuous heart rate difference, and pole amplitude of the preprocessed photoplethysmogram wave with set parameters for the good signal indicator to determine whether the photoplethysmogram wave is a good or bad signal. It can be determined whether

More specifically, in the case of ectopic beating confirmation, the determination unit 240 aligns the pole interval values, calculates the length of the 25% (q1)-75% (q3) section in the distribution of corresponding values, and calculates the heart rate. If there is an interval between poles whose interval (RRI) is less than q1-1.5*iqr or greater than q3+1.5*iqr, it can be determined as a bad signal.

In addition, in the case of the heart rate, the determination unit 240 may determine that the signal is defective when the measured bpm based on the pole point is greater than max(200) or less than min(30).

In the case of the continuous heart rate difference, the determination unit 240 may determine that the difference between the bpm measured at the systolic extreme point and the bpm measured at the pulsation start point exceeds 10 as a bad signal.

In the case of pole amplitude, the determination unit 240 may determine that the signal is defective when the differential difference between the amplitudes of the systolic poles is 0.75 or more or the differential difference between the pulsation start points is 0.75 or more.

Meanwhile, the determination unit 240 may determine whether the biosignal is a good signal by iteratively learning the biosignal through a predetermined machine learning-based signal quality evaluation model so as to improve the biosignal discrimination accuracy.

In other words, the determination unit 240 learns the data of the photoplethysmogram, which is determined as a good or bad signal, and analyzes the preprocessed one-dimensional signal image of the photoplethysmogram based on the learning result to determine whether the photoplethysmogram is a good or bad signal. It is possible to use a signal quality evaluation model capable of discriminating whether or not it is recognized, and accordingly, the processing speed of photoplethysmogram data can be increased, thereby further improving the real-time processing function of data.

More specifically, the signal quality assessment (SQR) model was obtained from photoplethysmograms (from the left finger to measurement) based on a validation data set (excellent: 14,644, acceptable: 31,413, unfit: 3,196, abnormal: 308) manually evaluated by an experienced researcher in Learning may be performed, and evaluation accuracy may be further improved by performing learning on good or bad signals of the photoplethysmogram determined according to the above-described method.

The signal quality evaluation model is a machine learning model, and may specifically be a convolution neural network (CNN), and in this case, the used architecture may be activation function=ReLU, dropout rate=0.5.

The signal quality evaluation model includes a matrix conversion unit 242 that converts an image of a biosignal into a matrix, a layer generation unit 243 that generates a layer by performing a convolution operation on the matrix converted by the matrix conversion unit 242, and a layer A normalizer 244 that normalizes the layer generated by the generation unit 243, an activation unit 245 that nonlinearly activates the layer normalized by the first normalizer 244, and a nonlinear activation by the activation unit 245 A subsampling unit 246 subsampling the subsampled layer, an integrating unit 247 generating an integration matrix by integrating and connecting the layers subsampled by the subsampling unit 246, and the integration generated by the integrating unit 247. A connection unit 248 for normalizing and connecting the matrices, and a determination unit 249 for determining whether the biosignal is a good signal according to a probability derived by applying a probability function to the integration matrix normalized and connected by the connection unit 248. can

Referring to FIG. 10 and FIG. 12, the signal quality evaluation process of the signal quality evaluation model is described in detail.

1) The matrix conversion unit 242 converts the preprocessed signal image of the optical pulse wave into a matrix, and 2) the layer generation unit 243 performs convolution operation on the matrices to generate (convolution) layers. 3) After batch normalization in the normalization unit 244, 4) nonlinear activation (rectified linear unit activation; ReLU) in the activation unit 245, and 5) non-linear activation in the subsampling unit 246 After subsampling (max pooling) the activated layer (at this time, steps 2) to 5) can be repeated multiple times, preferably twice, and 6) merged in the integration unit 247. An integration matrix is generated by connection (dense; connection), 7) normalization connection (dropout, dropout rate=0.5) is performed on the integration matrix in the connection unit 248 (step 7), and then step 6) can be additionally performed), 8) Depending on the probability derived by applying the regression probability function (softmax) in the decision unit 249, it is possible to determine whether the photoplethysmogram is a good or bad signal.

As a result of the signal quality evaluation test through this signal quality evaluation model, it was confirmed that the accuracy increased as the learning process for the training data set and the verification data set was repeated (epoch increase).

Specifically, as shown in FIG. 13, bad and good signals were discriminated with high accuracy of maximum AUC (area under curve) = 0.998 in the training data set and maximum AUC = 0.994 in the test data set, On the test data set, accuracy (probability of hitting both good and bad signals)=0.975, sensitivity (probability of hitting a bad signal between good and bad signals)=0.964, specificity (probability of hitting a good signal between good and bad signals)=0.987, PPV (positive predictivity value; positive predictive value) = 0.848, showing high prediction accuracy.

In addition, it was confirmed that such a signal quality evaluation model exhibits improved performance compared to previous studies, as shown in FIG. 14 .

The indicator derivation unit may derive a set indicator from the biosignal determined as a good signal by the determination unit.

In this case, the setting index may include an index related to an area, an index related to a length, an index related to a time, an index related to an amplitude, and an index related to a slope of the photoplethysmogram waveform.

The set index is derived in relation to the waveform shape of the photoplethysmogram wave, as shown in FIG. 15, based on the fact that the blood volume change can be estimated through the photoplethysmogram waveform and the autonomic nerve activity can be predicted accordingly. Detailed indicators organized in the table of FIG. 3 may be included.

In relation to these setting indicators, a detailed look is as follows.

Photoplethysmography has an arbitrary unit that is difficult to compare relative to due to the influence of various factors such as skin color, anatomical factor, ambient light, sensor sensitivity, and measurement environment. , the area-related index can be normalized (spatial/temporal normalization) by dividing the entire waveform area or amplitude, and the length-related index by PPI or calculating a ratio.

In addition, as shown in FIG. 17 , the set index may include a time-related index and an amplitude-related index derived by calculating a time interval and an amplitude interval having a high expected pain correlation in the first-differentiated photoplethysmogram waveform, and may also include a beat-to-beat interval. Or, it may include non-dimensionalized and normalized indicators with amplitude.

The comparison and analysis unit may compare and analyze the set indicator derived by the indicator derivation unit with reference data to derive the degree of pain or change in biosignals before and after the procedure.

In detail, the reference data may include a set index value derived from the photoplethysmogram before the user's procedure or data serving as an index or standard of the degree of pain.

Subsequently, with reference to FIGS. 4 to 7 , the pain treatment unit 100 will be described in detail.

The pain treatment unit 100 includes a body portion 110 having a plurality of penetration holes 112 to allow light transmission, a battery 120 disposed inside the body portion 110 to supply electrical energy, and a battery A circuit board 130 electrically connected to the body 110 and disposed inside the body 110, and a low-power laser that is electrically connected to the circuit board 130 and irradiates a low-power laser to the treatment site through the penetration hole 112. The light irradiation unit 140, the ultrasonic transmission unit 150 electrically connected to the circuit board 130 to deliver ultrasonic waves to the treatment site, and the circuit board 130 electrically connected to the optical volume for pain analysis of the treatment site A life signal detection sensor 160 for detecting a pulse wave may be included, and each component will be described in detail below.

The body part 110 may have a plurality of transmission holes 112 to allow light transmission.

Here, the penetration hole 112 may serve as a passage through which the low-power light irradiation unit 140 disposed inside the body portion 110 irradiates a low-power laser beam to the treatment site, as will be described later.

An adjustment button 114 may be disposed on the body 110 to allow a user to adjust the degree of ultrasound transmission and/or low-power laser irradiation.

The control button 114 is electrically connected to a circuit board 130 to be described later, and may be installed on the body 110 in the form of a physical button or provided on the body 110 in a touch recognition method.

Meanwhile, a band part 116 that can be worn on the body by being combined with the body part 110 may be further included.

A space capable of coupling the body portion 110 may be provided in the band portion 116, and the band portion 116 may have a strap shape as in a wrist watch, as well as an armband or belt shape so as to be coupled to a thicker body part. there is.

The battery 120 may be disposed inside the body 110 to supply electrical energy.

The battery 120 may be rechargeable, and in this case, a charging terminal 118 capable of connecting a cable for externally charging the battery 120 may be disposed in the body 110 .

The circuit board 130 may be electrically connected to the battery 120 and disposed inside the body 110 .

That is, the circuit board 130 is electrically connected to the battery 120 to operate the low-power light irradiation unit 140, the ultrasonic transmission unit 150, and other components electrically connected to the circuit board 130 to be described later. And, more specifically, the circuit board 130 may be a printed circuit board (PCB).

The low-power light irradiation unit 140 is electrically connected to the circuit board 130 and can irradiate a low-power laser beam to the treatment area through the transmission hole 112 .

Low-power light irradiation unit 140 Also, since the low-power light irradiation unit 140 generates a large amount of heat while irradiating the low-power laser, a heat sink is provided between the laser element and the body 110 to dissipate this heat. (146) may be intervened.

Regarding low-power laser, the results of clinical trials conducted to evaluate the therapeutic effect of low-power laser on pain are as follows.

(1) Production of postoperative pain model: Incisional model (see Fig. 8(a))

1) 250-300g male Sprague-Dawley rat foot soles were incised under sevoflurane anesthesia.

2) Find the plantaris muscle of the sole of the experimental rat, leave the origin and stop, and make a longitudinal incision along the muscle fiber direction to induce postoperative pain.

(2) Creation of pain model by peripheral nerve damage: Chung model (spinal nerve ligation)

- Neuropathic pain was induced by ligating the spinal nerves (L5, 6) of 160-180 g male Sprague-Dawley rats under sevoflurane anesthesia.

(3) Behavioral evaluation of the presence or absence of pain after surgery (see Fig. 8(b))

1) Assessment of avoidance responses to mechanical stimuli

2) Evaluate avoidance responses to mechanical stimuli using von Frey filaments.

3) After adapting the rats to be used for the experiment for 30 minutes, von Frey filaments (0.4, 0.7, 1.2, 2.0, 3.6, 5.5, 8.5, 15g) were stimulated on the soles of the rats, and the avoidance response by moving the soles was up-down. evaluated using

4) The case where there is no response to 15g stimulation is set as the cut-off value

5) Statistically significant mechanical allodynia was induced at the site of the incision (***P<0.001) (see Fig. 8(c))

(4) Evaluation of therapeutic effect on pain after surgery (see Fig. 8(d))

1) Assessment of pain after laser irradiation

2) Group 1: Evaluation after 30 minutes of laser irradiation on the soles of pain-induced rats and 30 minutes of rest

Group 2: Evaluation after 60 minutes of laser irradiation on the soles of pain-induced rats and 30 minutes of rest

3) Pain was significantly reduced according to the laser irradiation time (*P<0.05, **P<0.01, ***P<0.001) (see FIG. 8(e))

The ultrasonic transmitter 150 may be electrically connected to the circuit board 130 to deliver ultrasonic waves to the treatment site.

The ultrasonic transmitter 150 is coupled to the body 110 and disposed in the case 152 having an internal space, and disposed in the internal space of the case 152 and is electrically connected to the circuit board 130 to have a piezoelectric effect. Accordingly, a piezoelectric element 154 for generating ultrasonic waves by converting electrical energy into vibration energy may be included.

At this time, the piezoelectric element 154 may be made of quartz, tourmaline, Rochelle salt, barium titanate, ammonium dihydrogen phosphate, ethylenediamine tartrate, artificial ceramics, and the like.

One surface of the case 152 is in contact with the skin to transmit ultrasonic waves generated by the piezoelectric element 154 disposed in the internal space to the skin. In this case, the case 152 can transmit ultrasonic waves to the treatment site more effectively It may protrude beyond the body portion 110 .

The bio-signal detection sensor 160 is electrically connected to the circuit board 130 and can detect photoplethysmograms to analyze pain in the treatment site.

The bio-signal detection sensor 160 may detect a photoplethysmogram by radiating light to an area to be measured and then receiving the returned light.

Accordingly, unlike the aforementioned low-output light irradiation unit 140, the bio-signal detection sensor 160 may not have the heat sink 146 interposed between the body unit 110 and the body unit 110 in order to receive the returned light.

The control unit (not shown) selects one of the low-power light irradiation unit 140 and the ultrasonic transmission unit 150 according to the degree of pain analyzed using the optical pulse wave detected by the bio-signal detection sensor 160 or the change in the bio-signal before and after the procedure. At least one operation can be controlled.

In this case, the control unit may receive data on the degree of pain analyzed by the pain measurement unit 200 or changes in biological signals before and after the procedure from the communication unit, which will be described later, and the low-power light irradiation unit 140 and/or Alternatively, the intensity of low-power laser irradiation and/or ultrasonic transmission may be adjusted by controlling the ultrasonic transmission unit 150 .

The low power light irradiation unit 140 may have an output value of 100 mW or less.

The low power light emitter 140 may include a plurality of light devices having different detailed condition values such as wavelength, output, wide angle, and output time.

In this case, the optical element may include a laser diode or an LED, and may also include various other light emitting elements.

Specifically, the low-power light irradiator 140 may include a first element 142 for irradiating a low-power laser in a wavelength range of 630 nm to 680 nm, and a second element 144 for irradiating a low-power laser in a wavelength range of 800 nm to 850 nm. .

In addition, the low-power light emitter 140 may further include a third element (not shown) for irradiating a low-power laser in a wavelength range of 900 nm to 1000 nm.

In the case of the low-power laser irradiated from the first element 142 in the wavelength range of 630 nm to 680 nm, effects such as pain relief according to the cell regeneration principle of the low-power laser can be generated.

In addition, in the case of a low-power laser irradiated from the second element 144 in a wavelength range of 800 nm to 850 nm, effects such as blood flow improvement, blood vessel dilation, and blood flow speed improvement can be generated.

In the case of a low-power laser of a wavelength range of 900 nm to 1000 nm irradiated from the third element, effects such as pain relief due to nerve stimulation and disturbance may be generated.

In contrast, the ultrasound transmission unit 150 may generate effects such as pain relief by transmitting ultrasound in a frequency band of 1 to 5 MHz that can reach superficial and deep muscles to the treatment site.

The first element 142 and the second element 144 may be operated alternately or simultaneously.

In this case, by irradiating the low-power lasers of different wavelength bands alone or in combination, the above-described effects in each wavelength band can occur, as well as complex treatment effects by combining them.

The first element 142 and the second element 144 are formed in plurality, respectively, the ultrasonic transmission unit 150 is disposed in the central region of the body part 110, and the first element 142 and the second element ( 144 may be alternately disposed along the peripheral area of the body portion 110 .

Accordingly, the ultrasonic transmission unit 150 transmits ultrasonic waves to the treatment site, and the plurality of first elements 142 are disposed to irradiate the low-power laser, so that the pain relief effect can be further enhanced.

In addition, since the plurality of second elements 144 are arranged and operated between the first elements 142, blood flow improvement, cell regeneration, and/or cell activation effects are generated in the treatment area, and thus the pain treatment effect can be further improved. there is.

The biological signal detection sensor 160 includes a light emitting unit 162 that radiates light to a measurement site through a transmission hole 112, and a light receiving unit 164 that receives light transmitted or reflected after being irradiated by the light emitting unit 162. ), and a light blocking unit 166 interposed between the light receiving unit 164 and the low power light irradiation unit 140 to prevent the low power laser light irradiated from the low power light irradiation unit 140 from entering the light receiving unit 164.

At this time, the light blocking portion 166 is formed as a dark film and is interposed between the circuit board 130 and the body portion 110 to spatially block and divide the low power light irradiation portion 140 and the light receiving portion 164 .

The inflow of the low-power laser light into the receiving unit is prevented through the light-blocking unit 166, thereby increasing the accuracy of the PDP measurement, and accordingly, the accuracy of the pain analysis result using the PDP can be improved.

The circuit board 130 may include an ambient light cancellation (ALC) circuit module and a low noise power supply circuit module.

In this case, the ALC circuit module may avoid interference between light received by the biosignal detection sensor 160 and ambient light.

In addition, the low-noise power supply circuit module can prevent noise generated by power switching from interfering with the optical pulse wave detection of the biological signal detection sensor 160 .

Therefore, since the accuracy of photoplethysmogram detection of the bio-signal detection sensor 160 is improved through the ALC circuit module and the low-noise power supply circuit module, a more accurate pain analysis result can be derived.

A communication unit (not shown) that is electrically connected to the circuit board 130 and transmits the photoplethysmogram detected by the biosignal detection sensor 160 to the user terminal 400 may be further included.

In this case, the communication unit may be a short-distance communication module, and in this case, Bluetooth may be used.

The user terminal 400 may transmit the photoplethysmogram received from the communication unit to the pain measurement unit 200, and the pain measurement unit 200 may use the photoplethysmogram wave to analyze the degree of pain or changes in biosignals before and after the procedure. can

The resultant data analyzed by the pain measurer 200 is transmitted to the user terminal 400 again, and the communication unit may receive the corresponding data from the user terminal 400 .

Referring to FIGS. 9 and 10 , a body-wearable pain management process based on bio-signal collection and analysis will be described.

In step 110, the bio-signal detection sensor 160 may detect a photoplethysmogram to analyze pain in the treatment area.

In step 120, the communication unit may transmit the photoplethysmogram detected by the biological signal detection sensor 160 to the user terminal 400.

In step 130, the user terminal 400 may transmit the photoplethysmogram received from the communication unit to the pain measuring unit 200.

In this case, in detail, the photoplethysmogram transmitted to the pain measurement unit 200 may be collected and stored in the biosignal collection unit 210 .

In step 140, the first processing unit 232 may remove signal noise by sequentially applying a nonlinear filter and a linear filter to the optical pulse wave.

Specifically, step 140 includes removing baseline fluctuation noise by firstly applying a nonlinear filter to the photoplethysmogram wave, and secondarily applying a linear filter to the photoplethysmogram wave to include a low-frequency component, a respiratory component, and a motion component. It may include passing a signal of a set frequency band other than a low-band frequency band to be used.

In step 150, the second processing unit 234 may remove ectopic beats from the OPTIC wave from which signal noise has been removed by the first processing unit 232, and may perform interpolation.

Step 150 includes, in detail, steps of extracting poles from the preprocessed photoplethysmogram and detecting pole intervals, determining ectopic beat intervals through the detected pole intervals and removing ectopic beats, and determining the removed ectopic beats. It may include interpolating the position.

In step 160, the determining unit 240 may determine whether the preprocessed photoplethysmogram is a good or bad signal.

More specifically, in step 160, good signal indices including ectopic beats, heart rate, continuous heart rate difference and pole amplitude of the preprocessed PPD are compared with set parameters for the good signal indices, so that the PPD is a good or bad signal. It can be determined whether or not

In other words, the determination unit 240 learns the data of the photoplethysmogram, which is determined as a good or bad signal, and analyzes the preprocessed one-dimensional signal image of the photoplethysmogram based on the learning result to determine whether the photoplethysmogram is good or bad. A signal quality evaluation model capable of discriminating whether it is a signal may be used.

More specifically, in the signal quality evaluation process (S170) of the signal quality evaluation model, as shown in FIG. 10, the matrix conversion unit 242 performs matrix conversion of the preprocessed signal image of the optical pulse wave (S171); The layer generation unit 243 performs a convolution operation on the matrices to generate (convolution) layers (S172), and the normalization unit 244 performs batch normalization on the layers (S173) , the activation unit 245 nonlinearly activating the normalized layer (rectified linear unit activation; ReLU) (S174), the subsampling unit 246 subsampling the nonlinearly activated layer (subsampling; max pooling) ( S175) (At this time, steps 172 to 175 may be repeated multiple times, preferably performed twice), the integrator 247 generates an integration matrix by integrating and connecting the subsampled layers. Step S176, step S177 of normalizing the integration matrix (dropout, dropout rate = 05) (step 176 can be additionally performed after step 177), and the decision unit 249 adds a regression probability function ( It may include determining whether the photoplethysmogram is a good or bad signal according to the probability derived by applying softmax (S178).

In step 180, the index derivation unit 250 may derive a setting index from the photoplethysmogram determined as a good signal.

In step 185, the comparison and analysis unit 260 compares and analyzes the set index with reference data to derive the degree of pain or changes in biosignals before and after the procedure.

Specifically, the reference data may include set index values of photoplethysmograms before the procedure or index or reference data of the degree of pain.

In addition, the analysis result by the comparison and analysis unit 260 may be transmitted to the user terminal 400 , where the degree of pain improvement or the change in biosignals before and after the procedure may be included.

In step 190, the communication unit may receive the level of pain analyzed by the pain measurement unit 200 or changes in bio signals before and after the procedure from the user terminal 400.

In step 200, the control unit controls the operation of at least one of the low-power light irradiation unit 140 and the ultrasonic transmission unit 150 according to the degree of pain analyzed by the pain measurer 200 received by the communication unit or changes in bio signals before and after the procedure. You can control it.

In step 210, the user terminal 400 may receive a satisfaction index for ultrasound and/or low-power laser treatment from the user.

In step 220, the control unit may calculate an improvement index by determining whether the degree of pain improvement analyzed by the pain measurer 200 is equal to or greater than a reference value.

In step 220, if the degree of pain improvement analyzed by the pain measurer 200 is greater than or equal to the reference value, calculating an improvement index by adding an addition index to the satisfaction index, and the degree of pain improvement analyzed by the pain measurer 200 is the reference value or less, calculating the satisfaction index as an improvement index (satisfaction index = improvement index).

In step 230, the control unit compares the improvement index with the reference index, and when the improvement index is less than the reference index, the control unit may control the operation of at least one of the low-power light irradiation unit 140 and the ultrasonic transmission unit 150, and also improve When the index is equal to or greater than the reference index, first user data including the degree of pain improvement, the satisfaction index, and the improvement index may be transmitted to the user terminal 400 .

In step 250, the pain measurer 200 may receive first user data from the user terminal 400 and store it in a database.

In step 260, the pain measurer 200 may learn the first user data stored in the database to generate and provide a recommended treatment mode for low-power laser and/or ultrasound, thereby providing a personalized treatment to the user. be able to

Meanwhile, in step 270, the pain measurer 200 may receive second user data including body information and indications from the user terminal 400 and store them in a database.

Also, in step 280, the pain measurer 200 may generate and update population statistical information by receiving and analyzing first user data and second user data from the user terminals 400 of a plurality of users.

In step 290, the pain measurer 200 may generate and provide a recommended treatment mode for low-power laser and/or ultrasound according to the first user data of the population statistical information corresponding to the user's second user information. Accordingly, the user can receive and use an effective treatment mode in the population included in the user.

In step 300, the pain measurer 200 may transmit the analyzed degree of pain or changes in biosignals before and after the procedure to a medical institution information server so that the pain measurer 200 can use it as first-diagnosis data when the user visits the hospital.

Here, the medical institution information server may include an electronic medical record (EMR), a personal health record (PHR), and the like.

In step 310, the pain measuring unit 200 compares the analyzed pain level or biosignal change before and after the procedure with a reference value, and if the value is greater than or equal to the reference value, a visit recommendation message may be sent to the user terminal 400.

As described above, according to the wearable pain management system 300 based on biosignal collection and analysis according to the present invention, personalized and accurate pain evaluation based on user's individual biosignal collection and analysis is possible.

According to the body-wearable pain management system 300 based on bio-signal collection and analysis according to the present invention, the accuracy of pain evaluation can be improved through repetitive learning of bio-signals through an artificial intelligence model.

According to the body-wearable pain management system 300 based on bio-signal collection and analysis according to the present invention, it is possible to evaluate, diagnose, and manage self-pain in a non-face-to-face manner in daily life without a user visiting a hospital.

According to the body-wearable pain management system 300 based on bio-signal collection and analysis according to the present invention, customized pain treatment according to the result of a user's self-diagnosis of pain is possible.

According to the body-wearable pain management system 300 based on bio-signal collection and analysis according to the present invention, pain diagnosis and corresponding pain treatment can be performed in-line.

On the other hand, the components of the above-described embodiment can be easily grasped from a process point of view. That is, each component can be identified as each process. In addition, the process of the above-described embodiment can be easily grasped from the viewpoint of components of the device.

In addition, the technical contents described above may be implemented in the form of program instructions that can be executed through various computer means and recorded in a computer readable medium. The computer readable medium may include program instructions, data files, data structures, etc. alone or in combination. Program commands recorded on the medium may be specially designed and configured for the embodiments or may be known and usable to those skilled in computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floptical disks. - includes hardware devices specially configured to store and execute program instructions, such as magneto-optical media, and ROM, RAM, flash memory, and the like. Examples of program instructions include high-level language codes that can be executed by a computer using an interpreter, as well as machine language codes such as those produced by a compiler. A hardware device may be configured to act as one or more software modules to perform the operations of the embodiments and vice versa.

Although one embodiment of the present invention has been described above, those skilled in the art can add, change, delete, or add components within the scope not departing from the spirit of the present invention described in the claims. The present invention can be variously modified and changed by the like, and this will also be said to be included within the scope of the present invention.

Claims

Pain treatment unit that performs a procedure for pain relief or pain treatment; and

A pain measurement unit that collects and analyzes changes in biosignals of a user using the pain treatment unit, derives the degree of pain of the user or changes in biosignals before and after the procedure, and provides the result to the user;

The pain measuring unit,

a biosignal collecting unit collecting the biosignal; and

An analyzer configured to analyze the biosignal collected by the biosignal collection unit to derive the degree of pain or a change in the biosignal before and after the procedure;

The analysis unit,

a pre-processing unit that pre-processes the bio-signal;

a determining unit determining whether the biosignal preprocessed by the preprocessing unit is a good signal;

an indicator derivation unit for deriving a set indicator from the biosignal determined as a good signal by the determination unit; and

Body wearable pain management based on biosignal collection and analysis, including a comparison and analysis unit that compares and analyzes the set index derived by the indicator derivation unit with reference data to derive the degree of pain or the biosignal change before and after the procedure. system.
According to claim 1,

The determination unit,

Body wear based on biosignal collection and analysis capable of determining whether the biosignal is a good signal by iteratively learning the biosignal through a predetermined machine learning-based signal quality evaluation model to improve the biosignal discrimination accuracy. type pain management system.
According to claim 1,

The determination unit,

a matrix conversion unit for matrix conversion of the bio-signal image;

a layer generator for generating a layer by performing a convolution operation on the matrix converted by the matrix converter;

a normalization unit normalizing the layer created by the layer creation unit;

an activation unit for non-linearly activating the layer normalized by the first normalization unit;

a subsampling unit for subsampling the layer nonlinearly activated by the activation unit;

an integration unit generating an integration matrix by integrating and connecting the layers subsampled by the subsampling unit;

a connection unit normalizing and connecting the integration matrix generated by the integration unit; and

A body-wearable pain management system based on biosignal collection and analysis, including a determination unit determining whether the biosignal is a good signal according to a probability derived by applying a probability function to the integration matrix normalized and connected by the connection unit.
According to claim 1,

The pain measuring unit,

Body wearable type based on collection and analysis of biosignals provided to enable bi-directional communication with the pain treatment unit and providing the user with the degree of pain or changes in the biosignals before and after the procedure in-line to the user. pain management system.
According to claim 1,

The biosignal includes a photoplethysmogram,

The pre-processing unit,

a first processor removing signal noise by sequentially applying a nonlinear filter and a linear filter to the photoplethysmogram collected by the biosignal collector; and

A body-wearable pain management system based on biological signal collection and analysis, comprising a second processing unit for removing and interpolating ectopic beats from the photoplethysmogram wave from which the signal noise has been removed by the first processing unit.
According to claim 1,

The biosignal includes a photoplethysmogram,

The pre-processing unit includes a first processing unit that removes signal noise by sequentially applying a non-linear filter and a linear filter to the photoplethysmogram collected by the bio-signal collection unit;

The first processing unit,

First, the nonlinear filter is applied to the optical pulse wave to remove baseline fluctuation noise, and secondarily, the linear filter is applied to a set frequency band excluding a low-pass frequency band including a low-frequency component, a respiratory component, and a motion component. A body-wearable pain management system based on bio-signal collection and analysis that passes the signal of
According to claim 1,

The biosignal includes a photoplethysmogram,

The pre-processing unit,

a first processor removing signal noise by sequentially applying a nonlinear filter and a linear filter to the photoplethysmogram collected by the biosignal collector; and

A method for extracting poles from the optical pulse wave from which the signal noise has been removed by the first processing unit, detecting the pole intervals, determining the ectopic beat intervals through the pole intervals, removing the ectopic beats, and performing interpolation. A body-wearable pain management system based on bio-signal collection and analysis, including two processing units.