CN116831595B - Pain grade assessment method - Google Patents
Pain grade assessment method Download PDFInfo
- Publication number
- CN116831595B CN116831595B CN202310739017.8A CN202310739017A CN116831595B CN 116831595 B CN116831595 B CN 116831595B CN 202310739017 A CN202310739017 A CN 202310739017A CN 116831595 B CN116831595 B CN 116831595B
- Authority
- CN
- China
- Prior art keywords
- layer
- pain
- value
- signal
- input end
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 17
- 210000004556 brain Anatomy 0.000 claims abstract description 23
- 238000012545 processing Methods 0.000 claims abstract description 8
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 6
- 230000008533 pain sensitivity Effects 0.000 claims abstract description 6
- 238000012216 screening Methods 0.000 claims abstract description 6
- 230000004927 fusion Effects 0.000 claims description 27
- 238000000605 extraction Methods 0.000 claims description 22
- 238000010606 normalization Methods 0.000 claims description 15
- 230000035945 sensitivity Effects 0.000 claims description 12
- 238000013210 evaluation model Methods 0.000 claims description 8
- 239000011159 matrix material Substances 0.000 claims description 6
- 230000004913 activation Effects 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 description 9
- 230000009286 beneficial effect Effects 0.000 description 7
- 239000000284 extract Substances 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000011176 pooling Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 210000003205 muscle Anatomy 0.000 description 1
- 210000000653 nervous system Anatomy 0.000 description 1
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
- A61B5/369—Electroencephalography [EEG]
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
- A61B5/369—Electroencephalography [EEG]
- A61B5/372—Analysis of electroencephalograms
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/48—Other medical applications
- A61B5/4824—Touch or pain perception evaluation
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/48—Other medical applications
- A61B5/4824—Touch or pain perception evaluation
- A61B5/4827—Touch or pain perception evaluation assessing touch sensitivity, e.g. for evaluation of pain threshold
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7203—Signal processing specially adapted for physiological signals or for diagnostic purposes for noise prevention, reduction or removal
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Surgery (AREA)
- Public Health (AREA)
- Pathology (AREA)
- Veterinary Medicine (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Molecular Biology (AREA)
- Psychiatry (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Pain & Pain Management (AREA)
- Artificial Intelligence (AREA)
- Signal Processing (AREA)
- Psychology (AREA)
- Hospice & Palliative Care (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physiology (AREA)
- Evolutionary Computation (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The invention discloses a pain level assessment method, which belongs to the technical field of electric digital signal processing, and comprises the steps of starting from brain wave signals, collecting brain wave sequences, carrying out orthogonal decomposition on the brain wave sequences, splitting the brain wave sequences, thus obtaining various signal components, screening out signals with highest pain sensitivity, discarding other insensitive or less sensitive signal components, selecting out effective signals, obtaining a pain target signal, extracting significant characteristic values from the pain target signal, combining with non-significant characteristics in the pain target signal, assessing the pain level based on a pain level assessment model, and improving the accuracy of assessing the pain level.
Description
Technical Field
The invention relates to the technical field of electric digital signal processing, in particular to a pain level assessment method.
Background
The existing pain level evaluation method obtains a target low-frequency coefficient and a target high-frequency coefficient by collecting an electromyographic signal and decomposing the electromyographic signal, and extracts a signal characteristic value from the target low-frequency coefficient, thereby obtaining an objective pain level according to the signal characteristic value. The electromyographic signals are signals transmitted from the nervous system to the body where the body is stimulated, and reflect the muscle activity, not the pain level perceived by the brain, and the pain is perceived by the brain.
Disclosure of Invention
Aiming at the defects in the prior art, the pain level evaluation method provided by the invention solves the problem of low evaluation precision in the existing method for evaluating the pain level by the electromyographic signals.
In order to achieve the aim of the invention, the invention adopts the following technical scheme: a method of pain class assessment comprising the steps of:
s1, sampling brain waves to obtain an electroencephalogram sequence;
s2, carrying out orthogonal decomposition on the electroencephalogram sequence to obtain a plurality of signal components;
s3, screening out the signal with the highest pain sensitivity degree from the plurality of signal components to be used as a pain target signal;
s4, extracting a significant characteristic value from the pain target signal, and constructing a significant characteristic value sequence;
and S5, obtaining the pain level based on the pain level evaluation model according to the significant characteristic value sequence and the pain target signal.
Further, the expression of the signal component in S2 is:
,
wherein,is->Signal components>For the length of the brain electrical sequence, < > a->Is the +.>The data of the plurality of data,as an exponential function based on natural constants, < +.>Is imaginary.
The beneficial effects of the above further scheme are: the invention adopts the exponential factor to decompose the electroencephalogram sequence, so that the electroencephalogram sequence is divided into a plurality of signal components, thereby splitting the signal components and facilitating the screening of the most effective signals.
Further, the step S3 includes the following sub-steps:
s31, calculating a sensitivity value of the signal component to pain;
s32, selecting the signal component with the largest sensitivity value as a pain target signal.
Further, the formula for calculating the sensitivity value in S31 is:
,
wherein,is->Sensitivity value of the individual signal components,/>Is->First of the signal components>Numerical value->Stores the normal brain electrical sequence +.>First of the signal components>Numerical value->Is the number of values in the signal component.
The beneficial effects of the above further scheme are: according to the invention, the difference between the numerical value in each signal component and the numerical value in each signal component of the electroencephalogram sequence under normal conditions is calculated, when the difference is large, the signal component is more sensitive to pain, and the most sensitive signal component is selected, so that on one hand, noise can be removed, and meanwhile, the interference of other insensitive signals can be removed, and the evaluation accuracy is improved.
Further, the significant eigenvalue sequence in S4 isWherein->For the 1 st salient feature value, +.>For the 2 nd salient feature value, +.>For the 3 rd salient feature value, +.>Is the 4 th salient feature value;
the 1 st salient feature valueThe expression of (2) is:
,
wherein,to take the maximum value of the sequence, +.>Is the +.>Numerical value->Is the number of values in the pain target signal;
the 2 nd significant bitSign valueThe expression of (2) is:
,
the 3 rd salient feature valueThe expression of (2) is:
,
the 4 th salient feature valueThe expression of (2) is:
。
the beneficial effects of the above further scheme are: the invention passes through the 1 st significant characteristic valueExpress max, 2 nd salient feature value +.>Express mean, 3 rd salient feature value +.>Expressing the degree of deviation from the average value, expressing the fluctuation distribution of the data, 4 th salient feature value +.>The difference between the values is enhanced by taking the 4 th power of each value, and the overall distribution of the values is further expressed.
Further, the pain level assessment model in S5 includes: the device comprises a plurality of feature extraction networks, a feature fusion layer, a normalization layer, a first hiding layer, a second hiding layer and an output layer;
the input end of the characteristic extraction network is used for inputting the segmented pain target signal; the first input end of the feature fusion layer is used for being connected with the output end of the feature extraction network, and the second input end of the feature fusion layer is used for inputting a significant feature value sequence; the input end of the normalization layer is connected with the output end of the feature fusion layer, and the output end of the normalization layer is connected with the input end of the first hiding layer; the input end of the second hiding layer is connected with the output end of the first hiding layer; the input end of the output layer is connected with the output end of the second hidden layer, and the output end of the output layer is used as the output end of the pain class evaluation model.
Further, the feature extraction network includes: a Maxpool layer, an Avgpool layer, a Concat layer, a first Conv layer, a second Conv layer, a Softmax layer, a multiplier, and a sigmoid layer;
the input end of the first Conv layer is respectively connected with the input end of the Maxpool layer and the input end of the Avgpool layer and is used as the input end of the feature extraction network; the input end of the Concat layer is respectively connected with the output end of the Maxpool layer and the output end of the Avgpool layer, and the output end of the Concat layer is connected with the first input end of the multiplier; the second input end of the multiplier is connected with the output end of the Softmax layer, and the output end of the multiplier is connected with the input end of the second Conv layer; the input end of the Softmax layer is connected with the output end of the first Conv layer; the input end of the sigmoid layer is connected with the output end of the second Conv layer, and the output end of the sigmoid layer is used as the output end of the feature extraction network.
The beneficial effects of the above further scheme are: according to the invention, a first Conv layer is set to extract characteristics, a weight value is calculated through a Softmax layer and is applied to a multiplier, a maximum value characteristic is extracted through a Maxpool layer, an average value characteristic is extracted through an Avgpool layer, a Concat layer splices the maximum value characteristic and the average value characteristic to obtain a spliced signal, and the spliced signal is multiplied by the weight value, so that self-adaptive application attention degree is realized, the expression of characteristics is enhanced or weakened, the characteristics are extracted again through a second Conv layer, the condition of the output characteristics of the second Conv layer is comprehensively considered through a sigmoid layer, and the fine characteristics are extracted.
Further, the expression of the sigmoid layer is:
,
wherein,is a fine feature value->For S-type activation function, +.>Output of the +.>Numerical value->Is the +.>Personal weight(s)>The number of values is output for the second Conv layer.
The beneficial effects of the above further scheme are: in the sigmoid layer, weighting is carried out on all values output by the second Conv layer, so that a fine characteristic value of each target sub-signal is obtained.
Further, the step S5 includes the following sub-steps:
s51, segmenting the pain target signal to obtain a target sub-signal;
s52, converting the target sub-signals into matrixes to obtain matrix signals;
s53, extracting fine characteristic values from the matrix signals by adopting a characteristic extraction network, and constructing a fine characteristic value sequence;
s54, fusing the fine characteristic value sequence and the remarkable characteristic value sequence by adopting a characteristic fusion layer to obtain a fusion signal;
s55, carrying out normalization processing on the fusion signal by adopting a normalization layer to obtain normalized data;
s56, adopting a first hidden layer and a second hidden layer to perform continuous linear transformation on the normalized data twice to obtain the output of the hidden layers;
and S57, processing the output of the hidden layer by adopting the output layer to obtain the pain level.
The beneficial effects of the above further scheme are: according to the invention, the pain target signal is segmented, so that the fine characteristic value of each part is extracted, the fine characteristic value sequence and the obvious characteristic value sequence are fused, the data characteristic is fully expressed, and then the data characteristic is input into the BP neural network for evaluation, so that the evaluation accuracy is improved.
Further, the expression of the fusion signal in S54 is:
,
wherein,for fusion signal, ++>For the 1 st minute characteristic value, +.>Is->Fine feature value->Is->Fine feature value->For the 1 st salient feature value, +.>For the 2 nd salient feature value, +.>For the 3 rd salient feature value, +.>For the 4 th salient feature value, +.>Is a transpose operation.
In summary, the invention has the following beneficial effects: the invention starts from brain wave signals, acquires brain wave sequences, carries out orthogonal decomposition on the brain wave sequences, splits the brain wave sequences, thereby obtaining various signal components, screens out signals with highest pain sensitivity, discards other insensitive or insensitive signal components, picks out effective signals, obtains pain target signals, extracts obvious characteristic values of the pain target signals, combines with non-obvious characteristics in the pain target signals, evaluates pain grades based on a pain grade evaluation model, and improves the accuracy of evaluating the pain grades.
Drawings
FIG. 1 is a flow chart of a method of pain class assessment;
FIG. 2 is a schematic diagram of a pain class assessment model;
fig. 3 is a schematic diagram of the structure of the feature extraction network.
Detailed Description
The following description of the embodiments of the present invention is provided to facilitate understanding of the present invention by those skilled in the art, but it should be understood that the present invention is not limited to the scope of the embodiments, and all the inventions which make use of the inventive concept are protected by the spirit and scope of the present invention as defined and defined in the appended claims to those skilled in the art.
As shown in fig. 1, a pain level assessment method includes the steps of:
s1, sampling brain waves to obtain an electroencephalogram sequence;
s2, carrying out orthogonal decomposition on the electroencephalogram sequence to obtain a plurality of signal components;
the expression of the signal component in S2 is:
,
wherein,is->Signal components>For the length of the brain electrical sequence, < > a->Is the +.>The data of the plurality of data,as an exponential function based on natural constants, < +.>Is imaginary.
The invention adopts the exponential factor to decompose the electroencephalogram sequence, so that the electroencephalogram sequence is divided into a plurality of signal components, thereby splitting the signal components and facilitating the screening of the most effective signals.
S3, screening out the signal with the highest pain sensitivity degree from the plurality of signal components to be used as a pain target signal;
the step S3 comprises the following substeps:
s31, calculating a sensitivity value of the signal component to pain;
the formula for calculating the sensitivity value in S31 is:
,
wherein,is->Sensitivity value of the individual signal components,/>Is->First of the signal components>Numerical value->Stores the normal brain electrical sequence +.>First of the signal components>Numerical value->Is the number of values in the signal component;
s32, selecting the signal component with the largest sensitivity value as a pain target signal.
According to the invention, the difference between the numerical value in each signal component and the numerical value in each signal component of the electroencephalogram sequence under normal conditions is calculated, when the difference is large, the signal component is more sensitive to pain, and the most sensitive signal component is selected, so that on one hand, noise can be removed, and meanwhile, the interference of other insensitive signals can be removed, and the evaluation accuracy is improved.
S4, extracting a significant characteristic value from the pain target signal, and constructing a significant characteristic value sequence;
the significant eigenvalue sequence in S4 isWherein->For the 1 st salient feature value, +.>For the 2 nd salient feature value, +.>For the 3 rd salient feature value, +.>Is the 4 th salient feature value;
the 1 st salient feature valueThe expression of (2) is:
,
wherein,to take the maximum value of the sequence, +.>Is the +.>Numerical value->Is the number of values in the pain target signal;
the 2 nd significant eigenvalueThe expression of (2) is:
,
the 3 rd salient feature valueThe expression of (2) is:
,
the 4 th salient feature valueThe expression of (2) is:
。
the invention passes through the 1 st significant characteristic valueExpress max, 2 nd salient feature value +.>Express mean, 3 rd salient feature value +.>Expressing the degree of deviation from the average value, expressing the fluctuation distribution of the data, 4 th salient feature value +.>The difference between the values is enhanced by taking the 4 th power of each value, and the overall distribution of the values is further expressed.
And S5, obtaining the pain level based on the pain level evaluation model according to the significant characteristic value sequence and the pain target signal.
As shown in fig. 2, the pain class assessment model in S5 includes: the device comprises a plurality of feature extraction networks, a feature fusion layer, a normalization layer, a first hiding layer, a second hiding layer and an output layer;
the input end of the characteristic extraction network is used for inputting the segmented pain target signal; the first input end of the feature fusion layer is used for being connected with the output end of the feature extraction network, and the second input end of the feature fusion layer is used for inputting a significant feature value sequence; the input end of the normalization layer is connected with the output end of the feature fusion layer, and the output end of the normalization layer is connected with the input end of the first hiding layer; the input end of the second hiding layer is connected with the output end of the first hiding layer; the input end of the output layer is connected with the output end of the second hidden layer, and the output end of the output layer is used as the output end of the pain class evaluation model.
As shown in fig. 3, the feature extraction network includes: a Maxpool layer, an Avgpool layer, a Concat layer, a first Conv layer, a second Conv layer, a Softmax layer, a multiplier, and a sigmoid layer;
the input end of the first Conv layer is respectively connected with the input end of the Maxpool layer and the input end of the Avgpool layer and is used as the input end of the feature extraction network; the input end of the Concat layer is respectively connected with the output end of the Maxpool layer and the output end of the Avgpool layer, and the output end of the Concat layer is connected with the first input end of the multiplier; the second input end of the multiplier is connected with the output end of the Softmax layer, and the output end of the multiplier is connected with the input end of the second Conv layer; the input end of the Softmax layer is connected with the output end of the first Conv layer; the input end of the sigmoid layer is connected with the output end of the second Conv layer, and the output end of the sigmoid layer is used as the output end of the feature extraction network.
The Maxpool layer is the largest pooling layer and the Avgpool layer is the average pooling layer.
According to the invention, a first Conv layer is set to extract characteristics, a weight value is calculated through a Softmax layer and is applied to a multiplier, a maximum value characteristic is extracted through a Maxpool layer, an average value characteristic is extracted through an Avgpool layer, a Concat layer splices the maximum value characteristic and the average value characteristic to obtain a spliced signal, and the spliced signal is multiplied by the weight value, so that self-adaptive application attention degree is realized, the expression of characteristics is enhanced or weakened, the characteristics are extracted again through a second Conv layer, the condition of the output characteristics of the second Conv layer is comprehensively considered through a sigmoid layer, and the fine characteristics are extracted.
The expression of the sigmoid layer is as follows:
,
wherein,is a fine feature value->For S-type activation function, +.>Output of the +.>Numerical value->Is the +.>Personal weight(s)>The number of values is output for the second Conv layer.
In the sigmoid layer, weighting is carried out on all values output by the second Conv layer, so that a fine characteristic value of each target sub-signal is obtained.
The step S5 comprises the following substeps:
s51, segmenting the pain target signal to obtain a target sub-signal;
s52, converting the target sub-signals into matrixes to obtain matrix signals;
s53, extracting fine characteristic values from the matrix signals by adopting a characteristic extraction network, and constructing a fine characteristic value sequence;
s54, fusing the fine characteristic value sequence and the remarkable characteristic value sequence by adopting a characteristic fusion layer to obtain a fusion signal;
the expression of the fusion signal in S54 is:
,
wherein,for fusion signal, ++>For the 1 st minute characteristic value, +.>Is->Fine feature value->Is->Fine feature value->For the 1 st salient feature value, +.>For the 2 nd salient feature value, +.>For the 3 rd salient feature value, +.>For the 4 th salient feature value, +.>Is a transpose operation.
S55, carrying out normalization processing on the fusion signal by adopting a normalization layer to obtain normalized data;
s56, adopting a first hidden layer and a second hidden layer to perform continuous linear transformation on the normalized data twice to obtain the output of the hidden layers;
and S57, processing the output of the hidden layer by adopting the output layer to obtain the pain level.
According to the invention, the pain target signal is segmented, so that the fine characteristic value of each part is extracted, the fine characteristic value sequence and the obvious characteristic value sequence are fused, the data characteristic is fully expressed, and then the data characteristic is input into the BP neural network for evaluation, so that the evaluation accuracy is improved.
The invention starts from brain wave signals, acquires brain wave sequences, carries out orthogonal decomposition on the brain wave sequences, splits the brain wave sequences, thereby obtaining various signal components, screens out signals with highest pain sensitivity, discards other insensitive or insensitive signal components, picks out effective signals, obtains pain target signals, extracts obvious characteristic values of the pain target signals, combines with non-obvious characteristics in the pain target signals, evaluates pain grades based on a pain grade evaluation model, and improves the accuracy of evaluating the pain grades.
The above is only a preferred embodiment of the present invention, and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (8)
1. A method of pain rating assessment, comprising the steps of:
s1, sampling brain waves to obtain an electroencephalogram sequence;
s2, carrying out orthogonal decomposition on the electroencephalogram sequence to obtain a plurality of signal components;
s3, screening out the signal with the highest pain sensitivity degree from the plurality of signal components to be used as a pain target signal;
s4, extracting a significant characteristic value from the pain target signal, and constructing a significant characteristic value sequence;
s5, obtaining a pain grade based on the pain grade evaluation model according to the significant characteristic value sequence and the pain target signal;
the step S5 comprises the following substeps:
s51, segmenting the pain target signal to obtain a target sub-signal;
s52, converting the target sub-signals into matrixes to obtain matrix signals;
s53, extracting fine characteristic values from the matrix signals by adopting a characteristic extraction network, and constructing a fine characteristic value sequence;
s54, fusing the fine characteristic value sequence and the remarkable characteristic value sequence by adopting a characteristic fusion layer to obtain a fusion signal;
s55, carrying out normalization processing on the fusion signal by adopting a normalization layer to obtain normalized data;
s56, adopting a first hidden layer and a second hidden layer to perform continuous linear transformation on the normalized data twice to obtain the output of the hidden layers;
s57, processing the output of the hidden layer by adopting the output layer to obtain a pain level;
the expression of the fusion signal in S54 is:
,
wherein,for fusion signal, ++>For the 1 st minute characteristic value, +.>Is->Fine feature value->Is->Fine feature value->Is 1 stSignificant eigenvalue(s)>For the 2 nd salient feature value, +.>For the 3 rd salient feature value, +.>For the 4 th salient feature value, +.>Is a transpose operation.
2. The pain level assessment method according to claim 1, wherein the expression of the signal component in S2 is:
,
wherein,is->Signal components>For the length of the brain electrical sequence, < > a->Is the +.>Data of->As an exponential function based on natural constants, < +.>Is imaginary.
3. The pain level assessment method according to claim 2, wherein S3 comprises the following substeps:
s31, calculating a sensitivity value of the signal component to pain;
s32, selecting the signal component with the largest sensitivity value as a pain target signal.
4. A pain level assessment method according to claim 3, wherein the formula for calculating the sensitivity value in S31 is:
,
wherein,is->Sensitivity value of the individual signal components,/>Is->First of the signal components>Numerical value->Stores the normal brain electrical sequence +.>First of the signal components>Numerical value->Is the number of values in the signal component.
5. The pain level assessment method according to claim 2, wherein the sequence of salient feature values in S4 isWherein->For the 1 st salient feature value, +.>For the 2 nd salient feature value, +.>For the 3 rd salient feature value, +.>Is the 4 th salient feature value;
the 1 st salient feature valueThe expression of (2) is:
,
wherein,to take the maximum value of the sequence, +.>Is the +.>Numerical value->Is the number of values in the pain target signal;
the 2 nd significant eigenvalueThe expression of (2) is:
,
the 3 rd salient feature valueThe expression of (2) is:
,
the 4 th salient feature valueThe expression of (2) is:
。
6. the pain level assessment method according to claim 1, wherein the pain level assessment model in S5 comprises: the device comprises a plurality of feature extraction networks, a feature fusion layer, a normalization layer, a first hiding layer, a second hiding layer and an output layer;
the input end of the characteristic extraction network is used for inputting the segmented pain target signal; the first input end of the feature fusion layer is used for being connected with the output end of the feature extraction network, and the second input end of the feature fusion layer is used for inputting a significant feature value sequence; the input end of the normalization layer is connected with the output end of the feature fusion layer, and the output end of the normalization layer is connected with the input end of the first hiding layer; the input end of the second hiding layer is connected with the output end of the first hiding layer; the input end of the output layer is connected with the output end of the second hidden layer, and the output end of the output layer is used as the output end of the pain class evaluation model.
7. The pain level assessment method according to claim 6, wherein the feature extraction network comprises: a Maxpool layer, an Avgpool layer, a Concat layer, a first Conv layer, a second Conv layer, a Softmax layer, a multiplier, and a sigmoid layer;
the input end of the first Conv layer is respectively connected with the input end of the Maxpool layer and the input end of the Avgpool layer and is used as the input end of the feature extraction network; the input end of the Concat layer is respectively connected with the output end of the Maxpool layer and the output end of the Avgpool layer, and the output end of the Concat layer is connected with the first input end of the multiplier; the second input end of the multiplier is connected with the output end of the Softmax layer, and the output end of the multiplier is connected with the input end of the second Conv layer; the input end of the Softmax layer is connected with the output end of the first Conv layer; the input end of the sigmoid layer is connected with the output end of the second Conv layer, and the output end of the sigmoid layer is used as the output end of the feature extraction network.
8. The pain level assessment method according to claim 7, wherein the sigmoid layer has the expression:
,
wherein,is a fine feature value->For S-type activation function, +.>Output of the +.>The number of the values to be used in the process,is the +.>Personal weight(s)>The number of values is output for the second Conv layer.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310739017.8A CN116831595B (en) | 2023-06-21 | 2023-06-21 | Pain grade assessment method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310739017.8A CN116831595B (en) | 2023-06-21 | 2023-06-21 | Pain grade assessment method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN116831595A CN116831595A (en) | 2023-10-03 |
CN116831595B true CN116831595B (en) | 2024-04-12 |
Family
ID=88166255
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310739017.8A Active CN116831595B (en) | 2023-06-21 | 2023-06-21 | Pain grade assessment method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116831595B (en) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107184204A (en) * | 2017-04-28 | 2017-09-22 | 北京易飞华通科技开发有限公司 | The extraction of nociceptive component and expression way in a kind of brain wave |
CN111493836A (en) * | 2020-05-31 | 2020-08-07 | 天津大学 | Postoperative acute pain prediction system based on brain-computer interface and deep learning and application |
CN213156536U (en) * | 2020-07-24 | 2021-05-11 | 中国人民解放军西部战区总医院 | Medical pain scoring indicating device |
CN112957014A (en) * | 2021-02-07 | 2021-06-15 | 广州大学 | Pain detection and positioning method and system based on brain waves and neural network |
CN113614751A (en) * | 2019-03-29 | 2021-11-05 | 新加坡科技研究局 | Electroencephalogram signal identification and extraction |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20110087125A1 (en) * | 2009-10-09 | 2011-04-14 | Elvir Causevic | System and method for pain monitoring at the point-of-care |
US20150201879A1 (en) * | 2012-07-24 | 2015-07-23 | Cerephex Corporation | Method and Apparatus for Diagnosing and Assessing Centralized Pain |
EP3948893A4 (en) * | 2019-04-02 | 2022-12-07 | Cerebral Diagnostics Canada Incorporated | Evaluation of pain disorders via expert system |
-
2023
- 2023-06-21 CN CN202310739017.8A patent/CN116831595B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107184204A (en) * | 2017-04-28 | 2017-09-22 | 北京易飞华通科技开发有限公司 | The extraction of nociceptive component and expression way in a kind of brain wave |
CN113614751A (en) * | 2019-03-29 | 2021-11-05 | 新加坡科技研究局 | Electroencephalogram signal identification and extraction |
CN111493836A (en) * | 2020-05-31 | 2020-08-07 | 天津大学 | Postoperative acute pain prediction system based on brain-computer interface and deep learning and application |
CN213156536U (en) * | 2020-07-24 | 2021-05-11 | 中国人民解放军西部战区总医院 | Medical pain scoring indicating device |
CN112957014A (en) * | 2021-02-07 | 2021-06-15 | 广州大学 | Pain detection and positioning method and system based on brain waves and neural network |
Non-Patent Citations (2)
Title |
---|
Machine learning on encephalographic activity may predict opioid analgesia;M. Gram 等;European Journal of pain;20151130;第19卷(第10期);1552-1561 * |
基于EEG的突发疼痛识别方法及实验研究;曹天傲;优秀硕士论文电子期刊-信息科技;20190115(第2019年第01期期);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN116831595A (en) | 2023-10-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107844755B (en) | Electroencephalogram characteristic extraction and classification method combining DAE and CNN | |
CN114201988B (en) | Satellite navigation composite interference signal identification method and system | |
CN108629413B (en) | Neural network model training and transaction behavior risk identification method and device | |
CN111753968A (en) | Non-invasive load monitoring intelligent ammeter and electricity decomposition method | |
CN110853668A (en) | Voice tampering detection method based on multi-feature fusion | |
CN107942214A (en) | A kind of feature extracting method of transformer partial discharge signal, device | |
CN106375780A (en) | Method and apparatus for generating multimedia file | |
CN113538037B (en) | Method, system, equipment and storage medium for monitoring charging event of battery car | |
CN113378160A (en) | Graph neural network model defense method and device based on generative confrontation network | |
CN110859616A (en) | Cognitive assessment method, device and equipment of object and storage medium | |
CN115884032B (en) | Smart call noise reduction method and system for feedback earphone | |
CN114841191A (en) | Epilepsia electroencephalogram signal feature compression method based on fully-connected pulse neural network | |
CN116831595B (en) | Pain grade assessment method | |
CN115899598A (en) | Heat supply pipe network state monitoring method and system integrating auditory and visual characteristics | |
CN112183582A (en) | Multi-feature fusion underwater target identification method | |
CN113850013B (en) | Ship radiation noise classification method | |
CN113707175B (en) | Acoustic event detection system based on feature decomposition classifier and adaptive post-processing | |
CN112200221B (en) | Epilepsy prediction system and method based on electrical impedance imaging and electroencephalogram signals | |
CN113052099A (en) | SSVEP classification method based on convolutional neural network | |
CN113208632A (en) | Attention detection method and system based on convolutional neural network | |
CN112205990A (en) | Wrist angle prediction method and device based on sEMG under different loads | |
CN117079005A (en) | Optical cable fault monitoring method, system, device and readable storage medium | |
CN106580319A (en) | Electroencephalogram relaxation degree identification method and device based on wavelet transformation | |
CN115951167A (en) | Power distribution network fault type judgment method | |
CN104102834A (en) | Method for identifying sound recording locations |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |