CN115886815A - Emotional pressure monitoring method and device and intelligent wearable device - Google Patents
Emotional pressure monitoring method and device and intelligent wearable device Download PDFInfo
- Publication number
- CN115886815A CN115886815A CN202211410640.0A CN202211410640A CN115886815A CN 115886815 A CN115886815 A CN 115886815A CN 202211410640 A CN202211410640 A CN 202211410640A CN 115886815 A CN115886815 A CN 115886815A
- Authority
- CN
- China
- Prior art keywords
- ppg signal
- prv
- emotional stress
- frequency
- data
- 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.)
- Pending
Links
- 230000002996 emotional effect Effects 0.000 title claims abstract description 118
- 238000000034 method Methods 0.000 title claims abstract description 48
- 238000012544 monitoring process Methods 0.000 title claims abstract description 45
- 238000013145 classification model Methods 0.000 claims abstract description 28
- 238000003062 neural network model Methods 0.000 claims description 23
- 238000004364 calculation method Methods 0.000 claims description 12
- 238000013528 artificial neural network Methods 0.000 claims description 11
- 238000012806 monitoring device Methods 0.000 claims description 10
- 238000007781 pre-processing Methods 0.000 claims description 8
- 238000005070 sampling Methods 0.000 claims description 6
- 230000009466 transformation Effects 0.000 claims description 4
- 230000001131 transforming effect Effects 0.000 claims description 3
- 238000012545 processing Methods 0.000 abstract description 10
- 230000007774 longterm Effects 0.000 abstract description 9
- 238000013186 photoplethysmography Methods 0.000 description 162
- 238000002565 electrocardiography Methods 0.000 description 12
- 238000012549 training Methods 0.000 description 11
- 230000006870 function Effects 0.000 description 6
- 238000001914 filtration Methods 0.000 description 5
- 238000000605 extraction Methods 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000004590 computer program Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 208000030523 mesoaxial synostotic syndactyly with phalangeal reduction Diseases 0.000 description 2
- 230000000737 periodic effect Effects 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000009189 diving Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 229910021389 graphene Inorganic materials 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000003183 myoelectrical effect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000029058 respiratory gaseous exchange Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Landscapes
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention provides an emotional stress monitoring method and device and intelligent wearable equipment. The emotional stress monitoring method comprises the steps that a PPG sensor on intelligent wearable equipment is adopted to obtain a PPG signal within a preset time period of pulse of a user, the PPG signal within the preset time period is a first PPG signal, and baseline drift and high-frequency noise removal processing are carried out on the first PPG signal to obtain a second PPG signal; searching all wave crests in the second PPG signal, calculating PRV data containing a plurality of RR intervals in the second PPG signal, wherein the RR intervals are time differences between two adjacent wave crests, calculating a time domain index and a frequency domain index of the PRV data, and inputting an emotional stress classification model to obtain the emotional stress state of the user. Compare to adopt ECG mode monitoring emotional stress to compare among the prior art, the scheme in this application can be based on intelligent wearing equipment, collects the first PPG signal of user's pulse in real time, real-time and long-term monitoring user's emotional stress state.
Description
Technical Field
The invention relates to the technical field of intelligent wearing, in particular to an emotional pressure monitoring method and device and intelligent wearing equipment.
Background
At present, the pace of life of people is accelerated, the work is busy, and the long-term psychological pressure can influence the physical state of human bodies. In the prior art, electrocardiosignals of a human body are collected and processed to find the position of an R peak (wave crest) of an ECG (electrocardiography) electrocardio waveform, an adjacent R peak time interval is calculated to obtain HRV (heart rate variability) data, time domain and frequency domain analysis is carried out on the HRV data, various indexes reflecting the HRV data are calculated, and medical staff analyze the emotional stress state of the human body through the indexes.
The ECG method is used for monitoring emotional stress, and the test environment is harsh, so the ECG method is widely applied to hospital clinic. In recent years, some portable electrocardiograph monitors have appeared, and are small in size and convenient to carry. However, the electrocardio-electrode wire is needed to connect the electrode with the body or the electrodes are held by both hands, so that real-time measurement under the conditions of working, learning and the like cannot be realized, and long-term monitoring cannot be realized. In conclusion, the detection of ECG has many limitations, and the detection environment is harsh, so that it is impossible to monitor the ECG at any time and any place.
Disclosure of Invention
The invention provides an emotional stress monitoring method and device and intelligent wearable equipment, which can monitor the emotional stress state of a user in real time and for a long time.
In a first aspect, the invention provides an emotional stress monitoring method, which is applied to an intelligent wearable device. The emotional stress monitoring method comprises the following steps: acquiring a PPG signal within a preset time period of the pulse of a user through a PPG (photoplethysmography) sensor, wherein the PPG signal within the preset time period is a first PPG signal; removing baseline drift and high-frequency noise of the first PPG signal to obtain a second PPG signal; finding all peaks in the second PPG signal; calculating PRV (pulse rate variability) data comprising a plurality of RR intervals in the second PPG signal; the RR interval is the time difference between two adjacent wave crests; calculating a time domain index and a frequency domain index of the PRV data; and inputting the time domain index and the frequency domain index into the emotional stress classification model to obtain the emotional stress state of the user.
In the scheme, a PPG signal in a preset time period of pulse of a user is acquired by adopting a PPG sensor on intelligent wearable equipment, the PPG signal in the preset time period is a first PPG signal, and baseline drift and high-frequency noise removal processing are carried out on the first PPG signal to obtain a second PPG signal; searching all wave crests in a second PPG signal, and calculating PRV data of the second PPG signal containing a plurality of RR intervals, wherein the RR intervals are time differences of two adjacent wave crests; and calculating a time domain index and a frequency domain index of the PRV data, and inputting the time domain index and the frequency domain index into the emotional stress classification model to obtain the emotional stress state of the user. Compare to adopt ECG mode monitoring emotional stress among the prior art, the scheme in this application can be based on intelligent wearing equipment, collects the PPG signal of user's pulse in real time, real-time and long-term monitoring user's emotional stress state.
In a specific embodiment, removing baseline drift and high frequency noise of the first PPG signal comprises: removing baseline drift of the first PPG signal by using an integer coefficient filter of a linear phase; a low pass filter is used to remove high frequency noise of the first PPG signal. The whole coefficient filter through adopting linear phase gets rid of the baseline drift of first PPG signal, satisfies real-time operation on wearing formula equipment for filtering is simple, high-efficient, prevents PPG waveform distortion after the filtering, has guaranteed the integrality of PPG waveform, also promotes the operation rate simultaneously.
In a specific embodiment, finding all peaks in the second PPG signal comprises: setting a sliding window with the time length of a multiplied by fs multiplied by f, wherein fs represents a sampling rate, and f represents a preset frequency; sliding on the second PPG signal using a sliding window, and finding a maximum and a minimum of the second PPG signal within the slid window; judging whether the maximum value meets the amplitude value of 0.7-1.3 times of the previous wave crest and whether the minimum value meets the amplitude value of 0.7-1.3 times of the previous wave trough; if so, the maximum value is the peak of the second PPG signal in the sliding window; if not, the sliding window continues to slide until all peaks in the second PPG signal have been sought. All wave crests in the second PPG signal are searched by adopting a sliding window threshold value method, the PPG wave crests can be accurately found under the condition of motion interference, the PRV data can be conveniently and accurately calculated subsequently, and the later time-frequency domain extraction features are more accurate.
In one specific embodiment, the preset frequency f is determined as follows: transforming the second PPG signal to the frequency domain by fourier transformation; finding a peak with the maximum amplitude in a second PPG signal transformed to a frequency domain through Fourier transform, and judging whether the peak with the maximum amplitude has harmonic waves; if the wave crest with the maximum amplitude has harmonic waves, taking the frequency of the wave crest with the maximum amplitude as a preset frequency f; and if no harmonic exists in the peak with the maximum amplitude, finding the peak with the second maximum amplitude in the second PPG signal which is transformed into the frequency domain through Fourier transform, and taking the frequency of the peak with the second maximum amplitude as a preset frequency f. By determining the size of the preset frequency f in the above manner, the stepping time length of the sliding window can be dynamically adjusted, so that the sliding window is suitable for PPG signals in different states.
In a specific embodiment, before the time domain index and the frequency domain index of the PRV data are calculated, the singular point data in the PRV data are removed, so that the accuracy of the subsequent time-frequency domain extraction features is prevented from being influenced by the singular point data in the PRV data.
In a particular embodiment, removing singularity data from the PRV data comprises: determining whether each RR interval in a plurality of RR intervals meets the following condition: 0.8 × P (i) is less than 1.2 × P (i); wherein P (i) and P (i + 1) denote an ith RR interval and an i +1RR interval of the plurality of RR intervals, respectively; if so, retaining P (i + 1); if not, P (i + 1) is removed. And each singular point is convenient to find accurately.
In a specific embodiment, the calculating the time domain index and the frequency domain index of the PRV data includes: calculating the root mean square value of the PRV mean value, the PRV standard deviation and the difference between adjacent RR intervals of the PRV data as the time domain index of the PRV data; the low frequency power, the high frequency power, and the low frequency power/high frequency power of the PRV data are calculated as frequency domain indices of the PRV data. The time-frequency domain characteristics of the PRV data can be extracted accurately, and the accuracy of the time-frequency domain index is improved.
In a specific embodiment, the emotional stress classification model is a BP neural network model obtained by training a neural network (a multi-layer feedforward neural network trained according to an error back propagation algorithm); the BP neural network model comprises six input nodes, wherein the six input nodes are respectively used for inputting a PRV mean value, a PRV standard deviation, a root mean square value of a difference between adjacent RR intervals, low-frequency power, high-frequency power and low-frequency power/high-frequency power; the BP neural network model comprises four output nodes for outputting a certain emotional stress state in the emotional stress state category; wherein the emotional stress state categories at least comprise: non-pressure state, light pressure state, medium pressure state, high pressure state. Inputting the time domain index and the frequency domain index into the emotional stress classification model to obtain the emotional stress state of the user, wherein the method comprises the following steps: and inputting the calculated PRV mean value, PRV standard deviation and root mean square value of the difference between adjacent RR intervals, low-frequency power, high-frequency power and low-frequency power/high-frequency power of the PRV data into a BP neural network model to obtain the emotional stress state of the user. The pressure state is classified through the BP neural network, so that the pressure state is accurately measured, the pressure state can be monitored for a long time, and the pressure monitoring is more convenient and intelligent.
In a second aspect, the invention further provides an emotional stress monitoring device, and the emotional stress monitoring device is applied to the intelligent wearable equipment. This emotional stress monitoring device includes: the system comprises an acquisition module, a preprocessing module, a peak searching module, a first calculating module, a second calculating module and an emotional stress classification model. The acquisition module is used for acquiring a PPG signal within a preset time period of the pulse of the user through a PPG sensor, and the PPG signal within the preset time period is a first PPG signal. The preprocessing module is used for removing the baseline drift and the high-frequency noise of the first PPG signal to obtain a second PPG signal. The peak searching module is used for searching all peaks in the second PPG signal. The first calculation module is used for calculating PRV data of a second PPG signal containing a plurality of RR intervals; the RR interval is the time difference between two adjacent peaks. The second calculation module is used for calculating the time domain index and the frequency domain index of the PRV data. The emotional stress classification model is used for receiving the time domain index and the frequency domain index and outputting the emotional stress state of the user according to the received time domain index and the frequency domain index.
In the scheme, a PPG signal in a preset time period of pulse of a user is acquired by adopting a PPG sensor on intelligent wearable equipment, the PPG signal in the preset time period is a first PPG signal, and baseline drift and high-frequency noise removal processing are carried out on the first PPG signal to obtain a second PPG signal; searching all wave crests in a second PPG signal, and calculating PRV data of the second PPG signal containing a plurality of RR intervals, wherein the RR intervals are time differences of two adjacent wave crests; and calculating a time domain index and a frequency domain index of the PRV data, and inputting the time domain index and the frequency domain index into the emotional stress classification model to obtain the emotional stress state of the user. Compare to adopt ECG mode monitoring emotional stress among the prior art, the scheme in this application can be based on intelligent wearing equipment, collects the PPG signal of user's pulse in real time, real-time and long-term monitoring user's emotional stress state.
In a third aspect, the present invention further provides an intelligent wearable device, including: a PPG sensor and a processor connected with the PPG sensor; the PPG sensor is used for acquiring a PPG signal within a preset time period of the pulse of the user, the PPG signal within the preset time period is a first PPG signal, and the first PPG signal is sent to the processor; the processor is used for executing the emotional stress monitoring method.
Acquiring a first PPG signal of the pulse of a user by adopting a PPG sensor on intelligent wearable equipment, and removing baseline drift and high-frequency noise processing on the first PPG signal to obtain a second PPG signal; searching and calculating time periods of all peaks and two adjacent peaks in the second PPG signal to obtain PRV data comprising a plurality of RR intervals; and calculating a time domain index and a frequency domain index of the PRV data, and inputting the time domain index and the frequency domain index into the emotional stress classification model to obtain the emotional stress state of the user. Compare to adopt ECG mode monitoring emotional stress to compare among the prior art, the scheme in this application can be based on intelligent wearing equipment, collects the first PPG signal of user's pulse in real time, real-time and long-term monitoring user's emotional stress state.
Drawings
Fig. 1 is a flowchart of an emotional stress monitoring method according to an embodiment of the present invention;
FIG. 2 is a flow chart of another method of emotional stress monitoring provided by embodiments of the invention;
fig. 3 is a flowchart of another emotional stress monitoring method according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In order to facilitate understanding of the emotional stress monitoring method provided by the embodiment of the present invention, an application scenario of the emotional stress monitoring method provided by the embodiment of the present invention is first described below, and the emotional stress monitoring method is applied to an intelligent wearable device to monitor an emotional stress state of a user wearing the intelligent wearable device in real time. The intelligent wearable device can be a wearable device such as but not limited to a diving computer. The emotional stress monitoring method is described in detail below with reference to the accompanying drawings.
Referring to fig. 1, an emotional stress monitoring method provided by an embodiment of the invention includes: acquiring a PPG signal within a preset time period of the pulse of a user through a PPG sensor, wherein the PPG signal within the preset time period is a first PPG signal; removing baseline drift and high-frequency noise of the first PPG signal to obtain a second PPG signal; finding all peaks in the second PPG signal; calculating PRV data comprising a plurality of RR intervals in the second PPG signal, wherein the RR intervals are the time difference of two adjacent peaks; calculating a time domain index and a frequency domain index of the PRV data; and inputting the time domain index and the frequency domain index into the emotional stress classification model to obtain the emotional stress state of the user.
In the scheme, a PPG signal in a preset time period of the pulse of the user is obtained by adopting a PPG sensor on the intelligent wearable device, the PPG signal is a first PPG signal, the time period does not include boundary time, in actual application, the value of the preset time period is 30s, and about 60 wave peaks are approximately included in the time period 30s, the data volume is moderate, the processing of a processor on the intelligent wearable device is facilitated, the accuracy is not reduced, and the baseline drift and high-frequency noise removal processing is carried out on the first PPG signal to obtain a second PPG signal; searching all peaks in a second PPG signal, and calculating PRV data of the second PPG signal containing a plurality of RR intervals, wherein the RR intervals are time differences of two adjacent peaks; and calculating a time domain index and a frequency domain index of the PRV data, and inputting the time domain index and the frequency domain index into the emotional stress classification model to obtain the emotional stress state of the user. Compare to adopt ECG mode monitoring emotional stress among the prior art, the scheme in this application can be based on intelligent wearing equipment, collects the PPG signal of user's pulse in real time, real-time and long-term monitoring user's emotional stress state. The above steps will be described in detail with reference to the accompanying drawings.
First, referring to fig. 1 and 2, a PPG signal of a pulse of a user within a preset time period is acquired by a PPG sensor. Specifically, as shown in fig. 2, a PPG sensor may be installed on the smart wearable device, and the PPG sensor may be specifically a photoelectric sensor.
Next, as shown in fig. 1 and fig. 2, the baseline drift and the high-frequency noise of the first PPG signal are removed to obtain a second PPG signal. Due to the respiration, movement, shaking, myoelectric interference and other reasons of a human body, the first PPG signal directly acquired under normal conditions is often accompanied by more burrs (high frequency), baseline drift (low frequency) and motion artifacts. And preprocessing the first PPG signal and filtering out an interference signal. In particular to remove the baseline drift and high frequency noise of the first PPG signal, referring to fig. 2, an integer coefficient filter of linear phase may be first used to remove the baseline drift of the first PPG signal. The baseline drift of first PPG signal is removed through the integer coefficient filter of adopting linear phase place, satisfies real-time operation on wearing formula equipment for filtering is simple, high-efficient, prevents the PPG wave form distortion after the filtering, has guaranteed the integrality of PPG wave form, also promotes the operation rate simultaneously. The prior art filters out interfering signals by using a high-pass filter. A general high-pass filter, such as a butterworth filter, is a nonlinear phase filter, which causes distortion of the filtered PPG waveform, and the filter coefficients are floating point type, which reduces the operating efficiency. This application has adopted the integer coefficient filter of a linear phase place to get rid of the baseline drift, because the filter coefficient is the integer, can promote the operation rate, satisfies real-time operation on wearing formula equipment, guarantees simultaneously that the PPG waveform is undistorted. The linear high-pass filter is composed of a low-pass filter and an all-pass filter with the same delay as the low-pass filter, and the specific principle is as follows:
Y 1 (Z)＝H 1 (Z)×X(Z)
wherein f is pm Is the 3db cut-off frequency point, h 1 Is the maximum fluctuation amplitude in the passband, K, M all integers, Y1 (Z), Y2 (Z), Y3 (Z) are the outputs of the all-pass filter, the low-pass filter and the high-pass filter, H 1 (Z)，H 2 (Z)，H 3 (Z) is the transfer function of the all-pass filter, the low-pass filter and the high-pass filter. y is 3 (n) is the inverse transform of Y3 (Z), X (n) is the inverse transform of X (Z), X (n) is the input signal, Y is the inverse transform of Y3 (Z) 1 (n) is the inverse transformation of Y1 (Z), f s2 Is 50Hz, f s1 Is the sampling frequency and L is the length of the filter.
And then, a low-pass filter is adopted to remove the high-frequency noise of the first PPG signal, wherein the low-pass filter can be specifically an FIR low-pass filter, so that the high-frequency noise can be effectively removed, meanwhile, the PPG waveform is not distorted, and the calculation is simple. It should be understood that the above only shows one way to remove baseline wander and high frequency noise, and that other ways may be used in addition to this.
Next, referring to fig. 1 and 2, all peaks in the second PPG signal are sought. In particular, when searching for all peaks in the second PPG signal, a sliding window threshold method may be employed to search for all peaks in the second PPG signal. Because wearable equipment is worn on the hand, more motion artifacts can be generated, and PPG peak false detection can be caused by the traditional difference and extreme method under the condition that the motion artifacts exist, so that PRV calculation is inaccurate. According to the method, all peaks in the second PPG signal can be searched through a sliding window threshold method, so that motion artifacts can be effectively avoided, and the PPG peaks can be accurately found.
The principle of finding all peaks in the second PPG signal using a sliding window threshold method is as follows: referring to fig. 1-3, a sliding window having a time length of a × fs × f may be set, where fs denotes a sampling rate and f denotes a preset frequency. And sliding the sliding window on the second PPG signal, and searching the maximum value and the minimum value of the second PPG signal in the sliding window after each sliding. And judging whether the maximum value meets the amplitude value of 0.7-1.3 times of the previous wave crest or not and whether the minimum value meets the amplitude value of 0.7-1.3 times of the previous wave trough or not. And if the maximum value meets the amplitude value of 0.7-1.3 times of the previous wave peak and the minimum value meets the amplitude value of 0.7-1.3 times of the previous wave trough, the maximum value is the wave peak of the PPG signal in the sliding window. If one of the maximum and minimum values does not satisfy the aforementioned condition, the sliding window continues to slide until all peaks in the second PPG signal have been found. If motion interference exists, the collected PPG signal is extremely unstable, the peak or trough amplitude is large, the maximum value is used as the peak amplitude only under the condition that the maximum value and the minimum value in the sliding window meet the conditions, all peaks in the second PPG signal are searched by adopting a sliding window threshold value method, the PPG peak point can be accurately found under the condition of motion interference, the PRV data can be conveniently and accurately calculated subsequently, and the later time-frequency domain extraction features are more accurate.
Referring to fig. 3, the predetermined frequency f may be determined as follows: transforming the second PPG signal to the frequency domain by fourier transformation; finding a peak with the maximum amplitude in the second PPG signal which is transformed to a frequency domain through Fourier transform, and judging whether the peak with the maximum amplitude has harmonic waves; if the wave crest with the maximum amplitude has harmonic waves, taking the frequency of the wave crest with the maximum amplitude as a preset frequency f; and if no harmonic exists in the peak with the maximum amplitude, finding the peak with the second maximum amplitude in the second PPG signal transformed to the frequency domain through Fourier transform, and taking the frequency of the peak with the second maximum amplitude as the preset frequency f. By determining the size of the preset frequency f in the above manner, the step time length of the sliding window can be dynamically adjusted, so that the sliding window is suitable for PPG signals in different states.
Next, referring to fig. 1 and 2, PRV data may be calculated for a second PPG signal comprising a plurality of RR intervals. The RR interval is the time difference between two adjacent peaks.
In addition, after the PRV data including a plurality of RR intervals is obtained, referring to fig. 2, before the time domain index and the frequency domain index of the PRV data are calculated, singular point data in the PRV data may be removed first, so as to prevent the singular point data in the PRV data from affecting the accuracy of subsequent time-frequency domain extraction features. Specifically, when the singular point data in the PRV data is removed, the following method may be adopted. Determining whether each RR interval in a plurality of RR intervals meets the following condition: 0.8 × P (i) is less than 1.2 × P (i); wherein P (i) and P (i + 1) denote an ith RR interval and an i +1RR interval, respectively, of the plurality of RR intervals. If so, the i +1RR interval, i.e., P (i + 1), is retained. If not, the i +1RR interval is removed from the PRV data as singular point data. The method is convenient for accurately finding each singular point.
Next, as shown in fig. 1 and 2, a time domain index and a frequency domain index of the PRV data are calculated. When calculating the time domain index and the frequency domain index of the PRV data, the root mean square value of the PRV mean, the PRV standard deviation, and the difference between adjacent RR intervals of the PRV data may be calculated first, and used as the time domain index of the PRV data. Specifically, the following formula may be adopted to calculate the time domain index of the PRV data:
where Mean represents the PRV Mean of the calculated PRV data, SDNN represents the standard deviation of the calculated PRV data, r-MSSD represents the root Mean square value of the difference between adjacent RR intervals, RR i Is the time difference between the ith and i +1 PPG peaks, and RR is the time difference of the adjacent 2 PPG peaks.
Then, the low frequency power, the high frequency power, and the low frequency power/high frequency power of the PRV data are calculated as the frequency domain indices of the PRV data. The time domain signal can be transformed to the frequency domain by using the analysis of the Lomb-Scargle algorithm, so that the accuracy of the frequency domain index is higher. Specifically, the Lomb-Scargle can be used for analyzing the PRV frequency domain index to obtain power spectrum information. It should be explained that the Lomb-Scargle periodogram method is based on the principle of discrete fourier transform, decomposes a time domain signal into linear combinations of some sine functions y = acos ω t + bsin ω t, converts the time domain signal into a frequency domain, performs model curve fitting on data by using a least square method through a sine function, and judges the implicit periodic variation trend of the data by using a mean square error, so that the fourier transform can be applied to non-uniform signals. The Lomb-scope has a good effect on analyzing the non-uniform signals, can effectively extract weak periodic signals and can weaken false signals generated by the non-uniform signals in the time domain, and the Lomb-scope mathematical expression is as follows.
Wherein f represents frequency, P x (f) Is the power spectrum of each frequency bin, τ is the time offset, t j For the sampling time, X (t) j ) Is t j Input data of time of day.
Then, the low frequency power, the high frequency power, and the low frequency power/high frequency power of the PRV data are calculated using the following formulas.
Where LF denotes the low frequency power of the PRV data, which is the total power of 0.04-0.15 Hz. Where HF represents the high frequency power of the PRV data, which is the total power of 0.15-0.4 Hz. Where LH denotes the low frequency power/high frequency power of the PRV data, LF/HF.
By adopting the method to calculate the time-frequency domain index of the PRV data, the time-frequency domain characteristic of the PRV data is conveniently and accurately extracted, and the accuracy of the time-frequency domain index is improved.
Next, referring to fig. 1 and fig. 2, the time domain index and the frequency domain index are input into the emotional stress classification model, so as to obtain the emotional stress state of the user. The emotional stress classification model can be a BP neural network model obtained by training a neural network. It should be explained that the BP neural network is a multi-layer feedforward network trained by an error inverse propagation algorithm, and model training includes forward propagation and backward calculation. And forward propagation, namely input data are transmitted from the input layer, processed by the hidden layer and transmitted to the output layer, the network minimizes the mean square error between the output and an expected value, and the weight value and the threshold value of each network layer are reversely calculated by a gradient descent method to obtain model parameters. The BP neural network model in the application comprises six input nodes, and the six input nodes are respectively used for inputting a PRV mean value, a PRV standard deviation, a root mean square value of a difference between adjacent RR intervals, low-frequency power, high-frequency power and low-frequency power/high-frequency power. The BP neural network model comprises four output nodes for outputting a certain emotional stress state in the emotional stress state category; wherein the emotional stress state categories at least comprise: non-pressure state, light pressure state, medium pressure state, high pressure state. Each output node corresponds to a category of emotional stress states.
The neural network training process of the BP neural network model can adopt the following modes: and acquiring training set data and training the BP neural network model. The training set data obtains PPG data of different pressure states through pressure tests of different degrees (no pressure, light pressure, medium pressure and high pressure), calculates PRV and extracts features including Mean, SDNN, r-MSSD, LF, HF and LH. And then training the neural network model by taking the training set data as the input characteristics of the neural network. And then, taking the trained model as a BP neural network model for clear stress state judgment and outputting emotional stress states (no stress, mild stress, moderate stress and high stress).
When the BP neural network model obtained by the neural network training is used as the emotional stress classification model, the time domain index and the frequency domain index are input into the emotional stress classification model, and the emotional stress state of the user is obtained by the following specific steps: and inputting the calculated PRV mean value, PRV standard deviation and root mean square value of the difference between adjacent RR intervals, low-frequency power, high-frequency power and low-frequency power/high-frequency power of the PRV data into a BP neural network model to obtain the emotional stress state of the user. The pressure state is classified through the BP neural network, so that the pressure state is accurately measured, the pressure state can be monitored for a long time, and the pressure monitoring is more convenient and intelligent.
It should be noted that the emotional stress classification model is not limited to the BP neural network model, and other manners may be adopted. For example, an SVM multi-classification model may be used instead of the BP neural network model.
In the various embodiments shown above, a PPG sensor on the smart wearable device is used to obtain a PPG signal within a preset time period of a pulse of a user, where the PPG signal within the preset time period is a first PPG signal, and the first PPG signal is subjected to baseline drift removal and high-frequency noise processing to obtain a second PPG signal; searching all peaks in a second PPG signal, and calculating PRV data of the second PPG signal containing a plurality of RR intervals, wherein the RR intervals are time differences of two adjacent peaks; and calculating a time domain index and a frequency domain index of the PRV data, and inputting the time domain index and the frequency domain index into the emotional stress classification model to obtain the emotional stress state of the user. Compare to adopt ECG mode monitoring emotional stress to compare among the prior art, the scheme in this application can be based on intelligent wearing equipment, collects the PPG signal of user's pulse in real time, real-time and long-term monitoring user's emotional stress state.
In addition, the embodiment of the invention also provides an emotional stress monitoring device, and the emotional stress monitoring device is applied to intelligent wearable equipment. This emotional stress monitoring device includes: the system comprises an acquisition module, a preprocessing module, a peak searching module, a first calculating module, a second calculating module and an emotional stress classification model. The acquisition module is used for acquiring a PPG signal within a preset time period of the pulse of the user through a PPG sensor, and the PPG signal within the preset time period is a first PPG signal. The preprocessing module is used for removing the baseline drift and the high-frequency noise of the first PPG signal to obtain a second PPG signal. The peak searching module is used for searching all peaks in the second PPG signal. The first calculation module is used for calculating PRV data of a second PPG signal containing a plurality of RR intervals, and the RR intervals are time differences of two adjacent wave crests. The second calculation module is used for calculating the time domain index and the frequency domain index of the PRV data. The emotional stress classification model is used for receiving the time domain index and the frequency domain index and outputting the emotional stress state of the user according to the received time domain index and the frequency domain index.
In the scheme, a PPG signal in a preset time period of pulse of a user is acquired by adopting a PPG sensor on intelligent wearable equipment, the PPG signal in the preset time period is a first PPG signal, and baseline drift and high-frequency noise removal processing are carried out on the first PPG signal to obtain a second PPG signal; searching all peaks in a second PPG signal, and calculating PRV data of the second PPG signal containing a plurality of RR intervals, wherein the RR intervals are time differences of two adjacent peaks; and calculating a time domain index and a frequency domain index of the PRV data, and inputting the time domain index and the frequency domain index into the emotional stress classification model to obtain the emotional stress state of the user. Compare to adopt ECG mode monitoring emotional stress among the prior art, the scheme in this application can be based on intelligent wearing equipment, collects the PPG signal of user's pulse in real time, real-time and long-term monitoring user's emotional stress state.
It should be explained that the acquisition module, the preprocessing module, the peak searching module, the first calculating module, the second calculating module and the emotional stress classification model are functional modules for realizing corresponding functions. Each functional module is embodied as an aggregate of hardware and software. The hardware is specifically hardware with logical operation and storage functions, such as a processor and a memory, the software is specifically software codes stored in the memory, and when the processor reads and executes the software codes, the corresponding functions of the corresponding functional modules can be executed. It should be noted that the functional modules in the emotional stress monitoring device are not limited to the above-described modules, and other functional modules may be included.
For example, the preprocessing module may include a first removal module configured to remove a baseline shift of the first PPG signal using a linear-phase integer coefficient filter, and a second removal module configured to remove a high-frequency noise of the first PPG signal using a low-pass filter.
The peak finding module may comprise a sliding window module. The sliding window module is used for setting a sliding window with the time length of a multiplied by fs multiplied by f, wherein fs represents a sampling rate, and f represents a preset frequency. The sliding window module is further configured to slide on the second PPG signal using a sliding window, and find a maximum and a minimum of the second PPG signal within the sliding window after each sliding. The sliding window module is also used for judging whether the maximum value meets the amplitude value of 0.7-1.3 times of the previous wave crest and whether the minimum value meets the amplitude value of 0.7-1.3 times of the previous wave trough; if the maximum value meets the amplitude value of 0.7-1.3 times of the previous wave crest and the minimum value meets the amplitude value of 0.7-1.3 times of the previous wave trough, the sliding window module takes the maximum value as the wave crest of the PPG signal in the sliding window; if one of the maximum and minimum values does not satisfy the aforementioned condition, the sliding window continues to slide.
The peak searching module may further include a preset frequency module, and the preset frequency module is configured to determine the preset frequency f in the following manner. The preset frequency module transforms the second PPG signal to a frequency domain through Fourier transform; the preset frequency module finds a peak with the maximum amplitude in the second PPG signal transformed to the frequency domain through Fourier transform, and judges whether the peak with the maximum amplitude has harmonic waves; if the wave crest with the maximum amplitude has harmonic waves, the preset frequency module takes the frequency of the wave crest with the maximum amplitude as a preset frequency f; and if no harmonic exists in the peak with the maximum amplitude, finding the peak with the second maximum amplitude in the second PPG signal transformed to the frequency domain through Fourier transform, and taking the frequency of the peak with the second maximum amplitude as the preset frequency f.
The emotional stress monitoring device can also comprise a singular point data removing module, wherein the singular point data removing module removes the singular point data in the PRV data before calculating the time domain index and the frequency domain index of the PRV data.
The singular point data removal module may remove the singular point data in the PRV data as follows. The singular point data removing module judges whether each RR interval in the RR intervals meets the following conditions: 0.8 × P (i) is less than 1.2 × P (i); wherein P (i) and P (i + 1) denote an ith RR interval and an i +1RR interval of the plurality of RR intervals, respectively; if yes, the singular point data removing module reserves an i +1RR interval, namely P (i + 1); if not, the singularity data removal module removes the i +1RR interval as singularity data from the PRV data.
The second calculating module may calculate the time domain index and the frequency domain index of the PRV data in the following manner. The second calculation module calculates the root mean square value of the PRV mean value, the PRV standard deviation and the difference between adjacent RR intervals of the PRV data as the time domain index of the PRV data. The second calculation module calculates low frequency power, high frequency power, and low frequency power/high frequency power of the PRV data as frequency domain indices of the PRV data.
The emotional stress classification model can be a BP neural network model obtained by adopting neural network training; the BP neural network model comprises six input nodes, wherein the six input nodes are respectively used for inputting a PRV mean value, a PRV standard deviation, a root mean square value of a difference between adjacent RR intervals, low-frequency power, high-frequency power and low-frequency power/high-frequency power; the BP neural network model comprises four output nodes for outputting a certain emotional stress state in the emotional stress state category; wherein the emotional stress state categories at least comprise: no pressure state, light pressure state, medium pressure state, high pressure state. And when the time domain index and the frequency domain index which are calculated by the second calculation module are input into the emotional stress classification model to obtain the emotional stress state of the user, the PRV mean value, the PRV standard deviation and the root mean square value of the difference between adjacent RR intervals, the low-frequency power, the high-frequency power and the low-frequency power/high-frequency power of the PRV data which are calculated are input into the BP neural network model to obtain the emotional stress state of the user.
In addition, the embodiment of the invention also provides intelligent wearable equipment, which comprises a PPG sensor and a processor connected with the PPG sensor; the PPG sensor is used for acquiring a PPG signal within a preset time period of the pulse of the user, the PPG signal within the preset time period is a first PPG signal, and the first PPG signal is sent to the processor; the processor is configured to perform the aforementioned emotional stress monitoring method.
The embodiment of the invention also provides an intelligent wearing device, which comprises: a body. The body specifically can be intelligent wearing equipment's dial plate and watchband, and the integration has intelligent wearing equipment's various functional device in the dial plate. The body is integrated with the emotional pressure monitoring device.
In an embodiment, a computer-readable storage medium is provided, on which a computer program is stored which, when being executed by a processor, carries out the steps of the above-mentioned method embodiments.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, database, or other medium used in the embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high-density embedded nonvolatile Memory, resistive Random Access Memory (ReRAM), magnetic Random Access Memory (MRAM), ferroelectric Random Access Memory (FRAM), phase Change Memory (PCM), graphene Memory, and the like. Volatile Memory can include Random Access Memory (RAM), external cache Memory, and the like. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others. The databases referred to in various embodiments provided herein may include at least one of relational and non-relational databases. The non-relational database may include, but is not limited to, a block chain based distributed database, and the like. The processors referred to in the various embodiments provided herein may be, without limitation, general purpose processors, central processing units, graphics processors, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, or the like.
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above description is only for the specific embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. The utility model provides an emotional stress monitoring method, is applied to intelligent wearing equipment, its characterized in that includes:
acquiring a PPG signal within a preset time period of the pulse of a user through a PPG sensor, wherein the PPG signal within the preset time period is a first PPG signal;
removing baseline drift and high-frequency noise of the first PPG signal to obtain a second PPG signal;
finding all peaks in the second PPG signal;
calculating PRV data comprising a plurality of RR intervals in the second PPG signal; the RR interval is the time difference between two adjacent peaks;
calculating a time domain index and a frequency domain index of the PRV data;
and inputting the time domain index and the frequency domain index into an emotional stress classification model to obtain the emotional stress state of the user.
2. The method of emotional stress monitoring of claim 1, wherein the removing baseline drift and high frequency noise of the first PPG signal comprises:
removing baseline drift of the first PPG signal by using an integer coefficient filter of a linear phase;
removing high frequency noise of the first PPG signal by using a low pass filter.
3. The method of emotional stress monitoring of claim 1, wherein the finding all peaks in the second PPG signal comprises:
setting a sliding window with the time length of a multiplied by fs multiplied by f, wherein fs represents a sampling rate, and f represents a preset frequency;
sliding on the second PPG signal using the sliding window, and finding a maximum value and a minimum value of the second PPG signal within the sliding window after sliding;
judging whether the maximum value meets the amplitude value of 0.7-1.3 times of the previous wave crest or not and whether the minimum value meets the amplitude value of 0.7-1.3 times of the previous wave trough or not;
if so, the maximum value is the peak of the second PPG signal within the sliding window;
if not, the sliding window continues to slide until all peaks in the second PPG signal have been found.
4. The method of emotional stress monitoring of claim 3, wherein the predetermined frequency f is determined by:
transforming the second PPG signal to the frequency domain by Fourier transformation;
finding a peak with the maximum amplitude in the second PPG signal which is transformed to a frequency domain through Fourier transform, and judging whether the peak with the maximum amplitude has harmonic waves;
if the peak with the maximum amplitude has harmonic waves, taking the frequency of the peak with the maximum amplitude as the preset frequency f;
and if no harmonic exists in the peak with the maximum amplitude, finding a peak with the second maximum amplitude in the second PPG signal transformed to the frequency domain through Fourier transform, and taking the frequency of the peak with the second maximum amplitude as the preset frequency f.
5. The method of emotional stress monitoring of claim 1, wherein prior to the computing the time-domain indicators and the frequency-domain indicators of the PRV data, comprising:
and removing singular point data in the PRV data.
6. The method of emotional stress monitoring of claim 5, wherein the removing singularity data from the PRV data comprises:
determining whether each RR interval of the plurality of RR intervals satisfies the following condition: 0.8 × P (i) is less than 1.2 × P (i); wherein P (i) and P (i + 1) represent the ith RR interval and the (i + 1) RR interval, respectively, of the RR intervals;
if so, retaining P (i + 1);
if not, P (i + 1) is removed.
7. The emotional stress monitoring method of claim 1, wherein the calculating the time-domain index and the frequency-domain index of the PRV data comprises:
calculating a root mean square value of a PRV mean value, a PRV standard deviation and a difference between adjacent RR intervals of the PRV data as a time domain index of the PRV data;
and calculating the low-frequency power, the high-frequency power and the low-frequency power/high-frequency power of the PRV data as the frequency domain index of the PRV data.
8. The method of monitoring emotional stress of claim 7, wherein the emotional stress classification model is a BP neural network model trained using a neural network; the BP neural network model comprises six input nodes, and the six input nodes are respectively used for inputting the PRV mean value, the PRV standard deviation, the root mean square value of the difference between the adjacent RR intervals, the low-frequency power, the high-frequency power and the low-frequency power/high-frequency power; the BP neural network model comprises four output nodes for outputting a certain emotional stress state in the emotional stress state category; wherein the emotional stress state categories include at least: a no-pressure state, a light pressure state, a medium pressure state, and a high pressure state;
the step of inputting the time domain indexes and the frequency domain indexes into an emotional stress classification model to obtain the emotional stress state of the user comprises the following steps: and inputting the calculated PRV mean value, PRV standard deviation, root mean square value of the difference between adjacent RR intervals, low-frequency power, high-frequency power and low-frequency power/high-frequency power of the PRV data into the BP neural network model to obtain the emotional stress state of the user.
9. The utility model provides an emotional stress monitoring device is applied to intelligent wearing equipment, a serial communication port, include:
the acquisition module is used for acquiring a PPG signal within a preset time period of the pulse of the user through a PPG sensor, wherein the PPG signal within the preset time period is a first PPG signal;
the preprocessing module is used for removing baseline drift and high-frequency noise of the first PPG signal to obtain a second PPG signal;
a peak searching module, configured to search all peaks in the second PPG signal;
a first calculation module to calculate PRV data for which the second PPG signal comprises a plurality of RR intervals; the RR interval is the time difference between two adjacent peaks;
the second calculation module is used for calculating a time domain index and a frequency domain index of the PRV data;
and the emotional stress classification model is used for receiving the time domain index and the frequency domain index and outputting the emotional stress state of the user according to the received time domain index and the frequency domain index.
10. An intelligence wearing equipment which characterized in that includes: a PPG sensor and a processor connected to the PPG sensor;
the PPG sensor is used for acquiring a PPG signal within a preset time period of the pulse of a user, the PPG signal within the preset time period is a first PPG signal, and the first PPG signal is sent to the processor;
the processor is configured to perform the emotional stress monitoring method of any of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211410640.0A CN115886815A (en) | 2022-11-10 | 2022-11-10 | Emotional pressure monitoring method and device and intelligent wearable device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211410640.0A CN115886815A (en) | 2022-11-10 | 2022-11-10 | Emotional pressure monitoring method and device and intelligent wearable device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115886815A true CN115886815A (en) | 2023-04-04 |
Family
ID=86490701
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211410640.0A Pending CN115886815A (en) | 2022-11-10 | 2022-11-10 | Emotional pressure monitoring method and device and intelligent wearable device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115886815A (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105943015A (en) * | 2016-06-04 | 2016-09-21 | 浙江大学 | Wearable heart rate variability (HRV) monitoring device with active noise reduction function |
| WO2017160186A1 (en) * | 2016-03-18 | 2017-09-21 | Валерий Викторович КУРЫШЕВ | Method for continuously monitoring a person's stress level |
| CN108809463A (en) * | 2017-04-28 | 2018-11-13 | 工业和信息化部电信研究院 | A kind of multiple reference clock weighting synthetic method |
| CN110090029A (en) * | 2018-01-28 | 2019-08-06 | 北京师范大学 | Emotional measurement system, Intelligent bracelet and portable terminal |
| CN110353704A (en) * | 2019-07-12 | 2019-10-22 | 东南大学 | Mood assessments method and apparatus based on wearable ECG monitoring |
| CN111419250A (en) * | 2020-04-08 | 2020-07-17 | 恒爱高科(北京)科技有限公司 | Emotion recognition method based on pulse waves |
| GB202016726D0 (en) * | 2020-10-21 | 2020-12-02 | August Int Ltd | Improvements in or relating to wearable sensor apparatus |
| CN113017623A (en) * | 2021-03-04 | 2021-06-25 | 江苏盖睿健康科技有限公司 | Measuring method, device and storage medium for hypoperfusion blood oxygen saturation |
| CN113040771A (en) * | 2021-03-01 | 2021-06-29 | 青岛歌尔智能传感器有限公司 | Emotion recognition method and system, wearable device and storage medium |
| WO2021139310A1 (en) * | 2020-07-30 | 2021-07-15 | 平安科技（深圳）有限公司 | Emotion recognition method, apparatus, computer device, and computer-readable storage medium |
| US20220071535A1 (en) * | 2020-09-10 | 2022-03-10 | Frictionless Systems, LLC | Mental state monitoring system |
| CN114615926A (en) * | 2019-09-30 | 2022-06-10 | 马里技术 | Snore detection device and method based on sound analysis |
-
2022
- 2022-11-10 CN CN202211410640.0A patent/CN115886815A/en active Pending
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017160186A1 (en) * | 2016-03-18 | 2017-09-21 | Валерий Викторович КУРЫШЕВ | Method for continuously monitoring a person's stress level |
| CN105943015A (en) * | 2016-06-04 | 2016-09-21 | 浙江大学 | Wearable heart rate variability (HRV) monitoring device with active noise reduction function |
| CN108809463A (en) * | 2017-04-28 | 2018-11-13 | 工业和信息化部电信研究院 | A kind of multiple reference clock weighting synthetic method |
| CN110090029A (en) * | 2018-01-28 | 2019-08-06 | 北京师范大学 | Emotional measurement system, Intelligent bracelet and portable terminal |
| CN110353704A (en) * | 2019-07-12 | 2019-10-22 | 东南大学 | Mood assessments method and apparatus based on wearable ECG monitoring |
| CN114615926A (en) * | 2019-09-30 | 2022-06-10 | 马里技术 | Snore detection device and method based on sound analysis |
| CN111419250A (en) * | 2020-04-08 | 2020-07-17 | 恒爱高科(北京)科技有限公司 | Emotion recognition method based on pulse waves |
| WO2021139310A1 (en) * | 2020-07-30 | 2021-07-15 | 平安科技（深圳）有限公司 | Emotion recognition method, apparatus, computer device, and computer-readable storage medium |
| US20220071535A1 (en) * | 2020-09-10 | 2022-03-10 | Frictionless Systems, LLC | Mental state monitoring system |
| GB202016726D0 (en) * | 2020-10-21 | 2020-12-02 | August Int Ltd | Improvements in or relating to wearable sensor apparatus |
| CN113040771A (en) * | 2021-03-01 | 2021-06-29 | 青岛歌尔智能传感器有限公司 | Emotion recognition method and system, wearable device and storage medium |
| CN113017623A (en) * | 2021-03-04 | 2021-06-25 | 江苏盖睿健康科技有限公司 | Measuring method, device and storage medium for hypoperfusion blood oxygen saturation |
Non-Patent Citations (2)
| Title |
|---|
| 吴雅琴;师兰兰;: "基于LoRa的火灾救援现场人员定位算法研究", 计算机应用与软件, no. 06, 12 June 2020 (2020-06-12) * |
| 李振宇;: "基于IMU和红外传感器网络的室内人体定位方法", 传感器与微系统, no. 12, 5 December 2018 (2018-12-05) * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113749630B (en) | Blood pressure monitoring system and method based on ECG signal and PPG signal | |
| CN113349752A (en) | Wearable device real-time heart rate monitoring method based on sensing fusion | |
| CN106618542A (en) | Denoising heart rate detecting device and method | |
| CN106889981B (en) | A kind of intelligent terminal for being used to extract fetal heart frequency | |
| CN112998690B (en) | Pulse wave multi-feature fusion-based respiration rate extraction method | |
| Kelwade et al. | Prediction of cardiac arrhythmia using artificial neural network | |
| CN109938719B (en) | Driver fatigue detection method based on physiological parameters | |
| CN107361764A (en) | A kind of rapid extracting method of electrocardiosignal signature waveform R ripples | |
| CN110876615B (en) | Real-time dynamic heart rate monitoring device and monitoring method | |
| Almalchy et al. | Noise removal from ECG signal based on filtering techniques | |
| Frei et al. | Least squares acceleration filtering for the estimation of signal derivatives and sharpness at extrema [and application to biological signals] | |
| Elbuni et al. | ECG parameter extraction algorithm using (DWTAE) algorithm | |
| CN117379021B (en) | Old person health index monitoring system based on intelligent wearing equipment | |
| Lin et al. | A characteristic filtering method for pulse wave signal quality assessment | |
| CN113033358A (en) | Cuff-free blood pressure modeling method based on deep neural network | |
| Chang et al. | Cubic spline interpolation with overlapped window and data reuse for on-line hilbert huang transform biomedical microprocessor | |
| CN115886815A (en) | Emotional pressure monitoring method and device and intelligent wearable device | |
| CN115836847A (en) | Blood pressure prediction device and equipment | |
| Outram | Intelligent pattern analysis of the foetal electrocardiogram | |
| EP3652750A1 (en) | Systems, methods, and media for efficient real-time embedded processing of physiological signals using s transforms | |
| CN114098690A (en) | Lightweight pulse wave signal noise removal and heart rate detection method | |
| CN113576438A (en) | Non-invasive blood pressure extraction method and system | |
| Meddah et al. | New system cardiac arrhythmia detection on FPGA | |
| Yazdanpanah et al. | Design and comparison of digital IIR filters for reduction of artifacts from electrocardiogram waveform | |
| Zhu et al. | An ECG detection algorithm using wavelet and autocorrelation transform |
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 |