CN112957013B - Dynamic vital sign signal acquisition system, monitoring device and equipment - Google Patents
Dynamic vital sign signal acquisition system, monitoring device and equipment Download PDFInfo
- Publication number
- CN112957013B CN112957013B CN202110164206.8A CN202110164206A CN112957013B CN 112957013 B CN112957013 B CN 112957013B CN 202110164206 A CN202110164206 A CN 202110164206A CN 112957013 B CN112957013 B CN 112957013B
- Authority
- CN
- China
- Prior art keywords
- vital sign
- signal
- module
- dynamic
- sign signal
- 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
- 238000012806 monitoring device Methods 0.000 title claims abstract description 29
- 238000012545 processing Methods 0.000 claims abstract description 47
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 claims description 52
- 239000008280 blood Substances 0.000 claims description 52
- 210000004369 blood Anatomy 0.000 claims description 52
- 229910052760 oxygen Inorganic materials 0.000 claims description 52
- 239000001301 oxygen Substances 0.000 claims description 52
- 108010076504 Protein Sorting Signals Proteins 0.000 claims description 29
- 238000001914 filtration Methods 0.000 claims description 29
- 230000008030 elimination Effects 0.000 claims description 18
- 238000003379 elimination reaction Methods 0.000 claims description 18
- 238000012549 training Methods 0.000 claims description 18
- 238000012544 monitoring process Methods 0.000 claims description 14
- 230000009466 transformation Effects 0.000 claims description 13
- 238000001514 detection method Methods 0.000 claims description 9
- 238000009966 trimming Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 5
- 230000002159 abnormal effect Effects 0.000 claims description 4
- 238000004891 communication Methods 0.000 claims description 3
- 238000013138 pruning Methods 0.000 claims description 2
- 238000000034 method Methods 0.000 description 11
- 238000005259 measurement Methods 0.000 description 10
- 230000008569 process Effects 0.000 description 8
- 238000010586 diagram Methods 0.000 description 5
- 230000009471 action Effects 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000007670 refining Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 239000000126 substance Substances 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- QVFWZNCVPCJQOP-UHFFFAOYSA-N chloralodol Chemical compound CC(O)(C)CC(C)OC(O)C(Cl)(Cl)Cl QVFWZNCVPCJQOP-UHFFFAOYSA-N 0.000 description 1
- 210000001061 forehead Anatomy 0.000 description 1
- 238000009532 heart rate measurement Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000003205 muscle Anatomy 0.000 description 1
- 230000003183 myoelectrical effect Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 208000010540 rapid respiration Diseases 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Images
Classifications
-
- 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
- A61B5/7207—Signal processing specially adapted for physiological signals or for diagnostic purposes for noise prevention, reduction or removal of noise induced by motion artifacts
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0002—Remote monitoring of patients using telemetry, e.g. transmission of vital signals via a communication network
- A61B5/0015—Remote monitoring of patients using telemetry, e.g. transmission of vital signals via a communication network characterised by features of the telemetry system
- A61B5/0022—Monitoring a patient using a global network, e.g. telephone networks, internet
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
- A61B5/0205—Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/145—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue
- A61B5/14542—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue for measuring blood gases
-
- 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
- A61B5/7207—Signal processing specially adapted for physiological signals or for diagnostic purposes for noise prevention, reduction or removal of noise induced by motion artifacts
- A61B5/721—Signal processing specially adapted for physiological signals or for diagnostic purposes for noise prevention, reduction or removal of noise induced by motion artifacts using a separate sensor to detect motion or using motion information derived from signals other than the physiological signal to be measured
-
- 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
- A61B5/7253—Details of waveform analysis characterised by using transforms
-
- 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
- A61B5/7253—Details of waveform analysis characterised by using transforms
- A61B5/7257—Details of waveform analysis characterised by using transforms using Fourier transforms
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Abstract
The invention relates to a dynamic vital sign signal acquisition system, a monitoring device and equipment, wherein the system comprises a dynamic vital sign signal acquisition module, a dynamic vital sign signal processing module and a dynamic vital sign signal processing module, wherein the dynamic vital sign signal acquisition module is used for acquiring a vital sign signal of an individual in a motion state to obtain a vital sign signal containing motion noise; the inertial navigation signal acquisition module is used for synchronously acquiring inertial navigation signals of an individual in a motion state; the denoising module is used for analyzing and processing the inertial navigation signal and the vital sign signal containing the motion noise which are synchronously acquired based on a denoising network model trained in advance to obtain the vital sign signal without the motion noise. According to the invention, the denoising module based on the denoising network model is combined with the inertial navigation signal to perform motion denoising processing on the individual vital sign signals in motion, so that accurate vital sign signals can be obtained, and the influence of motion on accurate acquisition of vital sign signals is overcome.
Description
Technical Field
The invention relates to the field of vital sign signal monitoring, in particular to a dynamic vital sign signal acquisition system, a monitoring device and equipment.
Background
Degree of blood oxygen saturation (SpO) 2 ) Heart rate (BP) is a key indicator of constant human vital signs. Degree of blood oxygen saturation (SpO) 2 ) Existing measurement schemes related to heart rate (BP) tend to fall into three modes: i.e., the sensor is placed at the forehead, ear, or finger tip. The common measurement scheme is finger measurement, and the muscle at the finger end is less, so that the myoelectric interference in the measurement process is less, and the stability is better. The measurement results are often displayed directly on the measurement device, and are not transmitted to a remote terminal, nor are they analyzed. In the measuring process, the person is required to be in a static state and cannot move, otherwise, the measuring effect is very inaccurate. The existing measurement scheme cannot solve the blood oxygen and heart rate measurement of people under the dynamic condition. At present, there is no oneThe product can carry out the measurement of blood oxygen under the motion state, and its root cause lies in the photoelectric tube that blood oxygen device adopted in order to accomplish to measure accurately, often makes very high with its perception sensitivity, and such benefit is that it can measure the small change of blood oxygen value, and then improves measurement accuracy. However, a side effect of this arrangement is that it is very sensitive to the effects of other disturbances, such as motion disturbances. The multiplicative noise caused by the motion disturbance to the blood oxygen signal is often very large, and the useful blood oxygen signal is submerged under the noise. These noises often do not have a fixed frequency or other characteristics and are therefore difficult to filter by conventional signal processing means.
Disclosure of Invention
The invention aims to provide a dynamic vital sign signal acquisition system, a monitoring device and equipment, which can accurately acquire vital sign signals of individuals in motion and monitor vital sign values and overcome the influence of motion on the accurate acquisition of the vital sign signals.
The technical scheme for solving the technical problems is as follows: a dynamic vital sign signal acquisition system comprises the following modules,
the dynamic vital sign signal acquisition module is used for acquiring vital sign signals of an individual in a motion state to obtain the vital sign signals containing motion noise;
the inertial navigation signal acquisition module is used for synchronously acquiring inertial navigation signals of an individual in a motion state;
and the denoising module is used for analyzing and processing the synchronously acquired inertial navigation signals and the vital sign signals containing the motion noise based on a pre-trained denoising network model to obtain the vital sign signals without the motion noise.
Based on the dynamic vital sign signal acquisition system, the invention also provides a dynamic vital sign monitoring device.
A dynamic vital signs monitoring device comprises a dynamic vital signs signal processing system and the dynamic vital signs signal acquisition system as described above;
the dynamic vital sign signal processing system is used for processing the vital sign signals which are collected by the dynamic vital sign signal collecting system and are removed of the motion noise to obtain vital sign values;
the vital sign signals with the motion noise removed comprise blood oxygen signals and heart rate signals with the motion noise removed, and correspondingly, the vital sign values comprise blood oxygen values and heart rate values;
the dynamic vital sign signal processing system comprises in particular the following modules,
the wavelet transformation module is used for performing wavelet transformation processing on the original sequence of the vital sign signals with the motion noise removed to obtain a wavelet transformation signal sequence;
a Fourier transform module for obtaining a heart rate value by performing Fourier transform on the wavelet transform signal sequence;
the difference processing module is used for carrying out difference processing on the wavelet transformation signal sequence to obtain a difference signal sequence;
the sliding window filtering module is used for performing sliding window filtering on the differential signal sequence to obtain a sliding window filtering signal sequence;
the zero-crossing detection module is used for carrying out zero-crossing detection on the sliding window filtering signal sequence to obtain an extreme point set;
the redundancy elimination module is used for eliminating abnormal extreme points in the extreme point set to obtain a redundancy elimination extreme point set;
an extreme value fine-trimming module, configured to perform fine trimming on positions of the redundant elimination extreme points in the redundant elimination extreme point set to obtain a fine-trimmed extreme point set;
the R value calculation module is used for calculating an initial blood oxygen value according to the refined extreme point set;
and the Kalman filtering module is used for carrying out Kalman filtering on the initial blood oxygen value to obtain a final blood oxygen value after filtering.
Based on the dynamic vital sign monitoring device, the invention also provides dynamic vital sign monitoring equipment.
A dynamic vital sign monitoring device comprises an intelligent gateway, a remote monitoring terminal and the dynamic vital sign monitoring device;
the dynamic vital sign monitoring device is in communication connection with the remote monitoring terminal through the intelligent gateway.
The invention has the beneficial effects that: in the dynamic vital sign signal acquisition system, the denoising module based on the denoising network model is combined with the inertial navigation signal to perform denoising processing on the vital sign signal of an individual in motion, so that an accurate vital sign signal can be obtained, the influence of motion on accurate acquisition of the vital sign signal is overcome, a guarantee can be provided for a dynamic vital sign monitoring device to accurately calculate the vital sign value, and the vital sign value of the individual in motion is accurately monitored; in addition, the dynamic vital sign monitoring device transmits the monitored vital sign value to the remote monitoring terminal through the intelligent gateway, so that the vital signs can be remotely monitored.
Drawings
Fig. 1 is a block diagram of a dynamic vital sign signal acquisition system according to the present invention;
FIG. 2 is a block diagram of a denoising network model training module;
FIG. 3 is a frame diagram of an initial denoised network model;
FIG. 4 is a block diagram of a dynamic vital signs monitoring device according to the present invention;
fig. 5 is a block diagram of a dynamic vital signs monitoring device according to the present invention.
Detailed Description
The principles and features of this invention are described below in conjunction with the following drawings, which are set forth to illustrate, but are not to be construed to limit the scope of the invention.
As shown in fig. 1, a dynamic vital sign signal acquisition system comprises the following modules,
the dynamic vital sign signal acquisition module is used for acquiring vital sign signals of an individual in a motion state to obtain the vital sign signals containing motion noise;
the inertial navigation signal acquisition module is used for synchronously acquiring inertial navigation signals of the individual in a motion state;
and the denoising module is used for analyzing and processing the synchronously acquired inertial navigation signals and the vital sign signals containing the motion noise based on a pre-trained denoising network model to obtain the vital sign signals without the motion noise.
It should be noted that: in the signal acquisition process, the acquisition actions of the dynamic vital sign signal acquisition module and the inertial navigation signal acquisition module are strictly synchronous so as to overcome the influence caused by the decoupling motion of a subsequent denoising network model.
In this particular embodiment: the dynamic vital sign signal acquisition system further comprises a denoising network model training module, wherein the denoising network model training module is used for training the initial denoising network model to obtain a trained denoising network model;
as shown in fig. 2, the denoised network model training module specifically includes the following units,
the dynamic vital sign signal sample acquisition unit is used for acquiring a vital sign signal of a sample in a motion state to obtain a vital sign signal sample containing motion noise;
the inertial navigation signal sample acquisition unit is used for synchronously acquiring an inertial navigation signal sample of the sample in a motion state;
the real vital sign signal sample acquisition unit is used for synchronously acquiring the vital sign signals of the samples at the moment when the motion state stops to obtain real vital sign signal samples;
and the denoising network model training unit is used for inputting the real vital sign signal sample, the inertial navigation signal sample and the vital sign signal sample containing the motion noise which are synchronously acquired into the initial denoising network model for training to obtain the trained denoising network model.
The denoising network model adopted by the invention is a deep neural network, so a large number of labeled signal samples are needed to train the deep neural network, the acquisition of the signal samples is extremely troublesome, the real vital sign signals of an individual in the motion process need to be accurately measured, and the problem to be solved by the invention is that no special equipment exists at present. The blood oxygen signal interfered in the motion state and the corresponding real value are not existed at present. In order to solve the problem of acquiring enough data, the invention provides a new acquisition method. Firstly: the method comprises the following steps that a person is in a motion state including walking, running, shaking and the like, and at the moment, a dynamic vital sign signal sample collecting unit starts to collect vital sign signal samples (including motion noise); then, at some point, the individual is allowed to stop momentarily, such as lying in a comfortable place, while still collecting vital sign signal samples (containing no motion noise). Since the vital signs of the individual cannot change suddenly at the moment of stopping the exercise, when the stopping time is very short, such as 10 seconds or 30 seconds, the vital signs of the individual, including the blood oxygen value, can be the same as the value at the moment of the previous exercise state, so that the purpose of acquiring the real vital sign signal sample of the individual in the exercise state is achieved. In order to collect the vital sign signal samples, two collecting units, namely a dynamic vital sign signal sample collecting unit and a real vital sign signal sample collecting unit, can be deployed on the sample body, the dynamic vital sign signal sample collecting unit is disposed at the head end, and the real vital sign signal sample collecting unit is disposed at the finger end. When the finger tip is still, the influence caused by rapid respiration is small, so that the finger tip can be closer to the real vital sign. In the process of collecting the vital sign signal samples, the time of the motion state collection process is long, and the stop state can only be kept for a short time, such as 10s or even shorter, in order to ensure that the measured vital sign signal samples are close to the real motion state.
It should be noted that: in the signal acquisition process, the acquisition actions of the dynamic vital sign signal sample acquisition unit, the inertial navigation signal sample acquisition unit and the real vital sign signal sample acquisition unit are strictly synchronous, so that the training precision of the denoising network model is improved.
In addition, the dynamic vital sign signal sample acquisition unit and the dynamic vital sign signal acquisition module can be the same device, and the inertial navigation signal sample acquisition unit and the inertial navigation signal acquisition module can be the same device.
As shown in fig. 3, the initial denoising network model is specifically a UNet network with skip connection (residual connection) (UNet is a well-known image segmentation network in medical image processing), the UNet network includes a plurality of stacked coding layers and a plurality of stacked decoding layers, and further includes an output layer, the bottommost coding layer is connected to the inertial navigation signal samples and the vital sign signal samples containing motion noise, and the bottommost decoding layer is connected to the output layer; the encoding layer at the top layer is connected with the decoding layer at the top layer through an embedding layer, and the embedding layer is specifically an LSTM layer (the LSTM is a long-short term memory network and is a time cycle neural network); the LSTM layer is accessed with a feature loss layer, and the feature loss layer is accessed with a real vital sign signal sample synchronously acquired through a pre-training network layer; the output layer has access to a loss of vital sign (R, BP) layer.
Specifically, the feature loss signal generated in the feature loss layer is obtained by a distance L1 between the first low-dimensional signal and the second low-dimensional signal; wherein the first low-dimensional signal is obtained by directly extracting the real vital sign signal sample through the pre-training network layer, and the second low-dimensional signal is generated by the inertial navigation signal sample and the vital sign signal sample containing motion noise;
the vital sign value loss signal generated in the vital sign value loss layer is obtained from the distance L2 between the real vital sign signal sample and the output of the output layer at the calculated vital sign signal (in other words, the vital sign value loss signal generated in the vital sign value loss layer is obtained from the distance L2 between a and B, where a is the real vital sign signal sample and B is the calculated vital sign signal at the output of the output layer).
In other embodiments, the denoising network model may also use a cGAN network, and may also use other networks commonly used for images based on STFT input, such as a complex network.
In this particular embodiment: the vital sign signals are specifically blood oxygen signals and heart rate signals;
the dynamic vital sign signal sample acquisition unit is specifically used for acquiring a blood oxygen signal and a heart rate signal of a head end of a sample in a motion state to obtain a vital sign signal sample containing motion noise;
the real vital sign signal sample acquisition unit is specifically used for synchronously acquiring blood oxygen signals and heart rate signals of finger ends of the samples at the moment when the motion state stops, so as to obtain real vital sign signal samples.
Further, the dynamic vital sign signal acquisition module is specifically configured to acquire an blood oxygen signal and a heart rate signal of a head end of an individual in a motion state, so as to obtain a vital sign signal including motion noise.
Further, the expression of the entire loss signal composed of the feature loss signal and the vital sign value loss signal is,
wherein R is F And f F Respectively, a blood oxygen value and a heart rate value, R, obtained from the real vital sign signal samples H And f H Respectively a blood oxygen value and a heart rate value obtained after a vital sign signal sample containing motion noise passes through a Generator (a term of a denoising network model),
and
respectively calculating the ith element of the embedding layer obtained by the real vital sign signal sample and the vital sign signal sample containing the motion noise; alpha and beta are respectively hyper-parameters (hyper-parameters are parameters which can be adjusted), and L is the whole ios signal.
In this particular embodiment: the dynamic vital sign signal acquisition system further comprises a denoising network model integration module, wherein the denoising network model embedding module is used for compressing the network weight by quantizing, pruning and knowledge distilling the network weight of the trained denoising network model, and integrating the trained denoising network model with the compressed network weight into an embedded network to obtain the denoising module;
the denoising module is specifically used for operating the trained denoising network model with the compressed network weight in the embedded network, and analyzing and processing the inertial navigation signal and the vital sign signal containing the motion noise which are synchronously acquired, so as to obtain the vital sign signal without the motion noise.
Since the vital sign signal output by the denoising network model and subjected to motion noise removal is a waveform, and is not the final required vital sign values, such as the R value (blood oxygen value) and the BP value (heart rate value), further processing is required on the vital sign signal subjected to motion noise removal. At this time, the motion noise in the vital sign signal from which the motion noise is removed has been filtered, so that the following dynamic vital sign monitoring apparatus may be adopted to further process the vital sign signal from which the motion noise is removed, so as to obtain the vital sign value.
As shown in fig. 4, a dynamic vital signs monitoring device comprises a dynamic vital signs signal processing system and the dynamic vital signs signal acquisition system as described above;
the dynamic vital sign signal processing system is used for processing the vital sign signals which are collected by the dynamic vital sign signal collecting system and are removed of the motion noise to obtain vital sign values; the dynamic vital sign signal processing system can adopt an ultra-low power consumption singlechip or other embedded processors.
The vital sign signals with the motion noise removed comprise blood oxygen signals and heart rate signals with the motion noise removed, and correspondingly, the vital sign values comprise blood oxygen values and heart rate values;
the dynamic vital sign signal processing system comprises in particular the following modules,
the wavelet transformation module is used for performing wavelet transformation processing on the original sequence of the vital sign signals with the motion noise removed to obtain a wavelet transformation signal sequence;
a Fourier transform module for obtaining a heart rate value by performing Fourier transform on the wavelet transform signal sequence;
the difference processing module is used for carrying out difference processing on the wavelet transformation signal sequence to obtain a difference signal sequence;
the sliding window filtering module is used for performing sliding window filtering on the differential signal sequence to obtain a sliding window filtering signal sequence;
the zero-crossing detection module is used for carrying out zero-crossing detection on the sliding window filtering signal sequence to obtain an extreme point set;
the redundancy elimination module is used for eliminating abnormal extreme points in the extreme point set to obtain a redundancy elimination extreme point set;
an extreme value fine-trimming module, configured to perform fine trimming on positions of the redundant elimination extreme points in the redundant elimination extreme point set to obtain a fine-trimmed extreme point set;
the R value calculation module is used for calculating an initial blood oxygen value according to the refined extreme point set;
and the Kalman filtering module is used for carrying out Kalman filtering on the initial blood oxygen value to obtain a final blood oxygen value after filtering.
In this particular embodiment:
the dynamic vital sign signal processing system further comprises a first-in first-out queue module,
the first-in first-out queue module is used for storing the detected vital sign signals with the motion noise removed to obtain an original sequence of the vital sign signals with the motion noise removed, and transmitting the original sequence of the vital sign signals with the motion noise removed to the wavelet transform module in a first-in first-out mode after the original sequence is fully stored;
wherein, the length of the first-in first-out queue in the first-in first-out queue module is N, then the first-in first-out queue moduleThe motion noise-removed vital sign signal original sequence obtained by storing the motion noise-removed vital sign signal in the de-queue module is represented as
Original sequence of vital sign signals for removing motion noise, x n For the nth motion noise removed vital sign signal in the original sequence of motion noise removed vital sign signals, N =1,2, ·, N;
the method is characterized in that a small amount of high-frequency noise often exists in an original sequence of the vital sign signal for removing the motion noise under the action of electromyographic signals or external electromagnetic interference, and the specific expression is that a plurality of burrs exist in the signal, if the burr signals are not eliminated, the subsequent peak detection is influenced, and therefore the finally calculated blood oxygen value is greatly influenced. In order to eliminate the high-frequency interference, the original sequence of the vital sign signals with the motion noise removed enters a wavelet transform module to be subjected to wavelet transform processing.
In the wavelet transform module, the wavelet transform signal sequence obtained after the wavelet transform processing is performed on the original sequence x of the vital sign signals without the motion noise is represented as
And is
x′ n Is the n-th wavelet transform signal in the wavelet transform signal sequence, specifically, x' n Vital sign signal x for removing motion noise n And obtaining a wavelet transform signal after wavelet transform processing.
After the original sequence of the vital sign signals without the motion noise is subjected to wavelet transform processing by the wavelet transform module, most high-frequency noise is filtered out, and a wavelet transform signal is obtained. Because the heart rate signal is a periodic signal, the wavelet transform signal obtained after filtering is subjected to Fourier transform at the moment, and the heart rate value can be obtained.
In the Fourier transform module, the heart rate value is calculated by the formula,
BP＝arg max(X)，s.t.L＜X＜H；
wherein, BP is the heart rate value, X is a Fourier transform signal obtained by fast Fourier transform of the wavelet transform signal sequence, and
f () represents a fast fourier transform, and L and H represent a boundary minimum value and a boundary maximum value to be sought in the fast fourier transform, respectively. The normal fluctuation range of the human body is recorded as 60-100 times/minute,
where N represents the number of points used in the calculation of the FFT (fast Fourier transform), i.e. the length of the FIFO queue, F s Representing the sampling rate.
To obtain blood oxygen values, wavelet transformed signals
Firstly, the difference processing is carried out in a difference processing module.
In the difference processing module, the formula for performing difference processing on the wavelet transform signal sequence is as follows,
wherein the content of the first and second substances,
for the purpose of the sequence of differential signals,
and is a differential signal in the sequence of differential signals.
In the sliding window filtering module, the obtained sliding window filtering signal sequence is represented as
The sliding window filtering of the differential signal sequence is formulated as,
wherein the content of the first and second substances,
filtering the signal sequence for the sliding window
The kth sliding window of (b) filters the signal, W being the size of the sliding window.
In the zero-crossing detection module, the formula for performing zero-crossing detection on the sliding window filtering signal sequence is as follows,
wherein ZP is the extreme point set.
The extreme point set ZP contains all the extreme points, and due to the existence of noise, these extreme points may contain some local extreme values, and do not represent that it is an extreme value, and a local maximum or a local minimum value that is required for calculating the blood oxygen value, so it is necessary to further filter the extreme points in the extreme point set ZP. The extreme points of these anomalies are often more concentrated and less spaced by a large number of measurements, so these local extreme values are removed by using this information of spacing.
In the redundancy elimination module, the formula for eliminating the abnormal extreme points in the extreme point set is as follows,
wherein ZP' is the set of redundancy elimination extrema, I k Is composed of
Corresponding serial numbers, i.e.
In the position of I k -I k-1 To represent
And with
H is a preset first threshold.
The extreme point in the redundancy elimination extreme point set obtained after the processing of the redundancy elimination module is an extreme point which may be needed and is also a maximum point; however, since the sliding window filtering is adopted in the foregoing step, the positions of these extreme points need to be further refined.
In the extreme value refining module, the formula for refining the position of the redundancy elimination extreme point in the redundancy elimination extreme point set is as follows,
wherein ZP' is the finishing extreme point set, M is a preset second threshold value,
represents the sequence number I k The corresponding blood oxygen value is measured by the blood oxygen measuring instrument,
represent
Is a local extremum.
And the refined extreme points in the refined extreme point set obtained after the processing of the extreme value refining module are used as the extreme points for finally calculating the blood oxygen value.
In the R value calculation module, the formula for calculating the initial blood oxygen value is as follows,
wherein R is i Is the initial blood oxygen value, x i，max For the local pole, maximum, x, obtained from the IR sensor in the refined extreme point set ZP ″ i，min For the local pole, minimum, y from the IR sensor in the refined extreme point set ZP ″ i，max For the local pole, maximum, y, from the RED sensor in the refined extreme point set ZP ″ i，min Local poles and minimum values obtained from the RED sensor in the fine extreme point set ZP' are obtained; RED and IR sensors are sensors used to acquire vital sign signals containing motion noise.
In the Kalman filtering module, the final blood oxygen value is calculated by the formula,
R(k)＝R′(k)+K(R z (k)-H*R′(k))；
wherein R (k) is the final blood oxygen value calculated at the time k, R z (k) For the blood oxygen value measured at time K, R' (K) is the blood oxygen value updated at time K, K is kalman gain and K =1,H is the transformation matrix and H =1; r' (K) = AR (K-1), K = HP/(HPH) T +R)，P＝AP′A T + Q, P '= P-HKP, P is estimated state variable, P' is updated state variable, R is the initial blood oxygen value R i Obtaining a blood oxygen value after Kalman filtering, wherein A is an update matrix and A =1,Q is an observed covariance;
in particular, the alpha scale factor, here relates the covariance to the value of the change in the observed variable.
Based on the dynamic vital sign monitoring device, the invention also provides dynamic vital sign monitoring equipment.
As shown in fig. 5, a dynamic vital signs monitoring device includes an intelligent gateway, a remote monitoring terminal, and the dynamic vital signs monitoring apparatus as described above;
the dynamic vital sign monitoring device is in communication connection with the remote monitoring terminal through the intelligent gateway.
In the dynamic vital sign monitoring device of the present invention, the dynamic vital sign monitoring device is provided with a wireless transmission module, such as LoRa, NBIoT or a low power consumption high speed module, such as bluetooth, and the dynamic vital sign monitoring device transmits the measured vital sign value to the intelligent gateway through the wireless transmission module. The intelligent gateway is used for receiving the vital sign values transmitted by different dynamic vital sign monitoring devices and forwarding the vital sign values to the remote monitoring terminal for analysis and early warning.
In the dynamic vital sign signal acquisition system, the denoising module based on the denoising network model is combined with the inertial navigation signal to perform denoising processing on the vital sign signals of the individual in motion, so that accurate vital sign signals can be obtained, a guarantee can be provided for a dynamic vital sign monitoring device to accurately calculate the vital sign values, and the vital sign values of the individual in motion can be accurately monitored; in addition, the dynamic vital sign monitoring device transmits the monitored vital sign value to the remote monitoring terminal through the intelligent gateway, so that the vital signs can be remotely monitored.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (6)
1. A dynamic vital sign signal acquisition system, characterized by: comprises the following modules which are used for realizing the functions of the system,
the dynamic vital sign signal acquisition module is used for acquiring vital sign signals of an individual in a motion state to obtain the vital sign signals containing motion noise;
the inertial navigation signal acquisition module is used for synchronously acquiring inertial navigation signals of the individual in a motion state;
the denoising module is used for analyzing and processing the inertial navigation signal and the vital sign signal containing the motion noise which are synchronously acquired based on a pre-trained denoising network model to obtain the vital sign signal without the motion noise;
the system also comprises a denoising network model training module, wherein the denoising network model training module is used for training the initial denoising network model to obtain a trained denoising network model;
the denoised network model training module comprises in particular the following elements,
the dynamic vital sign signal sample acquisition unit is used for acquiring a vital sign signal of a sample in a motion state to obtain a vital sign signal sample containing motion noise;
the inertial navigation signal sample acquisition unit is used for synchronously acquiring an inertial navigation signal sample of the sample in a motion state;
the real vital sign signal sample acquisition unit is used for synchronously acquiring the vital sign signals of the samples at the moment when the motion state stops to obtain real vital sign signal samples;
the denoising network model training unit is used for inputting a real vital sign signal sample, an inertial navigation signal sample and a vital sign signal sample containing motion noise which are synchronously acquired into the initial denoising network model for training to obtain a trained denoising network model;
the initial denoising network model is specifically a UNet network with skip connection, the UNet network comprises a plurality of stacked coding layers, a plurality of stacked decoding layers and an output layer, the coding layer at the bottom layer is connected with an inertial navigation signal sample and a vital sign signal sample containing motion noise, which are synchronously acquired, and the decoding layer at the bottom layer is connected with the output layer; the encoding layer at the top layer is connected with the decoding layer at the top layer through an embedding layer, and the embedding layer is specifically an LSTM layer; the LSTM layer is accessed with a featurelos layer, and the featurelos layer is accessed with real vital sign signal samples which are synchronously acquired through a pre-training network layer; the output layer is accessed with a life characteristic value loss layer;
the featurelos signal generated in the featurelos layer is obtained by the distance L1 between the first low-dimensional signal and the second low-dimensional signal; wherein the first low-dimensional signal is obtained by directly extracting the real vital sign signal sample through the pre-training network layer, and the second low-dimensional signal is generated by the inertial navigation signal sample and the vital sign signal sample containing motion noise;
the vital sign value loss signal generated in the vital sign value loss layer is obtained by the distance L2 between the real vital sign signal sample and the output of the output layer in the calculated vital sign signal;
the expression of the entire loss signal formed by the featureless signal and the loss signal is,
wherein R is F And f F Respectively, a blood oxygen value and a heart rate value, R, obtained from the real vital sign signal samples H And f H Respectively is a blood oxygen value and a heart rate value obtained after a vital sign signal sample containing motion noise passes through the denoising network model,
and
respectively calculating the ith element of the embedding layer obtained by the real vital sign signal sample and the vital sign signal sample containing the motion noise; both alpha and beta are hyperparameters, and L is the entire loss signal.
2. The dynamic vital sign signal acquisition system of claim 1, wherein: the vital sign signals are specifically blood oxygen signals and heart rate signals;
the dynamic vital sign signal sample acquisition unit is specifically used for acquiring a blood oxygen signal and a heart rate signal of a head end of a sample in a motion state to obtain a vital sign signal sample containing motion noise;
the real vital sign signal sample acquisition unit is specifically used for synchronously acquiring blood oxygen signals and heart rate signals of finger ends of the samples at the moment when the motion state stops, so as to obtain real vital sign signal samples.
3. The dynamic vital sign signal acquisition system of claim 1, wherein: the dynamic vital sign signal acquisition module is specifically used for acquiring blood oxygen signals and heart rate signals of the head end of an individual in a motion state to obtain vital sign signals containing motion noise.
4. Dynamic vital sign signal acquisition system according to any one of claims 1 to 3, wherein: the denoising network model embedding module is used for compressing the network weight by quantizing, pruning and knowledge distilling the network weight of the trained denoising network model, and integrating the trained denoising network model with the compressed network weight into an embedded network to obtain the denoising module;
the denoising module is specifically used for operating the trained denoising network model with the compressed network weight in the embedded network, and analyzing and processing the inertial navigation signal and the vital sign signal containing the motion noise which are synchronously acquired, so as to obtain the vital sign signal without the motion noise.
5. A dynamic vital sign monitoring device, comprising: comprising a dynamic vital signs signal processing system and a dynamic vital signs signal acquisition system as claimed in any one of claims 1 to 4;
the dynamic vital sign signal processing system is used for processing the vital sign signals which are collected by the dynamic vital sign signal collecting system and are removed of the motion noise to obtain vital sign values;
the vital sign signals with the motion noise removed comprise blood oxygen signals and heart rate signals with the motion noise removed, and correspondingly, the vital sign values comprise blood oxygen values and heart rate values;
the dynamic vital sign signal processing system comprises in particular the following modules,
the wavelet transformation module is used for performing wavelet transformation processing on the original sequence of the vital sign signals with the motion noise removed to obtain a wavelet transformation signal sequence;
a Fourier transform module for obtaining a heart rate value by performing Fourier transform on the wavelet transform signal sequence;
the differential processing module is used for carrying out differential processing on the wavelet transformation signal sequence to obtain a differential signal sequence;
the sliding window filtering module is used for performing sliding window filtering on the differential signal sequence to obtain a sliding window filtering signal sequence;
the zero-crossing detection module is used for carrying out zero-crossing detection on the sliding window filtering signal sequence to obtain an extreme point set;
the redundancy elimination module is used for eliminating abnormal extreme points in the extreme point set to obtain a redundancy elimination extreme point set;
an extreme value fine-trimming module, configured to perform fine trimming on positions of the redundant elimination extreme points in the redundant elimination extreme point set to obtain a fine-trimmed extreme point set;
the R value calculation module is used for calculating an initial blood oxygen value according to the refined extreme point set;
and the Kalman filtering module is used for carrying out Kalman filtering on the initial blood oxygen value to obtain a filtered final blood oxygen value.
6. A dynamic vital signs monitoring device, characterized by: comprising an intelligent gateway and a remote monitoring terminal and a dynamic vital signs monitoring device according to claim 5;
the dynamic vital sign monitoring device is in communication connection with the remote monitoring terminal through the intelligent gateway.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110164206.8A CN112957013B (en) | 2021-02-05 | 2021-02-05 | Dynamic vital sign signal acquisition system, monitoring device and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110164206.8A CN112957013B (en) | 2021-02-05 | 2021-02-05 | Dynamic vital sign signal acquisition system, monitoring device and equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112957013A CN112957013A (en) | 2021-06-15 |
| CN112957013B true CN112957013B (en) | 2022-11-11 |
Family
ID=76274836
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110164206.8A Active CN112957013B (en) | 2021-02-05 | 2021-02-05 | Dynamic vital sign signal acquisition system, monitoring device and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112957013B (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101803925A (en) * | 2010-03-31 | 2010-08-18 | 上海交通大学 | Monitoring device of blood oxygen saturation in motion state |
| CN105534517A (en) * | 2016-02-29 | 2016-05-04 | 浙江铭众科技有限公司 | Method for removing vehicle motion noise in three-lead electrocardiosignal |
| CN107423376A (en) * | 2017-07-10 | 2017-12-01 | 上海交通大学 | One kind has the quick picture retrieval method of supervision depth Hash and system |
| CN108742538A (en) * | 2018-06-20 | 2018-11-06 | 大国创新智能科技(东莞)有限公司 | Sign measurement method and medical robotic system based on big data and artificial intelligence |
| CN109086726A (en) * | 2018-08-10 | 2018-12-25 | 陈涛 | A kind of topography's recognition methods and system based on AR intelligent glasses |
| CN110084216A (en) * | 2019-05-06 | 2019-08-02 | 苏州科达科技股份有限公司 | Human face recognition model training and face identification method, system, equipment and medium |
| CN110163347A (en) * | 2019-05-24 | 2019-08-23 | 刘斌 | A kind of underground coal mine human body attitude monitoring method |
| LU100925B1 (en) * | 2018-09-10 | 2020-03-10 | Iee Sa | Removing noise caused by vehicular movement from sensor signals using Deep Neural Networks |
| CN111695673A (en) * | 2020-05-09 | 2020-09-22 | 华为技术有限公司 | Method for training neural network predictor, image processing method and device |
| CN112041856A (en) * | 2018-03-01 | 2020-12-04 | 皇家飞利浦有限公司 | Cross-modal neural network for prediction |
| CN112057066A (en) * | 2020-09-03 | 2020-12-11 | 青岛歌尔智能传感器有限公司 | Heart rate detection method, wearable device and computer storage medium |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070219059A1 (en) * | 2006-03-17 | 2007-09-20 | Schwartz Mark H | Method and system for continuous monitoring and training of exercise |
| US8725669B1 (en) * | 2010-08-02 | 2014-05-13 | Chi Yung Fu | Signal processing method and apparatus |
| WO2018071591A1 (en) * | 2016-10-12 | 2018-04-19 | The Brigham And Women's Hospital, Inc. | Deployable local magnetic resonance imaging coil and methods for use thereof |
| US10607319B2 (en) * | 2017-04-06 | 2020-03-31 | Pixar | Denoising monte carlo renderings using progressive neural networks |
| US10664979B2 (en) * | 2018-09-14 | 2020-05-26 | Siemens Healthcare Gmbh | Method and system for deep motion model learning in medical images |
| US20200388287A1 (en) * | 2018-11-13 | 2020-12-10 | CurieAI, Inc. | Intelligent health monitoring |
| US11055575B2 (en) * | 2018-11-13 | 2021-07-06 | CurieAI, Inc. | Intelligent health monitoring |
| CN111179961B (en) * | 2020-01-02 | 2022-10-25 | 腾讯科技（深圳）有限公司 | Audio signal processing method and device, electronic equipment and storage medium |
| CN112150442A (en) * | 2020-09-25 | 2020-12-29 | 帝工(杭州)科技产业有限公司 | New crown diagnosis system based on deep convolutional neural network and multi-instance learning |
-
2021
- 2021-02-05 CN CN202110164206.8A patent/CN112957013B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101803925A (en) * | 2010-03-31 | 2010-08-18 | 上海交通大学 | Monitoring device of blood oxygen saturation in motion state |
| CN105534517A (en) * | 2016-02-29 | 2016-05-04 | 浙江铭众科技有限公司 | Method for removing vehicle motion noise in three-lead electrocardiosignal |
| CN107423376A (en) * | 2017-07-10 | 2017-12-01 | 上海交通大学 | One kind has the quick picture retrieval method of supervision depth Hash and system |
| CN112041856A (en) * | 2018-03-01 | 2020-12-04 | 皇家飞利浦有限公司 | Cross-modal neural network for prediction |
| CN108742538A (en) * | 2018-06-20 | 2018-11-06 | 大国创新智能科技(东莞)有限公司 | Sign measurement method and medical robotic system based on big data and artificial intelligence |
| CN109086726A (en) * | 2018-08-10 | 2018-12-25 | 陈涛 | A kind of topography's recognition methods and system based on AR intelligent glasses |
| LU100925B1 (en) * | 2018-09-10 | 2020-03-10 | Iee Sa | Removing noise caused by vehicular movement from sensor signals using Deep Neural Networks |
| CN110084216A (en) * | 2019-05-06 | 2019-08-02 | 苏州科达科技股份有限公司 | Human face recognition model training and face identification method, system, equipment and medium |
| CN110163347A (en) * | 2019-05-24 | 2019-08-23 | 刘斌 | A kind of underground coal mine human body attitude monitoring method |
| CN111695673A (en) * | 2020-05-09 | 2020-09-22 | 华为技术有限公司 | Method for training neural network predictor, image processing method and device |
| CN112057066A (en) * | 2020-09-03 | 2020-12-11 | 青岛歌尔智能传感器有限公司 | Heart rate detection method, wearable device and computer storage medium |
Non-Patent Citations (10)
| Title |
|---|
| Arsalan Mosenia等.Wearable Medical Sensor-Based System Design: A Survey.《IEEE Transactions on Multi-Scale Computing Systems》.2017,第3卷(第2期), * |
| Guillaume Cathelain等.U-Net Neural Network for Heartbeat Detection in Ballistocardiography.《2020 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC)》.2020, * |
| Yi Wang等.Online Subspace Learning from Gradient Orientations for Robust Image Alignment.《IEEE Transactions on Image Processing》.2019,第28卷(第7期), * |
| 张伟.消防现场信息记录系统关键技术研究.《中国优秀硕士学位论文全文数据库 (工程科技Ⅱ辑)》.2020,(第3期), * |
| 李佳珍.基于传感器数据和深度学习的日常活动识别方法研究.《中国优秀硕士学位论文全文数据库 (信息科技辑)》.2019,(第1期), * |
| 王书燕.基于CNN的多生理信号动脉硬化检测方法研究.《中国优秀硕士学位论文全文数据库 (医药卫生科技辑)》.2020,(第9期), * |
| 蔡超等.一种基于人脸图像的年龄估计方法.《计算机与数字工程》.2012,第40卷(第4期), * |
| 赵丰等.LAC-DGLU:基于CNN和注意力机制的命名实体识别模型.《计算机科学》.2020,第47卷(第11期), * |
| 陈文强等.国有控股、市场竞争与股权激励效应——基于倾向得分匹配法的实证研究.《浙江大学学报(人文社会科学版)》.2015,第45卷(第5期), * |
| 陈泽宁.基于穿戴视觉的人机交互技术.《中国优秀硕士学位论文全文数据库 (信息科技辑)》.2015,(第7期), * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112957013A (en) | 2021-06-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10825569B2 (en) | Universal non-invasive blood glucose estimation method based on time series analysis | |
| CN108606798B (en) | Non-contact atrial fibrillation intelligent detection system based on deep convolution residual error network | |
| US20120184862A1 (en) | System and method for patient monitoring | |
| CN108388912A (en) | Sleep stage method based on multisensor feature optimization algorithm | |
| CN104042191A (en) | Wrist watch type multi-parameter biosensor | |
| CN110664390A (en) | Heart rate monitoring system and method based on wrist strap type PPG and deep learning | |
| CN102988036B (en) | Method for measuring pulse rate | |
| EP3231368B1 (en) | Method and system for identifying respiratory events | |
| CN110811647B (en) | Multi-channel hidden lie detection method based on ballistocardiogram signal | |
| CN114376564A (en) | Sleep staging method, system, device and medium based on cardiac shock signal | |
| Zhao et al. | An IoT-based wearable system using accelerometers and machine learning for fetal movement monitoring | |
| CN112998690B (en) | Pulse wave multi-feature fusion-based respiration rate extraction method | |
| DE112015005804T5 (en) | Breath condition estimator, portable device, body wearable device, program, medium, breath state estimation method, and breath condition estimator | |
| CN114916913B (en) | Portable sleep breathing state real-time monitoring system and method | |
| CN109523545A (en) | A kind of contactless heart rate detection method and system | |
| CN113017559A (en) | Vital sign extraction algorithm and system based on piezoelectric film sensor | |
| CN112957013B (en) | Dynamic vital sign signal acquisition system, monitoring device and equipment | |
| CN110384490B (en) | BCG signal heart rate extraction method based on phase space | |
| JP2008253727A (en) | Monitor device, monitor system and monitoring method | |
| CN113033358A (en) | Cuff-free blood pressure modeling method based on deep neural network | |
| KR20140086182A (en) | Apparatus for measuring heart rate | |
| CN113100776B (en) | Fatigue monitoring system and method for fusing myoelectricity and electrocardiosignal | |
| KR20160096895A (en) | Lying position classification system and method using ballistocardiography | |
| Kang et al. | Pulse Rate Estimation using PPG Affected with Motion Artifacts Base d on VMD and Hilbert Transform | |
| CN113171098A (en) | Detection device and detection method for magnetoencephalography signals in cerebral oscillation |
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 | ||
| CB03 | Change of inventor or designer information |
Inventor after: Chen Wenqiang Inventor after: Li Dongliang Inventor after: Zhang Shian Inventor after: Wu Haicong Inventor after: Ling Jingyi Inventor after: Liu Xuping Inventor after: Cai Chao Inventor after: Wang Yi Inventor before: Chen Wenqiang Inventor before: Liu Xuping Inventor before: Cai Chao Inventor before: Wang Yi |
|
| CB03 | Change of inventor or designer information |