CN114052692B - Heart rate analysis method and equipment based on millimeter wave radar - Google Patents
Heart rate analysis method and equipment based on millimeter wave radar Download PDFInfo
- Publication number
- CN114052692B CN114052692B CN202111248081.3A CN202111248081A CN114052692B CN 114052692 B CN114052692 B CN 114052692B CN 202111248081 A CN202111248081 A CN 202111248081A CN 114052692 B CN114052692 B CN 114052692B
- Authority
- CN
- China
- Prior art keywords
- millimeter wave
- wave radar
- heart rate
- entropy
- waveform 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.)
- Active
Links
- 238000004458 analytical method Methods 0.000 title claims abstract description 31
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 24
- 238000005070 sampling Methods 0.000 claims abstract description 21
- 238000001914 filtration Methods 0.000 claims abstract description 15
- 238000012544 monitoring process Methods 0.000 claims abstract description 13
- 238000012545 processing Methods 0.000 claims abstract description 11
- 238000000034 method Methods 0.000 claims description 22
- 230000009466 transformation Effects 0.000 claims description 13
- 230000006870 function Effects 0.000 claims description 9
- 238000013461 design Methods 0.000 claims description 6
- 238000005259 measurement Methods 0.000 claims description 5
- 238000012549 training Methods 0.000 claims description 5
- 238000009532 heart rate measurement Methods 0.000 claims description 4
- 238000013528 artificial neural network Methods 0.000 claims description 3
- 230000004044 response Effects 0.000 claims description 3
- 230000000452 restraining effect Effects 0.000 claims description 2
- 238000012163 sequencing technique Methods 0.000 claims description 2
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 9
- 230000008569 process Effects 0.000 description 7
- 230000008901 benefit Effects 0.000 description 5
- 238000001514 detection method Methods 0.000 description 4
- 230000036387 respiratory rate Effects 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 238000004590 computer program Methods 0.000 description 3
- 230000007547 defect Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000008602 contraction Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000000737 periodic effect Effects 0.000 description 2
- 238000002310 reflectometry Methods 0.000 description 2
- 230000000241 respiratory effect Effects 0.000 description 2
- 230000029058 respiratory gaseous exchange Effects 0.000 description 2
- 238000013526 transfer learning Methods 0.000 description 2
- 210000001015 abdomen Anatomy 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 230000004397 blinking Effects 0.000 description 1
- 230000001427 coherent effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 210000001035 gastrointestinal tract Anatomy 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 210000000876 intercostal muscle Anatomy 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 230000005012 migration 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
- 210000002569 neuron Anatomy 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 230000008855 peristalsis Effects 0.000 description 1
- 230000001902 propagating effect Effects 0.000 description 1
- 230000023718 relaxation of cardiac muscle Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
-
- 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
- A61B5/024—Detecting, measuring or recording pulse rate or heart rate
-
- 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
- A61B5/024—Detecting, measuring or recording pulse rate or heart rate
- A61B5/02438—Detecting, measuring or recording pulse rate or heart rate with portable devices, e.g. worn by the patient
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7203—Signal processing specially adapted for physiological signals or for diagnostic purposes for noise prevention, reduction or removal
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/725—Details of waveform analysis using specific filters therefor, e.g. Kalman or adaptive filters
-
- 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
- A61B5/726—Details of waveform analysis characterised by using transforms using Wavelet transforms
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Biophysics (AREA)
- General Health & Medical Sciences (AREA)
- Veterinary Medicine (AREA)
- Public Health (AREA)
- Physiology (AREA)
- Animal Behavior & Ethology (AREA)
- Surgery (AREA)
- Molecular Biology (AREA)
- Medical Informatics (AREA)
- Pathology (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Artificial Intelligence (AREA)
- Signal Processing (AREA)
- Psychiatry (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Cardiology (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- Fuzzy Systems (AREA)
- Measuring Pulse, Heart Rate, Blood Pressure Or Blood Flow (AREA)
Abstract
The invention relates to a heart rate analysis method and a technical scheme of equipment based on millimeter wave radar, comprising the following steps: acquiring heartbeat waveform data of a front living body including noise through a millimeter wave radar device; sampling and filtering the heartbeat waveform data; sampling the heartbeat waveform data subjected to filtering processing, and calculating time-frequency characteristics; creating a convolutional neural network, taking the time-frequency characteristic as the input of the convolutional neural network, and outputting the entropy of heart rate and heart beat waveform data; and determining the optimal heart rate monitoring position of the millimeter wave radar device through entropy. The beneficial effects of the invention are as follows: the heart rate is measured accurately, which can guide the user to use the device correctly.
Description
Technical Field
The invention relates to the field of computers, in particular to a heart rate analysis method and device based on millimeter wave radar.
Background
Millimeter wave (mmWave) is a special radar technology that uses short wavelength electromagnetic waves, and millimeter wave radars can emit signals with wavelengths on the order of millimeters. In the electromagnetic spectrum, such wavelengths are considered short wavelengths, and are also one of the advantages of this technology. Indeed, the size of the system components (e.g., antennas) required to process millimeter wave signals is quite small. Another advantage of short wavelength is high accuracy. Millimeter wave systems operating at 76-81GHz (corresponding to wavelengths of about 4 mm) will be able to detect movements as small as a fraction of a millimeter.
The breathing and heartbeat of the human body are macroscopic and are mechanical movements, the breathing movement is periodic movement generated by contraction and relaxation of diaphragm and intercostal muscles, the abdomen has a fluctuation of 1 to 12mm, and the back has a fluctuation of 0.1 to 0.5 mm. The heart beat depends on the periodic motion generated by the contraction and the relaxation of cardiac muscle, the chest around the heart has a fluctuation of 0.1 to 0.5mm, and the back around the heart has a fluctuation of 0.01 to 0.2 mm. It is apparent that millimeter wave radar can detect these small variations. Currently, the millimeter wave radar is applied to monitoring the respiratory rate, such as monitoring the respiratory rate in sleeping, monitoring the respiratory rate of a driver, and the like.
There are four difficulties with millimeter wave radar monitoring respiratory heart rate:
(1) The sensitivity is insufficient, the millimeter wave radar can detect movements of a few tenths of a millimeter and the heart beat produces movements of 0.1 to 0.5mm (chest), 0.01 to 0.2mm (back).
(2) The human body has a plurality of interference and strong interference, and the movement of the trunk and limbs of the human body is strong interference for the detection of the respiratory heart rate, and the talking and blinking of eyes are also strong interference, and also the peristalsis of intestinal tracts, the shake of muscles and the like.
(3) The non-contact measurement has the defects that the non-contact measurement is originally the advantage of the millimeter wave radar, but the non-contact measurement has the defects that radar waves are emitted from a transmitting antenna, reflected after reaching a measured object through a section of space, and then reach a receiving antenna through a section of space, and various interferences are introduced in the whole propagation process, and the influences of reflectivity, reflection area and the like are also caused.
(4) Software and hardware technology is still immature.
The method aims to solve four difficulties of monitoring the respiratory rate by the millimeter wave radar, and is difficult to achieve by software or hardware.
Disclosure of Invention
The invention aims to at least solve one of the technical problems in the prior art, provides a heart rate analysis method and device based on millimeter wave radar, and solves the defects in the prior art.
The technical scheme of the invention comprises a heart rate analysis method based on millimeter wave radar, which is characterized by comprising the following steps: acquiring heartbeat waveform data of a front living body including noise through a millimeter wave radar device; performing first sampling and filtering processing on the heartbeat waveform data; performing second sampling on the heartbeat waveform data subjected to the filtering processing, and calculating time-frequency characteristics; creating a convolutional neural network, taking the time-frequency characteristic as the input of the convolutional neural network, and outputting the entropy of heart rate and heartbeat waveform data; and determining the optimal heart rate monitoring position of the millimeter wave radar device through the entropy.
The heart rate analysis method based on millimeter wave radar, wherein the step of acquiring the heart beat waveform data of the front living body including noise through the millimeter wave radar device comprises the following steps: and restraining radar waves emitted by the millimeter wave radar device, and calling an interface to analyze the acquired data to obtain distance, position and heartbeat waveform data of the front living body.
The heart rate analysis method based on millimeter wave radar, wherein performing first sampling and filtering processing on the heartbeat waveform data comprises: sampling the heartbeat waveform data for the first time to obtain a time sequence S, and filtering the time sequence S, wherein the filter is a finite length unit impulse response filter, and the mathematical expression of the filter is as follows。
According to the heart rate analysis method based on the millimeter wave radar, the filter is configured into a band-pass filter with the equal ripple wave design and the order of 200, the cut-off frequency of the band-pass filter is 0.6HZ and 5.8HZ, and the minimum attenuation in a stop band is 40dB.
The heart rate analysis method based on millimeter wave radar, wherein the performing the second sampling on the heartbeat waveform data subjected to the filtering processing, and calculating the time-frequency characteristic comprises: continuously sampling the filtered time sequence data to obtain a time sequence S1, and continuously performing wavelet transformation on the S1, wherein a transformation formula is as follows
Where φ is the fundamental wavelet and a is the scale parameter a>0, b is a position parameter, a wavelet function cluster is obtained by continuously changing values of a scale parameter a and a position parameter b, and similarity operation is carried out on the wavelet function cluster and a time sequence S1 to obtain a continuous wavelet transformation coefficient +.>For continuous wavelet transform coefficientsAnd calculating the absolute value to obtain the time-frequency characteristic of the time sequence S1.
The heart rate analysis method based on millimeter wave radar, wherein a convolutional neural network is created, the time-frequency characteristic is used as the input of the convolutional neural network, and the entropy of the output heart rate and heartbeat waveform data comprises the following steps: acquiring a pre-trained GoogLeNet convolutional neural network, and converting a time-frequency characteristic obtained by continuous wavelet transformation into an RGB image 224 x 3 as an input; setting the random discard probability of the pool5-drop_7x7_s1 layer to 0.6; replacing the loss 3-classifer by a new full-connection layer, and resetting the learning rate of the full-connection layer; setting the output of the GoogLeNet convolutional neural network as 135 classes, wherein the 135 classes comprise entropy of 27 classes of heart rate and 5 classes of heart beat waveform data; wherein the google net convolutional neural network is a pre-trained deep neural network.
The heart rate analysis method based on millimeter wave radar, wherein determining the optimal heart rate monitoring position of the millimeter wave radar device through the entropy comprises: generating a plurality of heart rate measurement positions through horizontal and vertical rotation of two stepping motors, calculating entropy of each measurement position through a convolution network, sequencing the calculated entropy from small to large, and determining the optimal search position when the minimum entropy is smaller than a set threshold value.
According to the heart rate analysis method based on millimeter wave radar, the method further comprises the following steps: the heart rate is continuously monitored while the optimal search position is stayed, and when the entropy of the position exceeds the set threshold value, the search is restarted.
The technical scheme of the invention also comprises analysis equipment based on the millimeter wave radar, which comprises an operation processor, a millimeter wave radar sensing device, a power supply circuit, a motor driving circuit and a stepping motor; the millimeter wave radar sensing device is provided with a receiving and transmitting antenna; the millimeter wave radar sensing device is connected with the operation processor through an SPI bus; the power supply circuit is used for respectively supplying power to the operation processor and the millimeter wave radar sensing device; the motor driving circuit is used for driving the stepping motor; the arithmetic processor executes any heart rate analysis method based on millimeter wave radar.
The millimeter wave radar-based analysis device of claim, wherein the transceiver antenna is configured to constrain the transmitted millimeter radar waves, the transceiver antenna comprising a horn antenna or a lens.
The beneficial effects of the invention are as follows: (1) The millimeter wave radar integrated chip is used on hardware, and the high integration has the advantages of simplifying circuit design, being good in consistency, strong in interference resistance and the like; (2) The lens or the horn antenna is additionally arranged, and two first enhancement signals and second enhancement millimeter wave radars are acted on the lens or the horn antenna; (3) The circuit board provided with the radar chip is driven to rotate in the horizontal and vertical directions by two stepping motors to find the optimal heart rate measurement position; (4) Feature extraction, namely converting the one-dimensional heartbeat waveform obtained after filtering into time-frequency features through Continuous Wavelet Transform (CWT), wherein the time-frequency features can more describe the essence of data; (5) The method comprises the steps of constructing a convolutional neural network, outputting entropy of one heart rate and two heart beat waveform data of two dimensions, and driving a circuit board provided with a radar chip to rotate in the horizontal and vertical directions through an entropy adjusting stepping motor to search the optimal heart rate measurement position, wherein the input of the convolutional neural network is a time-frequency characteristic.
Drawings
The invention is further described below with reference to the drawings and examples;
fig. 1 is a flowchart showing heart rate analysis of a millimeter wave radar according to an embodiment of the present invention.
Fig. 2 is a diagram showing a device connection relationship of the millimeter wave radar apparatus according to the embodiment of the present invention.
Fig. 3 is a schematic diagram of millimeter wave radar sensing according to an embodiment of the present invention.
Fig. 4 is a schematic diagram showing different constraint angle acquisition heart rate of a millimeter wave radar according to an embodiment of the present invention.
Fig. 5 is a schematic diagram illustrating operation of a stepper motor according to an embodiment of the present invention.
Fig. 6 is a schematic diagram of a convolutional neural network according to an embodiment of the present invention.
Fig. 7 is a graph showing heart rate waveforms versus different signal to noise ratios in accordance with an embodiment of the present invention.
Detailed Description
Reference will now be made in detail to the present embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein the accompanying drawings are used to supplement the description of the written description so that one can intuitively and intuitively understand each technical feature and overall technical scheme of the present invention, but not to limit the scope of the present invention.
In the description of the present invention, a number means one or more, a number means two or more, and greater than, less than, exceeding, etc. are understood to not include the present number, and above, below, within, etc. are understood to include the present number.
In the description of the present invention, the continuous reference numerals of the method steps are used for facilitating examination and understanding, and by combining the overall technical scheme of the present invention and the logic relationships between the steps, the implementation sequence between the steps is adjusted without affecting the technical effect achieved by the technical scheme of the present invention.
In the description of the present invention, unless explicitly defined otherwise, terms such as arrangement and the like should be construed broadly, and those skilled in the art can reasonably determine the specific meaning of the terms in the present invention in combination with the specific contents of the technical scheme.
Fig. 1 is a flowchart showing heart rate analysis of a millimeter wave radar according to an embodiment of the present invention. The process comprises the following steps: acquiring heartbeat waveform data of a front living body including noise through a millimeter wave radar device; performing first sampling and filtering processing on the heartbeat waveform data; performing second sampling on the heartbeat waveform data subjected to the filtering processing, and calculating time-frequency characteristics; creating a convolutional neural network, taking the time-frequency characteristic as the input of the convolutional neural network, and outputting the entropy of heart rate and heart beat waveform data; and determining the optimal heart rate monitoring position of the millimeter wave radar device through entropy.
In one embodiment, for filter design, a FIR filter is used, the data of a front living body (heartbeat waveform data containing noise) is acquired by calling services provided by a millimeter wave radar chip library, then the data is continuously sampled, the sampling rate is 50HZ, a time sequence S is obtained, then the S is filtered, the filter is a finite length unit impulse response Filter (FIR), and the math expression of the FIR filter is as followsSpecific parameters of the filter include: the equal ripple design, the order is 200, the band-pass filter (cut-off frequency is 0.6HZ and 5.8 HZ), the minimum attenuation in the stop band is 40dB.
In one embodiment, extracting the time-frequency characteristic includes continuously sampling the filtered time-series data at a sampling rate of 50HZ to obtain a time sequence S1, and continuously performing wavelet transform (CWT) on the S1, wherein the transform formula is
，
Wherein phi is basic wavelet, a is a scale parameter a >0, b is a position parameter, a wavelet function cluster is obtained by continuously changing values of the scale parameter a and the position parameter b, the wavelet function cluster and the time sequence S1 are subjected to similarity operation to obtain a continuous wavelet transformation coefficient (W phi f) (a, b), and the continuous wavelet transformation coefficient (W phi f) (a, b) is subjected to absolute value obtaining to obtain the time-frequency characteristic of the time sequence S1.
And extracting time-frequency characteristics, wherein the time-frequency analysis comprises short-time Fourier transform and wavelet transform. The short time fourier transform time-frequency analysis window has a fixed size, and has a high time resolution using the short window, but has a poor frequency resolution capability. With a long window there is a higher frequency resolution but the time resolution is weak. In practice, a compromise between time window and window width must be made, and any compromise becomes meaningless when the signal being analyzed is of a type where transients and transients coexist. The wavelet transform is essentially a correlation of the original signal with a family of scaled wavelet functions. By adjusting the scale, wavelets with different time-frequency widths can be obtained to match different positions of the original signal, localized analysis of the signal is achieved, the window of the wavelet transform is an adjustable time-frequency window, a short window is used at high frequency, a wide window is used at low frequency, and the wavelet transform can better solve the contradiction between time and frequency resolution unlike short-time Fourier transform.
Continuously sampling the filtered time sequence data, wherein the sampling rate is 50HZ, and the length 1000 of the time sequence S1, S1 is obtained, and a matlab implementation code is as follows:
Fs = 50;
fb = cwtfilterbank('SignalLength',1000,'SamplingFrequency',Fs,'VoicesPerOctave',12);
[ cfs, frq ] = wt (fb, S1);% continuous wavelet transform
t = (0:999)/Fs;figure;
pcolor (t, frq, abs (cfs))%time-frequency characteristics are shown.
Fig. 2 is a diagram showing a device connection relationship of the millimeter wave radar apparatus according to the embodiment of the present invention, describing a positional relationship of the millimeter wave radar chip and the horn antenna on the circuit board. Comprising the following steps: an arithmetic processor, a millimeter wave radar sensor chip (device), a power supply circuit, a motor driving circuit and a stepping motor (not shown); the millimeter wave radar sensing chip is provided with a receiving and transmitting antenna; the millimeter wave radar sensing chip is connected with the operation processor through an SPI bus; the power supply circuit is used for respectively supplying power to the operation processor and the millimeter wave radar sensing chip; the motor driving circuit is used for driving the stepping motor.
In a specific embodiment, a millimeter wave radar integrated chip (also integrating a transceiver antenna) is used on hardware, so that the design of the whole circuit is simplified, a pulse coherent radar sensor is adopted by the millimeter wave radar chip, all circuits used by the millimeter wave radar are integrated, the transceiver antenna is included, a library running on an operation processor is provided except that the millimeter wave radar chip itself, the library is provided in a service form, and a developer can directly call the service to acquire the information of distance, speed and the like of an object in front of the radar. The motor driving circuit is a driving circuit of two stepping motors. The operation processor is a platform for the operation of the radar chip library and a platform for the operation of all the algorithms.
Referring to the embodiment of fig. 3, a horn antenna is added, and echo energy received by the millimeter wave radar chip is affected by multiple factors such as transmitting power, reflectivity, reflecting area, etc., the horn antenna is a method for optimizing these factors, and the horn antenna has two functions:
1. the signal is enhanced.
2. The directivity of the millimeter wave is enhanced, the millimeter wave radar has stronger directivity, and the directivity is further enhanced through the horn antenna. The horn antenna may also be replaced by a lens.
FIG. 4 is a schematic diagram showing the narrowing of the range of millimeter wave energy detection after the addition of the horn antenna to show more directionality, and for heart rate detection, the best position is the chest around the heart, which is a relatively small area, and if the millimeter wave radar can precisely point to this area, the signal to noise ratio can be improved well
Fig. 5 is a schematic diagram illustrating operation of a stepper motor according to an embodiment of the present invention. The two stepping motors are added for horizontal and vertical rotation, and people in the scene of monitoring heart rate by the millimeter wave radar can freely move, so that the optimal monitoring position of the heart is continuously changed, and the two stepping motors drive a circuit board provided with the millimeter wave radar chip to rotate in two directions of horizontal and vertical directions, so that the millimeter wave is accurately directed to the optimal heart rate detection position.
The work flow is that a plurality of heart rate measuring positions are generated through horizontal and vertical rotation of two stepping motors, entropy is calculated by each measuring position through a convolution network, the calculated entropy is ordered from small to large, and when the minimum entropy is smaller than a set threshold value, the optimal searching position is determined.
Fig. 6 is a schematic diagram of a convolutional neural network according to an embodiment of the present invention. The time-frequency characteristic is constructed as entropy of input heart rate and heartbeat waveform data and is used as an output convolution neural network, the architecture of the network is obtained by modifying GoogLeNet, the GoogLeNet is shown in figure 6, and the specific modification is as follows:
1. the google net network inputs 224×224×3, so the time-frequency characteristics obtained by continuous wavelet transformation need to be converted into 224×224×3 RGB images.
2. Dropout Layer is to prevent overfitting, and when propagating forward, the activation value of a certain neuron stops working with a certain probability p, so that the model generalization is stronger, the default probability of pool5-drop_7x7_s1 in the GoogleNet network is 0.5, and the Dropout Layer with the probability of 0.6 is used for substitution.
3. The convolutional layer of the network extracts image features, and the two layers 'loss 3-classification' and 'output' in google net contain information on how to combine the network extracted features into class probabilities and predictive labels. To retrain the google net to classify the RGB images, both layers are replaced with new layers that are data-compatible.
4. The google net network outputs 1000 classes, modifies the class into 135 classes, the calculation of heart rate is a regression problem, the classification problem is regarded as the advantage of the data to be convenient for labeling, because the heart rate of people changes every moment, the classification method is that the class is 45 times to 50 times per minute, the class is 50 times to 55 times per minute, the class is the second class, and the last class is 175 times to 180 times per minute, and the total class is 27. The entropy of the heartbeat waveform data represented by the time-frequency characteristics is classified into 5 categories, and the final network output is 135 categories.
The convolutional neural network constructed by the invention takes the time-frequency characteristic of the filtered heartbeat time sequence waveform as input and takes the entropy of the heart rate and the heartbeat time sequence waveform as output. The calculation of heart rate is a regression problem, and the classification problem is considered to be that the data is convenient to label, and the heart rate of a person changes every moment, and the classification method is that the heart rate of the person is classified into a class from 45 times to 50 times per minute, the heart rate of the person is classified into a second class from 50 times to 55 times per minute, and the last class of the heart rate of the person is classified into 175 times to 180 times per minute, and the classification method is 27 classes in total. The entropy is the quantity describing the disorder degree of time series data, the lower the value of the entropy is, the higher the self similarity of the sequence is, the higher the entropy value is, the higher the probability of generating a new mode is, the complexity of the sequence is, the entropy also reflects the signal-to-noise ratio of the data in a system for detecting the heart rate by the millimeter wave radar, and the lower the entropy is, the higher the signal-to-noise ratio is, and the higher the entropy is, the lower the signal-to-noise ratio is. According to the different degrees that the millimeter wave points to the position close to the heart, the obtained heartbeat waveforms are generally divided into three types shown in fig. 7, and the waveform 1 is nearest to the heart, and is characterized in that the waveform of the heartbeat period is clear and the signal-to-noise ratio is low. Waveform 2 is slightly far from the heart and is characterized by a slightly larger signal-to-noise ratio and slightly larger entropy than the waveform of the heart cycle. Waveform 3 is far from heart and features low signal-to-noise ratio and high entropy when the waveform is submerged by noise. The training data is (train_x, train_y), wherein train_x is a time-frequency representation of the heartbeat cycle waveform, and train_y is a heart rate value corresponding to the heartbeat cycle waveform and entropy of the heartbeat time sequence waveform. The training data collection process is as follows: the heart rate value of the heart rate value is recorded by the heart rate waveform, the heart rate value is recorded as the heart rate value in the train_y, the calculation step of train_y is to intercept two sections of adjacent heart rate waveforms to respectively carry out Fourier transformation to obtain amplitude-frequency characteristics, and then the pearson correlation coefficient of the amplitude-frequency characteristics is calculated. This coefficient is inversely related to entropy. The calculated entropy is divided into five classes, namely 5 classes, by setting a threshold.
The transfer learning (Transfer Learning) is a machine learning method, which transfers knowledge in one domain (i.e. source domain) to another domain (i.e. target domain), so that the target domain can obtain better learning effect. The source field data size is sufficient, the target field data size is smaller, the scene is very suitable for migration learning, and the invention utilizes the trained weight in the GoogLeNet network convolution layer.
The matlab code is as follows:
% acquisition google et network
net = googlenet;
% pool5-drop_7x7_s1 set random discard probability
newDropoutLayer = dropoutLayer(0.6,'Name','new_Dropout');
lgraph = replaceLayer(lgraph,'pool5-drop_7x7_s1',newDropoutLayer);
numClasses = numel(categories(imgsTrain.Labels));
% replacement of loss 3-classifer with new full connectivity layer
newConnectedLayer = fullyConnectedLayer(numClasses,'Name','new_fc',...
'WeightLearnRateFactor',5,'BiasLearnRateFactor',5);
lgraph = replaceLayer(lgraph,'loss3-classifier',newConnectedLayer);
newClassLayer = classificationLayer('Name','new_classoutput');
% alternative output layer
lgraph = replaceLayer(lgraph,'output',newClassLayer);
% training parameter settings
options = trainingOptions('sgdm',...
'MiniBatchSize',15,...
'MaxEpochs',20,...
'InitialLearnRate',1e-4,...
'ValidationData',imgsValidation,...
'ValidationFrequency',10,...
'Verbose',1,...
'ExecutionEnvironment','cpu',...
'Plots','training-progress');
rng default
% start training
trainedGN = trainNetwork(imgsTrain,lgraph,options);
It should be appreciated that the method steps in embodiments of the present invention may be implemented or carried out by computer hardware, a combination of hardware and software, or by computer instructions stored in non-transitory computer-readable memory. The method may use standard programming techniques. Each program may be implemented in a high level procedural or object oriented programming language to communicate with a computer system. However, the program(s) can be implemented in assembly or machine language, if desired. In any case, the language may be a compiled or interpreted language. Furthermore, the program can be run on a programmed application specific integrated circuit for this purpose.
Furthermore, the operations of the processes described herein may be performed in any suitable order unless otherwise indicated herein or otherwise clearly contradicted by context. The processes (or variations and/or combinations thereof) described herein may be performed under control of one or more computer systems configured with executable instructions, and may be implemented as code (e.g., executable instructions, one or more computer programs, or one or more applications), by hardware, or combinations thereof, collectively executing on one or more processors. The computer program includes a plurality of instructions executable by one or more processors.
Further, the method may be implemented in any type of computing platform operatively connected to a suitable computing platform, including, but not limited to, a personal computer, mini-computer, mainframe, workstation, network or distributed computing environment, separate or integrated computer platform, or in communication with a charged particle tool or other imaging device, and so forth. Aspects of the invention may be implemented in machine-readable code stored on a non-transitory storage medium or device, whether removable or integrated into a computing platform, such as a hard disk, optical read and/or write storage medium, RAM, ROM, etc., such that it is readable by a programmable computer, which when read by a computer, is operable to configure and operate the computer to perform the processes described herein. Further, the machine readable code, or portions thereof, may be transmitted over a wired or wireless network. When such media includes instructions or programs that, in conjunction with a microprocessor or other data processor, implement the steps described above, the invention described herein includes these and other different types of non-transitory computer-readable storage media. The invention also includes the computer itself when programmed according to the methods and techniques of the present invention.
The computer program can be applied to the input data to perform the functions described herein, thereby converting the input data to generate output data that is stored to the non-volatile memory. The output information may also be applied to one or more output devices such as consumers. In a preferred embodiment of the invention, the transformed data represents physical and tangible objects, including specific visual depictions of physical and tangible objects produced on the consumer.
The embodiments of the present invention have been described in detail with reference to the accompanying drawings, but the present invention is not limited to the above embodiments, and various changes can be made within the knowledge of one of ordinary skill in the art without departing from the spirit of the present invention.
Claims (9)
1. The heart rate analysis method based on the millimeter wave radar is characterized by comprising the following steps of:
acquiring heartbeat waveform data of a front living body including noise through a millimeter wave radar device;
performing first sampling and filtering processing on the heartbeat waveform data;
performing second sampling on the heartbeat waveform data subjected to the filtering processing, and calculating time-frequency characteristics;
creating a convolutional neural network, taking the time-frequency characteristic as the input of the convolutional neural network, and outputting the entropy of heart rate and heartbeat waveform data;
determining an optimal heart rate monitoring position of the millimeter wave radar device through the entropy;
the creating the convolutional neural network, taking the time-frequency characteristic as the input of the convolutional neural network, outputting the entropy of heart rate and heartbeat waveform data, and comprises the following steps:
acquiring a pre-trained GoogLeNet convolutional neural network, and converting a time-frequency characteristic obtained by continuous wavelet transformation into an RGB image 224 x 3 as an input;
setting the random discard probability of the pool5-drop_7x7_s1 layer to 0.6;
replacing the loss 3-classifer by a new full-connection layer, and resetting the learning rate of the full-connection layer;
setting the output of the GoogLeNet convolutional neural network as 135 classes, wherein the 135 classes comprise entropy of 27 classes of heart rate and 5 classes of heart beat waveform data;
wherein the google net convolutional neural network is a pre-trained deep neural network;
the training data of the convolutional neural network are (train_x, train_y), the train_x is time-frequency representation of a heartbeat cycle waveform, the train_y is a heart rate value corresponding to the heartbeat cycle waveform and entropy of a heartbeat time sequence waveform, the train_y obtains amplitude-frequency characteristics by intercepting two sections of adjacent heartbeat cycle waveforms to respectively perform Fourier transformation, and pearson correlation coefficients of the amplitude-frequency characteristics are calculated to obtain the pearson correlation coefficients, and the pearson correlation coefficients are in inverse relation with the entropy.
2. The millimeter wave radar-based heart rate analysis method according to claim 1, wherein the acquiring, by the millimeter wave radar device, the front living body including noisy heartbeat waveform data includes:
and restraining radar waves emitted by the millimeter wave radar device, and calling an interface to analyze the acquired data to obtain distance, position and heartbeat waveform data of the front living body.
3. The millimeter wave radar-based heart rate analysis method according to claim 1, wherein the performing first sampling and filtering processing on the heartbeat waveform data comprises:
sampling the heartbeat waveform data for the first time to obtain a time sequence S, and filtering the time sequence S, wherein the filter is a finite length unit impulse response filter, and the mathematical expression of the filter is as follows
Wherein->For the filter coefficients +.>Is time-series waveform data.
4. A method of heart rate analysis based on millimeter wave radar according to claim 3, wherein the filter is configured as an equiripple design, a bandpass filter with an order of 200, a cut-off frequency of the bandpass filter being 0.6HZ and 5.8HZ, a minimum attenuation of 40dB in the stop band.
5. The millimeter wave radar-based heart rate analysis method according to claim 1, wherein the performing of the second sampling of the heartbeat waveform data subjected to the filtering processing and calculating the time-frequency characteristics include:
continuously sampling the filtered time sequence data to obtain a time sequence S1, and continuously performing wavelet transformation on the S1, wherein a transformation formula is as follows
，
Where φ is the basic wavelet and a is the scale parameter a>0, b is a position parameter, and the values of the scale parameter a and the position parameter b are continuously changed to obtain a wavelet function cluster, and the wavelet function cluster and the time sequence S1 are subjected to similarity operation to obtain a continuous wavelet transformation coefficientFor continuous wavelet transform coefficient +.>And calculating the absolute value to obtain the time-frequency characteristic of the time sequence S1.
6. The millimeter wave radar-based heart rate analysis method of claim 1, wherein the determining an optimal heart rate monitoring location of the millimeter wave radar device by the entropy comprises:
generating a plurality of heart rate measurement positions through horizontal and vertical rotation of two stepping motors, calculating entropy of each measurement position through a convolution network, sequencing the calculated entropy from small to large, and determining the optimal search position when the minimum entropy is smaller than a set threshold value.
7. The millimeter wave radar-based heart rate analysis method of claim 6, further comprising:
the heart rate is continuously monitored while the optimal search position is stayed, and when the entropy of the position exceeds the set threshold value, the search is restarted.
8. The analysis equipment based on the millimeter wave radar is characterized by comprising an operation processor, a millimeter wave radar sensing device, a power supply circuit, a motor driving circuit and a stepping motor;
the millimeter wave radar sensing device is provided with a receiving and transmitting antenna;
the millimeter wave radar sensing device is connected with the operation processor through an SPI bus;
the power supply circuit is used for respectively supplying power to the operation processor and the millimeter wave radar sensing device;
the motor driving circuit is used for driving the stepping motor;
the arithmetic processor performs the millimeter wave radar-based heart rate analysis method of any one of claims 1 to 7.
9. The millimeter wave radar-based analysis device of claim 8, wherein the transceiver antenna is configured to constrain the emitted millimeter radar waves, the transceiver antenna comprising a horn antenna or a lens.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111248081.3A CN114052692B (en) | 2021-10-26 | 2021-10-26 | Heart rate analysis method and equipment based on millimeter wave radar |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111248081.3A CN114052692B (en) | 2021-10-26 | 2021-10-26 | Heart rate analysis method and equipment based on millimeter wave radar |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114052692A CN114052692A (en) | 2022-02-18 |
| CN114052692B true CN114052692B (en) | 2024-01-16 |
Family
ID=80235569
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111248081.3A Active CN114052692B (en) | 2021-10-26 | 2021-10-26 | Heart rate analysis method and equipment based on millimeter wave radar |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114052692B (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| AU2125899A (en) * | 1994-10-07 | 1999-05-13 | Masimo Corporation | Physiological monitor and method of minimizing noise |
| CN110325929A (en) * | 2016-12-07 | 2019-10-11 | 阿瑞路资讯安全科技股份有限公司 | System and method for detecting the waveform analysis of cable network variation |
| CN110598549A (en) * | 2019-08-07 | 2019-12-20 | 王满 | Convolutional neural network information processing system based on cardiac function monitoring and training method |
| KR20200006447A (en) * | 2018-07-10 | 2020-01-20 | 재단법인 아산사회복지재단 | Method and program for stroke volume acquisition using arterial pressure waveform |
| CN111329455A (en) * | 2020-03-18 | 2020-06-26 | 南京润楠医疗电子研究院有限公司 | Non-contact cardiovascular health assessment method |
| CN111358435A (en) * | 2020-03-13 | 2020-07-03 | 珠海向量科技有限公司 | Data enhancement method for improving precision of deep neural network |
| WO2021026400A1 (en) * | 2019-08-06 | 2021-02-11 | Neuroenhancement Lab, LLC | System and method for communicating brain activity to an imaging device |
| CN112716468A (en) * | 2020-12-14 | 2021-04-30 | 首都医科大学 | Non-contact heart rate measuring method and device based on three-dimensional convolution network |
| CN112716462A (en) * | 2020-12-09 | 2021-04-30 | 北京航空航天大学 | Narrow-beam millimeter wave human body heartbeat/respiration sign monitoring device capable of controlling irradiation direction |
| WO2021171091A2 (en) * | 2020-02-27 | 2021-09-02 | Takuya Sakamoto | Vital information acquisition apparatus and method |
| CN113420624A (en) * | 2021-06-11 | 2021-09-21 | 华中师范大学 | Non-contact fatigue detection method and system |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10580131B2 (en) * | 2017-02-23 | 2020-03-03 | Zebra Medical Vision Ltd. | Convolutional neural network for segmentation of medical anatomical images |
| US11298017B2 (en) * | 2019-06-27 | 2022-04-12 | Bao Tran | Medical analysis system |
-
2021
- 2021-10-26 CN CN202111248081.3A patent/CN114052692B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| AU2125899A (en) * | 1994-10-07 | 1999-05-13 | Masimo Corporation | Physiological monitor and method of minimizing noise |
| CN110325929A (en) * | 2016-12-07 | 2019-10-11 | 阿瑞路资讯安全科技股份有限公司 | System and method for detecting the waveform analysis of cable network variation |
| KR20200006447A (en) * | 2018-07-10 | 2020-01-20 | 재단법인 아산사회복지재단 | Method and program for stroke volume acquisition using arterial pressure waveform |
| WO2021026400A1 (en) * | 2019-08-06 | 2021-02-11 | Neuroenhancement Lab, LLC | System and method for communicating brain activity to an imaging device |
| CN110598549A (en) * | 2019-08-07 | 2019-12-20 | 王满 | Convolutional neural network information processing system based on cardiac function monitoring and training method |
| WO2021171091A2 (en) * | 2020-02-27 | 2021-09-02 | Takuya Sakamoto | Vital information acquisition apparatus and method |
| CN111358435A (en) * | 2020-03-13 | 2020-07-03 | 珠海向量科技有限公司 | Data enhancement method for improving precision of deep neural network |
| CN111329455A (en) * | 2020-03-18 | 2020-06-26 | 南京润楠医疗电子研究院有限公司 | Non-contact cardiovascular health assessment method |
| CN112716462A (en) * | 2020-12-09 | 2021-04-30 | 北京航空航天大学 | Narrow-beam millimeter wave human body heartbeat/respiration sign monitoring device capable of controlling irradiation direction |
| CN112716468A (en) * | 2020-12-14 | 2021-04-30 | 首都医科大学 | Non-contact heart rate measuring method and device based on three-dimensional convolution network |
| CN113420624A (en) * | 2021-06-11 | 2021-09-21 | 华中师范大学 | Non-contact fatigue detection method and system |
Non-Patent Citations (5)
| Title |
|---|
| Detection of Congestive Heart Failure Based on LSTM-Based Deep Network via Short-Term RR Intervals;Wang, LD and Zhou, XG;SENSORS;第19卷(第7期);全文 * |
| 中西医结合治疗黄牛大叶性肺炎;金宝财, 张强, 耿金繁等;畜牧兽医科技信息(第9期);46页 * |
| 基于CNN与心跳信号的胚蛋成活性检测方法研究;胡玉舟;《中国优秀硕士学位论文全文数据库》(第7期);全文 * |
| 基于模糊思想的熵方法及在心血管信号分析中的应用;赵莉娜;《中国优秀硕士学位论文全文数据库》(第9期);全文 * |
| 救援机器人的非接触式人体体征视觉测量方法研究;黄如强, 苏卫华, 张世月;医疗卫生装备;第41卷(第6期);26-31页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114052692A (en) | 2022-02-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11896380B2 (en) | Medical decision support system | |
| KR20190060606A (en) | Apparatus and method for diagnosis of medical image | |
| WO2021066918A1 (en) | Systems and methods of determining heart-rate and respiratory rate from a radar signal using machine learning methods | |
| US11207055B2 (en) | Ultrasound Cardiac Doppler study automation | |
| CN111178331B (en) | Radar image recognition system, method, apparatus, and computer-readable storage medium | |
| CN109416729A (en) | Feature is extracted from physiological signal | |
| Afandizadeh Zargari et al. | An accurate non-accelerometer-based ppg motion artifact removal technique using cyclegan | |
| KR20190046471A (en) | Method for processing of medical image and medical image processing apparatus thereof | |
| US20220318548A1 (en) | Method, apparatus, computing device and computer-readable storage medium for identifying signal | |
| CN110520935A (en) | Learn sleep stage from radio signal | |
| Al-Assaf | Surface myoelectric signal analysis: dynamic approaches for change detection and classification | |
| CN114305355B (en) | Breathing heartbeat detection method, system and device based on millimeter wave radar | |
| CN115343704A (en) | Gesture recognition method of FMCW millimeter wave radar based on multi-task learning | |
| CN114052693B (en) | Heart rate analysis method, device and equipment | |
| WO2020219991A1 (en) | Medical decision support system | |
| CN114052692B (en) | Heart rate analysis method and equipment based on millimeter wave radar | |
| CN106419884B (en) | A kind of rate calculation method and system based on wavelet analysis | |
| CN115754956A (en) | Millimeter wave radar gesture recognition method based on envelope data time sequence | |
| Sahoo et al. | Recognition of ocular artifacts in EEG signal through a hybrid optimized scheme | |
| CN114847931A (en) | Human motion tracking method, device and computer-readable storage medium | |
| Cadart et al. | Classification in C-band of doppler signatures of human activities in indoor environments | |
| CN114052694B (en) | Radar-based heart rate analysis method and device | |
| WO2022032041A1 (en) | Medical decision support system | |
| Li et al. | Advancing IR-UWB Radar Human Activity Recognition With Swin-Transformers and Supervised Contrastive Learning | |
| US20230301599A1 (en) | Method and System for Detection of Inflammatory Conditions |
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 |