CN113397519A - Cardiovascular health state detection device - Google Patents
Cardiovascular health state detection device Download PDFInfo
- Publication number
- CN113397519A CN113397519A CN202110898085.XA CN202110898085A CN113397519A CN 113397519 A CN113397519 A CN 113397519A CN 202110898085 A CN202110898085 A CN 202110898085A CN 113397519 A CN113397519 A CN 113397519A
- Authority
- CN
- China
- Prior art keywords
- average
- value
- poincare
- ippg signal
- cardiovascular health
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 51
- 230000036996 cardiovascular health Effects 0.000 title claims abstract description 37
- 238000007781 pre-processing Methods 0.000 claims abstract description 9
- 238000012545 processing Methods 0.000 claims description 37
- 230000000737 periodic effect Effects 0.000 claims description 20
- 238000000034 method Methods 0.000 claims description 19
- 238000010586 diagram Methods 0.000 claims description 14
- 230000008569 process Effects 0.000 claims description 13
- 238000012216 screening Methods 0.000 claims description 8
- 239000000284 extract Substances 0.000 claims description 5
- 238000000926 separation method Methods 0.000 claims description 5
- 238000001914 filtration Methods 0.000 claims description 4
- 230000036387 respiratory rate Effects 0.000 claims description 4
- 238000009499 grossing Methods 0.000 claims description 3
- 230000001815 facial effect Effects 0.000 claims 1
- 238000005070 sampling Methods 0.000 description 5
- 238000001228 spectrum Methods 0.000 description 5
- 238000012549 training Methods 0.000 description 5
- 230000003595 spectral effect Effects 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 230000035565 breathing frequency Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000036541 health Effects 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000029058 respiratory gaseous exchange Effects 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 210000001124 body fluid Anatomy 0.000 description 1
- 239000010839 body fluid Substances 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 210000000748 cardiovascular system Anatomy 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000036581 peripheral resistance Effects 0.000 description 1
- 238000013186 photoplethysmography Methods 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000036391 respiratory frequency Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000001052 transient 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
- A61B5/02—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
- A61B5/02405—Determining heart rate variability
-
- 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
- A61B5/02416—Detecting, measuring or recording pulse rate or heart rate using photoplethysmograph signals, e.g. generated by infrared radiation
Abstract
The invention provides a cardiovascular health state detection device, which is characterized in that original IPPG signals of a face area of a tested person are obtained; preprocessing the original IPPG signal; extracting HRV characteristics and average period waveform characteristics according to the preprocessed IPPG signal; inputting the HRV characteristics and the average period waveform characteristics into a pre-trained classification detection model to obtain a detection result of the cardiovascular health state of the detected person; therefore, the detection result is associated with the physiological characteristics with higher dimension, and the accuracy of the detection result is higher.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a cardiovascular health state detection device.
Background
At present, when the cardiovascular health state is monitored, an electrocardiograph is generally used for obtaining accurate electrocardiographic data for analysis, and the method has the defects of complex operation process and high professional knowledge required for a user. Therefore, Pavlidis et al propose a non-contact heart rate detection method using a common camera, which is called imaging photoplethysmography (IPPG), to obtain an IPPG signal of a face region, and then conveniently and quickly extract various physiological index features related to functional changes of a cardiovascular system, so as to determine cardiovascular health states according to the physiological index features. However, at present, only low-dimensional HRV features (heart rate variability features) are generally extracted according to IPPG signals for analysis and judgment, and the accuracy of a judgment result cannot meet the requirements of practical application.
Disclosure of Invention
In view of the foregoing disadvantages of the prior art, an object of the embodiments of the present application is to provide a method, an apparatus, an electronic device and a storage medium for detecting a cardiovascular health status, which have high accuracy of a detection result of the cardiovascular health status.
The embodiment of the application provides a cardiovascular health state detection device, which comprises a camera device and a processing device;
the camera device is used for collecting a face video of the tested person;
the processing device is used for acquiring an original IPPG signal of a face area of a tested person according to the face video, preprocessing the original IPPG signal, extracting HRV (high resolution vector) features and average period waveform features according to the preprocessed IPPG signal, and inputting the HRV features and the average period waveform features into a pre-trained classification detection model to obtain a detection result of the cardiovascular health state of the tested person.
The cardiovascular health state detection device extracts the HRV characteristic and the average period waveform characteristic according to the IPPG signal, and uses the HRV characteristic and the average period waveform characteristic together for cardiovascular health state detection.
Preferably, when the processing device acquires the original IPPG signal of the face area of the subject according to the face video:
acquiring a face video of the tested person;
performing RGB channel separation on each frame of image of the face video, and extracting a green channel image;
selecting an ROI (region of interest) region of each green channel image;
and calculating the pixel value mean value of each ROI area to obtain an original IPPG signal.
Preferably, the frame rate of the image pickup device for picking up the face video of the person to be tested is 30 fps.
Preferably, the preprocessing includes a trending process, a moving average process, and a band-pass filtering denoising process.
Preferably, the processing means, when extracting the HRV features from the pre-processed IPPG signal:
carrying out peak point positioning on the preprocessed IPPG signal;
calculating the interval between adjacent peak points to obtain an RR interval sequence;
calculating an HRV signature from the RR interval sequence; the HRV features include heart rate, respiratory rate, time domain features, frequency domain features, and Poincare scattergram distribution features.
Preferably, the Poincare scattergram distribution characteristics include a quantified description of the Poincare scattergram for a major axis of the ellipse, a quantified description of the Poincare scattergram for a minor axis of the ellipse, and a quantified description of the Poincare scattergram for a ratio of the major axis to the minor axis of the ellipse;
when the processing device calculates the distribution characteristics of the Poincare scatter diagram:
sequentially taking each RR interval value as an abscissa value of a Poincare point, and taking the next RR interval value as an ordinate value of the Poincare point to generate a Poincare scatter diagram;
and taking the minimum ellipse surrounding all Poincare points in the Poincare scatter diagram as a quantitative description ellipse, acquiring the length of the long axis and the length of the short axis of the quantitative description ellipse, and calculating the ratio of the length of the long axis and the length of the short axis.
Preferably, the average periodic waveform characteristics include a main peak amplitude Hb, a descending branch amplitude Hc, a reflection peak amplitude Hd, an average duration T, a main wave occurrence time Tab, and a main wave-to-dicrotic wave peak interval Tbd; when the processing device extracts the average periodic waveform characteristics according to the preprocessed IPPG signal:
carrying out waveform inversion processing and trend removing processing on the IPPG signal;
dividing the IPPG signal subjected to the negation processing and the detrending processing into a plurality of pulse waves by taking a wave valley point as a dividing point, and calculating the average value of the duration of the plurality of pulse waves to obtain the average duration T;
removing the deformed pulse waves according to preset screening conditions;
adjusting the rest pulse waves into pulse waves with the duration equal to the average duration T in a mode of amplifying or intercepting signal segments;
linearly average superposing all the adjusted pulse waves to obtain an average periodic waveform of the IPPG signal;
after interpolation and smoothing processing are carried out on the average periodic waveform, a second derivative waveform is obtained;
detecting the wave crest and the wave trough of the second derivative waveform, and taking an IPPG signal value in an average periodic waveform corresponding to a first minimum value of the second derivative waveform as a main peak amplitude Hb; taking an IPPG signal value in the average periodic waveform corresponding to the second minimum value of the second derivative waveform as a reflection peak amplitude Hd; taking an IPPG signal value in an average periodic waveform corresponding to a maximum value between a first minimum value and a second minimum value of the second derivative waveform as a descending branch amplitude Hc; taking the time corresponding to the first minimum value as a main wave appearance time Tab; and taking the time interval from the first minimum value to the second minimum value as the interval time Tbd from the main wave to the dicrotic wave peak.
The average cycle waveform characteristics comprise main peak amplitude, descending branch amplitude, reflection peak amplitude, average duration, main wave appearance time and interval time from the main wave to the dicrotic wave peak.
Preferably, the device further comprises an illuminating device, wherein the illuminating device is used for illuminating the face area of the testee when the camera device collects the face video of the testee.
Has the advantages that:
the cardiovascular health state detection device provided by the embodiment of the application obtains the original IPPG signal of the face area of a tested person; preprocessing the original IPPG signal; extracting HRV characteristics and average period waveform characteristics according to the preprocessed IPPG signal; inputting the HRV characteristics and the average period waveform characteristics into a pre-trained classification detection model to obtain a detection result of the cardiovascular health state of the detected person; therefore, the detection result is associated with the physiological characteristics with higher dimension, and the accuracy of the detection result is higher.
Drawings
Fig. 1 is a schematic structural diagram of a cardiovascular health status detection device according to an embodiment of the present disclosure.
Fig. 2 is a schematic view of the use state of the cardiovascular health state detection device.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the accompanying drawings are illustrative only for the purpose of explaining the present invention, and are not to be construed as limiting the present invention.
In the description of the present invention, it is to be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", and the like, indicate orientations and positional relationships based on those shown in the drawings, and are used only for convenience of description and simplicity of description, and do not indicate or imply that the device or element being referred to must have a particular orientation, be constructed and operated in a particular orientation, and thus, should not be considered as limiting the present invention. Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, features defined as "first", "second", may explicitly or implicitly include one or more of the described features. In the description of the present invention, "a plurality" means two or more unless specifically defined otherwise.
The following disclosure provides embodiments or examples for implementing different configurations of the invention. To simplify the disclosure of the present invention, the components and arrangements of specific examples are described below. Of course, they are merely examples and are not intended to limit the present invention. Furthermore, the present invention may repeat reference numerals and/or letters in the various examples, such repetition is for the purpose of simplicity and clarity and does not in itself dictate a relationship between the various embodiments and/or configurations discussed. In addition, the present invention provides examples of various specific processes and materials, but those of ordinary skill in the art will recognize applications of other processes and/or uses of other materials.
Referring to fig. 1, a device for detecting cardiovascular health status according to an embodiment of the present disclosure includes an image capturing device 1 and a processing device 2;
the camera device 1 is used for collecting a face video of a detected person;
the processing device 2 is used for acquiring an original IPPG signal of a face area of a tested person according to the face video, preprocessing the original IPPG signal, extracting HRV (high resolution vector) features and average period waveform features according to the preprocessed IPPG signal, and inputting the HRV features and the average period waveform features into a pre-trained classification detection model to obtain a detection result of the cardiovascular health state of the tested person.
According to the detection device for the cardiovascular health state, the HRV characteristic and the average period waveform characteristic are extracted according to the IPPG signal, and the HRV characteristic and the average period waveform characteristic are jointly used for detecting the cardiovascular health state.
Preferably, when acquiring the original IPPG signal of the face area of the subject according to the face video, the processing device 2:
acquiring a face video of the tested person;
performing RGB channel separation on each frame of image of the face video, and extracting a green channel image;
selecting an ROI (region of interest) for each green channel image;
and calculating the pixel value mean value of each ROI area to obtain an original IPPG signal.
The video images are collected through the camera device 1, RGB channel separation is carried out on each frame of image, the green channel image is high in sensitivity to changes of HRV characteristics such as heart rate and respiratory frequency, the green channel image is extracted to be analyzed and processed, a human face skin region can be selected as an ROI region according to a skin detection algorithm, finally, the pixel value mean value of the ROI region is calculated for each frame of green channel image, and a pixel value mean value sequence is obtained and is an original IPPG signal. Wherein the mean of the pixel values of the ROI region can be calculated according to the following formula:
wherein the content of the first and second substances,
is the pixel value mean of the ROI region of the i-th frame green channel image,
the pixel value of the jth pixel point of the ROI area of the ith frame of green channel image is obtained, and N is the total number of the pixel points of the ROI area of the ith frame of green channel image.
The frame rate of the video image acquisition can be set according to actual needs, generally 25 fps to 40 fps, and preferably, the frame rate of the video image acquisition is 30 fps.
In some preferred embodiments, the preprocessing includes a trending process, a moving average process, and a band-pass filtering denoising process.
The influence of the respiration rate on the extraction of the wave crest and the wave trough of the original waveform single pulse wave can be effectively reduced through trend processing.
The extraction of short-time stable signals can be realized through the moving average processing, and the interference of sampling abnormal values is reduced.
The noise in the IPPG signal can be effectively removed through the band-pass filtering denoising processing.
In some preferred embodiments, the processing means 2, when extracting HRV features from the pre-processed IPPG signal:
carrying out peak point positioning on the preprocessed IPPG signal;
calculating the interval between adjacent peak points to obtain an RR interval sequence;
calculating an HRV signature from the RR interval sequence; the HRV features include heart rate, respiratory rate, time domain features, frequency domain features, and Poincare scattergram distribution features.
The method for positioning the peak point of the preprocessed IPPG signal comprises the following steps: finding the position of all maximum elements (IPPG signal values), i.e. the element amplitude at this position is larger than the values of two adjacent elements; while excluding small spike signals due to the influence of noise according to a screening condition (which may be referred to as a first screening condition) including a limitation on the characteristics of the extracted wave, for example, the width of the extracted peak is not less than 5/60 for the estimated heart rate.
And subtracting the corresponding time of the adjacent peak points to obtain the RR interval sequence.
Wherein the heart rate can be calculated by the following formula:
Bpm=60*K/T；
wherein Bpm is the heart rate, K represents the number of peak points in the preprocessed IPPG signal, and T represents the duration of the preprocessed IPPG signal. The heart rate Bpm represents the number of beats per minute at rest.
In practice, the preprocessed IPPG signal may also be fast fourier transformed to obtain its power spectrum, and then the maximum power spectrum value is extracted from it, and the heart rate is calculated by the following formula:
Bpm=60*T1*Fs/M；
wherein, T1 is the maximum power spectrum value, Fs is the acquisition frame rate of the video image, and M is the total number of sampling points of the preprocessed IPPG signal. Because the heart rate is calculated by the method, the accuracy of the calculation result is influenced by factors such as the acquisition frame rate of the video image, the sampling length (total number of sampling points), the window length selected during fast Fourier transform and the like, and compared with the former method, the reliability of the calculation result is relatively low.
In some embodiments, two heart rate values (a first heart rate and a second heart rate) may be calculated according to the above two manners, and then a deviation rate between the two heart rate values (a ratio of a difference between the first heart rate and the second heart rate and the first heart rate) is determined, if the deviation rate does not exceed a preset deviation rate threshold, an average value of the two heart rate values is used as a final heart rate value, otherwise, the heart rate value calculated in the first manner (the first heart rate) is used as the final heart rate value.
Further, the preprocessed IPPG signal may be fast fourier transformed to obtain a power spectrum thereof, and then a second high power spectrum value is extracted therefrom, and the respiratory rate is calculated by the following formula:
Br=60* T2*Fs/M；
wherein Br is the breathing frequency, T2 is the second high-power spectral value, Fs is the acquisition frame rate of the video image, and M is the total number of sampling points of the preprocessed IPPG signal. The breathing frequency Br represents the number of breaths per minute.
In this embodiment, the temporal features in the HRV features include the features in the following table:
| serial number | (symbol) | Means of |
| 1 | IBI | Mean value of |
| 2 | SDNN | Standard deviation of |
| 3 | SDSD | Standard deviation of difference between |
| 4 | RMSSD | Root mean square of difference between |
| 5 | pNN20 | The ratio of the number of the difference between adjacent RR intervals greater than 20ms in total number |
| 6 | pNN50 | The ratio of the number of the difference between adjacent RR intervals greater than 50ms in total number |
The frequency domain features in the HRV features include features in the following table:
| serial number | (symbol) | Means of |
| 1 | LF | Area corresponding to 0.04-0.15Hz in power spectral density diagram of |
| 2 | HF | Area corresponding to 0.15-0.4Hz in power spectral density diagram of |
| 3 | LF/HF | Ratio of area of low frequency to high frequency power spectral density |
Preferably, the Poincare scattergram distribution characteristics include a quantified description of the Poincare scattergram for a major axis of the ellipse, a quantified description of the Poincare scattergram for a minor axis of the ellipse, and a quantified description of the Poincare scattergram for a ratio of the major axis to the minor axis of the ellipse;
when the processing device 2 calculates the distribution characteristics of the Poincare scatter diagram:
sequentially taking each RR interval value as an abscissa value of a Poincare point, and taking the next RR interval value as an ordinate value of the Poincare point to generate a Poincare scatter diagram;
and taking the minimum ellipse surrounding all Poincare points in the Poincare scatter diagram as a quantitative description ellipse, acquiring the major axis length SD1 and the minor axis length SD2 of the quantitative description ellipse, and calculating the ratio SD1/SD2 of the major axis length and the minor axis length.
The Poincare scatter diagram distribution characteristics reflect the characteristics of heart state dynamics, such as interaction effects of heart frequency with nerves, body fluid and respiration, so that the acquisition of Poincare scatter diagram distribution characteristics as part of HRV characteristics can realize the description and research of heart rate change based on the nonlinear motion mode.
In some preferred embodiments, the average periodic waveform characteristic comprises a main peak amplitude Hb, a descending branch amplitude Hc, a reflection peak amplitude Hd, an average duration T, a main wave occurrence time Tab, and a main wave to dicrotic peak separation time Tbd; when the processing device 2 extracts the average period waveform feature according to the preprocessed IPPG signal:
carrying out waveform inversion processing and trend removing processing on the IPPG signal;
dividing the IPPG signal subjected to the negation processing and the detrending processing into a plurality of pulse waves by taking a wave valley point as a dividing point, and calculating the average value of the duration of the plurality of pulse waves to obtain the average duration T;
removing the deformed pulse wave according to a preset screening condition (which can be called as a second screening condition); wherein the preset screening conditions (second screening conditions) are: if the variance of the target pulse wave is not more than m1 times of the average value of the variances of all the pulse waves, and the ratio of the duration of the target pulse wave to the average duration T is within the tolerance range, determining that the target pulse wave is a non-deformed pulse wave, otherwise, determining that the target pulse wave is a deformed pulse wave. Wherein, the value of m1 can be set according to actual needs, such as m1= 1.75; wherein, the tolerance range can be set according to actual needs, such as 0.8-1.2;
adjusting the rest pulse waves into pulse waves with the duration equal to the average duration T in a mode of amplifying or intercepting signal segments; for example, if the duration of the target pulse wave is less than the average duration T, a data segment may be inserted in the last of the target pulse waves by a fitting interpolation method, or a data segment may be inserted in the target pulse wave every several data segments in a linear interpolation manner, so as to expand the duration of the target pulse wave to T; if the duration of the target pulse wave is longer than the average duration T, deleting the signal segment at the end of the target pulse wave, and intercepting the duration of the target pulse wave as T;
linearly average superposing all the adjusted pulse waves to obtain an average periodic waveform of the IPPG signal;
after the average periodic waveform is subjected to interpolation and smoothing processing, a second derivative waveform (obtained through second-order difference calculation) is obtained;
performing peak and trough detection on the second derivative waveform, and taking an IPPG signal value in an average periodic waveform corresponding to a first minimum value (namely a first minimum value) of the second derivative waveform as a main peak amplitude Hb; taking the IPPG signal value in the average periodic waveform corresponding to the second minimum value (namely the second minimum value) of the second derivative waveform as the reflection peak amplitude Hd; taking an IPPG signal value in an average periodic waveform corresponding to a maximum value between a first minimum value and a second minimum value of the second derivative waveform as a descending branch amplitude Hc; taking the time corresponding to the first minimum value as a main wave appearance time Tab; and taking the time interval from the first minimum value to the second minimum value as the interval time Tbd from the main wave to the dicrotic wave peak.
The average period waveform characteristics reflect information such as arterial stiffness, peripheral resistance and elastic change, and health state information mining based on non-transient stationary pulse wave characteristics can be realized by considering the average period waveform characteristics when cardiovascular health state detection is carried out.
Through the steps, the high-dimensional HRV characteristics can be obtained to detect the cardiovascular health state of the detected person, so that the accuracy and reliability of the detection result are improved.
In some embodiments, the classification detection model is a classification detection model based on combination of bayesian hyper-parameter optimization and a random forest classifier, and can perform two-classification detection on the cardiovascular health state and the non-health state of the measured person. The classification detection model is obtained by training in the following way: training with a training set comprising samples and labels, performing model training by back-propagating minimal errors during the training process, and performing cross validation and evaluation on the trained models with a test set comprising samples and labels. The classified detection model can obtain the detection result that the cardiovascular health state of the detected person is a healthy state or an unhealthy state, and the detection result is accurate and reliable.
In some embodiments, see fig. 1 and 2, the cardiovascular health status detection apparatus further includes an illumination device 3, where the illumination device 3 is configured to illuminate a face area of a subject when the image capturing device 1 captures a face video of the subject. Therefore, the brightness of the face area can be improved, a clearer face video frame image can be obtained, and the detection accuracy is improved. Preferably, the light emitted by the illuminating device 3 is green light, for example, green light with a wavelength of 530nm, because the light with the wavelength of 530nm can penetrate through the skin to reach the capillary vessels, the absorption rate of blood to the light is higher than that of other wavelengths, so that the information of the reflected light signal and other changes can be easily captured, and the pulse wave with high signal-to-noise ratio and its characteristics can be extracted.
In some embodiments, see fig. 2, the cardiovascular health status detecting device comprises a cylinder-like detecting frame 4, a positioning and supporting part 5 for supporting the lower jaw of the testee is arranged at the lower part of the front side opening of the detecting frame 4, the camera device 1 is a camera arranged at the middle part of the rear side of the detecting frame 4, and the optical axis of the camera is coaxial with the detecting frame 4; irradiation device 3 is for centering on the lamp plate that camera device 1 set up, the lamp plate front side evenly is provided with the light source (like LED lamp pearl) to guarantee evenly to shine the people's face region that is surveyed. The distance between the camera and the front opening of the detection frame 4 can be set according to actual needs, for example, in fig. 2, the distance is 20 cm.
Therefore, the cardiovascular health state detection device acquires the original IPPG signal of the face area of the tested person; preprocessing the original IPPG signal; extracting HRV characteristics and average period waveform characteristics according to the preprocessed IPPG signal; inputting the HRV characteristics and the average period waveform characteristics into a pre-trained classification detection model to obtain a detection result of the cardiovascular health state of the detected person; therefore, the detection result is associated with the physiological characteristics with higher dimension, and the accuracy of the detection result is higher.
In summary, although the present invention has been described with reference to the preferred embodiments, the above-described preferred embodiments are not intended to limit the present invention, and those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention, which are substantially the same as the present invention.
Claims (8)
1. The cardiovascular health state detection device is characterized by comprising a camera device and a processing device;
the camera device is used for collecting a face video of the tested person;
the processing device is used for acquiring an original IPPG signal of a face area of a tested person according to the face video, preprocessing the original IPPG signal, extracting HRV (high resolution vector) features and average period waveform features according to the preprocessed IPPG signal, and inputting the HRV features and the average period waveform features into a pre-trained classification detection model to obtain a detection result of the cardiovascular health state of the tested person.
2. The apparatus for detecting cardiovascular health status of claim 1, wherein the processing means, when acquiring the raw IPPG signal of the face region of the subject according to the face video:
acquiring a face video of the tested person;
performing RGB channel separation on each frame of image of the face video, and extracting a green channel image;
selecting an ROI (region of interest) region of each green channel image;
and calculating the pixel value mean value of each ROI area to obtain an original IPPG signal.
3. The apparatus according to claim 1, wherein the image capturing device captures a facial video of the subject at a frame rate of 30 fps.
4. The cardiovascular health status detection apparatus of claim 1, wherein the preprocessing comprises a trending process, a moving average process, and a band-pass filtering denoising process.
5. The cardiovascular health status detection apparatus of claim 1, wherein the processing apparatus, when extracting the HRV features from the pre-processed IPPG signal:
carrying out peak point positioning on the preprocessed IPPG signal;
calculating the interval between adjacent peak points to obtain an RR interval sequence;
calculating an HRV signature from the RR interval sequence; the HRV features include heart rate, respiratory rate, time domain features, frequency domain features, and Poincare scattergram distribution features.
6. The cardiovascular health status detection apparatus of claim 5, wherein the Poincare scattergram distribution features comprise a quantified description of a Poincare scattergram of a major axis of an ellipse, a quantified description of a Poincare scattergram of a minor axis of an ellipse, and a quantified description of a Poincare scattergram of a ratio of the major axis to the minor axis of an ellipse;
when the processing device calculates the distribution characteristics of the Poincare scatter diagram:
sequentially taking each RR interval value as an abscissa value of a Poincare point, and taking the next RR interval value as an ordinate value of the Poincare point to generate a Poincare scatter diagram;
and taking the minimum ellipse surrounding all Poincare points in the Poincare scatter diagram as a quantitative description ellipse, acquiring the length of the long axis and the length of the short axis of the quantitative description ellipse, and calculating the ratio of the length of the long axis and the length of the short axis.
7. The cardiovascular health status detection apparatus of claim 1, wherein the average periodic waveform characteristics comprise a main peak amplitude Hb, a descending branch amplitude Hc, a reflection peak amplitude Hd, an average duration T, a main wave occurrence time Tab, and a main wave to dicrotic wave peak interval Tbd; when the processing device extracts the average periodic waveform characteristics according to the preprocessed IPPG signal:
carrying out waveform inversion processing and trend removing processing on the IPPG signal;
dividing the IPPG signal subjected to the negation processing and the detrending processing into a plurality of pulse waves by taking a wave valley point as a dividing point, and calculating the average value of the duration of the plurality of pulse waves to obtain the average duration T;
removing the deformed pulse waves according to preset screening conditions;
adjusting the rest pulse waves into pulse waves with the duration equal to the average duration T in a mode of amplifying or intercepting signal segments;
linearly average superposing all the adjusted pulse waves to obtain an average periodic waveform of the IPPG signal;
after interpolation and smoothing processing are carried out on the average periodic waveform, a second derivative waveform is obtained;
detecting the wave crest and the wave trough of the second derivative waveform, and taking an IPPG signal value in an average periodic waveform corresponding to a first minimum value of the second derivative waveform as a main peak amplitude Hb; taking an IPPG signal value in the average periodic waveform corresponding to the second minimum value of the second derivative waveform as a reflection peak amplitude Hd; taking an IPPG signal value in an average periodic waveform corresponding to a maximum value between a first minimum value and a second minimum value of the second derivative waveform as a descending branch amplitude Hc; taking the time corresponding to the first minimum value as a main wave appearance time Tab; and taking the time interval from the first minimum value to the second minimum value as the interval time Tbd from the main wave to the dicrotic wave peak.
8. The apparatus for detecting cardiovascular health status of claim 1, further comprising an illuminating device, wherein the illuminating device is used for illuminating the face area of the subject when the image capturing device captures the face video of the subject.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110898085.XA CN113397519A (en) | 2021-08-05 | 2021-08-05 | Cardiovascular health state detection device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110898085.XA CN113397519A (en) | 2021-08-05 | 2021-08-05 | Cardiovascular health state detection device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113397519A true CN113397519A (en) | 2021-09-17 |
Family
ID=77688472
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110898085.XA Pending CN113397519A (en) | 2021-08-05 | 2021-08-05 | Cardiovascular health state detection device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113397519A (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105476630A (en) * | 2015-11-26 | 2016-04-13 | 北京期颐科技有限公司 | Method for detecting traditional Chinese medicine visceral functions |
| CN106037694A (en) * | 2016-05-13 | 2016-10-26 | 吉林大学 | Continuous blood pressure measuring device based on pulse waves |
| CN106539580A (en) * | 2016-11-21 | 2017-03-29 | 西安交通大学 | A kind of continuous monitoring method of autonomic nervous system dynamic change |
| CN106691425A (en) * | 2016-12-30 | 2017-05-24 | 北京工业大学 | Wrist heart rate monitoring method of sports bracelet |
| CN109480808A (en) * | 2018-09-27 | 2019-03-19 | 深圳市君利信达科技有限公司 | A kind of heart rate detection method based on PPG, system, equipment and storage medium |
| CN109770885A (en) * | 2019-03-25 | 2019-05-21 | 合肥工业大学 | A kind of examing heartbeat fastly method based on preview frame |
| CN111714144A (en) * | 2020-07-24 | 2020-09-29 | 长春理工大学 | Mental stress analysis method based on video non-contact measurement |
| CN111839491A (en) * | 2020-06-29 | 2020-10-30 | 北京曙光易通技术有限公司 | Heart beat function monitoring method, heart beat function continuous monitoring method and electronic equipment |
| CN111870235A (en) * | 2020-08-04 | 2020-11-03 | 杭州艺兴科技有限公司 | Drug addict screening method based on IPPG |
| CN112040846A (en) * | 2018-04-23 | 2020-12-04 | 赢创运营有限公司 | Method for estimating blood pressure and degree of arteriosclerosis based on photoplethysmography (PPG) signal |
| CN112890792A (en) * | 2020-11-25 | 2021-06-04 | 合肥工业大学 | Cloud computing cardiovascular health monitoring system and method based on network camera |
| CN113080917A (en) * | 2021-03-04 | 2021-07-09 | 浙江慧养科技有限公司 | Method and device for monitoring abnormal heart rate |
-
2021
- 2021-08-05 CN CN202110898085.XA patent/CN113397519A/en active Pending
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105476630A (en) * | 2015-11-26 | 2016-04-13 | 北京期颐科技有限公司 | Method for detecting traditional Chinese medicine visceral functions |
| CN106037694A (en) * | 2016-05-13 | 2016-10-26 | 吉林大学 | Continuous blood pressure measuring device based on pulse waves |
| CN106539580A (en) * | 2016-11-21 | 2017-03-29 | 西安交通大学 | A kind of continuous monitoring method of autonomic nervous system dynamic change |
| CN106691425A (en) * | 2016-12-30 | 2017-05-24 | 北京工业大学 | Wrist heart rate monitoring method of sports bracelet |
| CN112040846A (en) * | 2018-04-23 | 2020-12-04 | 赢创运营有限公司 | Method for estimating blood pressure and degree of arteriosclerosis based on photoplethysmography (PPG) signal |
| CN109480808A (en) * | 2018-09-27 | 2019-03-19 | 深圳市君利信达科技有限公司 | A kind of heart rate detection method based on PPG, system, equipment and storage medium |
| CN109770885A (en) * | 2019-03-25 | 2019-05-21 | 合肥工业大学 | A kind of examing heartbeat fastly method based on preview frame |
| CN111839491A (en) * | 2020-06-29 | 2020-10-30 | 北京曙光易通技术有限公司 | Heart beat function monitoring method, heart beat function continuous monitoring method and electronic equipment |
| CN111714144A (en) * | 2020-07-24 | 2020-09-29 | 长春理工大学 | Mental stress analysis method based on video non-contact measurement |
| CN111870235A (en) * | 2020-08-04 | 2020-11-03 | 杭州艺兴科技有限公司 | Drug addict screening method based on IPPG |
| CN112890792A (en) * | 2020-11-25 | 2021-06-04 | 合肥工业大学 | Cloud computing cardiovascular health monitoring system and method based on network camera |
| CN113080917A (en) * | 2021-03-04 | 2021-07-09 | 浙江慧养科技有限公司 | Method and device for monitoring abnormal heart rate |
Non-Patent Citations (1)
| Title |
|---|
| JINGJING LUO, ET AL.: "An iPPG-Based Device for Pervasive Monitoring of Multi-Dimensional Cardiovascular Hemodynamics", SENSORS, vol. 21, no. 3, pages 3 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Li et al. | The obf database: A large face video database for remote physiological signal measurement and atrial fibrillation detection | |
| EP3414739B1 (en) | Device, system and method for pulsatility detection | |
| JP6194105B2 (en) | Improved signal selection for acquiring remote photoplethysmographic waveforms | |
| Tasli et al. | Remote PPG based vital sign measurement using adaptive facial regions | |
| Trenta et al. | Advanced motion-tracking system with multi-layers deep learning framework for innovative car-driver drowsiness monitoring | |
| WO2019161609A1 (en) | Method for analyzing multi-parameter monitoring data and multi-parameter monitor | |
| CN111466876B (en) | Alzheimer's disease auxiliary diagnosis system based on fNIRS and graph neural network | |
| US20220151504A1 (en) | Smart windowing to reduce power consumption of a head-mounted camera used for iPPG | |
| KR101752873B1 (en) | Method and system for extracting heart information of time domain | |
| CN111387959A (en) | Non-contact physiological parameter detection method based on IPPG | |
| Wedekind et al. | Automated identification of cardiac signals after blind source separation for camera-based photoplethysmography | |
| CN112294282A (en) | Self-calibration method of emotion detection device based on RPPG | |
| KR102482796B1 (en) | Device and method for non-contact hrv analysis using remote ppg | |
| KR101996027B1 (en) | Method and system for extracting Heart Information of Frequency domain by using pupil size variation | |
| KR20170004549A (en) | Method and system for extracting Heart Information of Frequency domain | |
| Pal et al. | Improved heart rate detection using smart phone | |
| Cherif et al. | Detection of artifacts on photoplethysmography signals using random distortion testing | |
| Sharma et al. | Non-contact measurement of human heart rate using low cost video camera | |
| US20220039753A1 (en) | Modulation model of photoplethysmography signal for vital sign extraction | |
| CN113397519A (en) | Cardiovascular health state detection device | |
| KR20170004548A (en) | Method and system for extracting heart information of time domain | |
| Zhang et al. | Photoplethysmogram signal quality assessment using support vector machine and multi-feature fusion | |
| Banerjee et al. | Design of a Realtime Photoplethysmogram Signal Quality Checker for Wearables and Edge Computing | |
| Nagpal | Contactless measurement of Heart Rate Variability by Extracting PPG Signals from Video | |
| Debjyoti et al. | Evaluating Visual Photoplethysmography Method |
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 |