CN117649403B - Video-derived facial blood flow pulsation quantitative evaluation method, device and equipment for migraine - Google Patents
Video-derived facial blood flow pulsation quantitative evaluation method, device and equipment for migraine Download PDFInfo
- Publication number
- CN117649403B CN117649403B CN202410114861.6A CN202410114861A CN117649403B CN 117649403 B CN117649403 B CN 117649403B CN 202410114861 A CN202410114861 A CN 202410114861A CN 117649403 B CN117649403 B CN 117649403B
- Authority
- CN
- China
- Prior art keywords
- blood flow
- flow pulsation
- facial
- time
- signal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000010349 pulsation Effects 0.000 title claims abstract description 109
- 230000017531 blood circulation Effects 0.000 title claims abstract description 98
- 230000001815 facial effect Effects 0.000 title claims abstract description 81
- 206010027599 migraine Diseases 0.000 title claims abstract description 41
- 208000019695 Migraine disease Diseases 0.000 title claims abstract description 40
- 238000000034 method Methods 0.000 title claims abstract description 28
- 238000011158 quantitative evaluation Methods 0.000 title claims abstract description 27
- 201000010099 disease Diseases 0.000 claims abstract description 40
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims abstract description 40
- 238000012545 processing Methods 0.000 claims abstract description 17
- 238000013186 photoplethysmography Methods 0.000 claims description 34
- 230000010354 integration Effects 0.000 claims description 20
- 238000004590 computer program Methods 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 claims description 11
- 238000004458 analytical method Methods 0.000 claims description 10
- 238000011156 evaluation Methods 0.000 claims description 8
- 230000002146 bilateral effect Effects 0.000 claims description 7
- 238000012935 Averaging Methods 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 4
- 230000008569 process Effects 0.000 claims description 3
- 238000003384 imaging method Methods 0.000 abstract description 11
- 238000005516 engineering process Methods 0.000 abstract description 7
- 238000013473 artificial intelligence Methods 0.000 abstract description 6
- 238000001514 detection method Methods 0.000 abstract description 5
- 238000003745 diagnosis Methods 0.000 abstract description 5
- 238000004422 calculation algorithm Methods 0.000 abstract description 4
- 238000012544 monitoring process Methods 0.000 abstract description 4
- 230000000007 visual effect Effects 0.000 abstract description 4
- 210000004204 blood vessel Anatomy 0.000 abstract description 3
- 238000013461 design Methods 0.000 abstract description 3
- 230000000694 effects Effects 0.000 abstract description 3
- 230000003287 optical effect Effects 0.000 abstract description 3
- 238000004393 prognosis Methods 0.000 abstract description 3
- 230000008326 skin blood flow Effects 0.000 abstract description 3
- 230000009466 transformation Effects 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 10
- 239000008280 blood Substances 0.000 description 9
- 210000004369 blood Anatomy 0.000 description 9
- 230000006870 function Effects 0.000 description 7
- 210000001061 forehead Anatomy 0.000 description 6
- 230000002792 vascular Effects 0.000 description 6
- 230000001052 transient effect Effects 0.000 description 5
- 230000002159 abnormal effect Effects 0.000 description 4
- 238000004891 communication Methods 0.000 description 4
- 230000000541 pulsatile effect Effects 0.000 description 4
- 208000024891 symptom Diseases 0.000 description 4
- 230000005856 abnormality Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000013500 data storage Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000001121 heart beat frequency Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 230000010287 polarization Effects 0.000 description 2
- 230000004895 regional blood flow Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000012800 visualization Methods 0.000 description 2
- 208000006561 Cluster Headache Diseases 0.000 description 1
- 208000002193 Pain Diseases 0.000 description 1
- 206010043269 Tension headache Diseases 0.000 description 1
- 208000008548 Tension-Type Headache Diseases 0.000 description 1
- 230000003321 amplification Effects 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 239000000969 carrier Substances 0.000 description 1
- 208000018912 cluster headache syndrome Diseases 0.000 description 1
- 230000001149 cognitive effect Effects 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000001667 episodic effect Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 210000000887 face Anatomy 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
- 230000008506 pathogenesis Effects 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Abstract
The invention discloses a video derived facial blood flow pulsation quantitative evaluation method, a device and equipment for migraine, and the main design concept of the invention is that an optical plethysmography imaging technology based on image data is used for acquiring a tested facial image by using a video acquisition system with excellent imaging effect, extracting facial feature points by using artificial intelligence and an automatic algorithm, dividing facial skin blood vessel corpuscle areas to extract pulse waveforms, analyzing and processing the waveforms by continuous wavelet transformation and other methods, establishing a space-time variation model of facial skin blood flow pulsation amplitude and phase, realizing visual imaging of facial blood flow pulsation caused by specific diseases, and dynamically monitoring blood flow pulsation asymmetry and asynchronism in occurrence and progress of diseases. The invention provides objective, sensitive, quantifiable and automatic detection indexes for the prediction, identification and prognosis of the blood flow pulsation related diseases, and provides reliable and accurate auxiliary support for the individual accurate diagnosis and treatment of the diseases.
Description
Technical Field
The invention relates to the field of artificial intelligence technology application, in particular to a video derived facial blood flow pulsation quantitative evaluation method, device and equipment for migraine.
Background
Among the various diseases associated with blood flow pulsations, migraine is a typical and common disease, with a prevalence of about 15% among the population, most commonly occurring among the young and middle-aged population, and a prevalence of about 3 times that of men among women in the 20 to 50 year old population, which makes migraine the third most common disease worldwide. Because of the complexity of its pathogenesis, the flow regulation shows a lower diagnosis rate for migraine, only 50% for episodic migraine and 40% for chronic migraine; the misdiagnosis rate is high, and migraine is easily confused with cluster headache or tension headache.
The general clinical treatment methods for migraine mainly comprise two methods, namely a scale inquiry mode and an electroencephalogram event related potential (event related potentials, ERP). The scale mode requires clinical manifestation inquiry of a patient by a clinician, migraine screening and pain degree scoring by comparing the scale, missed diagnosis and misdiagnosis are easy to occur, and the accuracy rate is only 50%; the electroencephalogram event related potential technology objectively quantifies clinically observed abnormal cognitive activities so as to evaluate the cortex state and the functions of migraine, but potentiometer equipment does not generally provide standardized test tasks and needs to be developed and designed by combining actual experiment requirements, so that the standardized test tasks and paradigms are difficult to provide, and multi-dimensional evaluation indexes are lacked.
With the development of artificial intelligence, especially machine vision technology, an attempt is made in the industry to apply the vascular pulsation visualization technology to the field of monitoring diseases such as migraine, etc., but after practice and analysis, the vascular pulsation visualization scheme at the present stage is found to have at least two restriction factors: 1) Lower detection accuracy: because the signal is subject to more external interference, and the signal extraction is highly dependent on subjective, simple and manual facial region selection, a large measurement error can be generated due to the lower photoplethysmography (PPG) signal quality, so the prior art cannot ensure high-precision identification of the symptoms; 2) Lower temporal resolution: the prior art performs analysis processing by extracting and comparing the difference of PPG signals between two areas over a period of time, but since diseases such as migraine are clinically often manifested as intermittent attacks, the existing time-period-based evaluation route easily submerges the instantaneous difference signal in the time-period signal, resulting in lower time resolution of the measurement results.
Disclosure of Invention
In view of the foregoing, the present invention aims to provide a method, apparatus and device for video-derived facial blood flow pulsatility quantitative evaluation for migraine, which solve the above-mentioned specific problems.
The technical scheme adopted by the invention is as follows:
in a first aspect, the present invention provides a method for video-derived facial blood flow pulsation quantitative assessment of migraine, comprising:
collecting a facial image of a tested person;
extracting facial feature nodes from the facial image, and dividing a plurality of facial target areas in the facial image according to the facial feature nodes and the vascularity characteristics of the target diseases;
acquiring photoplethysmography signals of each face target area, and carrying out continuous wavelet analysis on the photoplethysmography signals to obtain a time-frequency chart of the corresponding photoplethysmography signals, wherein the time-frequency chart comprises time-frequency distribution of blood flow pulsation amplitude values and blood flow pulsation phases;
based on the blood flow pulsation amplitude and the time-frequency distribution of the blood flow pulsation phase, respectively correspondingly calculating asymmetry and asynchronism of blood flow pulsation serving as quantitative evaluation indexes;
and identifying and evaluating the target diseases by utilizing the quantified asymmetry and/or asynchronism.
In at least one possible implementation, the acquiring photoplethysmographic signals of each of the face target areas includes: and extracting the photoplethysmograph signals of the face target areas and carrying out space integration to obtain the photoplethysmograph signals averaged by the face target areas.
In at least one possible implementation manner, the quantitative evaluation method further comprises: before quantitative evaluation index calculation is carried out, a higher signal quality frequency range related to blood flow pulsation in the photoplethysmography signal corresponding to the tested person is adaptively selected according to a preset threshold value in the time-frequency chart.
In at least one possible implementation thereof, the means for obtaining asymmetry of the blood flow pulsations comprises:
in the higher signal quality frequency range, carrying out frequency domain integration on the time-frequency distribution of the blood flow pulsation amplitude to obtain a first time domain signal with enhanced signal-to-noise ratio and normalized treatment;
subtracting the first time domain signals of the symmetrical face target areas and taking an absolute value to obtain an absolute amplitude difference curve;
and performing time integration on the absolute amplitude difference curve to obtain bilateral absolute difference values of the blood flow pulsation amplitude used for representing the asymmetry.
In at least one possible implementation thereof, the means for obtaining asynchrony of blood flow pulsations includes:
in the higher signal quality frequency range, performing frequency domain integration on time-frequency distribution of blood flow pulsation phase to obtain a second time domain signal with enhanced signal-to-noise ratio and normalized treatment;
subtracting the second time domain signals of the symmetrical face target area and taking an absolute value to obtain an absolute phase difference curve;
and performing time integration on the absolute amplitude difference curve to obtain a bilateral absolute difference value of the blood flow pulsation phase for representing the asynchronism.
In at least one possible implementation manner, the normalization processing includes: the signal-to-noise ratio enhanced time domain signal is normalized using the original photoplethysmograph signal.
In a second aspect, the present invention provides a video derived facial blood flow pulsation quantitative assessment device for migraine, comprising:
the video stream acquisition unit is used for acquiring the facial image of the tested person;
a facial region dividing unit, configured to extract facial feature nodes from the facial image, and divide a plurality of facial target regions in the facial image according to the facial feature nodes and the vascularity characteristics of the target diseases;
a photoplethysmograph signal time-frequency processing unit, configured to obtain photoplethysmograph signals of each face target area, and perform continuous wavelet analysis on the photoplethysmograph signals to obtain a time-frequency chart of corresponding photoplethysmograph signals, where the time-frequency chart includes a blood flow pulsation amplitude and a time-frequency distribution of a blood flow pulsation phase;
a quantitative evaluation index calculation unit for correspondingly calculating asymmetry and asynchronism of the blood flow pulsation serving as quantitative evaluation indexes based on the blood flow pulsation amplitude and the time-frequency distribution of the blood flow pulsation phase;
and the disease evaluation unit is used for identifying and evaluating the target diseases by utilizing the quantified asymmetry and/or asynchronism.
In at least one possible implementation thereof, the photoplethysmograph signal time-frequency processing unit comprises: a signal averaging component;
the signal averaging component is used for extracting the photoplethysmography signals of the face target areas and performing space integration to obtain the photoplethysmography signals averaged by the face target areas.
In at least one possible implementation manner, the quantitative evaluation index calculating unit includes: a signal frequency range selection component;
the signal frequency range selection component is used for adaptively selecting a higher signal quality frequency range which corresponds to the photoplethysmography signal of the tested person and is related to blood flow pulsation in the time-frequency chart according to a preset threshold before quantitative evaluation index calculation is carried out.
In a third aspect, the present invention provides a video-derived facial blood flow pulsation quantitative assessment device for migraine, comprising:
one or more processors, a memory, and one or more computer programs, the memory may employ a non-volatile storage medium, wherein the one or more computer programs are stored in the memory, the one or more computer programs comprising instructions which, when executed by the device, cause the device to perform the method as in the first aspect or any of the possible implementations of the first aspect.
The main design concept of the invention is that an optical plethysmographic imaging technology based on image data is used for acquiring the image of a tested face by using a video acquisition system with excellent imaging effect, extracting facial feature points by using artificial intelligence and an automatic algorithm, dividing a facial skin blood vessel small body area to extract pulse waveforms, analyzing and processing the waveforms by continuous wavelet transformation and other methods, establishing a space-time variation model of the amplitude and the phase of facial skin blood flow pulsation, realizing the visual imaging of facial blood flow pulsation caused by specific diseases, and dynamically monitoring the asymmetry and the asynchronism of blood flow pulsation in the occurrence and the progress of the diseases. The invention provides objective, sensitive, quantifiable and automatic detection indexes for the prediction, identification and prognosis of the blood flow pulsation related diseases, and provides reliable and accurate auxiliary support for the individual accurate diagnosis and treatment of the diseases.
Drawings
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the present invention will be further described with reference to the accompanying drawings, in which:
FIG. 1 is a flow chart of a method for quantitatively evaluating video-derived facial blood flow pulsatility for migraine provided by an embodiment of the present invention;
FIG. 2 is a schematic diagram of the use of a photoplethysmographic imaging video system provided by an embodiment of the present invention;
FIG. 3 is a time-frequency diagram of wavelet transform of blood flow pulsation amplitude provided by an embodiment of the present invention;
FIG. 4 is a time-frequency diagram of a wavelet transform of the pulsatile phase of blood flow provided by an embodiment of the present invention;
fig. 5 is a schematic diagram of a video derived facial blood flow pulsation quantitative evaluation device for migraine provided by an embodiment of the invention.
Detailed Description
Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to like or similar elements or elements having like or similar functions throughout. The embodiments described below by referring to the drawings are illustrative only and are not to be construed as limiting the invention.
The present invention proposes an embodiment of a method for quantitatively evaluating the pulsatility of a video-derived face for migraine, specifically, as shown in fig. 1 (note that the schematic diagram is a comprehensive flow and does not include the following step numbers), which includes:
s1, collecting a facial image of a tested person;
in some preferred embodiments of the invention, a photoplethysmographic imaging video system as shown in fig. 2 is preferably employed, which may mainly comprise a light source module, an image acquisition module, in particular provided with a polarization module. The light source module is used for providing uniform and stable illumination of the face; the image acquisition module adopts a high-bit-depth camera and is used for recording high-bit-depth video data of the face scattered light signals; the polarization module is used for eliminating the influence of the face specular reflection light on the scattered light signal, so that the signal-to-noise ratio of image acquisition is improved.
S2, extracting facial feature nodes from the facial image, and dividing a plurality of facial target areas corresponding to anatomic vascular small areas in the facial image according to the facial feature nodes and the vascular distribution characteristics of the target diseases;
the link can be implemented by using a mature artificial intelligence algorithm, and by combining the vascular distribution characteristics (which are indicated by migraine below) related to the facial and target diseases (which belong to the prior knowledge in the medical field), six areas of left forehead, right forehead, left cheek and right cheek can be automatically divided by a machine based on facial feature nodes (it can be understood that the method is not limited to the six areas in practical application, all skin areas related to the target diseases can be used for monitoring and quantification calculation), and each area corresponds to a relatively independent anatomic vascular small area.
S3, acquiring photoplethysmography signals of each face target area, and carrying out continuous wavelet analysis on the photoplethysmography signals to obtain a time-frequency diagram of the corresponding photoplethysmography signals, wherein the time-frequency diagram comprises time-frequency distribution of blood flow pulsation amplitude and blood flow pulsation phase;
specifically, for example, the photoplethysmographic signals of pixels in the six face target regions in the above example can be extracted and spatially integrated to obtain the PPG signals averaged for the respective regions, which are defined as、、、、、Corresponding to left forehead, right forehead, left cheek and right cheek. The advantage of this operation is: the region-averaged PPG signal has a higher signal-to-noise ratio than a single-pel PPG signal.
For signalsRespectively performing continuous wavelet analysis to obtain pairsThe corresponding time-frequency distribution, the horizontal axis being the time axis and the vertical axis being the frequency, the value of each point on the time-frequency (time-frequency) diagram can be expressed as. Since each point on the time-frequency diagram is complex, the pairThe amplitude and the phase characteristics can be obtained by splitting, and the time-frequency characteristics of the amplitude can be expressed asAs shown in fig. 3 a), represents the distribution of regional blood flow pulsation amplitude over time and frequency; the time-frequency characteristic of the phase can be expressed asAs shown in fig. 4 b), represents the distribution of regional blood flow pulsatile phase over time and frequency.
Step S4, respectively correspondingly calculating asymmetry and asynchronism of the blood flow pulsation serving as quantitative evaluation indexes based on the time-frequency distribution of the blood flow pulsation amplitude and the blood flow pulsation phase;
in some preferred embodiments of the present invention, before performing the quantitative evaluation index calculation, preferably, the higher signal quality frequency range [ f ] with high correlation with the heartbeat frequency (i.e. high correlation with the blood flow pulsation) in the PPG signal of the tested person can be adaptively selected according to the superficial blood flow pulsation difference of different tested persons through reasonable amplitude threshold (which can be a proportional value) set by expert experience in a time-frequency chart of the blood flow pulsation amplitude (as in a fig. 3 a) l ,f h ]For example, the amplitude reaches the peak amplitude70% (which can be optimally adjusted according to the actual condition of the sample to be tested), corresponding to a low frequency f l High frequency f h I.e. higher signal quality frequency range f l ,f h ]Corresponding to six faces as shown in c) of FIG. 3And (5) marking areas.
Based on this concept, the time-frequency characteristic of the amplitude of the blood flow pulsationsOf [ f ] l , f h ]In the frequency range, frequency domain integration is carried out to obtain a time domain signal with enhanced signal to noise ratioThe process and the calculation of the average PPG signal in each face target area in the previous step can be jointly applied, so that the signal-to-noise ratio of the signal can be remarkably improved;
the original photoplethysmographic signal can then be usedFor a pair ofPerforming normalization processing, which can refer to the following formula (1) to obtain normalized amplitude signalThe time domain plot of the amplitude signal is shown in e) of fig. 3;(1) Then, the normalized amplitude signal of the symmetrical face target region (which can be preset in the step of dividing the region) is divided (e.g.、，、，、) Subtracting and taking absolute values to obtain absolute amplitude difference curves in a higher time resolution rangeThe signal is also a time domain signal and contains transient changes in the magnitude of the blood flow pulsations caused by migraine symptoms, as shown in g) of fig. 3;
finally, for absolute amplitude difference curvesAnd (5) performing time integration to obtain bilateral absolute difference values of the blood flow pulsation amplitude, so as to represent asymmetry of the blood flow pulsation.
Similarly, the way to find the asynchrony index can be referred to as follows:
based on the time-frequency distribution of the blood flow pulsatile phase (b) of fig. 4), at [ f ] l , f h ]As shown in the frequency range (d) of fig. 4), frequency domain integration is performed to obtain a time domain signal with enhanced signal-to-noise ratioAnd through similar normalization processing mode as before, normalized phase signals are obtainedAs shown in f) of fig. 4;
then, the normalized signal of the symmetric region (e.g、，、，、) Subtracting and taking absolute value to obtain absolute phase difference curve in higher time resolution rangeThe signal is also a time domain signal and contains transient changes in the phase of the pulsatile flow caused by migraine symptoms, as shown in h) of fig. 4;
finally, for absolute phase difference curveAnd (3) performing time integration to obtain bilateral absolute difference values of the blood flow pulsation phases, so as to represent the asynchronism of the blood flow pulsation.
And S5, identifying and evaluating the target diseases by utilizing the quantified asymmetry and/or asynchronism.
The main task of the present invention has been completed in the foregoing embodiments, and finally, the two quantitative indexes can be used in combination with medical priori knowledge to identify and evaluate the target disease, which is not limited by how to evaluate the target disease, for example, but not limited by, selecting one or more indexes from the two indexes based on the characteristics of the target disease to evaluate, for example, adopting a weighted manner of the two indexes, or using only the index value with obvious abnormality as the main evaluation basis, which is not described herein, and it can be supplemented that the evaluation of the disease symptoms of the target disease by using the quantitative indexes can be completed manually by doctors with special experience, without using computers.
Aiming at the technical problems mentioned above:
firstly, the invention has higher detection precision of identifying abnormal blood flow pulsation, and after a time-frequency diagram of PPG signals is acquired, signals with high correlation degree with heart beat frequency (namely, high correlation degree with blood pulsation) in the PPG signals are adaptively selected through reasonable thresholds according to the superficial blood pulsation differences of different testees, and extraction and amplification are carried out. The operation can greatly improve the signal to noise ratio of the blood pulsation signal, thereby obviously improving the recognition accuracy of facial blood pulsation abnormality and reducing the misjudgment rate;
secondly, the invention can obtain the blood pulsation abnormality observation with high time resolution and transient state. As described above, in the prior art, a PPG time domain signal with a long time is required to be intercepted, a comprehensive difference of blood pulsation in the period is calculated, the time point where the transient change of the blood pulsation is located cannot be expressed, but after the time-frequency decomposition is performed by wavelet analysis, the result is separated into an amplitude component and a phase component, both components are curves which change along with time, an instantaneous signal is completely reserved in the curves, not only information of the change of intensity and phase along with time is contained, but also the dissimilarity value can respectively represent the asymmetry and asynchronism of facial blood pulsation after the time integration of the amplitude and phase curves. Therefore, the invention has the advantage of high time resolution which is not possessed by the prior art, and is more suitable for identifying and detecting the diseases which can occur transient and abnormal changes of blood pulsation.
For this reason, the proposed solution of the invention was also validated on the basis of clinical samples (migraine attack period group, migraine attack interval group, control group), analyzed: the phase asynchrony of each area of the seizure period group is significantly greater than that of the normal group, the forehead and lateral forehead areas of the seizure period group are significantly greater than that of the normal group, and most (75%) of the cases show an increase in the pulsation intensity and asymmetry of each area from the seizure period to the seizure period. The verification conclusion shows that the abnormal blood flow pulsation of the migraine sufferer can be effectively observed in an intuitive and visual mode, and quantitative objective evaluation can be carried out.
In summary, the main design concept of the present invention is to collect the image of the tested face by using the video collection system with excellent imaging effect based on the photoplethysmography imaging technology of the image data, extract the facial feature points by using the artificial intelligence and the automation algorithm, divide the facial skin blood vessel small body area to extract the pulse waveform, analyze the waveform by the continuous wavelet transformation and other methods, establish the space-time variation model of the facial skin blood flow pulsation amplitude and phase, realize the visual imaging of the facial blood flow pulsation caused by the specific disease, and dynamically monitor the asymmetry and asynchronism of the blood flow pulsation in the occurrence and the progress of the disease. The invention provides objective, sensitive, quantifiable and automatic detection indexes for the prediction, identification and prognosis of the blood flow pulsation related diseases, and provides reliable and accurate auxiliary support for the individual accurate diagnosis and treatment of the diseases.
Corresponding to the above embodiments and preferred solutions, the present invention further provides an embodiment of a video derived facial blood flow pulsation quantitative evaluation device for migraine, as shown in fig. 5, which may specifically include the following components:
the video stream acquisition unit 1 is used for acquiring facial images of a detected person;
a facial region dividing unit 2, configured to extract facial feature nodes from the facial image, and divide a plurality of facial target regions in the facial image according to the facial feature nodes and the vascularity characteristics of the target diseases;
a photoplethysmogram signal time-frequency processing unit 3, configured to obtain photoplethysmogram signals of each of the face target areas, and perform continuous wavelet analysis on the photoplethysmogram signals to obtain a time-frequency chart of corresponding photoplethysmogram signals, where the time-frequency chart includes a blood flow pulsation amplitude and a time-frequency distribution of a blood flow pulsation phase;
a quantitative evaluation index calculation unit 4 for correspondingly calculating asymmetry and asynchronism of the blood flow pulsation as quantitative evaluation indexes based on the blood flow pulsation amplitude and the time-frequency distribution of the blood flow pulsation phase, respectively;
and the disease evaluation unit 5 is used for identifying and evaluating the target disease by utilizing the quantified asymmetry and/or asynchronism.
In at least one possible implementation thereof, the photoplethysmograph signal time-frequency processing unit comprises: a signal averaging component;
the signal averaging component is used for extracting the photoplethysmography signals of the face target areas and performing space integration to obtain the photoplethysmography signals averaged by the face target areas.
In at least one possible implementation manner, the quantitative evaluation index calculating unit includes: a signal frequency range selection component;
the signal frequency range selection component is used for adaptively selecting a higher signal quality frequency range which corresponds to the photoplethysmography signal of the tested person and is related to blood flow pulsation in the time-frequency chart according to a preset threshold before quantitative evaluation index calculation is carried out.
It should be understood that the division of the components in the video derived facial blood flow pulsation quantitative evaluation device for migraine shown in fig. 5 is merely a division of logic functions, and may be fully or partially integrated into a physical entity or may be physically separated. And these components may all be implemented in software in the form of a call through a processing element; or can be realized in hardware; it is also possible that part of the components are implemented in the form of software called by the processing element and part of the components are implemented in the form of hardware. For example, some of the above units may be individually set up processing elements, or may be integrated in a chip of the electronic device. The implementation of the other components is similar. In addition, all or part of the components can be integrated together or can be independently realized. In implementation, each step of the above method or each component above may be implemented by an integrated logic circuit of hardware in a processor element or an instruction in the form of software.
For example, the above components may be one or more integrated circuits configured to implement the above methods, such as: one or more specific integrated circuits (Application Specific Integrated Circuit; hereinafter ASIC), or one or more microprocessors (Digital Signal Processor; hereinafter DSP), or one or more field programmable gate arrays (Field Programmable Gate Array; hereinafter FPGA), etc. For another example, these components may be integrated together and implemented in the form of a System-On-a-Chip (SOC).
In view of the foregoing examples and preferred embodiments thereof, it will be appreciated by those skilled in the art that in actual operation, the technical concepts of the present invention may be applied to various embodiments, and the present invention is schematically illustrated by the following carriers:
(1) A video derived facial blood flow pulsation quantitative assessment device for migraine. The device may specifically include: one or more processors, memory, and one or more computer programs, wherein the one or more computer programs are stored in the memory, the one or more computer programs comprising instructions, which when executed by the device, cause the device to perform the steps/functions of the foregoing embodiments or equivalent implementations.
The electronic device may be an electronic device related to a computer, such as, but not limited to, various computing terminals, electronic products, and the like.
Specifically, the device/terminal may be a computer device, and the hardware structure of the computer device may further specifically include: at least one processor, at least one communication interface, at least one memory and at least one communication bus; the processor, the communication interface and the memory can all communicate with each other through a communication bus. The processor may be a central processing unit CPU, DSP, microcontroller or digital signal processor, and may further include a GPU, an embedded Neural network processor (Neural-network Process Units; hereinafter referred to as NPU) and an image signal processor (Image Signal Processing; hereinafter referred to as ISP), and the processor may further include an ASIC (application specific integrated circuit) or one or more integrated circuits configured to implement embodiments of the present invention, and in addition, the processor may have a function of operating one or more software programs, and the software programs may be stored in a storage medium such as a memory; and the aforementioned memory/storage medium may include: nonvolatile Memory (Non-Volatile Memory), such as a Non-removable magnetic disk, a USB flash disk, a removable hard disk, an optical disk, and the like, and Read-Only Memory (ROM), random access Memory (Random Access Memory; RAM), and the like.
(2) A computer data storage medium having stored thereon a computer program or the above-mentioned means which, when executed, causes a computer to perform the steps/functions of the foregoing embodiments or equivalent implementations.
In several embodiments provided by the present invention, any of the functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer data storage medium. Based on such understanding, certain aspects of the present invention may be embodied in the form of a software product as described below, in essence, or as a part of, contributing to the prior art.
It is especially pointed out that the storage medium may refer to a server or a similar computer device, in particular, i.e. a storage means in the server or a similar computer device, in which the aforementioned computer program or the aforementioned means are stored.
(3) A computer program product (which may comprise the apparatus described above) which, when run on a terminal device, causes the terminal device to perform the video derived facial blood flow pulsation quantitative assessment method for migraine according to the previous example or equivalent implementation.
From the above description of embodiments, it will be apparent to those skilled in the art that all or part of the steps of the above described methods may be implemented in software plus necessary general purpose hardware platforms. Based on such understanding, the above-described computer program product may include, but is not limited to, an APP.
Claims (8)
1. A method for video-derived facial blood flow pulsation quantitative assessment of migraine, comprising:
collecting a facial image of a tested person;
extracting facial feature nodes from the facial image, and dividing a plurality of facial target areas in the facial image according to the facial feature nodes and the vascularity characteristics of the target diseases;
acquiring photoplethysmography signals of each face target area, and carrying out continuous wavelet analysis on the photoplethysmography signals to obtain a time-frequency chart of the corresponding photoplethysmography signals, wherein the time-frequency chart comprises time-frequency distribution of blood flow pulsation amplitude values and blood flow pulsation phases;
adaptively selecting a frequency range which corresponds to the photoplethysmography signal of the tested person and is related to blood flow pulsation according to a preset threshold in the time-frequency chart;
based on the blood flow pulsation amplitude and the time-frequency distribution of the blood flow pulsation phase, respectively correspondingly calculating asymmetry and asynchronism of blood flow pulsation serving as quantitative evaluation indexes; the way to obtain the asymmetry of the blood flow pulsation includes: performing frequency domain integration on time-frequency distribution of blood flow pulsation amplitude in the frequency range to obtain a first time domain signal with enhanced signal-to-noise ratio and normalized treatment; subtracting the first time domain signals of the symmetrical face target areas and taking an absolute value to obtain an absolute amplitude difference curve; performing time integration on the absolute amplitude difference curve to obtain bilateral absolute difference values of the blood flow pulsation amplitude used for representing the asymmetry;
and identifying and evaluating the target diseases by utilizing the quantified asymmetry and/or asynchronism.
2. The method of video derived facial blood flow pulsation quantitative assessment for migraine according to claim 1, wherein said acquiring photoplethysmographic signals of each of said facial target areas comprises: and extracting the photoplethysmograph signals of the face target areas and carrying out space integration to obtain the photoplethysmograph signals averaged by the face target areas.
3. The method for video-derived facial blood flow pulsation quantitative assessment of migraine according to claim 1, wherein the way to obtain asynchrony of blood flow pulsation comprises:
in the frequency range, performing frequency domain integration on time-frequency distribution of the blood flow pulsation phase to obtain a second time domain signal with enhanced signal-to-noise ratio and normalized treatment;
subtracting the second time domain signals of the symmetrical face target area and taking an absolute value to obtain an absolute phase difference curve;
and performing time integration on the absolute phase difference curve to obtain a bilateral absolute difference value of the blood flow pulsation phase for representing the asynchronism.
4. A method for video derived facial blood flow pulsation quantitative assessment of migraine according to claim 1 or 3, wherein said normalization process comprises: the signal-to-noise ratio enhanced time domain signal is normalized using the original photoplethysmograph signal.
5. A video-derived facial blood flow pulsation quantitative assessment device for migraine, comprising:
the video stream acquisition unit is used for acquiring the facial image of the tested person;
a facial region dividing unit, configured to extract facial feature nodes from the facial image, and divide a plurality of facial target regions in the facial image according to the facial feature nodes and the vascularity characteristics of the target diseases;
a photoplethysmograph signal time-frequency processing unit, configured to obtain photoplethysmograph signals of each face target area, and perform continuous wavelet analysis on the photoplethysmograph signals to obtain a time-frequency chart of corresponding photoplethysmograph signals, where the time-frequency chart includes a blood flow pulsation amplitude and a time-frequency distribution of a blood flow pulsation phase;
the quantitative evaluation index calculation unit is used for adaptively selecting a frequency range related to blood flow pulsation in the photoplethysmography signal corresponding to the tested person according to a preset threshold value in the time-frequency chart; based on the blood flow pulsation amplitude and the time-frequency distribution of the blood flow pulsation phase, respectively correspondingly calculating asymmetry and asynchronism of blood flow pulsation serving as quantitative evaluation indexes; the way to obtain the asymmetry of the blood flow pulsation includes: performing frequency domain integration on time-frequency distribution of blood flow pulsation amplitude in the frequency range to obtain a first time domain signal with enhanced signal-to-noise ratio and normalized treatment; subtracting the first time domain signals of the symmetrical face target areas and taking an absolute value to obtain an absolute amplitude difference curve; performing time integration on the absolute amplitude difference curve to obtain bilateral absolute difference values of the blood flow pulsation amplitude used for representing the asymmetry;
and the disease evaluation unit is used for identifying and evaluating the target diseases by utilizing the quantified asymmetry and/or asynchronism.
6. The video derived facial blood flow pulsation quantitative assessment device for migraine according to claim 5, wherein said photoplethysmography signal time-frequency processing unit comprises: a signal averaging component;
the signal averaging component is used for extracting the photoplethysmography signals of the face target areas and performing space integration to obtain the photoplethysmography signals averaged by the face target areas.
7. The video-derived facial blood flow pulsation quantitative evaluation device for migraine according to claim 5, wherein said quantitative evaluation index calculation unit comprises: a signal frequency range selection component;
the signal frequency range selection component is used for adaptively selecting a frequency range which corresponds to the photoplethysmography signal of the tested person and is related to blood flow pulsation in the time-frequency chart according to a preset threshold before quantitative evaluation index calculation is carried out.
8. A video-derived facial blood flow pulsation quantitative assessment device for migraine, comprising:
one or more processors, memory, and one or more computer programs, wherein the one or more computer programs are stored in the memory, the one or more computer programs comprising instructions that, when executed by the device, cause the device to perform the video-derived facial blood flow pulsation quantitative assessment method for migraine according to any of claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410114861.6A CN117649403B (en) | 2024-01-29 | 2024-01-29 | Video-derived facial blood flow pulsation quantitative evaluation method, device and equipment for migraine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410114861.6A CN117649403B (en) | 2024-01-29 | 2024-01-29 | Video-derived facial blood flow pulsation quantitative evaluation method, device and equipment for migraine |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117649403A CN117649403A (en) | 2024-03-05 |
| CN117649403B true CN117649403B (en) | 2024-04-12 |
Family
ID=90043565
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410114861.6A Active CN117649403B (en) | 2024-01-29 | 2024-01-29 | Video-derived facial blood flow pulsation quantitative evaluation method, device and equipment for migraine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117649403B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016110804A1 (en) * | 2015-01-06 | 2016-07-14 | David Burton | Mobile wearable monitoring systems |
| WO2017045976A1 (en) * | 2015-09-18 | 2017-03-23 | Koninklijke Philips N.V. | Device and method for migraine monitoring |
| CN111260634A (en) * | 2020-01-17 | 2020-06-09 | 天津工业大学 | Facial blood flow distribution extraction method and system |
| CN111728602A (en) * | 2020-08-21 | 2020-10-02 | 之江实验室 | Non-contact blood pressure measuring device based on PPG |
| CN114983414A (en) * | 2022-05-12 | 2022-09-02 | 华南理工大学 | Brain cognitive load quantification method based on facial video physiological index extraction |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8388543B2 (en) * | 2007-05-16 | 2013-03-05 | The Research Foundation Of State University Of New York | Photoplethysmography apparatus and method employing high resolution estimation of time-frequency spectra |
| US11903680B2 (en) * | 2015-06-14 | 2024-02-20 | Facense Ltd. | Wearable-based health state verification for physical access authorization |
-
2024
- 2024-01-29 CN CN202410114861.6A patent/CN117649403B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016110804A1 (en) * | 2015-01-06 | 2016-07-14 | David Burton | Mobile wearable monitoring systems |
| WO2017045976A1 (en) * | 2015-09-18 | 2017-03-23 | Koninklijke Philips N.V. | Device and method for migraine monitoring |
| CN111260634A (en) * | 2020-01-17 | 2020-06-09 | 天津工业大学 | Facial blood flow distribution extraction method and system |
| CN111728602A (en) * | 2020-08-21 | 2020-10-02 | 之江实验室 | Non-contact blood pressure measuring device based on PPG |
| CN114983414A (en) * | 2022-05-12 | 2022-09-02 | 华南理工大学 | Brain cognitive load quantification method based on facial video physiological index extraction |
Non-Patent Citations (1)
| Title |
|---|
| Asynchronicity of Facial Blood Perfusion in Migraine;Nina Zaproudina，et al.;PLoS ONE;20131204;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117649403A (en) | 2024-03-05 |
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 | |
| Wu et al. | Automatic epileptic seizures joint detection algorithm based on improved multi-domain feature of cEEG and spike feature of aEEG | |
| JP7429371B2 (en) | Method and system for quantifying and removing asynchronous noise in biophysical signals | |
| Faust et al. | Linear and non-linear analysis of cardiac health in diabetic subjects | |
| Lodder et al. | Automated EEG analysis: characterizing the posterior dominant rhythm | |
| US20240057888A1 (en) | Method And System For Determining Brain-State Dependent Functional Areas Of Unitary Pooled Activity And Associated Dynamic Networks With Functional Magnetic Resonance Imaging | |
| CN113995419B (en) | Atrial fibrillation risk prediction system based on heartbeat rhythm signal and application thereof | |
| CN111466876A (en) | Alzheimer's disease auxiliary diagnosis system based on fNIRS and graph neural network | |
| Chen et al. | Scalp EEG-based pain detection using convolutional neural network | |
| CN106974660B (en) | Method for realizing gender judgment based on blood oxygen characteristics in brain function activity detection | |
| 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 | |
| CN117649403B (en) | Video-derived facial blood flow pulsation quantitative evaluation method, device and equipment for migraine | |
| Gholamipour et al. | Estimation of the independent components reliability of EEG signal in a clinical application | |
| US11903742B2 (en) | Non-invasive non-contact system and method for measuring dyslipidemia condition using thermal imaging | |
| Kukreti | EEG-based Stress Detection using DEAP Dataset and Support Vector Machines | |
| Singhal et al. | Electrocardiogram Feature Based Heart Arrhythmia Detection Using Machine Learning and Apache Spark | |
| US20220280121A1 (en) | Non-invasive non-contact system and method for evaluating primary and secondary hypertension conditions using thermal imaging | |
| Prasad et al. | A novel electrocardiogram signal feature prediction technique using peak-to-peak sampling pattern for multiple abnormality level classification | |
| KR101870919B1 (en) | Method and apparatus for extracting Brain information | |
| García-Limón et al. | An Improved Estimation of Unsuitable Segments of Ballistocardiography Records Using Wavelet Transforms | |
| Malar et al. | FPGA Implementation of Heart Rate Variability using rPPG and Peak Detection | |
| CN117918856A (en) | INPH prediction method and iNPH prediction equipment based on new and old stimulation BCI paradigm | |
| Butkevičiūtė et al. | The unsupervised pattern recognition for the ECG signal features detection |
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 |