Detection method and device for assisting in determining blood pressure and cardiovascular health level

Abstract

The present invention provides a detection method and device for assisting in determining blood pressure and cardiovascular health level, which can be applied to the field of coherent optical imaging detection technology. The method includes: within a preset time length, continuously collecting diffuse coherent light emitted from tissue at a preset position to obtain multiple diffuse coherent images; according to the multiple diffuse coherent images, determining the dynamic coherent light signal corresponding to each pixel, and according to the dynamic coherent light signal and the reference coherent light signal, determining the tissue static signal and the tissue dynamic signal; calculating the information entropy of the tissue dynamic signal to obtain the flow velocity pulse signal; calculating the average intensity of the static signal to obtain the volume pulse signal; performing Fourier transform on the flow velocity pulse signal and the volume pulse signal respectively to obtain the flow velocity pulse spectrum and the volume pulse spectrum; performing cross-correlation operation on the flow velocity pulse spectrum and the volume pulse spectrum to obtain multi-dimensional fusion features, and determining the blood pressure and cardiovascular health level according to the multi-dimensional fusion features.

Description

Technical Field

[0001] The present invention relates to the field of coherent optical imaging detection technology, and more specifically, to a detection method and device for assisting in determining blood pressure and cardiovascular health level. Background Art

[0002] As one of the hemodynamic indicators, blood flow velocity is closely related to physiological health indicators such as human blood pressure and cardiovascular health. Since optical detection methods have advantages such as high resolution and fast dynamic response speed, they are widely used in non-invasive vascular and component detection.

[0003] In the process of realizing the concept of the present invention, it was found through research that the relevant technologies have at least the difficulty in realizing blood flow detection in tissues at different depths and the difficulty in distinguishing blood flow velocity and volume change due to limited measurement modalities and parameters, making it difficult to accurately detect in-body blood pressure and cardiovascular health level. Summary of the invention

[0004] In view of this, the present invention provides a detection method and device for assisting in determining blood pressure and cardiovascular health level.

[0005] One aspect of the present invention provides a detection method for assisting in determining blood pressure and cardiovascular health level, comprising:

[0006] In the case where reference coherent lights in different polarization states corresponding to a plurality of wavelengths are incident on tissue at a preset position of a target object at the same time, diffuse coherent lights emitted from the tissue at the preset position are continuously collected within a preset time length to obtain a plurality of diffuse coherent images, wherein the plurality of diffuse coherent images include features reflecting blood flow states in tissues at different depths; a dynamic coherent light signal corresponding to each pixel is determined based on the plurality of diffuse coherent images collected at each time period of the preset time length, and a tissue static signal and a reference coherent light signal corresponding to the pixel are determined based on the dynamic coherent light signal and the reference coherent light signal corresponding to the pixel. Tissue dynamic signal, wherein the reference coherent light signal is a signal used to characterize the reference coherent light corresponding to the pixel; the information entropy of the tissue dynamic signal of each pixel in the same time period is calculated to obtain the flow velocity pulse signal; the average intensity of the tissue static signal of each pixel in the same time period is calculated to obtain the volume pulse signal; the flow velocity pulse signal and the volume pulse signal are Fourier transformed respectively to obtain the flow velocity pulse spectrum and the volume pulse spectrum; the flow velocity pulse spectrum and the volume pulse spectrum are cross-correlated to obtain multi-dimensional fusion features; according to the multi-dimensional fusion features, the blood pressure and cardiovascular health level are determined.

[0007] According to an embodiment of the present invention, determining the blood pressure and cardiovascular health level based on multidimensional fusion features includes: extracting homogeneity features, energy features, contrast features and correlation features from the multidimensional fusion features; inputting the homogeneity features, energy features, contrast features and correlation features into the target regression analysis model to obtain the blood pressure and cardiovascular health level.

[0008] According to an embodiment of the present invention, a dynamic coherent light signal corresponding to each pixel is determined based on a plurality of diffuse coherent images collected in each time period of a preset time length, and a tissue static signal and a tissue dynamic signal are determined based on the dynamic coherent light signal and the reference coherent light signal corresponding to the pixel, including: performing correlation calculation on the dynamic coherent light signal and the reference coherent light signal to obtain initial blood flow state information; performing singular value decomposition on the initial blood flow state information to obtain a left singular value matrix, a diagonal matrix, and a right singular value matrix; determining the tissue static signal based on the first K row vectors of the left singular value matrix, the diagonal matrix, and the right singular value matrix; determining the tissue dynamic signal based on the K+1 to nth row vectors of the left singular value matrix, the diagonal matrix, and the right singular value matrix, wherein K is a positive integer and n is an integer greater than 1.

[0009] According to an embodiment of the present invention, extracting homogeneity features, energy features, contrast features, and correlation features from multidimensional fusion features includes extracting homogeneity features according to the following formula (I):

[0010] (one)

[0011] in, The homogeneity characteristic, Indicates the gray level is and The probability statistics of the pixels appearing simultaneously in each single-channel fusion feature included in the multi-dimensional fusion feature, and N is the maximum gray level of the pixels in the single-channel fusion feature.

[0012] According to an embodiment of the present invention, extracting homogeneity features, energy features, contrast features, and correlation features from multi-dimensional fusion features includes extracting energy features according to the following formula (ii):

[0013] (two)

[0014] in, Energy characteristics.

[0015] According to an embodiment of the present invention, extracting homogeneity features, energy features, contrast features, and correlation features from multidimensional fusion features includes extracting contrast features according to the following formula (III):

[0016] (three)

[0017] in, is the contrast feature.

[0018] According to an embodiment of the present invention, extracting homogeneity features, energy features, contrast features, and correlation features from multidimensional fusion features includes extracting correlation features according to the following formula (IV):

[0019] (Four)

[0020] in, is the correlation feature, , For The corresponding mean, , For The corresponding mean, , For The corresponding standard deviation, , For The corresponding standard deviation.

[0021] Another aspect of the present invention provides a detection device for assisting in determining blood pressure and cardiovascular health level, comprising:

[0022] A collection unit, configured to continuously collect diffuse coherent light emitted from the tissue at a preset position within a preset time length, and obtain a plurality of diffuse coherent images, each of which includes a feature reflecting the blood flow state in the tissue at different depths, when simultaneously incident reference coherent light in different polarization states corresponding to a plurality of wavelengths on the tissue at a preset position of the target object;

[0023] A processor is used to determine a dynamic coherent light signal corresponding to each pixel based on a plurality of diffuse coherent images collected in each time period of a preset time length, and to determine a tissue static signal and a tissue dynamic signal based on the dynamic coherent light signal and a reference coherent light signal corresponding to the pixel, wherein the reference coherent light signal is a signal used to characterize the reference coherent light corresponding to the pixel; calculate the information entropy of the tissue dynamic signal of each pixel in the same time period to obtain a flow velocity pulse signal; calculate the average intensity of the tissue static signal of each pixel in the same time period to obtain a volume pulse signal; perform Fourier transform on the flow velocity pulse signal and the volume pulse signal respectively to obtain a flow velocity pulse spectrum and a volume pulse spectrum; perform cross-correlation operation on the flow velocity pulse spectrum and the volume pulse spectrum to obtain a multi-dimensional fusion feature; determine the blood pressure and cardiovascular health level based on the multi-dimensional fusion feature.

[0024] According to an embodiment of the present invention, the acquisition unit includes: a coherent light source, which is used to emit reference coherent light in different polarization states corresponding to multiple wavelengths, and simultaneously incident the reference coherent light on tissue at a preset position of the target object; an image acquisition module: which is used to continuously acquire diffuse coherent light emitted from tissue at a preset position within a preset time length to obtain multiple diffuse coherent images.

[0025] According to an embodiment of the present invention, the image acquisition module includes: a polarization detection mask: used to divide the diffuse coherent light emitted from the tissue at a preset position into coherent lights corresponding to different polarization states; a wavelength partition detection mask, used to filter the coherent lights corresponding to different polarization states to obtain regional polarized coherent lights in different polarization states corresponding to multiple wavelengths; a photoelectric detection array, used to perform photoelectric conversion on the regional polarized coherent light within a preset time length to obtain multiple diffuse coherent images.

[0026] According to an embodiment of the present invention, when reference coherent light in different polarization states corresponding to multiple wavelengths is incident on tissue at a preset position of a target object at the same time, diffuse coherent light emitted from the tissue at the preset position is continuously collected within a preset time length, and multiple diffuse coherent images reflecting the characteristics of the blood flow state in tissues at different depths are obtained. Based on the multiple diffuse coherent images collected in each time period of the preset time length, the dynamic coherent light signal corresponding to each pixel is determined, and based on the dynamic coherent light signal and the reference coherent light signal corresponding to the pixel, the tissue static signal and the tissue dynamic signal are determined, and the information entropy of the tissue dynamic signal of each pixel in the same time period is calculated to obtain a flow velocity pulse signal; the average intensity of the tissue static signal of each pixel in the same time period is calculated to obtain a volume pulse signal, so as to achieve the resolution of blood flow velocity and blood flow volume in tissues at different depths. The flow velocity pulse signal and the volume pulse signal are respectively subjected to Fourier transform to obtain the flow velocity pulse spectrum and the volume pulse spectrum; based on the one-dimensional spectrum of the flow velocity pulse and the volume pulse, the flow velocity pulse spectrum and the volume pulse spectrum are cross-correlated to obtain multi-dimensional fusion features; based on the multi-dimensional fusion features, the blood pressure and cardiovascular health level are accurately predicted, and multi-modal detection of in vivo blood flow and pulse wave is realized. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The above and other objects, features and advantages of the present invention will become more apparent through the following description of the embodiments of the present invention with reference to the accompanying drawings, in which:

[0028] Figure 1 A flow chart of a detection method for assisting in determining blood pressure and cardiovascular health level according to an embodiment of the present invention is shown.

[0029] Figure 2A schematic diagram of the structure of a detection device for assisting in determining blood pressure and cardiovascular health level according to an embodiment of the present invention is shown.

[0030] Figure 3 A schematic structural diagram of a detection device for assisting in determining blood pressure and cardiovascular health level according to another embodiment of the present invention is shown. DETAILED DESCRIPTION

[0031] Below, embodiments of the present invention will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present invention. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of embodiments of the present invention. However, it is apparent that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of concepts of the present invention.

[0032] The terms used herein are only for describing specific embodiments and are not intended to limit the present invention. The terms "comprise", "include", etc. used herein indicate the existence of the features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components.

[0033] All terms (including technical and scientific terms) used herein have the meanings commonly understood by those skilled in the art unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0034] When using expressions such as "at least one of A, B, and C, etc.", they should generally be interpreted according to the meaning of the expression commonly understood by those skilled in the art (for example, "a system having at least one of A, B, and C" should include but is not limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, C, etc.).

[0035] In the embodiments of the present invention, the collection, updating, analysis, processing, use, transmission, provision, invention, storage and other aspects of the data involved (for example, including but not limited to user personal information) are in compliance with the provisions of relevant laws and regulations, are used for legitimate purposes, and do not violate public order and good morals. In particular, necessary measures are taken for user personal information to prevent illegal access to user personal information data and maintain the security of user personal information and network security.

[0036] In the embodiment of the present invention, before obtaining or collecting the user's personal information, the user's authorization or consent is obtained.

[0037] Blood circulation drives blood to circulate continuously in blood vessels through the rhythmic contraction and relaxation of the heart, transporting various nutrients to tissues and organs, and excreting metabolic waste from the body to ensure the body's metabolism and health. However, blood flow is affected by various factors such as myocardial contractility, ventricular filling, heart rate, resistance of the vascular system, blood viscosity, and blood volume. Therefore, when cardiovascular disease occurs, the heart's ability to transport blood will decrease, slowing the flow of blood in blood vessels and causing abnormal blood rheology, which is a precursor symptom of diseases such as stroke and myocardial infarction.

[0038] At present, the basic principles of clinical prevention and treatment of diseases caused by cardiovascular dysfunction are: early detection, early intervention, and long-term monitoring. Continuous blood flow detection in vivo can not only obtain the parameters required for daily health testing such as heart rate, pulse, blood pressure, etc., but also provide a reliable basis for the evaluation of cardiovascular health parameters. Therefore, it is the main research direction of wearable health monitoring.

[0039] The related analysis and measurement of blood vessels and blood flow in human tissues include laser Doppler measurement technology, laser speckle contrast imaging technology and diffusion correlation spectroscopy technology.

[0040] Laser Doppler technology is based on the Doppler effect of light. It detects the frequency shift signal generated by the scattering of laser light by moving red blood cells in the blood flow to obtain the blood flow velocity. However, the blood flow detection depth of laser Doppler technology is shallow, and its application scenarios have certain limitations. It is only suitable for the detection of blood flow in superficial tissues of several hundred microns. The accuracy and reliability of the measurement results are easily affected by the movement of surrounding tissues, respiratory movement, etc.

[0041] Laser speckle contrast imaging uses the spatial and temporal frequency characteristics of the two-dimensional speckle interference image of biological tissue to extract blood flow information in the tissue. After the coherent light is scattered by the biological tissue, the reflected or transmitted light has different scattering directions and optical paths. After meeting in space, random interference occurs, forming a two-dimensional speckle interference pattern. If the incident light is scattered by moving red blood cells, Doppler frequency shift will be generated, and the magnitude of the frequency shift is related to the speed of red blood cell movement. The dynamic speckle intensity will be modulated by the moving red blood cells. By analyzing the temporal-spatial modulation intensity of the speckle interference signal, it is suitable for detecting information such as the blood flow velocity of uniform tissue. The laser speckle imaging system mainly includes a laser light source, an imaging lens, and a camera. It has the advantages of a large imaging area, high spatial resolution, and fast imaging speed. At present, laser speckle contrast technology is widely used in fundus imaging, microcirculation status monitoring, brain science research, and trauma and surgical process monitoring, but it is difficult to apply to detection in other fields.

[0042] The existing diffusion correlation spectroscopy technology has the light penetration advantage of near-infrared spectroscopy technology, so it can detect blood flow in deep tissues. At the same time, it also has the advantages of low cost, high resolution, and real-time continuous monitoring, showing great advantages in daily blood flow velocity measurement. The diffusion correlation spectroscopy system is usually composed of a single-frequency laser in the near-infrared band, a high-speed photodetector, a high-speed acquisition card, and a data processing system. After the coherent light emitted by the laser enters the biological tissue, it will be scattered by the moving red blood cells. The amplitude and phase of the scattered light field change with the scattering direction and the cell movement speed. The dynamic scattered light in different directions produces fluctuations in light intensity after interference. The outgoing light is collected at a certain distance from the incident point, and the time series of light intensity changes is autocorrelated. The electric field autocorrelation function can be further calculated based on the Siegert relationship, a mathematical relationship used to describe the light intensity fluctuations in dynamic light scattering, and then the blood flow index representing the blood flow velocity is reconstructed by fitting the electric field autocorrelation function.

[0043] However, the relevant technology is only applicable to blood flow detection in uniform tissues, and the stratification of skin tissue is not considered in the model and technology. In addition, due to the influence of fiber lasers and single-photon counters in related equipment, the detection range and depth are limited, making it difficult to detect blood flow in tissues at different depths. At the same time, it is difficult to distinguish blood flow velocity and volume transformation, and the measurement mode and parameters are limited, making it difficult to accurately evaluate blood pressure and cardiovascular health in vivo.

[0044] In view of this, an embodiment of the present invention provides a detection method for assisting in determining blood pressure and cardiovascular health level, comprising:

[0045] In the case where reference coherent lights in different polarization states corresponding to a plurality of wavelengths are incident on tissue at a preset position of a target object at the same time, diffuse coherent lights emitted from the tissue at the preset position are continuously collected within a preset time length to obtain a plurality of diffuse coherent images, wherein the plurality of diffuse coherent images include features reflecting blood flow states in tissues at different depths; a dynamic coherent light signal corresponding to each pixel is determined based on the plurality of diffuse coherent images collected at each time period of the preset time length, and a tissue static signal and a reference coherent light signal corresponding to the pixel are determined based on the dynamic coherent light signal and the reference coherent light signal corresponding to the pixel. Tissue dynamic signal, wherein the reference coherent light signal is a signal used to characterize the reference coherent light corresponding to the pixel; the information entropy of the tissue dynamic signal of each pixel in the same time period is calculated to obtain the flow velocity pulse signal; the average intensity of the tissue static signal of each pixel in the same time period is calculated to obtain the volume pulse signal; the flow velocity pulse signal and the volume pulse signal are Fourier transformed respectively to obtain the flow velocity pulse spectrum and the volume pulse spectrum; the flow velocity pulse spectrum and the volume pulse spectrum are cross-correlated to obtain multi-dimensional fusion features; according to the multi-dimensional fusion features, the blood pressure and cardiovascular health level are determined.

[0046] Figure 1 A flow chart of a detection method for assisting in determining blood pressure and cardiovascular health level according to an embodiment of the present invention is shown.

[0047] like Figure 1 As shown, the detection method 100 for assisting in determining blood pressure and cardiovascular health level of this embodiment includes operations S110 to S170.

[0048] In operation S110, when reference coherent lights corresponding to multiple wavelengths and in different polarization states are simultaneously incident on tissue at a preset position of the target object, diffuse coherent lights emitted from the tissue at the preset position are continuously collected within a preset time period to obtain multiple diffuse coherent images.

[0049] In operation S120, a dynamic coherent light signal corresponding to each pixel is determined based on a plurality of diffuse coherent images collected in each time period of a preset time length, and a tissue static signal and a tissue dynamic signal are determined based on the dynamic coherent light signal and a reference coherent light signal corresponding to the pixel, wherein the reference coherent light signal is a signal for characterizing a reference coherent light corresponding to the pixel.

[0050] In operation S130, information entropy is calculated for the tissue dynamic signal of each pixel in the same time period to obtain a flow velocity pulse signal. The preset time period includes multiple time periods, and the multiple time periods occupy the same length of the preset time period.

[0051] In operation S140, the average intensity of the tissue static signal of each pixel in the same time period is calculated to obtain a volume pulse signal.

[0052] In operation S150, Fourier transform is performed on the flow velocity pulse signal and the volume pulse signal to obtain a flow velocity pulse spectrum and a volume pulse spectrum.

[0053] In operation S160, a cross-correlation operation is performed on the flow velocity pulse spectrum and the volume pulse spectrum to obtain a multi-dimensional fusion feature.

[0054] In operation S170, blood pressure and cardiovascular health level are determined based on the multi-dimensional fusion features.

[0055] According to an embodiment of the present invention, the reference coherent light can be a long coherent light wave source with multiple different wavelengths. The coherent light can represent light with the same frequency, non-vertical vibration directions and constant phase difference. The long coherent light can represent light with a relatively short coherence length within the coherence time.

[0056] For example, the reference coherent light may be coherent light of different polarization states corresponding to wavelengths of 520 nm, 575 nm, 650 nm, 732 nm, 780 nm and 830 nm, respectively.

[0057] According to an embodiment of the present invention, human tissue has a high scattering characteristic. After the reference coherent light is incident on the deep tissue at the preset position of the target object, it will undergo multiple scatterings and produce a depolarization phenomenon, which will change the polarization state of the reference coherent light and emit diffuse coherent light. In order to distinguish the output signals of tissues at different depths, the present invention can use reference coherent light with a polarization angle of 0° corresponding to multiple wavelengths to be incident on the tissue at the preset position of the target object. The detected signal is a signal that has been scattered less times by the shallow tissue, that is, multiple diffuse coherent images reflecting the characteristics of the blood flow state in the shallow tissue are obtained. At the same time, reference coherent light with a polarization angle of 90° corresponding to multiple wavelengths is used to be incident on the tissue at the preset position of the target object. The detected signal is a signal that has been scattered more times by the deep tissue, that is, multiple diffuse coherent images reflecting the characteristics of the blood flow state in the deep tissue are obtained. Therefore, the multiple diffuse coherent images obtained by the present invention include characteristics reflecting the blood flow state in tissues at different depths.

[0058] According to an embodiment of the present invention, the diffuse coherence image obtained includes a tissue dynamic signal caused by the change of blood flow in the body and a tissue static signal related to the blood flow volume in the body. The tissue static signal and the tissue dynamic signal in the dynamic coherent light signal can be decomposed based on a decomposition algorithm of a modulation signal correlation matrix.

[0059] According to an embodiment of the present invention, represents the reference coherent light signal corresponding to each pixel, Represents the dynamic coherent light signal corresponding to each pixel. The process of correlating the two signals to obtain the correlation matrix X can be expressed as the following formula (1).

[0060] (1)

[0061] Wherein, t represents a certain acquisition time in the preset time length, and T represents the matrix transposition calculation symbol.

[0062] According to an embodiment of the present invention, based on the dynamic scattering characteristics of diffuse coherent light, the blurriness of the diffuse coherent image after being scattered by tissue blood flow is inversely proportional to the blood flow velocity component. The greater the blurriness of the diffuse coherent image, the faster the blood flow velocity; conversely, the smaller the blurriness of the diffuse coherent image, the slower the blood flow velocity.

[0063] According to an embodiment of the present invention, in order to quantify the blur of the diffuse coherence image, the information entropy of the tissue dynamic signal corresponding to each pixel in the same time period can be calculated to reflect the blood flow velocity. The faster the blood flow velocity, the smaller the information entropy; the slower the blood flow velocity, the larger the information entropy. The calculation of the information entropy H can be expressed as the following formula (2).

[0064] (2)

[0065] Among them, m is the number of pixels, p i is the intensity of the tissue dynamic signal of the ith pixel in a certain period of time within a preset time length.

[0066] According to an embodiment of the present invention, the blood flow velocity of the human body is mainly affected by the pulse wave generated by the heart beat, so the flow velocity pulse signal H(t) can be obtained according to multiple information entropies corresponding to multiple time periods of preset time lengths.

[0067] According to an embodiment of the present invention, based on the optical properties of tissue, diffuse coherent light is easily absorbed by blood during transmission in tissue, thereby causing attenuation of tissue static signals, that is, the larger the blood volume, the greater the absorption attenuation.

[0068] According to an embodiment of the present invention, a signal M related to blood volume is obtained according to the average intensity of the tissue static signal of each pixel in the same period. The calculation of M can be expressed as the following formula (3).

[0069] (3)

[0070] Among them, q i is the tissue static signal intensity detected by the ith pixel in a certain period of time within a preset time length.

[0071] According to an embodiment of the present invention, the blood volume in the body is mainly affected by the pulse wave generated by the heart beat. Therefore, the volume pulse signal M(t) can be obtained according to a plurality of signals M related to the blood volume corresponding to a plurality of time periods of preset duration.

[0072] According to an embodiment of the present invention, the time series of the velocity pulse wave and the volume pulse wave (i.e., the velocity pulse signal and the volume pulse signal) obtained by the above algorithm is a one-dimensional signal, which contains amplitude and frequency information related to blood pressure and cardiovascular health. However, since the resolution of the one-dimensional signal is low, in order to improve the resolution of the features corresponding to blood pressure and cardiovascular health in the velocity pulse wave and the volume pulse wave, the velocity pulse wave and the volume pulse wave can be fused by feature dimensionality increase.

[0073] According to an embodiment of the present invention, the flow velocity pulse signal is subjected to Fourier transform to obtain a flow velocity pulse spectrum H(f1) which can be expressed as the following formula (4).

[0074] H(f1)=FFT(H(t)) (4)

[0075] Wherein, f1 is the frequency corresponding to the flow velocity pulse spectrum.

[0076] According to an embodiment of the present invention, the volume pulse signal is subjected to Fourier transform to obtain a volume pulse spectrum M(f2) which can be expressed as the following formula (5).

[0077] M(f2) =FFT(M(t)) (5)

[0078] Wherein, f2 is the frequency corresponding to the volume pulse spectrum.

[0079] According to an embodiment of the present invention, the cross-correlation operation on the flow velocity pulse spectrum and the volume pulse spectrum can be expressed as the following formula (6).

[0080] (6)

[0081] in, Represents multi-dimensional fusion features.

[0082] According to an embodiment of the present invention, the multidimensional fusion matrix obtained by performing a cross-correlation operation on two spectra is a two-dimensional matrix. The multidimensional fusion matrix can be input into a correlation analysis model to determine the blood pressure and cardiovascular health level, and further feature extraction can be performed on the multidimensional fusion matrix, and the obtained features can be input into a correlation analysis model to determine the blood pressure and cardiovascular health level, so as to assist relevant personnel in judging the blood pressure and cardiovascular health of the target object based on the blood pressure and cardiovascular health level.

[0083] According to an embodiment of the present invention, when reference coherent light in different polarization states corresponding to multiple wavelengths is incident on tissue at a preset position of a target object at the same time, diffuse coherent light emitted from the tissue at the preset position is continuously collected within a preset time length, and multiple diffuse coherent images reflecting the characteristics of the blood flow state in tissues at different depths are obtained. Based on the multiple diffuse coherent images collected in each time period of the preset time length, the dynamic coherent light signal corresponding to each pixel is determined, and based on the dynamic coherent light signal and the reference coherent light signal corresponding to the pixel, the tissue static signal and the tissue dynamic signal are determined, and the information entropy of the tissue dynamic signal of each pixel in the same time period is calculated to obtain a flow velocity pulse signal; the average intensity of the tissue static signal of each pixel in the same time period is calculated to obtain a volume pulse signal, so as to achieve the resolution of blood flow velocity and blood flow volume in tissues at different depths. The flow velocity pulse signal and the volume pulse signal are respectively subjected to Fourier transform to obtain the flow velocity pulse spectrum and the volume pulse spectrum; based on the one-dimensional spectrum of the flow velocity pulse and the volume pulse, the flow velocity pulse spectrum and the volume pulse spectrum are cross-correlated to obtain multi-dimensional fusion features; based on the multi-dimensional fusion features, the blood pressure and cardiovascular health level are accurately predicted, and multi-modal detection of in vivo blood flow and pulse wave is realized.

[0084] According to an embodiment of the present invention, a dynamic coherent light signal corresponding to each pixel is determined based on a plurality of diffuse coherent images collected in each time period of a preset time length, and a tissue static signal and a tissue dynamic signal are determined based on the dynamic coherent light signal and the reference coherent light signal corresponding to the pixel, including: performing correlation calculation on the dynamic coherent light signal and the reference coherent light signal to obtain initial blood flow state information; performing singular value decomposition on the initial blood flow state information to obtain a left singular value matrix, a diagonal matrix, and a right singular value matrix; determining the tissue static signal based on the first K row vectors of the left singular value matrix, the diagonal matrix, and the right singular value matrix; determining the tissue dynamic signal based on the K+1th to nth row vectors of the left singular value matrix, the diagonal matrix, and the right singular value matrix, wherein K is a positive integer and n is an integer greater than 1.

[0085] According to an embodiment of the present invention, the initial blood flow state information can be expressed as the correlation matrix X of the above formula (1). The process of performing singular value decomposition on the initial blood flow state information can be expressed as the following formula (7).

[0086] (7)

[0087] Among them, U represents the left singular value matrix, u k represents the corresponding k-th row vector in the left singular value matrix, S represents the diagonal matrix, s k represents the kth row vector in the diagonal matrix, V represents the right singular value matrix, v k represents the corresponding k-th row vector in the right singular value matrix, and n is the total number of row vectors.

[0088] According to an embodiment of the present invention, the tissue static signal is determined based on the first K row vectors of the left singular value matrix, the diagonal matrix and the right singular value matrix. The process can be expressed as the following formula (8).

[0089] (8)

[0090] According to an embodiment of the present invention, the tissue dynamic signal is determined based on the K+1th to nth row vectors of the left singular value matrix, the diagonal matrix and the right singular value matrix. The process can be expressed as the following formula (9).

[0091] (9)

[0092] According to an embodiment of the present invention, by decomposing the initial blood flow state information obtained by correlating the dynamic coherent light signal and the reference coherent light signal, a tissue dynamic signal and a tissue static signal are obtained respectively, thereby solving the technical problem that the prior art is difficult to distinguish between tissue dynamic signals and tissue static signals, and providing a certain basis for the subsequent detection of parameters of blood flow velocity and blood volume related to the pulse wave.

[0093] According to an embodiment of the present invention, blood pressure and cardiovascular health levels are determined based on multidimensional fusion features, including: extracting homogeneity features, energy features, contrast features and correlation features from the multidimensional fusion features; inputting the homogeneity features, energy features, contrast features and correlation features into a target regression analysis model to obtain blood pressure and cardiovascular health levels.

[0094] According to an embodiment of the present invention, after each dimension of the multi-dimensional fusion feature is spliced, homogeneity features, energy features, contrast features and correlation features can be extracted from the spliced ​​grayscale image. Homogeneity features, energy features, contrast features and correlation features can also be extracted for each dimension in the multi-dimensional fusion feature.

[0095] According to an embodiment of the present invention, the homogeneity feature (Homogeneity) can reflect the proximity between the grayscale value and the diagonal of the spectrum correlation matrix; the energy feature (energy) reflects the uniformity of the distribution of the image grayscale value and the coarseness of the texture; the contrast feature (contrast) reflects the clarity of the image and the depth of the texture; the correlation feature (correlation) reflects the similarity of the image in rows or columns.

[0096] According to an embodiment of the present invention, the target regression analysis model can analyze the extracted homogeneity features, energy features, contrast features and correlation features to obtain the blood pressure and cardiovascular health level corresponding to the collected diffuse coherent light image.

[0097] According to an embodiment of the present invention, the target regression analysis model may be a machine learning model based on a support vector machine (SVM) or a neural network model based on a back propagation (BP) algorithm.

[0098] According to an embodiment of the present invention, with the real-time extraction of multi-dimensional fusion features, the target regression analysis model can be optimized and trained to achieve accurate analysis of blood pressure and cardiovascular health level.

[0099] According to an embodiment of the present invention, by inputting the multidimensional fusion features into the target regression analysis model, blood pressure and cardiovascular health level are obtained. After multidimensional fusion is performed on the basis of the one-dimensional spectrum of the velocity pulse signal and the volume pulse signal, the multidimensional fusion features are analyzed using the target regression analysis model, and blood pressure and cardiovascular health level are determined based on the multidimensional fusion features. The technical problem that it is difficult to analyze multidimensional features at the same time in the prior art is solved, and accurate assessment of blood pressure and cardiovascular health level in vivo is achieved.

[0100] According to an embodiment of the present invention, extracting homogeneity features, energy features, contrast features, and correlation features from multidimensional fusion features includes extracting homogeneity features according to the following formula (I):

[0101] (one)

[0102] in, Represents homogeneity characteristics, Indicates the gray level is and The probability statistics of the pixels appearing simultaneously in each single-channel fusion feature included in the multi-dimensional fusion feature, and N is the maximum gray level of the pixels in the single-channel fusion feature.

[0103] According to an embodiment of the present invention, the higher the homogeneity feature value, the greater the probability that the pixel appears simultaneously in each single-channel fusion feature included in the multi-dimensional fusion feature, the smaller the texture change, and the smaller the blood flow feature change.

[0104] According to an embodiment of the present invention, extracting homogeneity features, energy features, contrast features, and correlation features from multi-dimensional fusion features includes extracting energy features according to the following formula (ii):

[0105] (two)

[0106] in, Energy characteristics.

[0107] According to an embodiment of the present invention, extracting homogeneity features, energy features, contrast features, and correlation features from multidimensional fusion features includes extracting contrast features according to the following formula (III):

[0108] (three)

[0109] in, is the contrast feature.

[0110] According to an embodiment of the present invention, the higher the contrast characteristic value is, the clearer the image is.

[0111] According to an embodiment of the present invention, it is characterized in that extracting homogeneity features, energy features, contrast features and correlation features from multi-dimensional fusion features includes extracting correlation features according to the following formula (IV):

[0112] (Four)

[0113] in, is the correlation feature, , For The corresponding mean, , For The corresponding mean, , For The corresponding standard deviation, , For The corresponding standard deviation.

[0114] Based on the above detection method for assisting in determining blood pressure and cardiovascular health level, an embodiment of the present invention provides a detection device for assisting in determining blood pressure and cardiovascular health level.

[0115] Figure 2 The schematic diagram of the structure of the detection device for assisting in determining the blood pressure and cardiovascular health level according to an embodiment of the present invention is shown. Figure 2 The detection device for assisting in determining blood pressure and cardiovascular health level in the embodiment corresponds to the detection method for assisting in determining blood pressure and cardiovascular health level mentioned above. For further description of the detection device for assisting in determining blood pressure and cardiovascular health level, refer to the detection method for assisting in determining blood pressure and cardiovascular health level mentioned above.

[0116] like Figure 2 As shown, the detection device 200 for assisting in determining blood pressure and cardiovascular health level includes a collection unit 210 and a processor 220 .

[0117] The acquisition unit 210 is used to continuously acquire diffuse coherent light emitted from the tissue at a preset position within a preset time length when the reference coherent light in different polarization states corresponding to multiple wavelengths is simultaneously incident on the tissue at a preset position of the target object, so as to obtain multiple diffuse coherent images, wherein each diffuse coherent image includes features reflecting the blood flow state in the tissue at different depths.

[0118] Processor 220 is used to determine the dynamic coherent light signal corresponding to each pixel based on multiple diffuse coherent images collected in each time period of a preset time length, and determine the tissue static signal and the tissue dynamic signal based on the dynamic coherent light signal and the reference coherent light signal corresponding to the pixel, wherein the reference coherent light signal is a signal used to characterize the reference coherent light corresponding to the pixel; calculate the information entropy of the tissue dynamic signal of each pixel in the same time period to obtain a flow velocity pulse signal; calculate the average intensity of the tissue static signal of each pixel in the same time period to obtain a volume pulse signal; perform Fourier transform on the flow velocity pulse signal and the volume pulse signal respectively to obtain a flow velocity pulse spectrum and a volume pulse spectrum; perform cross-correlation operation on the flow velocity pulse spectrum and the volume pulse spectrum to obtain a multi-dimensional fusion feature; determine the blood pressure and cardiovascular health level based on the multi-dimensional fusion feature.

[0119] According to an embodiment of the present invention, when the reference coherent light in different polarization states corresponding to multiple wavelengths is incident on the tissue at the preset position of the target object through the acquisition unit at the same time, the diffuse coherent light emitted from the tissue at the preset position is continuously acquired within the preset time length, and multiple diffuse coherent images reflecting the characteristics of the blood flow state in the tissue at different depths are obtained. The processor determines the dynamic coherent light signal corresponding to each pixel of the multiple diffuse coherent images collected in each time period of the preset time length, and determines the tissue static signal and the tissue dynamic signal based on the dynamic coherent light signal and the reference coherent light signal corresponding to the pixel, calculates the information entropy of the tissue dynamic signal of each pixel in the same time period, and obtains the flow velocity pulse signal; calculates the average intensity of the tissue static signal of each pixel in the same time period, and obtains the volume pulse signal, so as to achieve the resolution of blood flow velocity and blood flow volume in tissues of different depths. The flow velocity pulse signal and the volume pulse signal are respectively subjected to Fourier transform to obtain the flow velocity pulse spectrum and the volume pulse spectrum; based on the one-dimensional spectrum of the flow velocity pulse and the volume pulse, the flow velocity pulse spectrum and the volume pulse spectrum are cross-correlated to obtain multi-dimensional fusion features; based on the multi-dimensional fusion features, the blood pressure and cardiovascular health level are accurately predicted, and multi-modal detection of in vivo blood flow and pulse wave is realized.

[0120] According to an embodiment of the present invention, the acquisition unit includes: a coherent light source, which is used to emit reference coherent light in different polarization states corresponding to multiple wavelengths, and simultaneously incident the reference coherent light on tissue at a preset position of the target object; an image acquisition module: which is used to continuously acquire diffuse coherent light emitted from tissue at a preset position within a preset time length to obtain multiple diffuse coherent images.

[0121] According to an embodiment of the present invention, a coherent light source may be composed of narrowband lasers of different wavelengths, a polarization modulator, and a control circuit. Among them, the narrowband laser may use a VESCEL laser, the polarization modulator may modulate the output polarization state of the laser, for example, it may be selected to output linearly polarized light, and the control circuit may process reference coherent light of different frequencies, output a square wave modulation signal, to control the output signal amplitude and frequency of each laser, perform light source switching, and synchronously trigger the image acquisition module, providing a basis for realizing multi-wavelength multi-modal dynamic detection.

[0122] According to an embodiment of the present invention, the coherent light source may irradiate the reference coherent light to the tissue at a preset position of the target object through the illumination of the emission window.

[0123] According to the embodiment of the present invention, the image acquisition module can acquire the diffuse coherence image through the imaging surface of the CMOS sensor in the receiving window.

[0124] According to an embodiment of the present invention, the image acquisition module includes: a polarization detection mask, which is used to divide the diffuse coherent light emitted from the tissue at a preset position into coherent lights corresponding to different polarization states; a wavelength partitioning detection mask, which is used to filter the coherent lights corresponding to different polarization states to obtain regional polarized coherent lights in different polarization states corresponding to multiple wavelengths; and a photoelectric detection array, which is used to perform photoelectric conversion on the regional polarized coherent light within a preset time length to obtain multiple diffuse coherent images.

[0125] According to an embodiment of the present invention, in order to distinguish diffuse coherent light emitted from tissues at different depths, a polarization detection mask that divides corresponding polarization angles is set to divide the diffuse coherent light emitted from tissues at a preset position into coherent light corresponding to different polarization states.

[0126] For example, taking the linearly polarized diffuse coherent light as an example, the polarization angle of the first area in the polarization detection mask is set to 0 , which is used to detect coherent light with fewer scattering times through shallow tissues. Accordingly, the polarization angle of the second area in the polarization detection mask is set to 90 , used to detect coherent light that has been scattered more times through deep tissue.

[0127] According to an embodiment of the present invention, the wavelength partition detection mask may be divided into a plurality of regions, which respectively collect regional polarized coherent light in different polarization states of corresponding wavelengths.

[0128] For example, taking the polarization detection mask of two regions as an example, the wavelength partition detection mask can be set to 12 regions, wherein the first 6 regions overlap with the first region in the polarization detection mask, and narrow-band pass filters with central wavelengths of 520nm, 575nm, 650nm, 732nm, 780nm and 830nm are placed in sequence to collect the polarization angles of 0 corresponding to different wavelengths. The last six areas overlap with the second area in the polarization detection mask, and narrow bandpass filters with central wavelengths of 520nm, 575nm, 650nm, 732nm, 780nm and 830nm are placed in sequence to collect the polarization coherent light corresponding to different wavelengths at a polarization angle of 90 The area of ​​polarized coherent light.

[0129] According to an embodiment of the present invention, the photoelectric detection array converts the regional polarization coherent light passing through the polarization detection mask and the wavelength partition detection mask into an electrical signal, and converts it into a diffuse coherent image through the CMOS sensor imaging surface in the receiving window.

[0130] According to an embodiment of the present invention, in order to better understand the detection device for assisting in determining blood pressure and cardiovascular health level in an embodiment of the present invention, the following will be described. Figure 3 The above contents are explained.

[0131] Figure 3 A schematic structural diagram of a detection device for assisting in determining blood pressure and cardiovascular health level according to another embodiment of the present invention is shown.

[0132] like Figure 3 As shown, in the detection device 200 for assisting in determining blood pressure and cardiovascular health level, the acquisition unit 210 includes a coherent light source 211 and an image acquisition module 212. The coherent light source 211 includes a narrowband laser 211_1, a control circuit 211_2, and a polarization modulator; the image acquisition module 212 includes a polarization detection mask 212_1, a wavelength partition detection mask 212_2, and a photoelectric detection array 212_3. The upper part of the polarization detection mask 212_1 is the first area, and the lower part is the second area.

[0133] According to an embodiment of the present invention, two levels of masks are placed in sequence before the photoelectric detection array, which can synchronously detect diffuse coherent light in different polarization states and different wavelengths, solving the technical problem that the prior art is difficult to detect diffuse coherent light at different depths.

[0134] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. In this regard, each box in the flowchart or block diagram may represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box may also occur in an order different from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions. It can be understood by those skilled in the art that the features recorded in the various embodiments of the present invention can be combined and / or combined in various ways, even if such a combination or combination is not explicitly recorded in the present invention. In particular, without departing from the spirit and teachings of the present invention, the features recorded in the various embodiments of the present invention can be combined and / or combined in various ways. All these combinations and / or combinations fall within the scope of the present invention.

[0135] The embodiments of the present invention have been described above. However, these embodiments are only for the purpose of illustration, and are not intended to limit the scope of the present invention. Although each embodiment has been described above, this does not mean that the measures in each embodiment cannot be advantageously used in combination. The scope of the present invention is limited by the attached embodiments and their equivalents. Without departing from the scope of the present invention, those skilled in the art may make various substitutions and modifications, which should all fall within the scope of the present invention.

Claims

1. A detection method for assisting in determining blood pressure and cardiovascular health level, characterized in that: The detection method comprises: When reference coherent lights in different polarization states corresponding to a plurality of wavelengths are incident on tissue at a preset position of a target object at the same time, diffuse coherent lights emitted from the tissue at the preset position are continuously collected within a preset time period to obtain a plurality of diffuse coherent images, wherein the plurality of diffuse coherent images include features reflecting blood flow states in tissues at different depths; Determine a dynamic coherent light signal corresponding to each pixel according to a plurality of diffuse coherent images collected in each time period of the preset time length, and determine a tissue static signal and a tissue dynamic signal according to the dynamic coherent light signal and a reference coherent light signal corresponding to the pixel, wherein the reference coherent light signal is a signal for characterizing a reference coherent light corresponding to the pixel; The information entropy of the tissue dynamic signal of each pixel in the same period is calculated to obtain the flow velocity pulse signal; The average intensity of the tissue static signal of each pixel in the same period is calculated to obtain the volume pulse signal; Performing Fourier transformation on the flow velocity pulse signal and the volume pulse signal respectively to obtain a flow velocity pulse spectrum and a volume pulse spectrum; Performing a cross-correlation operation on the flow velocity pulse spectrum and the volume pulse spectrum to obtain a multi-dimensional fusion feature; According to the multi-dimensional fusion characteristics, the blood pressure and cardiovascular health level are determined including: Extracting homogeneity features, energy features, contrast features and correlation features from the multi-dimensional fusion features; The homogeneity feature, the energy feature, the contrast feature, and the correlation feature are input into a target regression analysis model to obtain the blood pressure and cardiovascular health level.

2. The detection method according to claim 1, characterized in that: The step of determining a dynamic coherent light signal corresponding to each pixel according to a plurality of diffuse coherent images collected in each time period of the preset time length, and determining a tissue static signal and a tissue dynamic signal according to the dynamic coherent light signal and a reference coherent light signal corresponding to the pixel, comprises: performing correlation calculation on the dynamic coherent optical signal and the reference coherent optical signal to obtain initial blood flow state information; Performing singular value decomposition on the initial blood flow state information to obtain a left singular value matrix, a diagonal matrix and a right singular value matrix; Determine the tissue static signal according to the left singular value matrix, the diagonal matrix and the first K row vectors of the right singular value matrix; The tissue dynamic signal is determined according to the K+1th to nth row vectors of the left singular value matrix, the diagonal matrix and the right singular value matrix, wherein K is a positive integer and n is an integer greater than 1.

3. The detection method according to claim 1, characterized in that: The extracting of homogeneity features, energy features, contrast features and correlation features from the multi-dimensional fusion features includes extracting the homogeneity features according to the following formula (I): (one) in, is the homogeneity characteristic, Indicates the gray level is and The probability statistics of the pixels appearing simultaneously in each single-channel fusion feature included in the multi-dimensional fusion feature, and N is the maximum grayscale level of the pixels in the single-channel fusion feature.

4. The detection method according to claim 3, characterized in that: The extracting of homogeneity features, energy features, contrast features and correlation features from the multi-dimensional fusion features includes extracting the energy features according to the following formula (ii): (two) in, is the energy characteristic.

5. The detection method according to claim 4, characterized in that: The step of extracting the homogeneity feature, the energy feature, the contrast feature and the correlation feature from the multi-dimensional fusion feature includes extracting the contrast feature according to the following formula (III): (three) in, is the contrast feature.

6. The detection method according to claim 5, characterized in that: The extracting of homogeneity features, energy features, contrast features and correlation features from the multi-dimensional fusion features includes extracting the correlation features according to the following formula (IV): (Four) in, is the correlation feature, , For The corresponding mean, , For The corresponding mean, , For The corresponding standard deviation, , For The corresponding standard deviation.

7. A detection device for assisting in determining blood pressure and cardiovascular health level, characterized in that: The detection device comprises: an acquisition unit, configured to, when simultaneously incident reference coherent lights in different polarization states corresponding to a plurality of wavelengths on tissue at a preset position of a target object, continuously acquire diffuse coherent lights emitted from the tissue at the preset position within a preset time length, to obtain a plurality of diffuse coherent images, wherein each diffuse coherent image includes features reflecting blood flow states in tissues at different depths; A processor, configured to determine a dynamic coherent light signal corresponding to each pixel according to a plurality of diffuse coherent images collected in each time period of the preset time length, and determine a tissue static signal and a tissue dynamic signal according to the dynamic coherent light signal and a reference coherent light signal corresponding to the pixel, wherein the reference coherent light signal is a signal for characterizing a reference coherent light corresponding to the pixel; The information entropy of the tissue dynamic signal of each pixel in the same period is calculated to obtain the flow velocity pulse signal; The average intensity of the tissue static signal of each pixel in the same period is calculated to obtain the volume pulse signal; Performing Fourier transformation on the flow velocity pulse signal and the volume pulse signal respectively to obtain a flow velocity pulse spectrum and a volume pulse spectrum; Performing a cross-correlation operation on the flow velocity pulse spectrum and the volume pulse spectrum to obtain a multi-dimensional fusion feature; According to the multi-dimensional fusion features, determining the blood pressure and cardiovascular health level includes: Extracting homogeneity features, energy features, contrast features and correlation features from the multi-dimensional fusion features; The homogeneity feature, the energy feature, the contrast feature, and the correlation feature are input into a target regression analysis model to obtain the blood pressure and cardiovascular health level.

8. The detection device according to claim 7, characterized in that: The acquisition unit comprises: A coherent light source, used to emit reference coherent light in different polarization states corresponding to a plurality of wavelengths, and simultaneously incident the reference coherent light onto tissue at a preset position of a target object; Image acquisition module: used for continuously acquiring diffuse coherent light emitted from the tissue at the preset position within a preset time period to obtain a plurality of diffuse coherent images.

9. The detection device according to claim 8, characterized in that: The image acquisition module comprises: A polarization detection mask, used to separate the diffuse coherent light emitted from the tissue at the preset position into coherent lights corresponding to different polarization states; The wavelength partition detection mask is used to filter the coherent light corresponding to different polarization states to obtain regional polarized coherent light in different polarization states corresponding to multiple wavelengths; The photoelectric detection array is used to perform photoelectric conversion on the regional polarized coherent light within a preset time length to obtain multiple diffuse coherent images.

Images