Vital sign monitoring equipment

Abstract

The invention discloses vital sign monitoring equipment, and relates to the technical field of computer vision. In the equipment, the processor determines the pulse wave signal based on the blood vessel-containing image acquired by the image acquisition device, and determines the information for representing the vital signs according to the pulse wave signal, that is, the vital signs are monitored based on a computer vision mode, and the vital signs can be monitored without contact with the human skin; the vital signs are monitored in a non-contact mode, the use limitation such as body surface damage is avoided, the use scene is greatly expanded, and the vital sign monitoring requirements of most people are met; environmental interference is not easily caused, and the detection precision is relatively high; according to the technical scheme, under the rotation of the first motor and the second motor with different rotating shaft directions, the rotating motion of the image acquisition device is realized, the accuracy of the acquired image is improved, and the accuracy of the determined vital sign information is improved when the vital sign information is determined based on the image containing the blood vessel.

Description

Technical Field

[0001] The present invention relates to the field of computer vision technology, and in particular to a vital sign monitoring device. Background Art

[0002] Common vital signs include heart rate, respiratory rate, blood pressure, body temperature, and state of consciousness. By continuously monitoring vital signs, medical staff can promptly detect abnormal conditions in patients and provide treatment.

[0003] Wearable monitoring devices (such as patch-type ECG monitors and smartwatches) are placed in contact with the skin. Light emitted by the wearable monitoring device is directed toward the skin, and the light reflected back through the skin tissue is used to monitor vital signs. However, because wearable monitoring devices require contact with the skin during monitoring, they cannot be used on patients with skin injuries.

[0004] It can be seen that providing a new vital signs monitoring device to meet the vital signs monitoring needs of most people is a technical problem that people in this field urgently need to solve. Summary of the Invention

[0005] The purpose of the present invention is to provide a vital signs monitoring device to solve the technical problem that patients with skin injuries cannot use wearable monitoring devices to monitor their vital signs.

[0006] To solve the above technical problems, the present invention provides a vital signs monitoring device, comprising: a processor, a first motor, a second motor, an image acquisition device, a first support component, and a second support component; the rotation axis direction of the first motor is different from the rotation axis direction of the second motor;

[0007] The image acquisition device is fixed to the first support component; the first support component and the first motor are located on one side of the second support component; the second motor is located on the other side of the second support component; the non-axial end of the first motor is fixed to the second support component; the axial end of the first motor is fastened to the first support component; and the axial end of the second motor is fastened to the second support component.

[0008] The processor is connected to the first motor and is used to control the first motor to rotate so as to drive the first support member and the image acquisition device to move; the processor is connected to the second motor and is used to control the second motor to rotate so as to drive the second support member to move;

[0009] The processor is connected to the image acquisition device and is used to determine a pulse wave signal based on an image containing blood vessels acquired by the image acquisition device; and to determine information used to characterize vital signs based on the pulse wave signal.

[0010] Exemplarily, it also includes a base; the processor is fixed on the base; and the non-axial end of the second motor is fixed on the base.

[0011] Exemplarily, the first supporting component is a U-shaped sheet metal component, and the second supporting component is a U-shaped seat;

[0012] The rotation axis direction of the first motor and the rotation direction of the second motor are perpendicular to each other, the rotation direction of the first motor is perpendicular to the ground, and the rotation direction of the second motor is parallel to the ground.

[0013] Exemplarily, the processor determining the pulse wave signal based on the image containing the blood vessel acquired by the image acquisition device includes:

[0014] Acquire an initial image containing blood vessels from the image acquisition device;

[0015] determining a target area of ​​the target detection part in the initial image containing the blood vessel according to the characteristics of the target detection part;

[0016] Separating the color channels of the image containing the blood vessels and obtaining features of the sub-regions contained in the target detection region;

[0017] The pixel grayscale value of the target area is determined according to the characteristics of the sub-region to determine the pulse wave signal.

[0018] Exemplarily, before determining information used to characterize vital signs based on the pulse wave signal, the processor further includes:

[0019] Determine the pulse wave signal of the target length to be acquired according to a preset image acquisition cycle and image acquisition rate;

[0020] Starting from receiving the image containing the blood vessel captured by the image acquisition device, obtaining the current length of the pulse wave signal;

[0021] In the case where it is detected that the current length is greater than or equal to the target length, entering the step of determining information used to characterize vital signs based on the pulse wave signal;

[0022] In the case where it is detected that the current length is less than the target length, the image containing the blood vessel acquired by the image acquisition device is received, and the process returns to the step of acquiring the current length of the pulse wave signal.

[0023] Exemplarily, after determining the pixel grayscale value of the target area according to the characteristics of the sub-region to determine the pulse wave signal, the processor further includes:

[0024] using the determined pulse wave signal as an initial pulse wave signal;

[0025] Using a planar orthogonal skin model, filtering motion artifacts and illumination changes in the initial pulse wave signal to obtain a final pulse wave signal;

[0026] The processor determines, based on the pulse wave signal, information for representing vital signs including:

[0027] The processor determines information for representing vital signs according to the final pulse wave signal.

[0028] Exemplarily, the processor determines, based on the final pulse wave signal, information for characterizing vital signs including:

[0029] Decomposing the final pulse wave signal and obtaining multiple intrinsic mode functions;

[0030] Obtain the frequency value of each eigenmode function;

[0031] According to the frequency value of each eigenmode function and the preset frequency reference range of each vital sign, each eigenmode function is divided into a corresponding vital sign group; wherein the vital sign group includes at least a heart rate group and a respiratory rate group;

[0032] Filtering out the intrinsic mode functions belonging to the noise group, and determining the extreme points of each vital sign group based on the frequency values ​​of the eigenmode functions in the remaining vital sign groups; wherein the eigenmode functions belonging to the noise group are the eigenmode functions remaining from all the eigenmode functions except the eigenmode functions belonging to the vital sign group;

[0033] The information used to characterize the vital signs is determined according to the situation of the extreme value points.

[0034] Exemplarily, the information used to characterize vital signs includes at least heart rate information, respiratory rate information, and heart rate variability information; and the information used to characterize vital signs determined based on the extreme point includes:

[0035] Obtaining a first number of maximum points or minimum points of the heart rate group within an image acquisition period; and determining that the number of heartbeats within the image acquisition period is equal to the first number to obtain heart rate information;

[0036] Obtaining a second number of maximum points or minimum points of the respiratory rate group within an image acquisition period; determining that the number of respirations within the image acquisition period is equal to the second number to obtain respiratory rate information;

[0037] The time intervals between all two adjacent maximum points or all two adjacent minimum points within the image acquisition period are obtained; and the heart rate variability information is determined according to the mean of all the time intervals.

[0038] Exemplarily, the information used to characterize the vital signs is blood oxygen saturation information; and the processor determining the blood oxygen saturation information according to the final pulse wave signal includes:

[0039] Acquire a first pulse wave signal of a light beam with a first channel wavelength and a second pulse wave signal of a light beam with a second channel wavelength within an image acquisition period;

[0040] determining a first maximum absorption amount of the light beam of the first channel wavelength by the blood, a first minimum absorption amount of the light beam of the first channel wavelength by the blood, a second maximum absorption amount of the light beam of the second channel wavelength by the blood, and a second minimum absorption amount of the light beam of the second channel wavelength by the blood based on a relationship between the pulse wave signal and the absorbance and a relationship between the light absorbance and the medium concentration determined by the Lambert-Beer law;

[0041] determining a first absorbance variation according to a difference between a first maximum absorption amount and a first minimum absorption amount of the blood for a light beam having a wavelength of the first channel, and determining a second absorbance variation according to a difference between a second maximum absorption amount and a second minimum absorption amount of the blood for a light beam having a wavelength of the second channel;

[0042] The blood oxygen saturation value is determined based on the first absorbance change and the second absorbance change, and in combination with a pre-established blood oxygen saturation calculation model; wherein the blood oxygen saturation calculation model is the ratio of oxygenated hemoglobin to total hemoglobin.

[0043] Exemplarily, determining the blood oxygen saturation value in combination with a pre-established blood oxygen saturation calculation model includes:

[0044] Determining a new calculation model for blood oxygen saturation by combining a pre-established calculation model for blood oxygen saturation;

[0045] Determine the value of blood oxygen saturation based on a new calculation model of blood oxygen saturation;

[0046] Among them, the expression of the new calculation model of blood oxygen saturation is:

[0047] ;

[0048] ;

[0049] ;

[0050] ;

[0051] in, Indicates blood oxygen saturation, A, B and R all represent numerical values; represents the absorption coefficient of reduced hemoglobin at the first channel wavelength; represents the absorption coefficient of oxygenated hemoglobin at the wavelength of the first channel; represents the absorption coefficient of reduced hemoglobin at the wavelength of the second channel; Indicates the first absorbance change, Indicates the second absorbance change.

[0052] In the vital signs monitoring device provided by the present invention, first, the processor determines a pulse wave signal based on an image containing blood vessels captured by the image acquisition device, and determines information used to represent the vital signs based on the pulse wave signal. That is, the vital signs are monitored using a computer vision method, and contact with human skin is not required for monitoring the vital signs. This achieves non-contact monitoring of vital signs, avoids usage restrictions such as body surface damage, greatly expands usage scenarios, and meets the vital signs monitoring needs of most people. Second, in the vital signs monitoring device, the image acquisition device is fixed to a first support member, and the shaft end of the first motor is tightly connected to the first support member. Therefore, when the first motor rotates, it drives the first support member and the image acquisition device to move. The shaft end of the second motor is tightly connected to the second support member. Therefore, when the second motor rotates, it drives the second support member to move. Since the non-shaft end of the first motor is fixed to the second support member, the first support member and the first motor are located on one side of the second support member. Therefore, when the second motor rotates and drives the second support member to move, the first support member, the first motor, and the image acquisition device fixed to the first support member on one side of the second support member also move. It can be seen that under the rotation of the first motor and the second motor with different rotation axis directions, the rotational movement of the image acquisition device is realized, which improves the accuracy of the image acquired by the image acquisition device, and further improves the accuracy of the determined vital signs information when determining vital signs information based on the image containing blood vessels; again, compared with non-wearable vital signs monitoring equipment such as millimeter wave radar, the non-wearable vital signs monitoring equipment provided by the present invention that monitors vital signs based on computer vision does not need to emit electromagnetic waves of a specific wavelength, so it is not easily affected by environmental interference and has relatively high detection accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the embodiments of the present invention, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0054] Figure 1 A schematic diagram of a vital signs monitoring device provided by an embodiment of the present invention;

[0055] Figure 2 A structural diagram of a vital sign detection system based on computer vision provided by an embodiment of the present invention;

[0056] Figure 3 This is a flowchart of a method for determining vital sign parameter values ​​based on computer vision provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0057] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0058] The core of the present invention is to provide a vital signs monitoring device to solve the technical problem that patients with skin injuries cannot use wearable monitoring devices to monitor their vital signs.

[0059] In order to enable those skilled in the art to better understand the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific implementation methods. Figure 1 A schematic diagram of a vital signs monitoring device provided by an embodiment of the present invention is shown in FIG. Figure 1 As shown, the device includes: a processor 10, a first motor 11, a second motor 12, an image acquisition device 13, a first support component 14 and a second support component 15; the rotation axis direction of the first motor 11 is different from the rotation axis direction of the second motor 12;

[0060] The image acquisition device 13 is fixed to the first support member 14; the first support member 14 and the first motor 11 are located on one side of the second support member 15; the second motor 12 is located on the other side of the second support member 15; the non-axial end of the first motor 11 is fixed to the second support member 15; the axial end of the first motor 11 is fastened to the first support member 14; and the axial end of the second motor 12 is fastened to the second support member 15.

[0061] The processor 10 is connected to the first motor 11 and is used to control the first motor 11 to rotate so as to drive the first support member 14 and the image acquisition device 13 to move; the processor 10 is connected to the second motor 12 and is used to control the second motor 12 to rotate so as to drive the second support member 15 to move;

[0062] The processor 10 is connected to the image acquisition device 13 and is used to determine a pulse wave signal based on an image containing blood vessels acquired by the image acquisition device 13; and to determine information used to represent vital signs based on the pulse wave signal.

[0063] The shapes of the first support member 14 and the second support member 15 are not limited, and may be U-shaped or L-shaped. To improve the stability of the vital signs monitoring device, the first support member 14 and the second support member 15 are both U-shaped. Specifically, the first support member 14 is a U-shaped sheet metal member, and the second support member 15 is a U-shaped seat.

[0064] The image acquisition device 13 is used to capture images of parts of the human body. For example, the image acquisition device 13 is a camera. To improve the accuracy of image capture by the image acquisition device 13, the present invention uses a motor to drive the image acquisition device 13. Specifically, the image acquisition device 13 is moved by a first motor 11 and a second motor 12 having different rotation axis directions. The specific rotation axis directions of the first motor 11 and the second motor 12 are not limited and are determined based on actual conditions. Preferably, the rotation axis direction of the first motor 11 and the rotation direction of the second motor 12 are perpendicular to each other, such as the first motor 11 rotating vertically and the second motor 12 rotating horizontally. When the rotation direction of the first motor 11 is perpendicular to the ground and the rotation direction of the second motor 12 is parallel to the ground, the vertical rotation of the first motor 11 can achieve vertical movement of the image acquisition device 13, while the horizontal rotation of the second motor 12 can achieve horizontal movement of the image acquisition device 13.

[0065] In order to improve the stability of the vital sign monitoring device, the vital sign monitoring device further includes a base; the processor 10 is fixed on the base; and the non-axial end of the second motor 12 is fixed on the base.

[0066] In order to enable those skilled in the art to better understand the vital signs monitoring device provided by the present invention, the above-mentioned vital signs monitoring device is further described in detail below, taking the first support component 14 as a U-shaped sheet metal component, the second support component 15 as a U-shaped seat, the first motor 11 as a vertical rotation motor, the second motor 12 as a horizontal rotation motor, the image acquisition device 13 as a Universal Serial Bus (USB) camera, and the processor 10 as a Raspberry Pi development board as an example. Figure 2 This is a structural diagram of a vital sign detection system based on computer vision provided by an embodiment of the present invention, such as Figure 2 As shown, the system mainly includes: base sheet metal 1, LED 2, motor control board 3, Raspberry Pi development board 4, horizontal rotation motor (5), U-shaped base 6, USB camera 7, U-shaped sheet metal 8, vertical rotation motor 9.

[0067] The motor control board 3 is connected to the Raspberry Pi development board 4, which is fixed to the base sheet metal 1. The USB camera 7 is fixed to the U-shaped sheet metal 8. The vertical rotation motor 9 is fixed to the U-shaped seat 6, with the shaft end and the U-shaped sheet metal 8 being tightly connected. The vertical rotation motor 9 rotates, driving the U-shaped sheet metal 8 and the USB camera 7 to achieve vertical rotation. The horizontal rotation motor 5 is fixed to the base sheet metal 1, with the shaft end and the U-shaped seat 6 being tightly connected. The horizontal rotation motor 5 rotates, driving the U-shaped seat 6 and other components connected thereto to achieve horizontal rotation. In summary, the combination of two axially perpendicular motors and the dual U-shaped member can drive the camera to achieve two degrees of freedom of rotation in space. The processing process of the processor 10 includes: positioning the target detection part, segmenting the target detection part area, extracting the photoplethysmography signal, signal processing, and calculating physiological indicator parameters (such as heart rate, respiratory rate, heart rate variability, and blood oxygen saturation). When the target detection part is a face part, the processing process of the processor 10 includes: face positioning, face area segmentation, photoplethysmography signal extraction, signal processing and physiological index parameter calculation.

[0068] This computer vision-based vital sign detection system uses an RGB color camera as a sensor to capture facial video. A set of vertically arranged motors forms a two-degree-of-freedom turntable, which drives the camera to locate faces in space. A Raspberry Pi development board serves as the core processor 10 for data and signal processing, calculating physiological indicators. This local operation protects user privacy. Status indicators are located on the device to facilitate user monitoring of the device's status. In practice, an on / off button can be added to the device for one-touch on / off functionality. A companion application (APP) can also be provided, allowing users to perform physiological indicator testing by following a few simple steps.

[0069] In implementation, based on the above-mentioned vital sign monitoring device, the processor 10 determines the pulse wave signal according to the image containing the blood vessels acquired by the image acquisition device 13, including:

[0070] Acquire an initial image containing blood vessels from the image acquisition device 13;

[0071] determining a target region of the target detection portion in an initial image containing blood vessels according to the characteristics of the target detection portion;

[0072] Separate the color channels of the image containing blood vessels and obtain the features of the sub-regions contained in the target detection region;

[0073] The pixel grayscale value of the target area is determined according to the characteristics of the sub-region to determine the pulse wave signal.

[0074] In practice, if vital signs monitoring is performed based on a relatively small number of images, the accuracy of vital sign monitoring is relatively low. In order to improve the accuracy of vital sign monitoring, it is necessary to obtain a sufficient number of images. Specifically, before determining the information used to represent the vital signs based on the pulse wave signal, the processor 10 further includes:

[0075] Determine the pulse wave signal of the target length to be acquired according to a preset image acquisition cycle and image acquisition rate;

[0076] Starting from receiving the image containing the blood vessel captured by the image acquisition device 13, obtaining the current length of the pulse wave signal;

[0077] When it is detected that the current length is greater than or equal to the target length, the step of determining information used to represent vital signs based on the pulse wave signal is entered;

[0078] When it is detected that the current length is less than the target length, the image containing the blood vessel acquired by the image acquisition device 13 is received, and the process returns to the step of acquiring the current length of the pulse wave signal.

[0079] There are no restrictions on the target detection area, such as the face. When the target detection area is the face, sub-areas within the target detection area include eyes, nose, mouth, and facial contour edges. The face detection module uses the Histogram of Oriented Gradients (HOG) feature descriptor and a cascade classifier to locate the face region and identify the eyes, nose, mouth, and facial contour edges. Based on this information, the grayscale values ​​of the facial region pixels are extracted. There are no restrictions on the pre-set image acquisition period or image acquisition rate. For example, if an image acquisition period (i.e., computation period) is 30 seconds and an image acquisition rate of 30 frames per second is used, the target pulse wave signal length is 900 pulse wave signals.

[0080] After obtaining a pulse wave signal, noise signals (such as motion artifacts and illumination changes) are often present in the pulse wave signal, which reduces the accuracy of information used to represent vital signs determined based on the pulse wave signal. Therefore, to improve the accuracy of vital sign monitoring, in some embodiments, after determining the pixel grayscale values ​​of the target area based on the characteristics of the sub-region to determine the pulse wave signal, the processor 10 further includes:

[0081] using the determined pulse wave signal as an initial pulse wave signal;

[0082] The motion artifacts and illumination changes in the initial pulse wave signal are filtered using a planar orthogonal skin model to obtain the final pulse wave signal.

[0083] The processor 10 determines, based on the pulse wave signal, information used to characterize vital signs, including:

[0084] The processor 10 determines information representing vital signs based on the final pulse wave signal.

[0085] It should be noted that in order to filter out noise signals, a variety of methods have been proposed to combine RGB signals into blood pulse signals, including: blind source separation algorithm based on principal component analysis (PCA) or independent component analysis (ICA), which uses different criteria to separate RGB signals into unrelated signals to retrieve blood pulse signals; CHROM (chromaticity) linearly combines chromaticity signals by assuming standardized skin color to white balance the image; photoplethysmography blood volume (PBV) uses blood volume change characteristics at different wavelengths to distinguish color changes caused by blood pulses from motion noise in RGB signals; two-step regression (2SR) measures the temporal rotation of the skin pixel space subspace for pulse extraction.

[0086] The present invention uses a Plane Orthogonal-to-Skin (POS) model (i.e., algorithm) to filter out noise signals. The POS algorithm is based on color space projection and uses orthogonal planes to suppress interfering signals, such as illumination changes and motion artifacts. Compared to the aforementioned methods, the POS algorithm has unique advantages:

[0087] (1) ICA requires iterative decomposition to find independent components, which has high computational complexity; while POS is based on linear projection, has low complexity and is suitable for real-time processing. ICA's assumption of signal source independence may not hold in real scenarios (e.g., motion artifacts are correlated with heart rate signals), resulting in separation failure. POS directly suppresses interference through a physical model (orthogonal projection) and is more robust. ICA requires the selection of the number of components and the convergence threshold; POS relies only on a fixed projection direction, and its parameter settings are simple.

[0088] (2) CHROM relies on chromaticity signals but is sensitive to head rotation or large-scale motion; POS uses orthogonal projection to separate noise that is inconsistent with the skin reflection direction, and has better motion robustness. CHROM is sensitive to changes in light color temperature; POS uses signal normalization and orthogonal projection to more effectively suppress light intensity / color temperature fluctuations. CHROM may lose some physiological information due to color space conversion; POS operates directly in RGB space, retaining more effective signal details.

[0089] (3) PBV needs to dynamically estimate the skin reflection direction (e.g., through PCA), with a relatively high computational cost; POS uses an empirical fixed projection direction (e.g., [1, -1, -1]), which simplifies the process. The dynamic projection direction update of PBV requires additional calculations; the fixed projection direction of POS does not require iteration and is suitable for low-latency scenarios. PBV has relatively high requirements for camera color calibration; POS reduces the dependence on device parameters through signal normalization.

[0090] (4) 2SR depends on non-linear system parameters (such as noise intensity, barrier height), making parameter tuning difficult; POS has no complex parameters and is more adaptable. 2SR needs to simulate the stochastic resonance process, with a large amount of calculation; POS is based on linear operations and has low resource occupancy. 2SR may fail at extremely low signal-to-noise ratios; POS suppresses noise through orthogonal projection and performs more stably under conventional interference.

[0091] The main steps of the POS algorithm include:

[0092] (1) Signal normalization: Normalize the luminance signal of each channel to eliminate the absolute difference in illumination intensity and retain the relative change. Usually, zero-mean unit-variance (Z-score) normalization is used. For the original BVP signal with length N, starting from the signal start value until the signal end value, taking 1 as the step size, successively select local signals with length E (where E < N) backward until E < N is not satisfied, obtaining several signal segments, and normalizing each signal segment:

[0093] ;

[0094] Where: represents the total signal length; represents the length of the intercepted signal segment; represents the starting position of the signal segment; represents the ending position of the signal segment; represents the number of the signal segment; represents the RGB three-channel signal with starting position m and ending position n; : represents the average value of; represents the time-normalized signal segment.

[0095] (2) Projection onto an orthogonal plane: Assume that the principal component of the light intensity change direction of skin reflection in the RGB space is the vector (empirical value). Project the normalized RGB signal onto the plane orthogonal to v to suppress interference (such as illumination change) consistent with the skin reflection direction.

[0096] 1. Construct the orthogonal plane basis vectors: Select two vectors and Construct an orthogonal matrix (e.g., by Gram-Schmidt orthogonalization) .

[0097] ;

[0098] 2. Projection calculation: The projected signal is the original signal and Linear combinations in directions:

[0099] ;

[0100] in: represents the projected signal;

[0101] ;

[0102] in: Represents projection to The signal on ; Represents projection to The signal on ; express The standard deviation of the signal; express The standard deviation of the signal; Indicates the maximum pulsation direction signal;

[0103] ;

[0104] in: Indicates the mean value of the maximum pulsation direction signal; Represents the result signal segment, starting at , the end position is .

[0105] Through the above processing, a pulse signal (ie, a final pulse wave signal) is obtained from the original blood volume pulse wave signal (ie, the initial pulse wave signal) after filtering out motion artifacts and illumination changes.

[0106] After obtaining the final pulse wave signal, the processor 10 determines information used to characterize vital signs based on the final pulse wave signal. The information used to characterize vital signs includes at least heart rate information, respiratory rate information, heart rate variability information, and blood oxygen saturation. In some embodiments, the information used to characterize vital signs determined by the processor 10 based on the final pulse wave signal includes:

[0107] Decomposing the final pulse wave signal and obtaining multiple intrinsic mode functions;

[0108] Obtain the frequency value of each eigenmode function;

[0109] According to the frequency value of each eigenmode function and the preset frequency reference range of each vital sign, each eigenmode function is divided into a corresponding vital sign group; wherein the vital sign group includes at least a heart rate group and a respiratory rate group;

[0110] Filtering out the intrinsic mode functions belonging to the noise group, and determining the extreme points of each vital sign group based on the frequency values ​​of the eigenmode functions in the remaining vital sign groups; wherein the eigenmode functions belonging to the noise group are the eigenmode functions remaining from all the eigenmode functions except the eigenmode functions belonging to the vital sign group;

[0111] The information used to characterize the vital signs is determined according to the situation of the extreme value points.

[0112] Among them, the pre-set frequency reference range of each vital sign is the medical detection range of human heart rate and respiratory rate. In addition, it should be noted that since the final pulse wave signal is a discrete signal, the multiple intrinsic mode functions obtained after the final pulse wave signal is decomposed and processed are also discrete signals. In a discrete signal, the determination of extreme points is usually achieved by comparing the values ​​of adjacent points. If the frequency value at the current moment is greater than the frequency value at the moment before the current moment and the frequency value at the moment after the current moment, the frequency value at the current moment is determined to be the maximum value; if the frequency value at the current moment is less than the frequency value at the moment before the current moment and the frequency value at the moment after the current moment, the frequency value at the current moment is determined to be the minimum value.

[0113] The information used to characterize vital signs includes at least heart rate information, respiratory rate information, and heart rate variability information. The information used to characterize vital signs determined based on the extreme point includes:

[0114] Obtaining a first number of maximum points or minimum points of a heart rate group during an image acquisition period; and determining that the number of heartbeats during the image acquisition period is equal to the first number to obtain heart rate information;

[0115] Obtaining a second number of maximum or minimum points of the respiratory rate group within the image acquisition period; determining that the number of respirations within the image acquisition period is equal to the second number to obtain respiratory rate information;

[0116] The time intervals between all two adjacent maximum points or all two adjacent minimum points within the image acquisition period are obtained; and the heart rate variability information is determined according to the mean of all the time intervals.

[0117] In order to enable those skilled in the art to better understand the above-mentioned process of determining heart rate information, respiratory rate information, and heart rate variability information, the process of determining heart rate information, respiratory rate information, and heart rate variability information based on empirical mode decomposition (EMD) is described below.

[0118] Empirical mode decomposition (EEMD) is a data-adaptive multiresolution technique used to decompose signals into physically meaningful components. It can be used to analyze nonlinear and nonstationary signals. The EEMD method recursively decomposes the original signal into components with varying resolutions. These components are also known as intrinsic mode functions (IMFs). By performing EEMD decomposition on a pulse wave signal, multiple IMFs are obtained. The frequencies of these IMFs are calculated, and based on the medical detection range of human heart rate and respiratory rate, the IMFs can be divided into heart rate, respiratory rate, and noise groups. The heart rate and respiratory rate groups are effective signal components for calculating heart rate and respiratory rate. Wavelet decomposition and reconstruction are used to further filter out noise, resulting in a uniform and stable photoplethysmography signal. Finally, extreme points are calculated for the heart rate and respiratory rate signals. The number of extreme points, representing the number of heartbeats and respiratory respirations during the detection period, can be used to determine the heart rate and respiratory rate. Heart rate variability, the time difference between heartbeats, can be easily calculated based on the extreme points of the heart rate signal.

[0119] The above describes the process of determining heart rate, respiratory rate, and heart rate variability. The vital signs monitoring device provided by the present invention can also monitor blood oxygen saturation. In some embodiments, the information used to represent vital signs is blood oxygen saturation information; the processor 10 determines the blood oxygen saturation information based on the final pulse wave signal, including:

[0120] Acquire a first pulse wave signal of a light beam with a first channel wavelength and a second pulse wave signal of a light beam with a second channel wavelength within an image acquisition period;

[0121] determining a first maximum absorption amount of the light beam of the first channel wavelength by the blood, a first minimum absorption amount of the light beam of the first channel wavelength by the blood, a second maximum absorption amount of the light beam of the second channel wavelength by the blood, and a second minimum absorption amount of the light beam of the second channel wavelength by the blood based on a relationship between the pulse wave signal and the absorbance and a relationship between the light absorbance and the medium concentration determined by the Lambert-Beer law;

[0122] determining a first absorbance variation according to a difference between a first maximum absorption amount and a first minimum absorption amount of the blood for a light beam having a wavelength of the first channel, and determining a second absorbance variation according to a difference between a second maximum absorption amount and a second minimum absorption amount of the blood for a light beam having a wavelength of the second channel;

[0123] The blood oxygen saturation value is determined according to the first absorbance change and the second absorbance change in combination with a pre-established blood oxygen saturation calculation model; wherein the blood oxygen saturation calculation model is the ratio of oxygenated hemoglobin to total hemoglobin.

[0124] Specifically, determining the value of blood oxygen saturation in combination with a pre-established blood oxygen saturation calculation model includes:

[0125] Determining a new calculation model for blood oxygen saturation by combining a pre-established calculation model for blood oxygen saturation;

[0126] Determine the value of blood oxygen saturation based on a new calculation model of blood oxygen saturation;

[0127] Among them, the expression of the new calculation model of blood oxygen saturation is:

[0128] ;

[0129] ;

[0130] ;

[0131] ;

[0132] in, Indicates blood oxygen saturation, A, B and R all represent numerical values; represents the absorption coefficient of reduced hemoglobin at the first channel wavelength; represents the absorption coefficient of oxygenated hemoglobin at the wavelength of the first channel; represents the absorption coefficient of reduced hemoglobin at the wavelength of the second channel; Indicates the first absorbance change, Indicates the second absorbance change.

[0133] The first channel used is the R channel, and the second channel is the G channel. In order to enable those skilled in the art to better understand the method of determining the blood oxygen saturation value, the method of determining the blood oxygen saturation value is described in detail below.

[0134] Hemoglobin in the human body consists of oxyhemoglobin, reduced hemoglobin, carboxyhemoglobin, and methemoglobin. Blood oxygen saturation refers to the ratio of oxyhemoglobin to total hemoglobin in the blood. In healthy people, the levels of carboxyhemoglobin and methemoglobin are very low, so the formula for calculating blood oxygen saturation is shown in (5-1):

[0135] ; (5-1)

[0136] in: Indicates the concentration of oxygenated hemoglobin; Indicates the concentration of reduced hemoglobin;

[0137] The expression of the Lambert-Beer law of light absorption is shown in (5-2):

[0138] ; (5-2)

[0139] in: Indicates the incident light intensity; Indicates the intensity of light after reflection; Indicates the medium concentration; Indicates optical path length; represents the attenuation coefficient; Indicates the wavelength of light;

[0140] The absorbance of the medium can be expressed as shown in (5-3):

[0141] ; (5-3)

[0142] in: Indicates the absorbance of the medium; It represents the absorption coefficient of media other than hemoglobin; indicates the concentration of mediators other than hemoglobin; It represents the distance that light travels in a medium other than hemoglobin; represents the absorption coefficient of hemoglobin; Indicates the medium concentration of hemoglobin; Represents the distance light travels through hemoglobin.

[0143] When a solution contains multiple solutes, the absorbance of the solution is equal to the sum of the absorbances of each solute individually. Since only oxygenated hemoglobin and reduced hemoglobin are considered, the new expression for the absorbance of the medium is shown in (5-4):

[0144] ; (5-4)

[0145] in: represents the absorption coefficient of oxygenated hemoglobin; Indicates the medium concentration of oxyhemoglobin; represents the absorption coefficient of reduced hemoglobin; Indicates the concentration of medium that reduces hemoglobin.

[0146] When the heart contracts, the blood volume in the blood vessels is the highest, the distance that light travels in the blood is the largest, and the blood absorbs the most light. The expression is shown in (5-5):

[0147] ; (5-5)

[0148] When the heart is in diastole, the blood volume in the blood vessels is the lowest, the distance that light travels in the blood is the smallest, and the blood absorbs the least amount of light. The expression is shown in (5-6):

[0149] ; (5-6)

[0150] The difference between the maximum and minimum light absorption in one heart cycle The expression is shown in (5-7):

[0151] ; (5-7)

[0152] in: Indicates the intensity of light reflected by media other than hemoglobin; Indicates the light reflection intensity of hemoglobin.

[0153] Furthermore, according to the definition of limit:

[0154] ;

[0155] The expression (5-8) can be obtained:

[0156] ; (5-8)

[0157] Among them, the optical path difference ;

[0158] It can be seen from the above formula that the change in absorbance is only related to the optical path difference. The wavelength used is and As the incident light, let:

[0159] ;

[0160] ;

[0161] Combining equations (5-1) and (5-2), we can get expression (5-9):

[0162] ; (5-9)

[0163] Let oxygenated hemoglobin and reduced hemoglobin react with the wavelength The light absorption of the beam is the same, that is: ; then we can get the expression (5-10):

[0164] ; (5-10)

[0165] make:

[0166] ;

[0167] ;

[0168] ;

[0169] Then, we can get expression (5-11):

[0170] ; (5-11)

[0171] By constructing a linear regression model, calculating the R value and substituting it into the model, the blood oxygen saturation value can be calculated.

[0172] In order to enable those skilled in the art to better understand the monitoring process of the above-mentioned vital signs monitoring device, the monitoring process of the vital signs monitoring device will be described below using the face as the target detection part. Figure 3 A flowchart of a method for determining vital sign parameter values ​​based on computer vision provided by an embodiment of the present invention is shown in FIG. Figure 3 As shown, the method includes:

[0173] S10: Image acquisition;

[0174] S11: face detection and positioning;

[0175] S12: facial region extraction;

[0176] S13: channel separation, calculation of ROI grayscale average;

[0177] S14: Determine whether the data volume is sufficient; if not, return to step S10; if so, proceed to step S15;

[0178] S15: obtaining the original pulse wave signal;

[0179] S16: POS algorithm processing to obtain effective pulse wave components;

[0180] S17: EEMD decomposition;

[0181] S18: Filter IMF based on the effective range of heart rate and respiratory rate;

[0182] S19: Get the respiratory group signal;

[0183] S20: obtaining a heart rate group signal;

[0184] S21: Wavelet decomposition and reconstruction, noise reduction;

[0185] S22: calculate extreme values;

[0186] S23: Calculate heart rate, heart rate variability, and respiratory rate;

[0187] S24: Calculates blood oxygen saturation using the Lambert-Beer law.

[0188] The vital signs monitoring device provided by the present invention adopts optical principles and can obtain richer and more stable signal data compared to infrared detection equipment and millimeter wave detection equipment, thereby obtaining more types and more accurate physiological indicator parameters; the non-contact usage mode avoids usage restrictions such as surface damage, greatly expanding the usage scenarios, and can be used in various scenarios such as physiological monitoring of the elderly, effectiveness evaluation and training program guidance of rehabilitation training for people with motor dysfunction, and physical status monitoring of office people during work.

[0189] The above is a detailed introduction to a vital signs monitoring device provided by the present invention. The various embodiments in the specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other. It should be pointed out that for ordinary technicians in this technical field, without departing from the principles of the present invention, several improvements and modifications can be made to the present invention, and these improvements and modifications also fall within the scope of protection of the present invention.

[0190] It should also be noted that, in this specification, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus comprising the element.

Claims

1. A vital signs monitoring device, characterized in that: include: a processor, a first motor, a second motor, an image acquisition device, a first supporting member, and a second supporting member; The rotation axis direction of the first motor is different from the rotation axis direction of the second motor; The image acquisition device is fixed on the first supporting component; the first supporting component and the first motor are located on one side of the second supporting component; the second motor is located on the other side of the second supporting component; the non-axial end of the first motor is fixed on the second supporting component; The shaft end of the first motor is tightly connected to the first supporting member; the shaft end of the second motor is tightly connected to the second supporting member; The processor is connected to the first motor and is used to control the first motor to rotate so as to drive the first supporting component and the image acquisition device to move; The processor is connected to the second motor and is used to control the second motor to rotate so as to drive the second supporting component to move; The processor is connected to the image acquisition device and is used to determine a pulse wave signal based on an image containing blood vessels acquired by the image acquisition device; and to determine information used to characterize vital signs based on the pulse wave signal.

2. The vital signs monitoring device according to claim 1, characterized in that: It also includes a base; the processor is fixed on the base; and the non-axial end of the second motor is fixed on the base.

3. The vital signs monitoring device according to claim 1 or 2, characterized in that: The first supporting component is a U-shaped sheet metal component, and the second supporting component is a U-shaped seat; The rotation axis direction of the first motor and the rotation direction of the second motor are perpendicular to each other, the rotation direction of the first motor is perpendicular to the ground, and the rotation direction of the second motor is parallel to the ground.

4. The vital signs monitoring device according to claim 3, characterized in that: The processor determining the pulse wave signal according to the image containing the blood vessel acquired by the image acquisition device includes: Acquire an initial image containing blood vessels from the image acquisition device; determining a target area of ​​the target detection part in the initial image containing the blood vessel according to the characteristics of the target detection part; Separating the color channels of the image containing the blood vessels and obtaining features of the sub-regions contained in the target detection region; The pixel grayscale value of the target area is determined according to the characteristics of the sub-region to determine the pulse wave signal.

5. The vital signs monitoring device according to claim 4, characterized in that: Before determining the information used to represent the vital signs based on the pulse wave signal, the processor further includes: Determine the pulse wave signal of the target length to be acquired according to a preset image acquisition cycle and image acquisition rate; Starting from receiving the image containing the blood vessel captured by the image acquisition device, obtaining the current length of the pulse wave signal; In the case where it is detected that the current length is greater than or equal to the target length, entering the step of determining information used to characterize vital signs based on the pulse wave signal; In the case where it is detected that the current length is less than the target length, the image containing the blood vessel acquired by the image acquisition device is received, and the process returns to the step of acquiring the current length of the pulse wave signal.

6. The vital signs monitoring device according to claim 5, characterized in that: After determining the pixel grayscale value of the target area according to the characteristics of the sub-region to determine the pulse wave signal, the processor further includes: using the determined pulse wave signal as an initial pulse wave signal; Using a planar orthogonal skin model, filtering motion artifacts and illumination changes in the initial pulse wave signal to obtain a final pulse wave signal; The processor determines, based on the pulse wave signal, information for representing vital signs including: The processor determines information for representing vital signs according to the final pulse wave signal.

7. The vital signs monitoring device according to claim 6, characterized in that: The processor determines, based on the final pulse wave signal, information used to characterize vital signs, including: Decomposing the final pulse wave signal and obtaining multiple intrinsic mode functions; Obtain the frequency value of each eigenmode function; According to the frequency value of each eigenmode function and the preset frequency reference range of each vital sign, each eigenmode function is divided into a corresponding vital sign group; wherein the vital sign group includes at least a heart rate group and a respiratory rate group; Filtering out the intrinsic mode functions belonging to the noise group, and determining the extreme points of each vital sign group based on the frequency values ​​of the eigenmode functions in the remaining vital sign groups; wherein the eigenmode functions belonging to the noise group are the eigenmode functions remaining from all the eigenmode functions except the eigenmode functions belonging to the vital sign group; The information used to characterize the vital signs is determined according to the situation of the extreme value points.

8. The vital signs monitoring device according to claim 7, characterized in that: The information used to represent vital signs includes at least heart rate information, respiratory rate information and heart rate variability information; The information used to characterize the vital signs determined according to the extreme point includes: Obtaining a first number of maximum points or minimum points of the heart rate group within an image acquisition period; and determining that the number of heartbeats within the image acquisition period is equal to the first number to obtain heart rate information; Obtaining a second number of maximum points or minimum points of the respiratory rate group within an image acquisition period; determining that the number of respirations within the image acquisition period is equal to the second number to obtain respiratory rate information; The time intervals between all two adjacent maximum points or all two adjacent minimum points within the image acquisition period are obtained; and the heart rate variability information is determined according to the mean of all the time intervals.

9. The vital signs monitoring device according to claim 6, characterized in that: The information used to represent the vital signs is blood oxygen saturation information; the processor determines the blood oxygen saturation information according to the final pulse wave signal, including: Acquire a first pulse wave signal of a light beam with a first channel wavelength and a second pulse wave signal of a light beam with a second channel wavelength within an image acquisition period; determining a first maximum absorption amount of the light beam of the first channel wavelength by the blood, a first minimum absorption amount of the light beam of the first channel wavelength by the blood, a second maximum absorption amount of the light beam of the second channel wavelength by the blood, and a second minimum absorption amount of the light beam of the second channel wavelength by the blood based on a relationship between the pulse wave signal and the absorbance and a relationship between the light absorbance and the medium concentration determined by the Lambert-Beer law; determining a first absorbance variation according to a difference between a first maximum absorption amount and a first minimum absorption amount of the blood for a light beam having a wavelength of the first channel, and determining a second absorbance variation according to a difference between a second maximum absorption amount and a second minimum absorption amount of the blood for a light beam having a wavelength of the second channel; The blood oxygen saturation value is determined based on the first absorbance change and the second absorbance change, and in combination with a pre-established blood oxygen saturation calculation model; wherein the blood oxygen saturation calculation model is the ratio of oxygenated hemoglobin to total hemoglobin.

10. The vital signs monitoring device according to claim 9, characterized in that: Determining the value of the blood oxygen saturation by combining a pre-established blood oxygen saturation calculation model includes: Determining a new calculation model for blood oxygen saturation by combining a pre-established calculation model for blood oxygen saturation; Determine the value of blood oxygen saturation based on a new calculation model of blood oxygen saturation; Among them, the expression of the new calculation model of blood oxygen saturation is: ； ； ； ； in, Indicates blood oxygen saturation, A, B and R all represent numerical values; represents the absorption coefficient of reduced hemoglobin at the first channel wavelength; represents the absorption coefficient of oxygenated hemoglobin at the wavelength of the first channel; represents the absorption coefficient of reduced hemoglobin at the wavelength of the second channel; Indicates the first absorbance change, Indicates the second absorbance change.

Images