Polarization-based single-band blood oxygen monitoring method, device, terminal and storage medium

Abstract

The application discloses a polarization-based single-waveband blood oxygen monitoring method and device, a terminal and a storage medium, and relates to the field of biomedical engineering.The method obtains a target offset compensation value corresponding to an infrared light mode and a target monitoring video of a to-be-monitored object, and infrared light of a lamp panel under the infrared light mode is in an alternating flashing mode of parallel light and vertical light.The method determines a calibrated amplitude ratio according to a target parallel direct current component, a target vertical direct current component and the target offset compensation value corresponding to the target monitoring video by using an amplitude ratio calibration formula, and determines a target blood oxygen saturation according to the calibrated amplitude ratio by using a first fitting formula.The method can effectively solve the problem that the existing polarization-based blood oxygen saturation monitoring method has a high requirement for the symmetry of an optical path and limits its popularization in actual application, because the method monitors in an alternating flashing mode of parallel light and vertical light and calibrates based on a target offset compensation value and a calibration model.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of biomedical engineering, and particularly relates to a single-band blood oxygen monitoring method, device, terminal and storage medium based on polarization. BACKGROUND

[0002] Blood oxygen saturation reflects the oxygen-carrying capacity of human blood and the oxygen supply state of tissues, so blood oxygen saturation monitoring has important application value in the field of clinical monitoring and health management. Recent studies have found that when polarized light is incident on human tissue, depolarization occurs, and different angles of polarizing plates can separate signals at different tissue depths, thereby extracting deep arterial signals rich in oxygen and shallow venous signals almost free of oxygen, and the ratio of the two signals changes with the change of blood oxygen saturation. This method requires a polarizing plate in front of the light source to produce polarized light, and by rotating the polarizing plate in front of the camera or using a vortex plate, the camera obtains two different polarized light signals, namely parallel and perpendicular, to predict blood oxygen saturation. However, this method has high symmetry requirements for the optical path, which limits its popularization and application in practical applications.

[0003] Therefore, the prior art still needs to be improved and developed. SUMMARY

[0004] The technical problem solved by the present application is to provide a single-band blood oxygen monitoring method, device, terminal and storage medium based on polarization to solve the problem that the prior art has high symmetry requirements for the optical path, which limits its popularization in practical applications.

[0005] The technical scheme adopted by the present application to solve the problem is as follows:

[0006] In a first aspect, the present application provides a single-band blood oxygen monitoring method based on polarization, wherein the method comprises:

[0007] obtaining a target offset compensation value, wherein the target offset compensation value is used to compensate the intensity of polarized light in an infrared light mode, and the infrared light of the lamp panel in the infrared light mode is in an alternating flashing mode of parallel light and perpendicular light;

[0008] obtaining a target monitoring video corresponding to the object to be monitored in the infrared light mode, and determining a target parallel direct current component and a target perpendicular direct current component according to the target monitoring video;

[0009] determining a calibrated amplitude ratio according to the target parallel direct current component, the target perpendicular direct current component and the target offset compensation value by using an amplitude ratio calibration formula, wherein the amplitude ratio calibration formula is used to calibrate the amplitude ratio in the infrared light mode according to the offset compensation value;

[0010] determining the target blood oxygen saturation according to the calibration amplitude ratio by using a first fitting formula, wherein the first fitting formula is used to represent a linear relationship between the amplitude ratio and the blood oxygen saturation under the infrared light mode.

[0011] In an implementation method, the target offset compensation value is determined by using the first fitting formula and the amplitude ratio calibration formula based on the second monitoring video and the first blood oxygen saturation, including:

[0012] The first monitoring video corresponding to the to-be-monitored object under the white light mode is acquired, and a first amplitude ratio is determined according to the first monitoring video, wherein the white light of the lamp panel under the white light mode is in a constant-on mode.

[0013] The first blood oxygen saturation is determined according to the first amplitude ratio by using a second fitting formula, wherein the second fitting formula is used to represent a linear relationship between the amplitude ratio and the blood oxygen saturation under the white light mode.

[0014] The second monitoring video corresponding to the to-be-monitored object under the infrared light mode is acquired, and the target offset compensation value is determined by using the first fitting formula and the amplitude ratio calibration formula based on the second monitoring video and the first blood oxygen saturation.

[0015] In an implementation method, the target offset compensation value is determined by using the first fitting formula and the amplitude ratio calibration formula based on the second monitoring video and the first blood oxygen saturation, including:

[0016] The first calibration amplitude ratio is determined according to the first blood oxygen saturation by using the first fitting formula.

[0017] The second parallel direct current component and the second vertical direct current component are determined according to the second monitoring video.

[0018] The target offset compensation value is determined according to the first calibration amplitude ratio, the second parallel direct current component and the second vertical direct current component by using the amplitude ratio calibration formula.

[0019] In an implementation method, the target parallel direct current component and the target vertical direct current component corresponding to each time period are determined according to the target monitoring video, including:

[0020] The target parallel polarization image and the target vertical polarization image in the target monitoring video are acquired.

[0021] The area of the region of interest is acquired, and the region of interest corresponding to the target parallel polarization image and the target vertical polarization image is determined based on the area of the region of interest.

[0022] The direct current component corresponding to the remote photoplethysmography signal is extracted based on the region of interest of the target parallel polarization image as the target parallel direct current component.

[0023] Extracting a direct current component corresponding to a remote photoplethysmography signal from a region of interest of the target vertical polarization image as the target vertical direct current component.

[0024] In an embodiment, the method for calculating the area of the region of interest comprises:

[0025] Obtaining a device parameter and an image size, and calculating a region of interest radius according to the device parameter and the image size;

[0026] Calculating the area of the region of interest according to the region of interest radius.

[0027] In an embodiment, the method for determining the first fitting formula comprises:

[0028] Obtaining a training set, wherein the training set comprises a plurality of training samples, and each training sample comprises a historical monitoring video obtained in an infrared light mode and a standard blood oxygen saturation corresponding to the historical monitoring video;

[0029] Determining an initial offset compensation value corresponding to each historical monitoring video according to the initial offset compensation value formula and each historical monitoring video;

[0030] Determining an initial calibrated amplitude ratio corresponding to each historical monitoring video according to the initial offset compensation value and each historical monitoring video using the amplitude ratio calibration formula;

[0031] Performing linear fitting on the initial calibrated amplitude ratio corresponding to each historical monitoring video and the standard blood oxygen saturation to determine the first fitting formula.

[0032] In an embodiment, determining an initial offset compensation value corresponding to each historical monitoring video according to the initial offset compensation value formula and each historical monitoring video comprises:

[0033] Calculating a first parallel direct current component and a first vertical direct current component corresponding to each time period when the blood oxygen saturation is 100% in each historical monitoring video;

[0034] Determining an amplitude ratio corresponding to each time period according to the first parallel direct current component and the first vertical direct current component corresponding to each time period, and determining an amplitude ratio mean value corresponding to the historical monitoring video according to each amplitude ratio;

[0035] Taking the smallest amplitude ratio mean value in each amplitude ratio mean value as a target amplitude ratio mean value;

[0036] Determining the initial offset compensation value according to the first parallel direct current component, the first vertical direct current component, and the target amplitude ratio mean value using the initial offset compensation value formula.

[0037] In a second aspect, the embodiments of the present application also provide a polarization-based single-band blood oxygen monitoring device, wherein the polarization-based single-band blood oxygen monitoring device comprises:

[0038] The device comprises a light plate and a computing device, the light plate comprises a camera, white lights and infrared lights arranged in parallel mode and vertical mode, the camera is configured to acquire video data and send the video data to the computing device, the white lights are configured to generate white light, the infrared lights are configured to generate infrared light, and the computing device comprises:

[0039] an offset compensation value acquisition module configured to acquire a target offset compensation value, wherein the target offset compensation value is configured to compensate for intensity of polarized light in an infrared light mode, and infrared light of the light plate in the infrared light mode is in an alternating flashing mode of parallel light and vertical light;

[0040] a monitoring data acquisition module configured to acquire a target monitoring video corresponding to a to-be-monitored object in the infrared light mode, and determine a target parallel direct current component and a target vertical direct current component according to the target monitoring video;

[0041] an amplitude ratio calibration module configured to determine a calibrated amplitude ratio according to the target parallel direct current component, the target vertical direct current component and the target offset compensation value by using an amplitude ratio calibration formula, wherein the amplitude ratio calibration formula is configured to calibrate the amplitude ratio in the infrared light mode according to the offset compensation value;

[0042] a blood oxygen saturation calculation module configured to determine a target blood oxygen saturation according to the calibrated amplitude ratio by using a first fitting formula, wherein the first fitting formula is configured to represent a linear relationship between the amplitude ratio and the blood oxygen saturation in the infrared light mode.

[0043] In an implementation method, the camera, the white lights and the infrared lights in the light plate are arranged in a nine-square grid form;

[0044] The camera is installed at a center position of the nine-square grid, and a front polarizer is arranged in the camera;

[0045] Each of the white lights is installed at a position of four corners of the nine-square grid;

[0046] On four sides of the nine-square grid, an infrared light is installed between every two white lights, a front polarizer of the infrared light above and below the camera is parallel to a polarizer of the camera, and a front polarizer of the infrared light left and right of the camera is perpendicular to the polarizer of the camera.

[0047] In a third aspect, an embodiment of the present application further provides a terminal, comprising a memory and one or more processors; the memory stores one or more programs; the programs contain instructions for executing the polarization-based single-band blood oxygen monitoring method according to any of the above; and the processors are configured to execute the programs.

[0048] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, which stores a plurality of instructions, wherein the instructions are adapted to be loaded and executed by a processor to implement the polarization-based single-band blood oxygen monitoring method according to any of the above.

[0049] The present application has the following beneficial effects: The embodiment of the present application obtains a target offset compensation value corresponding to an infrared light mode and a target monitoring video of a to-be-monitored object, the infrared light of the lamp panel in the infrared light mode is in an alternating flashing mode of parallel light and vertical light; a calibration amplitude ratio is determined according to a target parallel direct current component, a target vertical direct current component and the target offset compensation value corresponding to the target monitoring video by using an amplitude ratio calibration formula; and a target blood oxygen saturation is determined according to the calibration amplitude ratio by using a first fitting formula. Since the present application monitors in the alternating flashing mode of parallel light and vertical light, and calibrates based on the target offset compensation value and the calibration model, the problem that the existing polarization-based blood oxygen saturation monitoring method has a high requirement for the symmetry of the light path, which limits its popularization in practical applications, can be effectively solved. BRIEF DESCRIPTION OF DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor based on these drawings.

[0051] Figure 1 is a flowchart of the polarization-based single-band blood oxygen monitoring method provided by the embodiment of the present application.

[0052] Figure 2 is an internal module schematic diagram of the polarization-based single-band blood oxygen monitoring device provided by the embodiment of the present application.

[0053] Figure 3 is a principle block diagram of the terminal provided by the embodiment of the present application. DETAILED DESCRIPTION

[0054] The application discloses a polarization-based single-waveband blood oxygen monitoring method and device, a terminal and a storage medium. In order to make the purpose, technical scheme and effect of the application more clear and explicit, the application is further described in detail below with reference to the drawings and examples. It should be understood that the specific examples described herein are only used to explain the application and not to limit the application.

[0055] Those skilled in the art can understand that the singular forms "a," "an," and "the" used herein include plural forms unless specifically stated otherwise. It should be further understood that the use of the term "includes" in the specification of the application means that the stated features, integers, steps, operations, elements, and / or components are present, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. It should be understood that when we say an element is "connected" or "coupled" to another element, it can be directly connected or coupled to the other element, or there can be intermediate elements. In addition, "connected" or "coupled" used herein can include wireless connection or wireless coupling. The phrase "and / or" used herein includes all or any single unit and all combinations of the associated listed items.

[0056] Those skilled in the art can understand that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as that generally understood by those skilled in the art to which the application belongs. It should also be understood that terms such as those defined in a general dictionary should be understood to have meanings consistent with those in the context of the prior art, and should not be interpreted in an idealized or overly formal sense unless specifically defined as such.

[0057] Blood oxygen saturation reflects the oxygen-carrying capacity of human blood and the oxygen supply state of tissues, so blood oxygen saturation monitoring has important application value in the field of clinical monitoring and health management. Recent studies have found that when polarized light is incident on human tissue, depolarization occurs, and different angles of polarizing plates can separate signals at different tissue depths, thereby extracting deep arterial signals rich in oxygen and shallow venous signals almost free of oxygen, and the ratio of the two signals changes with the change of blood oxygen saturation. This method requires a polarizing plate in front of the light source to produce polarized light, and by rotating the polarizing plate in front of the camera or using a vortex plate, the camera obtains two different polarized light signals, namely parallel and perpendicular, to predict blood oxygen saturation. However, this method has high symmetry requirements for the optical path, which limits its popularization and application in practical applications.

[0058] In view of the above defects of the prior art, the present application provides a single-band blood oxygen monitoring method based on polarization, which obtains a target offset compensation value corresponding to an infrared light mode and a target monitoring video of a to-be-monitored object, and infrared light of a lamp panel in the infrared light mode is in an alternating flashing mode of parallel light and vertical light; a calibration amplitude ratio is determined according to a target parallel direct current component, a target vertical direct current component and the target offset compensation value corresponding to the target monitoring video by using an amplitude ratio calibration formula; and a target blood oxygen saturation is determined according to the calibration amplitude ratio by using a first fitting formula. Since the present application monitors in the infrared light mode of the alternating flashing mode of the parallel light and the vertical light, and calibrates based on the target offset compensation value and a calibration model, the problem that the existing blood oxygen saturation monitoring method based on polarization has a high requirement for the symmetry of an optical path and limits its popularization in practical applications can be effectively solved.

[0059] An example method includes:

[0060] As shown in Figure 1 the method includes:

[0061] In step S100, a target offset compensation value is obtained, wherein the target offset compensation value is used to compensate the intensity of polarized light in an infrared light mode, and infrared light of a lamp panel in the infrared light mode is in an alternating flashing mode of parallel light and vertical light.

[0062] In related technologies, a polarizer or a wave plate in front of a camera needs to be rotated so that the camera obtains two different polarized light signals of parallel and vertical, and then predicts blood oxygen saturation, and this method has a very high requirement for the symmetry of an optical path. In order to solve this problem, the present application sets the infrared light of the lamp panel to be in an alternating flashing mode of parallel light and vertical light, thereby avoiding the problem that the symmetry of the optical path cannot be met. The target offset compensation value is used to compensate the intensity of the polarized light in the infrared light mode, thereby eliminating the angle difference, so that the device has stronger robustness to to-be-monitored objects with different postures and different skin conditions.

[0063] The single-band blood oxygen monitoring device based on polarization is used to monitor the blood oxygen of a to-be-monitored object, and is divided into a calibration phase and a monitoring phase. In the calibration phase, a target offset compensation value is generated; in the monitoring phase, the mode of the lamp panel is set to an infrared light mode, and the blood oxygen saturation calculated according to the target monitoring video is calibrated based on the target offset compensation value and a calibration model in the infrared light mode, thereby realizing accurate monitoring of the change of the blood oxygen of the user. Generally, the white light of the lamp panel in the infrared light mode is in an off state.

[0064] In one implementation, the target offset compensation value is obtained, including:

[0065] In step S101, a first monitoring video corresponding to the to-be-monitored object in a white light mode is acquired, and a first amplitude ratio is determined according to the first monitoring video, wherein the white light of the lamp panel in the white light mode is in a constant-on mode.

[0066] In step S102, a first blood oxygen saturation is determined according to the first amplitude ratio by using a second fitting formula, wherein the second fitting formula is used to represent a linear relationship between the amplitude ratio and the blood oxygen saturation in the white light mode.

[0067] In step S103, a second monitoring video corresponding to the to-be-monitored object in an infrared light mode is acquired, and a target offset compensation value is determined by performing calculation based on the second monitoring video and the first blood oxygen saturation by using the first fitting formula and the amplitude ratio calibration formula.

[0068] Specifically, the to-be-monitored object is a user who needs to perform blood oxygen monitoring. In the calibration stage and the monitoring stage, the posture of the to-be-monitored object is basically the same, which can be lying or sitting. The shooting part of each monitoring video (such as the first monitoring video, the second monitoring video, or the target monitoring video) corresponding to the to-be-monitored object is a skin exposed part of the to-be-monitored object, such as the face or the hand. The main shooting part is the face of the to-be-monitored object. When the polarization-based single-band blood oxygen monitoring device is used for the first time, it needs to complete calibration through the calibration stage first, that is, the intensity of the polarized light is compensated by the target offset compensation value. After calibration, the blood oxygen saturation of the to-be-monitored object can be monitored through the monitoring stage. After the polarization-based single-band blood oxygen monitoring device performs a calibration operation once, it does not need to be calibrated again in the subsequent use process.

[0069] In the calibration stage, the first monitoring video corresponding to the to-be-monitored object is collected in the white light mode, such as 3 seconds of video data, and the first amplitude ratio is calculated based on the first monitoring video. In the white light mode, the white light of the lamp panel is in a constant-on mode, and the infrared light is in an off state. The first amplitude ratio is calculated according to the first monitoring video, which includes: extracting a remote photoplethysmography signal (rppg signal) according to each video frame corresponding to the first monitoring video, determining the first amplitude ratio according to a first direct current component and a first alternating current component corresponding to the remote photoplethysmography signal, and the amplitude ratio calculation formula is:

[0070] ,

[0071] wherein, , are the direct current component and the alternating current component (herein, the first direct current component and the first alternating current component are substituted, respectively) corresponding to the remote photoplethysmography signal, , are the direct current component and the alternating current component corresponding to the red channel, , respectively a direct current component and an alternating current component corresponding to the green channel, a signal representing a red channel, a signal representing a green channel.

[0072] When the first amplitude ratio is determined, the first amplitude ratio is substituted into a second fitting formula to calculate a first blood oxygen saturation. The second fitting formula is used to represent a linear relationship between the amplitude ratio and the blood oxygen saturation in the white light mode, and is expressed as:

[0073] ,

[0074] wherein, represents the blood oxygen saturation, represents the amplitude ratio, , respectively a slope and a constant term corresponding to the second fitting formula.

[0075] The determination method of the second fitting formula is as follows:

[0076] A training set (second training set) for training the second fitting formula is obtained, and the training set includes a plurality of training samples. Each training sample corresponds to one or more historical monitoring videos (second historical monitoring video) and a standard blood oxygen saturation (second standard blood oxygen saturation) corresponding to each historical monitoring video. Each monitoring video includes a plurality of video frames or image frames, and each monitoring video is obtained in a white light mode. In the white light mode, the region of interest is the entire forehead region of the user (or sample), and the region of interest can be used as a second region of interest. Based on the second region of interest, a remote photoplethysmography signal corresponding to each sample is extracted from the video frame or image frame corresponding to the sample; and the amplitude ratio is calculated according to the direct current component and the alternating current component corresponding to the remote photoplethysmography signal, to obtain the amplitude ratio corresponding to the sample.

[0077] The standard blood oxygen data corresponding to each sample in the training set is obtained, and the standard blood oxygen data is the blood oxygen saturation recorded by the monitor. The blood oxygen saturation and the amplitude ratio corresponding to each sample are linearly fitted to obtain the slope and the constant term in the second fitting formula, thereby obtaining the trained second fitting formula.

[0078] When the first blood oxygen saturation is obtained, the light plate is converted to an infrared light mode. In the infrared light mode, the infrared light of the light plate is in an alternating flashing mode of parallel light and vertical light, and the white light is in an off state. In the infrared light mode, a second monitoring video is obtained, and a target offset compensation value is calculated according to the second monitoring video and the first blood oxygen saturation through the first fitting formula and the calibration model.

[0079] In an implementation manner, the target offset compensation value is determined based on the second monitoring video and the first blood oxygen saturation by using the first fitting formula and the amplitude ratio calibration formula, including:

[0080] In step S1031, a first calibration amplitude ratio is determined according to the first blood oxygen saturation by using the first fitting formula.

[0081] In step S1032, a second parallel direct current component and a second vertical direct current component are determined according to the second monitoring video.

[0082] In step S1033, the target offset compensation value is determined according to the first calibration amplitude ratio, the second parallel direct current component and the second vertical direct current component by using the amplitude ratio calibration formula.

[0083] The first blood oxygen saturation is substituted into the first fitting formula to obtain a first calibration amplitude ratio corresponding to the first blood oxygen saturation. A parallel direct current component (second parallel direct current component) corresponding to a remote photoplethysmography signal is extracted from a parallel polarization image (second parallel polarization image) in the second monitoring video, and a vertical direct current component (second vertical direct current component) corresponding to the remote photoplethysmography signal is extracted from a vertical polarization image (second vertical polarization image) in the second monitoring video. The second parallel direct current component, the second vertical direct current component and the first calibration amplitude ratio are substituted into the calibration model to obtain the target offset compensation value. The amplitude ratio calibration formula is represented as:

[0084]

[0085] wherein, is a calibration amplitude ratio, is an offset compensation value, is a parallel direct current component, is a vertical direct current component. The first calibration amplitude ratio is substituted into the first fitting formula, the second parallel direct current component is substituted into the second parallel direct current component, and the second vertical direct current component is substituted into the second vertical direct current component, and the target offset compensation value is calculated. (this is the target offset compensation value).

[0086] In step S200, a target monitoring video corresponding to a to-be-monitored object in an infrared light mode is obtained, and a target parallel direct current component and a target vertical direct current component are determined according to the target monitoring video.

[0087] ​​​​Briefly, the present application monitors the blood oxygen saturation of the object to be monitored in infrared light mode, separates the deep and shallow layer remote photoplethysmography signals through polarized light, extracts the target parallel direct current component and the target vertical direct current component corresponding to each period, and realizes accurate tracking of the change of blood oxygen saturation by using the principle that the deep layer artery is rich in deoxyhemoglobin and the shallow layer vein is rich in oxyhemoglobin.

[0088] In an implementation manner, the target parallel direct current component and the target vertical direct current component corresponding to each period are determined according to the target monitoring video, and the method comprises the following steps.

[0089] Step S2011, acquiring a target parallel polarized image and a target vertical polarized image in the target monitoring video;

[0090] Step S2012, acquiring an area of a region of interest, and determining a region of interest corresponding to the target parallel polarized image and the target vertical polarized image respectively based on the area of the region of interest;

[0091] Step S2013, extracting a direct current component corresponding to a remote photoplethysmography signal as a target parallel direct current component based on the region of interest of the target parallel polarized image;

[0092] Step S2014, extracting a direct current component corresponding to a remote photoplethysmography signal as a target vertical direct current component based on the region of interest of the target vertical polarized image.

[0093] Specifically, the infrared light of the lamp panel in the infrared light mode is in an alternating flashing mode of parallel light and vertical light, and according to the different flashing modes of the infrared light, each image frame in the monitoring video is also different. The image frame of the target monitoring video comprises a plurality of target parallel polarized images and a plurality of target vertical polarized images. The area of the region of interest is acquired, the region of interest is located on each target parallel polarized image and each target vertical polarized image according to the area of the region of interest (generally in the forehead region), a direct current component corresponding to a remote photoplethysmography signal is extracted from the region of interest of the target parallel polarized image as a target parallel direct current component, and a direct current component corresponding to a remote photoplethysmography signal is extracted from the region of interest of the target vertical polarized image as a target vertical direct current component.

[0094] In an implementation manner, the method for calculating the area of the region of interest comprises the following steps.

[0095] Acquiring a device parameter and an image size, and calculating the radius of the region of interest according to the device parameter and the image size;

[0096] Calculating the area of the region of interest according to the radius of the region of interest.

[0097] Specifically, the region of interest in the infrared light mode (first region of interest) is different from the region of interest in the white light mode (second region of interest). Device parameters including object distance, focal length, infrared light (LED light) radius, image size (unit: pixel) and sensor size are obtained, and the region of interest radius in the infrared mode is calculated according to the device parameters and the image size. The calculation method is as follows:

[0098] ,

[0099] wherein, is the region of interest radius, is the object distance, is the focal length, is the infrared light radius, is the image size, is the sensor size. The area of the region of interest can be calculated according to the region of interest radius by using the area calculation formula of a circle.

[0100] In step S300, a calibration amplitude ratio is determined according to the target parallel direct current component, the target vertical direct current component and the target offset compensation value by using an amplitude ratio calibration formula, wherein the amplitude ratio calibration formula is used to calibrate the amplitude ratio in the infrared light mode according to the offset compensation value.

[0101] By substituting the target parallel polarized image into , substituting the target vertical polarized image into and substituting the target offset compensation value into , the following amplitude ratio calibration formula is used for calculation:

[0102] ,

[0103] , that is, the calibration amplitude ratio.

[0104] In step S400, a target blood oxygen saturation is determined according to the calibration amplitude ratio by using a first fitting formula, wherein the first fitting formula is used to represent the linear relationship between the amplitude ratio and the blood oxygen saturation in the infrared light mode.

[0105] The first fitting formula is used to represent the linear relationship between the amplitude ratio and the blood oxygen saturation in the infrared light mode, and the calibration amplitude ratio is substituted into the first fitting formula to obtain the target blood oxygen saturation. The first fitting formula is represented as:

[0106] ,

[0107] wherein, is the blood oxygen saturation calculated by the first fitting formula, is the calibration amplitude ratio,​ , respectively are the slope and constant term corresponding to the first fitting formula.

[0108] In an implementation manner, the method for obtaining the first fitting formula comprises:

[0109] obtaining a training set, wherein the training set comprises a plurality of training samples, and each training sample comprises a historical monitoring video acquired in an infrared light mode and a standard blood oxygen saturation corresponding to the historical monitoring video;

[0110] determining, according to each historical monitoring video, an initial offset compensation value corresponding to each historical monitoring video by using an initial offset compensation value formula;

[0111] determining, according to the initial offset compensation value and each historical monitoring video, an initial calibrated amplitude ratio corresponding to each historical monitoring video by using the amplitude ratio calibration formula;

[0112] performing linear fitting on the initial calibrated amplitude ratio corresponding to each historical monitoring video and the standard blood oxygen saturation to determine the first fitting formula.

[0113] obtaining a training set corresponding to an infrared light mode (a first training set), wherein each sample in the training set corresponds to one or more historical monitoring videos and a standard blood oxygen saturation corresponding to each historical monitoring video, and each historical monitoring video comprises a plurality of video frames or image frames. An initial offset compensation value corresponding to each historical monitoring video is determined according to each historical monitoring video by using an initial offset compensation value formula. The initial offset compensation value is calculated based on the training set, and the amplitude ratio calibration formula is used to calibrate the amplitude ratio calculated based on the video frames or image frames of each sample in the training set according to the initial offset compensation value, to obtain an initial calibrated amplitude ratio. The standard blood oxygen saturation corresponding to each sample (a first standard blood oxygen saturation) is obtained, and the first fitting formula is fitted according to the standard blood oxygen saturation corresponding to each sample and the initial calibrated amplitude ratio by using the first fitting formula, to obtain the slope and constant term corresponding to the first fitting formula, so as to determine the first fitting formula based on the slope and constant term corresponding to the first fitting formula.

[0114] In an implementation manner, the method for determining, according to each historical monitoring video, an initial offset compensation value corresponding to each historical monitoring video by using an initial offset compensation value formula comprises:

[0115] calculating a first parallel direct current component and a first vertical direct current component corresponding to each time period when the blood oxygen saturation is 100% in each historical monitoring video;

[0116] determining a magnitude ratio corresponding to each time period according to the first parallel direct current component and the first vertical direct current component corresponding to each time period respectively, and determining a magnitude ratio mean value corresponding to the historical monitoring video according to each magnitude ratio;

[0117] taking the smallest magnitude ratio mean value in each magnitude ratio mean value as a target magnitude ratio mean value;

[0118] determining the initial offset compensation value according to the first parallel direct current component, the first vertical direct current component and the target magnitude ratio mean value by using an initial offset compensation value formula.

[0119] calculating the first parallel direct current component and the first vertical direct current component corresponding to the historical monitoring video when the blood oxygen saturation is 100% in each training sample in the training set, specifically including: obtaining the first parallel polarization image and the first vertical polarization image corresponding to each time period when the blood oxygen saturation is 100% in the historical monitoring video of each training sample; extracting the direct current component corresponding to the remote photoplethysmography signal based on the first parallel polarization image as the first parallel direct current component; extracting the direct current component corresponding to the remote photoplethysmography signal based on the first vertical polarization image as the first vertical direct current component.

[0120] calculating each magnitude ratio according to each first parallel direct current component and each first vertical direct current component by using an infrared light mode magnitude ratio calculation formula; determining a magnitude ratio mean value by averaging each magnitude ratio. Wherein, the infrared light mode magnitude ratio calculation formula is expressed as:

[0121] ,

[0122] wherein, represents the magnitude ratio, represents the parallel direct current component (herein, the first parallel direct current component is substituted), represents the vertical direct current component (herein, the first vertical direct current component is substituted).

[0123] selecting the smallest magnitude ratio mean value from the magnitude ratio mean values corresponding to each sample in the training set as a target magnitude ratio mean value, and determining the initial offset compensation value according to the first parallel direct current component and the first vertical direct current component, the target magnitude ratio mean value and the initial offset compensation value formula, the first parallel direct current component and the first vertical direct current component being the same as the historical monitoring video corresponding to the smallest magnitude ratio mean value. Wherein, the initial offset compensation value formula is expressed as:

[0124] ,

[0125] wherein, is the initial offset compensation value, is a parallel direct current component (herein, the first parallel direct current component is substituted), is a target amplitude ratio mean value, is a vertical direct current component (herein, the first vertical direct current component is substituted).

[0126] In an implementation manner, the blood oxygen saturation monitoring can also utilize the physiological mechanism of different oxygen contents of deep and shallow blood or other forms of algorithms or formulas equivalent in mathematics to calibrate or monitor blood oxygen.

[0127] The above method of the present application has the following beneficial effects:

[0128] The polarized light separates the deep and shallow layer remote photoplethysmography signals, and utilizes the principle that the deep layer arteries are rich in deoxyhemoglobin and the shallow layer veins are rich in oxyhemoglobin to realize tracking of the change of blood oxygen saturation;

[0129] The calibration method is introduced, so that the quantitative monitoring of blood oxygen saturation is no longer limited to fixed optical points;

[0130] The redundancy of the camera is utilized, the blood oxygen saturation selection rules are formulated according to the imaging principle and device parameters, and the incident light compensation mechanism is introduced, so that the flexibility of the device is increased.

[0131] Based on the above embodiment, the present application further provides a single-band blood oxygen monitoring device based on polarization, as shown in Figure 2 The device includes a lamp panel 01 and a computing device 02, the lamp panel includes a camera 11, a white lamp 12, and infrared lamps 13 arranged in parallel and vertical modes, the camera 11 is used to acquire video data and send the video data to the computing device 02, the white lamp 12 is used to generate white light, and the infrared lamps 13 are used to generate infrared light, and the computing device includes:

[0132] An offset compensation value acquisition module 21 is configured to acquire a target offset compensation value, wherein the target offset compensation value is used to compensate the intensity of polarized light in an infrared light mode, and the infrared light of the lamp panel in the infrared light mode is in an alternating flashing mode of parallel light and vertical light;

[0133] A monitoring data acquisition module 22 is configured to acquire a target monitoring video corresponding to a to-be-monitored object in the infrared light mode, and determine a target parallel direct current component and a target vertical direct current component according to the target monitoring video;

[0134] An amplitude ratio calibration module 23 is configured to determine a calibrated amplitude ratio according to the initial amplitude ratio and the target offset compensation value by using an amplitude ratio calibration formula, wherein the amplitude ratio calibration formula is used to calibrate the amplitude ratio in the infrared light mode according to the offset compensation value;

[0135] The blood oxygen saturation calculation module 24 is configured to determine a target blood oxygen saturation according to the calibration amplitude ratio by using a first fitting formula, wherein the first fitting formula is used to represent a linear relationship between the amplitude ratio and the blood oxygen saturation in the infrared light mode.

[0136] In an implementation, the camera 11, the white light 12 and the infrared light 13 in the lamp panel 01 are arranged in a nine-square grid.

[0137] The camera 11 is installed at the center of the nine-square grid, and the camera 11 is provided with a front polarizer.

[0138] Each white light 12 is installed at a corner of the nine-square grid.

[0139] On the four sides of the nine-square grid, the infrared light 13 is installed between every two white lights 12, and the infrared light 13 above and below the camera is provided with a polarizer parallel to the polarizer of the camera, and the infrared light 13 left and right of the camera is provided with a polarizer perpendicular to the polarizer of the camera.

[0140] The white light 12, the infrared light 13 and the camera 11 in the lamp panel 01 are all aligned with the head of the human body, and the lamp panel 01 is powered by a direct current power supply.

[0141] Based on the above-mentioned embodiments, the application further provides a terminal, and a principle block diagram thereof can be shown in FIG. Figure 3 The terminal includes a processor, a memory, a network interface and a display screen connected through a system bus. The processor of the terminal is configured to provide computing and control capabilities. The memory of the terminal includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The network interface of the terminal is configured to communicate with external terminals through network connection. The computer program is executed by the processor to implement the polarization-based single-band blood oxygen monitoring method. The display screen of the terminal can be a liquid crystal display screen or an electronic ink display screen.

[0142] Those skilled in the art can understand that, Figure 3 The principle block diagram shown in FIG.

[0143] In an implementation, the memory of the terminal stores more than one program, and is configured to execute the more than one program by more than one processor, which includes instructions for performing the polarization-based single-band blood oxygen monitoring method.

[0144] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer readable storage medium. When the computer program is executed, it can include the processes of the above-mentioned embodiments of each method. Any reference to memory, storage, database or other medium used in each embodiment provided by the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0145] In summary, the present application discloses a single-band blood oxygen monitoring method and device based on polarization, a terminal and a storage medium. The method obtains a target offset compensation value corresponding to an infrared light mode and a target monitoring video of a to-be-monitored object. The infrared light of the lamp panel in the infrared light mode is in an alternating flashing mode of parallel light and vertical light. A calibration amplitude ratio formula is used to determine a calibration amplitude ratio according to a target parallel direct current component, a target vertical direct current component and the target offset compensation value corresponding to the target monitoring video. A first fitting formula is used to determine a target blood oxygen saturation according to the calibration amplitude ratio. Since the present application monitors in the alternating flashing mode of parallel light and vertical light, and calibrates based on the target offset compensation value and the calibration model, the problem that the existing blood oxygen saturation monitoring method based on polarization has a high requirement for the symmetry of the light path and limits its popularization in practical applications can be effectively solved.

[0146] It should be understood that the application of the present application is not limited to the above examples, and those skilled in the art can make improvements or changes according to the above description, and all these improvements and changes shall belong to the protection scope of the claims of the present application.

Claims

1. A polarization-based single-wavelength blood oxygen monitoring method, characterized in that, The method comprises: obtaining a target offset compensation value, wherein the target offset compensation value is used to compensate the intensity of polarized light in an infrared light mode, and the infrared light of the lamp panel in the infrared light mode is an alternating flashing mode of parallel light and vertical light; obtaining a target monitoring video corresponding to the to-be-monitored object in the infrared light mode, and determining a target parallel direct current component and a target vertical direct current component according to the target monitoring video; Based on the amplitude ratio calibration formula, the target parallel direct current component is substituted into The target vertical direct current component is substituted into The target offset compensation value is substituted into The calculation is performed to determine the calibrated amplitude ratio; based on the first fitting formula, substituting the calibration amplitude ratio into calculating to determine the target blood oxygen saturation; obtaining a target offset compensation value, comprising: obtaining a first monitoring video corresponding to the to-be-monitored object in a white light mode, and determining a first amplitude ratio according to the first monitoring video, wherein the white light of the lamp panel in the white light mode is a constant light mode; based on the second fitting formula, substituting the first amplitude ratio into performing a calculation to determine a first blood oxygen saturation; obtaining a second monitoring video corresponding to the to-be-monitored object in the infrared light mode, and determining a target offset compensation value based on the second monitoring video and the first blood oxygen saturation by using the first fitting formula and the amplitude ratio calibration formula; determining a target offset compensation value based on the second monitoring video and the first blood oxygen saturation by using the first fitting formula and the amplitude ratio calibration formula, comprising: based on the first fitting formula, substituting the first blood oxygen saturation into a first calibration amplitude ratio is determined by calculation; determining a second parallel direct current component and a second vertical direct current component according to the second monitoring video; Based on the amplitude ratio calibration formula, the first calibration amplitude ratio is substituted into The second parallel direct current component is substituted into The second vertical direct current component is substituted into The target offset compensation value is determined by calculation; the amplitude ratio calibration formula is represented as: ， wherein, is a calibration amplitude ratio, is an offset compensation value, is a parallel direct current component, is a perpendicular direct current component; the first fitting formula is represented as: ， wherein, SpO2 is the blood oxygen saturation, is the calibration amplitude ratio, , are the slope and constant term of the first fitting formula, respectively; determining a first amplitude ratio according to the first monitoring video, comprising: extracting a photoplethysmography signal from each video frame corresponding to the first monitoring video, and determining a first amplitude ratio according to a direct current component and an alternating current component corresponding to the photoplethysmography signal by using an amplitude ratio calculation formula; the amplitude ratio calculation formula is represented as: ， , are respectively the direct current component and the alternating current component corresponding to the red channel, , are respectively the direct current component and the alternating current component corresponding to the green channel, denotes the signal of the red channel, denotes the signal of the green channel; the second fitting formula is represented as: ， wherein, is the blood oxygen saturation, is the amplitude ratio, , are the slope and constant term of the second fitting formula, respectively.

2. The polarization-based single-wavelength blood oxygen monitoring method of claim 1, wherein, determining a target parallel direct current component and a target vertical direct current component corresponding to each period according to the target monitoring video, comprising: obtaining a target parallel polarization image and a target vertical polarization image in the target monitoring video; obtaining an area of a region of interest, and determining a region of interest corresponding to the target parallel polarization image and the target vertical polarization image based on the area of the region of interest; extracting a direct current component corresponding to a photoplethysmography signal based on the region of interest of the target parallel polarization image as the target parallel direct current component; extracting a direct current component corresponding to a photoplethysmography signal based on the region of interest of the target vertical polarization image as the target vertical direct current component.

3. The polarization-based single-wavelength blood oxygen monitoring method of claim 2, wherein, A method for calculating an area of a region of interest, comprising: obtaining a device parameter and an image size, and calculating a region of interest radius according to the device parameter and the image size; calculating the area of the region of interest according to the region of interest radius.

4. The polarization-based single-wavelength blood oxygen monitoring method of claim 1, wherein, A method for determining the first fitting formula, comprising: obtaining a training set, wherein the training set comprises a plurality of training samples, and each training sample comprises a historical monitoring video obtained in an infrared light mode and a standard blood oxygen saturation corresponding to the historical monitoring video; determining an initial offset compensation value corresponding to each historical monitoring video according to each historical monitoring video by using an initial offset compensation value formula; determining an initial calibration amplitude ratio corresponding to each historical monitoring video according to the initial offset compensation value and each historical monitoring video by using the amplitude ratio calibration formula; Linear fitting is performed on the initial calibration amplitude ratio corresponding to each historical monitoring video and the standard blood oxygen saturation to determine the first fitting formula.

5. The polarization-based single-wavelength blood oxygen monitoring method of claim 4, wherein, An initial offset compensation value corresponding to each historical monitoring video is determined according to the initial offset compensation value formula, including: A first parallel direct current component and a first vertical direct current component corresponding to each time period when the blood oxygen saturation is 100% in each historical monitoring video are calculated. An amplitude ratio corresponding to each time period is determined according to the first parallel direct current component and the first vertical direct current component corresponding to each time period, and an amplitude ratio mean value corresponding to the historical monitoring video is determined according to the amplitude ratio corresponding to each time period. The smallest amplitude ratio mean value among the amplitude ratio mean values is taken as a target amplitude ratio mean value. Based on the initial offset compensation value formula, the first parallel direct current component is substituted into the first vertical direct current component is substituted into the target amplitude ratio average is substituted into The calculation is performed to determine the initial offset compensation value; The initial offset compensation value formula is represented as: ， wherein, is the initial offset compensation value, is the parallel direct current component, is the target amplitude ratio mean value, is the perpendicular direct current component.

6. A polarization-based single-wavelength blood oxygen monitoring device, characterized by, The device is used to implement the polarization-based single-waveband blood oxygen monitoring method according to any one of claims 1-5, and the device includes a lamp panel and a computing device, the lamp panel includes a camera, a white lamp, and infrared lamps arranged in parallel and vertical modes, the camera is used to acquire video data and send the video data to the computing device, the white lamp is used to generate white light, and the infrared lamps are used to generate infrared light, and the computing device includes: An offset compensation value acquisition module is configured to acquire a target offset compensation value, wherein the target offset compensation value is used to compensate the intensity of polarized light in an infrared light mode, and the infrared light of the lamp panel in the infrared light mode is in an alternating flashing mode of parallel light and vertical light. A monitoring data acquisition module is configured to acquire a target monitoring video corresponding to a to-be-monitored object in the infrared light mode, and determine a target parallel direct current component and a target vertical direct current component according to the target monitoring video. The amplitude ratio calibration module substitutes the target parallel direct current component into an amplitude ratio calibration formula based on the amplitude ratio calibration formula substitutes the target vertical direct current component into the amplitude ratio calibration formula substitutes the target offset compensation value into the amplitude ratio calibration formula performs calculation to determine the calibrated amplitude ratio a blood oxygen saturation calculation module configured to determine a target blood oxygen saturation based on the calibrated amplitude ratio substituted into the first fitting formula performing the calculation to determine the target blood oxygen saturation.

7. A terminal, characterized by comprising: The terminal includes a memory and one or more processors, the memory stores one or more programs, the programs include instructions for executing the polarization-based single-waveband blood oxygen monitoring method according to any one of claims 1-5, and the processors are configured to execute the programs.

8. A computer-readable storage medium storing a plurality of instructions, characterized in that, The instructions are loaded and executed by the processor to implement the steps of the polarization-based single-waveband blood oxygen monitoring method according to any one of claims 1-5.

Images