Blood oxygen saturation detection method, device, terminal equipment and storage medium based on pulse signal calibration

Abstract

The present application relates to the field of biomedical engineering technology. The present application discloses a blood oxygen saturation detection method, apparatus, terminal device and storage medium based on pulsation signal calibration, which can improve the detection accuracy of blood oxygen saturation. The method includes obtaining a facial video of a user; extracting multiple cardiovascular signals from the facial video based on time sequence; using a preset regression model to perform deviation prediction processing on the pulsation signal and non-pulsation signal of the first target cardiovascular signal to obtain a deviation value, wherein the first target cardiovascular signal is any one of the multiple cardiovascular signals; calculating the blood oxygen saturation of the second target cardiovascular signal based on the deviation value to obtain the target blood oxygen saturation of the user, wherein the second target cardiovascular signal is any one of all cardiovascular signals except the first target cardiovascular signal.

Description

Technical Field

[0001] The present application relates to the field of biomedical engineering technology. More specifically, the present application relates to a blood oxygen saturation detection method, apparatus, terminal device, and storage medium based on pulse signal calibration. Background Art

[0002] Traditional methods for measuring blood oxygen saturation (BOS) obtain a pressure detection signal between a BOS device and the area being measured to determine a pressure measurement value. When the pressure measurement value is within a preset range, the BOS signal is obtained from the area being measured to determine the BOS. In other words, the determination of BOS depends on the pressure measurement value. Inaccurate pressure measurement values ​​directly impact the accuracy of BOS. Consequently, the accuracy of BOS remains an urgent issue in existing technologies. Summary of the Invention

[0003] The purpose of the embodiments of the present application is to provide a blood oxygen saturation detection method, apparatus, terminal device, and storage medium based on pulse signal calibration, which can improve the detection accuracy of blood oxygen saturation. The embodiments of the present application are mainly achieved through the following technical solutions:

[0004] A first aspect of an embodiment of the present application provides a blood oxygen saturation detection method based on pulse signal calibration, comprising:

[0005] Get the user's face video;

[0006] extracting a plurality of cardiovascular signals from the facial video based on a time sequence;

[0007] performing deviation prediction processing on a pulsatile signal and a non-pulsatile signal of a first target cardiovascular signal using a preset regression model to obtain a deviation value, wherein the first target cardiovascular signal is any one of the multiple cardiovascular signals;

[0008] The blood oxygen saturation of a second target cardiovascular signal is calculated based on the deviation value to obtain the target blood oxygen saturation of the user, where the second target cardiovascular signal is any one of all cardiovascular signals except the first target cardiovascular signal.

[0009] According to one embodiment of the present application, the step of obtaining a user's facial video includes:

[0010] Set up multiple different wavelengths of light;

[0011] The facial video is collected when the multiple light beams of different wavelengths illuminate the user's face in a time division multiplexing manner.

[0012] According to one embodiment of the present application, the step of extracting multiple cardiovascular signals from the facial video based on time sequence includes:

[0013] Dividing the facial video based on the time sequence and the multiple different wavelength information to obtain multiple sub-videos, wherein each sub-video corresponds to a type of wavelength information;

[0014] Cardiovascular pulse signal extraction processing is performed on each sub-video to obtain a cardiovascular signal corresponding to each sub-video.

[0015] According to one embodiment of the present application, the step of performing deviation prediction processing on the pulsatile signal and the non-pulsatile signal of the first target cardiovascular signal using a preset regression model to obtain the deviation value includes:

[0016] preprocessing both the pulsating signal and the non-pulsating signal of the first target cardiovascular signal to obtain a first preprocessed pulsating signal and a first preprocessed non-pulsating signal;

[0017] performing amplitude feature extraction processing on the first preprocessed pulsation signal to obtain a first pulsation amplitude;

[0018] performing amplitude feature extraction processing on the first preprocessed non-pulsating signal to obtain a first non-pulsating amplitude;

[0019] The preset regression model is used to perform deviation prediction processing on the first pulsating amplitude and the first non-pulsating amplitude to obtain the deviation value.

[0020] According to one embodiment of the present application, the step of performing deviation prediction processing on the first pulsating amplitude and the first non-pulsating amplitude using the preset regression model to obtain the deviation value includes:

[0021] performing amplitude feature extraction processing on the second target cardiovascular signal to obtain a second pulsating amplitude and a second non-pulsating amplitude;

[0022] calculating a first pulsation ratio based on the first pulsation amplitude and the second pulsation amplitude;

[0023] calculating a first non-pulsating ratio based on the first non-pulsating amplitude and the second non-pulsating amplitude;

[0024] The preset regression model is used to perform deviation prediction processing on the first pulsation ratio and the first non-pulsation ratio to obtain the deviation value.

[0025] According to one embodiment of the present application, the step of performing deviation prediction processing on the first pulsation ratio and the first non-pulsation ratio using the preset regression model to obtain the deviation value includes:

[0026] Using the pulse ratio regression model of the preset regression model to predict the first pulse ratio to obtain an initial blood oxygen saturation;

[0027] Performing reverse reasoning on the initial blood oxygen saturation using the first non-pulsating ratio regression model of the preset regression model to obtain a standard non-pulsating ratio;

[0028] A difference calculation process is performed on the standard non-pulsation ratio and the first non-pulsation ratio to obtain the deviation value.

[0029] According to one embodiment of the present application, the step of calculating the blood oxygen saturation of the second target cardiovascular signal based on the deviation value to obtain the target blood oxygen saturation of the user includes:

[0030] calculating a second non-pulsatile ratio of a third target cardiovascular signal, wherein the third target cardiovascular signal is any one of all cardiovascular signals except the first target cardiovascular signal and the second target cardiovascular signal;

[0031] summing the second non-pulsating ratio and the deviation value to obtain a calibrated non-pulsating ratio;

[0032] The second non-pulsating ratio regression model of the preset regression model is used to perform prediction processing on the calibrated non-pulsating ratio to obtain the target blood oxygen saturation corresponding to the third target cardiovascular signal.

[0033] A second aspect of the embodiments of the present application provides a blood oxygen saturation detection device based on pulse signal calibration, comprising:

[0034] A facial video acquisition module is used to acquire the user's facial video;

[0035] a signal extraction module, configured to extract a plurality of cardiovascular signals from the facial video based on a time sequence;

[0036] a deviation prediction module, configured to perform deviation prediction processing on a pulsatile signal and a non-pulsatile signal of a first target cardiovascular signal using a preset regression model to obtain a deviation value, wherein the first target cardiovascular signal is any one of the plurality of cardiovascular signals;

[0037] A target blood oxygen saturation calculation module is used to calculate the blood oxygen saturation of a second target cardiovascular signal based on the deviation value to obtain the target blood oxygen saturation of the user, where the second target cardiovascular signal is any one of all cardiovascular signals except the first target cardiovascular signal.

[0038] A third aspect of an embodiment of the present application provides a terminal device, comprising: a processor and a memory, the memory being used to store a computer program, the processor being used to call and run the computer program stored in the memory, and executing the steps of the blood oxygen saturation detection method based on pulse signal calibration provided in the first aspect of the embodiment of the present application.

[0039] In a fourth aspect of an embodiment of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium is used to store a computer program, and the computer program enables a computer to execute the steps of the blood oxygen saturation detection method based on pulse signal calibration provided in the first aspect of the embodiment of the present application.

[0040] The beneficial effects of the embodiments of the present application include:

[0041] The present embodiment combines the pulsatile and non-pulsatile signals of the same cardiovascular signal to calculate the deviation, and then uses the non-pulsatile signals and the deviation from the remaining cardiovascular signals to calculate the final blood oxygen saturation, thereby achieving the purpose of improving the detection accuracy of blood oxygen saturation. Specifically, the present embodiment obtains a user's facial video; extracts multiple cardiovascular signals from the facial video based on time sequence; uses a preset regression model to predict the deviation of the pulsatile and non-pulsatile signals of a first target cardiovascular signal to obtain a deviation value, wherein the first target cardiovascular signal is any one of the multiple cardiovascular signals; and calculates the blood oxygen saturation of a second target cardiovascular signal based on the deviation value to obtain the user's target blood oxygen saturation, wherein the second target cardiovascular signal is any one of the cardiovascular signals other than the first target cardiovascular signal. Therefore, compared with the prior art, the present embodiment can detect blood oxygen saturation without the need for additional pressure detection values, thereby simplifying the blood oxygen saturation detection steps, reducing the impact of additional data on blood oxygen saturation, and improving blood oxygen saturation detection accuracy.

[0042] In addition, the embodiments of the present application use time-division multiplexing to illuminate the user's face with light of multiple wavelengths to capture facial video. This time-division multiplexing technology not only reduces hardware complexity and equipment costs, but also improves signal quality and enhances the stability and anti-interference capabilities of blood oxygen detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the conventional technology, the following briefly introduces the drawings required for use in the embodiments or the conventional technology descriptions. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0044] Figure 1 A flowchart of a blood oxygen saturation detection method based on pulse signal calibration in some embodiments of the present application;

[0045] Figure 2 Schematic diagram of the flow of the blood oxygen saturation detection method based on pulse signal calibration in some embodiments of the present application;

[0046] Figure 3 This is a principle block diagram of a blood oxygen saturation detection device based on pulse signal calibration in some embodiments of the present application;

[0047] Figure 4 This is a principle block diagram of the terminal device of the present application in some embodiments. DETAILED DESCRIPTION

[0048] To make the above-mentioned objects, features, and advantages of the present application more clearly understood, the specific embodiments of the present application are described in detail below with reference to the accompanying drawings. The following description sets forth many specific details to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways than those described herein, and those skilled in the art can make similar improvements without violating the scope of the present application. Therefore, the present application is not limited to the specific embodiments disclosed below.

[0049] It should be noted that the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of such features. In the description of this application, "plurality" means at least two, for example, two, three, etc., unless otherwise specifically defined.

[0050] The terms "exemplary" or "for example" are used to indicate an example, illustration, or description. Any embodiment or design described as "exemplary" or "for example" in the embodiments of this application should not be construed as being preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0051] The terms "comprises," "comprising," or any other variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, product, or apparatus.

[0052] Blood oxygen saturation (SpO2) is a key physiological indicator for evaluating lung function and respiratory status. It is widely used in clinical detection of respiratory diseases, sleep quality assessment, and diagnosis and management of health problems such as chronic obstructive pulmonary disease.

[0053] Unless otherwise defined, all technical and scientific terms used in the specification of this application have the same meaning as commonly understood by those skilled in the art to which this application belongs. The terms used in the specification of this application are only for the purpose of describing specific embodiments and are not intended to limit this application. The term "and / or" used in the specification of this application includes any and all combinations of one or more of the relevant listed items.

[0054] The specific implementation of this application is further described below with reference to the accompanying drawings.

[0055] refer to Figure 1 FIG. 1 is a flow chart of a blood oxygen saturation detection method based on pulse signal calibration provided in the first aspect of the embodiment of the present application. Figure 1 In the method, the blood oxygen saturation detection method based on pulse signal calibration includes:

[0056] S1. Obtain the user's facial video.

[0057] Furthermore, step S1 includes:

[0058] S11. Set a variety of light beams with different wavelengths.

[0059] In the embodiments of the present application, the plurality of light beams of different wavelengths are 750 nm (nm stands for nanometers), 940 nm, and 800 nm. In other embodiments, the plurality of light beams of different wavelengths may also be other wavelengths, which may be determined by those skilled in the art based on actual needs.

[0060] S12. Collect the facial video when the light of the multiple different wavelengths illuminates the user's face in a time division multiplexing manner.

[0061] Specifically, the embodiment of the present application sequentially irradiates the user's face with the multiple light wavelengths in a timed sequence. For example, the embodiment of the present application repeatedly irradiates the user's face with 750nm light, 940nm light, and 800nm ​​light in that order. The exposure time for each wavelength is 13 milliseconds.

[0062] The camera used to capture the facial video had a capture frequency of 75 Hz and an exposure time of 8 milliseconds, which is shorter than the illumination time for each wavelength. To ensure that each frame in the facial video is precisely aligned with the corresponding light, a 2 millisecond phase delay is introduced between the camera trigger signal and the light activation signal to avoid brightness fluctuations during the initial illumination period.

[0063] The camera is equipped with an adjustable filter that blocks visible light, ensuring that the light entering the camera is above 700nm. When capturing facial video, the camera is placed 30cm away from the user's face to obtain a better quality video.

[0064] The above implementation not only reduces the complexity of hardware equipment but also improves signal quality.

[0065] S2. Extracting multiple cardiovascular signals from the facial video based on time sequence.

[0066] Furthermore, step S2 includes:

[0067] S21. Divide the facial video based on a time sequence and a plurality of different wavelength information to obtain a plurality of sub-videos, wherein each sub-video corresponds to a type of wavelength information.

[0068] Exemplarily, the first sub-video corresponds to 750nm, that is, the first sub-video is obtained by irradiating the face with 750nm light; the second sub-video corresponds to 940nm, that is, the second sub-video is obtained by irradiating the face with 940nm light; the third sub-video corresponds to 800nm, that is, the third sub-video is obtained by irradiating the face with 800nm ​​light; the fourth sub-video corresponds to 750nm, that is, the fourth sub-video is obtained by irradiating the face with 750nm light; the fifth sub-video corresponds to 940nm, that is, the fifth sub-video is obtained by irradiating the face with 940nm light; and so on, each sub-video corresponds to a wavelength information.

[0069] S22 , performing cardiovascular pulse signal extraction processing on each sub-video to obtain a cardiovascular signal corresponding to each sub-video.

[0070] When each sub-video corresponds to a type of wavelength information, each cardiovascular signal also corresponds to a type of wavelength information.

[0071] Furthermore, step S22 includes:

[0072] S221 . Perform spatial downsampling processing on each frame image of each sub-video to obtain a downsampled sub-video corresponding to each sub-video.

[0073] Specifically, each frame image of each sub-video is reduced by a factor of 10 in each dimension to obtain a downsampled sub-video corresponding to each sub-video. In other embodiments, the reduction factor can be set by those skilled in the art according to actual needs.

[0074] The implementation of step S221 can improve the computing efficiency of the embodiment of the present application.

[0075] S222: Extract the corresponding original signal from each downsampled sub-video.

[0076] Extracting the original signal from the downsampled sub-video can be achieved by referring to the existing technology, and this application will not go into details.

[0077] S223 . Convert each original signal into the frequency domain to obtain a frequency domain signal corresponding to each original signal.

[0078] S224 , performing filtering processing on each frequency domain signal using a bandpass filter to obtain an original pulsating component corresponding to each frequency domain signal.

[0079] In other implementations, other filters may be used to replace the bandpass filter, and may be specifically configured by those skilled in the art according to actual needs.

[0080] The original pulsating component is the original AC (alternating current) component.

[0081] S225. Filter each frequency domain signal using a low-pass filter to obtain an original non-pulsating component corresponding to each frequency domain signal. In other embodiments, other filters may be used to replace the low-pass filter, and the specific setting can be made by those skilled in the art according to actual needs.

[0082] The original non-pulsating component is an original DC (direct current) component.

[0083] S226. Calculate the ratio of the original pulsating component to the original non-pulsating component corresponding to each frequency domain signal, and normalize the ratio of the original pulsating component to the original non-pulsating component corresponding to each frequency domain signal to obtain a characteristic signal corresponding to each original signal.

[0084] S227. Divide the target characteristic signal into multiple frequency intervals, calculate the energy of each frequency interval, and use the frequency interval whose energy is within a preset range as the main frequency band of the target characteristic signal, where the target characteristic signal is one of all the characteristic signals.

[0085] The preset range is 0.6–3 Hz.

[0086] S228: Divide the target characteristic signal into local frequency bands with the main frequency band as the center, calculate the ratio of the energy of the local frequency band to the total energy of the target characteristic signal, and use the ratio as the signal-to-noise ratio of the sub-video corresponding to the target characteristic signal. Figure 2 The Signal-to-Noise Ratio Plot step.

[0087] S229: sorting the signal-to-noise ratios of all sub-videos, and selecting the sub-videos whose signal-to-noise ratios meet a preset condition as candidate regions.

[0088] The preset condition is that the signal-to-noise ratio ranks in the top 30%. In other embodiments, the preset condition may also be that the signal-to-noise ratio ranks in the bottom 30%, which can be specifically set by those skilled in the art according to actual needs.

[0089] The sorting process is from large to small. In other embodiments, it can be sorted from small to large. The specific sorting method can be set by those skilled in the art according to actual needs.

[0090] S2210: Select the connected region with the largest area from the candidate regions as the target region of interest. Figure 2 Target Region of Interest step in .

[0091] The target region of interest is the final region of interest. Obtaining the target region of interest can improve the stability of cardiovascular signals.

[0092] S2211 , performing cardiovascular pulse signal extraction processing on the target region of interest on each frame image in each sub-video to obtain a cardiovascular signal corresponding to each sub-video.

[0093] The implementation of steps S221 to S2211 utilizes the original pulsating component to realize automatic selection of the user's skin area. On the basis of ensuring the accuracy of the skin area, it further combines the advantages of pulsating signals and non-pulsating signals to complete blood oxygen detection. Among them, the pulsating signal is used to accurately predict the static blood oxygen level, while the non-pulsating signal is used to track the blood oxygen change trend, forming a two-stage prediction strategy.

[0094] S3. Use a preset regression model to perform deviation prediction processing on the pulsating signal and the non-pulsating signal of the first target cardiovascular signal to obtain a deviation value, wherein the first target cardiovascular signal is any one of the multiple cardiovascular signals.

[0095] The preset regression model includes a pulsating ratio regression model, a first non-pulsating ratio regression model and a second non-pulsating ratio regression model, wherein the pulsating ratio regression model is used to predict the initial blood oxygen saturation using the first pulsating ratio; the first non-pulsating ratio regression model is used to infer the standard non-pulsating ratio through the initial blood oxygen saturation; and the second non-pulsating ratio regression model is used to predict the target blood oxygen saturation using the calibrated non-pulsating ratio.

[0096] The first target cardiovascular signal may be the first signal among all cardiovascular signals. In other embodiments, the first target cardiovascular signal may be any one of all cardiovascular signals, which may be specifically set by those skilled in the art according to actual needs.

[0097] The pulsating signal described in the embodiment of the present application is an AC (alternating current) signal, and the non-pulsating signal is a DC (direct current) signal.

[0098] Non-pulsatile signals are more sensitive to changes in blood oxygen, while pulsatile signals provide a stable baseline value. The combination of the two can effectively reduce the impact of environmental interference and individual physiological differences on blood oxygen saturation detection results.

[0099] Furthermore, the step of performing deviation prediction processing on the pulsatile signal and the non-pulsatile signal of the first target cardiovascular signal using a preset regression model to obtain a deviation value includes:

[0100] S31 . Preprocess both the pulsating signal and the non-pulsating signal of the first target cardiovascular signal to obtain a first preprocessed pulsating signal and a first preprocessed non-pulsating signal.

[0101] Specifically, the embodiment of the present application uses a first filter to filter the pulsating signal of the first target cardiovascular signal to obtain the first pre-processed pulsating signal; and uses a second filter to filter the non-pulsating signal of the first target cardiovascular signal to obtain the first pre-processed non-pulsating signal. The first filter can refer to Figure 2 The Bandpass Filter step in .

[0102] The first filter is a fourth-order Butterworth bandpass filter. The frequency range of the fourth-order Butterworth bandpass filter is [0.6, 3] Hz, and is used to remove frequency components unrelated to heart rate. In other embodiments, the first filter can also be other filters, which can be specifically set by those skilled in the art according to actual needs.

[0103] The second filter is a fourth-order Butterworth low-pass filter. The cutoff frequency of the fourth-order Butterworth low-pass filter is 0.6 Hz, and is used to remove high-frequency noise of the non-pulsating signal of the first target cardiovascular signal. In other embodiments, the second filter can also be other filters, which can be set by those skilled in the art according to actual needs. The second filter can refer to Figure 2 The "Low-pass filter" step in

[0104] In some embodiments, the step of preprocessing the pulsation signal of the first target cardiovascular signal to obtain a first preprocessed pulsation signal includes:

[0105] S311: Filter the pulsation signal of the first target cardiovascular signal to obtain a first filtered signal. In step S311, the first filter is used to perform filtering.

[0106] S312: Perform normalization processing on the first filtered signal to obtain the first pre-processed pulsating signal. The normalization processing can reduce the influence of light intensity variation in the first filtered signal.

[0107] S32: Perform amplitude feature extraction processing on the first preprocessed pulsation signal to obtain a first pulsation amplitude.

[0108] Specifically, the standard deviation of the first pre-processed pulsation signal is calculated as the first pulsation amplitude. Figure 2 The "Standard Deviation" and "Pulse Amplitude" steps in the .

[0109] The first pulsation amplitude is the pulsation amplitude corresponding to the first target cardiovascular signal.

[0110] S33. Perform amplitude feature extraction processing on the first preprocessed non-pulsating signal to obtain a first non-pulsating amplitude.

[0111] Specifically, the mean value of the first pre-processed non-pulsating signal is used as the first non-pulsating amplitude. Figure 2 The "Mean" and "Non-Pulsatile Amplitude" steps in the .

[0112] The first non-pulsating amplitude is the non-pulsating amplitude corresponding to the first target cardiovascular signal.

[0113] S34. Use the preset regression model to perform deviation prediction processing on the first pulsating amplitude and the first non-pulsating amplitude to obtain the deviation value.

[0114] Furthermore, step S34 includes:

[0115] S341 . Perform amplitude feature extraction processing on the second target cardiovascular signal to obtain a second pulsating amplitude and a second non-pulsating amplitude.

[0116] The second target cardiovascular signal may be the last signal in all cardiovascular signals that is in the same time period as the first target cardiovascular signal. In other embodiments, the selection of the second target cardiovascular signal may be set by those skilled in the art according to actual needs.

[0117] Furthermore, step S341 includes:

[0118] S3411. Preprocess both the pulsating signal and the non-pulsating signal of the second target cardiovascular signal to obtain a second preprocessed pulsating signal and a second preprocessed non-pulsating signal.

[0119] Specifically, the embodiment of the present application uses the first filter to filter the pulsating signal of the second target cardiovascular signal to obtain the second preprocessed pulsating signal; and uses the second filter to filter the non-pulsating signal of the second target cardiovascular signal to obtain the second preprocessed non-pulsating signal.

[0120] In some embodiments, the step of preprocessing the pulsation signal of the second target cardiovascular signal to obtain a second preprocessed pulsation signal includes:

[0121] S34111: Filter the pulsation signal of the second target cardiovascular signal to obtain a second filtered signal. In step S34111, the first filter is used to perform filtering.

[0122] S34112: Perform normalization processing on the second filtered signal to obtain the second pre-processed pulsating signal. The normalization processing can reduce the influence of light intensity variation in the second filtered signal.

[0123] S3412: Perform amplitude feature extraction processing on the second preprocessed pulsation signal to obtain a second pulsation amplitude.

[0124] Specifically, the standard deviation of the second preprocessed pulsation signal is calculated as the second pulsation amplitude.

[0125] S3413. Perform amplitude feature extraction processing on the second preprocessed non-pulsating signal to obtain a second non-pulsating amplitude.

[0126] Specifically, the mean value of the second preprocessed non-pulsating signal is used as the second non-pulsating amplitude.

[0127] S342. Calculate a first pulsation ratio based on the first pulsation amplitude and the second pulsation amplitude.

[0128] Specifically, the ratio of the first pulsation amplitude to the second pulsation amplitude is calculated to obtain the first pulsation ratio.

[0129] The first pulsation ratio is a relative amplitude ratio calculated for different wavelength combinations, that is, the first pulsation amplitude and the second pulsation amplitude correspond to different wavelengths. For example, the first pulsation amplitude corresponds to 750nm and the second pulsation amplitude corresponds to 940nm; for another example, the first pulsation amplitude corresponds to 750nm and the second pulsation amplitude corresponds to 800nm; for another example, the first pulsation amplitude corresponds to 940nm and the second pulsation amplitude corresponds to 800nm. Figure 2 Pulse Ratio section.

[0130] S343. Calculate a first non-pulsation ratio based on the first non-pulsation amplitude and the second non-pulsation amplitude.

[0131] Specifically, the ratio of the first non-pulsating amplitude to the second non-pulsating amplitude is calculated to obtain the first non-pulsating ratio.

[0132] The first non-pulsating ratio is a relative amplitude ratio calculated for different wavelength combinations, that is, the first non-pulsating amplitude and the second non-pulsating amplitude correspond to different wavelengths. For example, the first non-pulsating amplitude corresponds to 750nm and the second non-pulsating amplitude corresponds to 940nm; for another example, the first non-pulsating amplitude corresponds to 750nm and the second non-pulsating amplitude corresponds to 800nm; for another example, the first non-pulsating amplitude corresponds to 940nm and the second non-pulsating amplitude corresponds to 800nm. Figure 2 Non-Pulsating Ratio section.

[0133] S344: Use the preset regression model to perform deviation prediction processing on the first pulsation ratio and the first non-pulsation ratio to obtain the deviation value.

[0134] Furthermore, step S344 includes:

[0135] S3441: Use the pulse ratio regression model of the preset regression model to predict the first pulse ratio to obtain the initial blood oxygen saturation. Figure 2 The "Pulse Ratio", "Pulse Ratio Regression Model", and "Initial Blood Oxygen Saturation" steps in the .

[0136] The prediction of the initial blood oxygen saturation can improve the accuracy and stability of the target blood oxygen saturation prediction. Based on the initial blood oxygen saturation, the high sensitivity of the non-pulsatile signal is used to track the blood oxygen change trend, thereby more accurately predicting the dynamic changes of blood oxygen.

[0137] Furthermore, the calculation formula of the pulsation ratio regression model is:

[0138] S1=108.9985-19.6983×RoR-AC;

[0139] Wherein, S1 is the initial blood oxygen saturation; RoR-AC is the first pulsation ratio.

[0140] S3442. Use the first non-pulsating ratio regression model of the preset regression model to perform reverse reasoning on the initial blood oxygen saturation to obtain a standard non-pulsating ratio.

[0141] Furthermore, the calculation formula of the first non-pulsating ratio regression model is:

[0142] S1=-557.5914+654.1909×RoR-DC1;

[0143] Wherein, S1 is the initial blood oxygen saturation; RoR-DC1 is the standard non-pulsatile ratio.

[0144] S3443. Perform difference calculation on the standard non-pulsation ratio and the first non-pulsation ratio to obtain the deviation value.

[0145] S4. Calculate the blood oxygen saturation of a second target cardiovascular signal based on the deviation value to obtain the target blood oxygen saturation of the user, where the second target cardiovascular signal is any one of all cardiovascular signals except the first target cardiovascular signal.

[0146] Furthermore, the step of calculating the blood oxygen saturation of the second target cardiovascular signal based on the deviation value to obtain the target blood oxygen saturation of the user includes:

[0147] S41. Calculate a second non-pulsatile ratio of a third target cardiovascular signal, where the third target cardiovascular signal is any one of all cardiovascular signals except the first target cardiovascular signal and the second target cardiovascular signal.

[0148] When the multiple light wavelengths are two different wavelengths, the third target cardiovascular signal is any one signal that is in a different time period than the first target cardiovascular signal. When the multiple light wavelengths are three or more different wavelengths, the third target cardiovascular signal can be a signal that is in the same time period as the first target cardiovascular signal among all cardiovascular signals and is the next signal after the first target cardiovascular signal. The third target cardiovascular signal can also be any one signal that is in a different time period than the first target cardiovascular signal among all cardiovascular signals. In other embodiments, the selection of the third target cardiovascular signal can be set by those skilled in the art based on actual needs.

[0149] Furthermore, step S41 includes:

[0150] S411 . Filter the non-pulsating signal of the third target cardiovascular signal to obtain a third pre-processed non-pulsating signal.

[0151] S412: Taking the mean value of the third preprocessed non-pulsating signal as the third non-pulsating amplitude.

[0152] S413: Calculate a second non-pulsation ratio based on the third non-pulsation amplitude and the second non-pulsation amplitude.

[0153] Specifically, the ratio of the third non-pulsating amplitude to the second non-pulsating amplitude is calculated to obtain the second non-pulsating ratio.

[0154] S42, summing the second non-pulsation ratio and the deviation value to obtain a calibrated non-pulsation ratio. Figure 2 Refer to the "Calibrating the Non-Pulsating Ratio" procedure in the .

[0155] The calibrated non-pulsation ratio is the calibrated non-pulsation ratio.

[0156] S43, using the second non-pulsating ratio regression model of the preset regression model to predict the calibrated non-pulsating ratio, and obtain the target blood oxygen saturation corresponding to the third target cardiovascular signal. Figure 2 Target Oxygen Saturation step in

[15] .

[0157] The calculation formula of the second non-pulsating ratio regression model is:

[0158] S2=-557.5914+654.1909×RoR-DC2;

[0159] Wherein, S2 is the target blood oxygen saturation; RoR-DC2 is the calibrated non-pulsatile ratio.

[0160] The present embodiment combines the pulsatile and non-pulsatile signals of the same cardiovascular signal to calculate the deviation, and then uses the non-pulsatile signals and the deviation in the remaining cardiovascular signals to calculate the final blood oxygen saturation, thereby achieving the purpose of improving the detection accuracy of blood oxygen saturation. Therefore, compared with the prior art, the present embodiment can detect blood oxygen saturation without the need for additional pressure detection values, thereby simplifying the blood oxygen saturation detection steps, reducing the impact of additional data on blood oxygen saturation, and improving the detection accuracy of blood oxygen saturation.

[0161] refer to Figure 3 FIG. 1 is a block diagram showing the principle of a blood oxygen saturation detection device based on pulse signal calibration according to the second aspect of the present application. Figure 3 In the embodiment, the blood oxygen saturation detection device 100 based on pulse signal calibration includes:

[0162] The facial video acquisition module 101 is used to acquire the user's facial video;

[0163] a signal extraction module 102 for extracting a plurality of cardiovascular signals from the facial video based on a time sequence;

[0164] a deviation prediction module 103 for performing deviation prediction processing on the pulsatile signal and the non-pulsatile signal of a first target cardiovascular signal using a preset regression model to obtain a deviation value, wherein the first target cardiovascular signal is any one of the multiple cardiovascular signals;

[0165] The target blood oxygen saturation calculation module 104 is used to calculate the blood oxygen saturation of a second target cardiovascular signal based on the deviation value to obtain the target blood oxygen saturation of the user, where the second target cardiovascular signal is any one of all cardiovascular signals except the first target cardiovascular signal.

[0166] The third aspect of the embodiment of the present application provides a terminal device, the principle block diagram of the terminal device can be as follows: Figure 4As shown. The terminal device includes a processor, a memory, a network interface, a display screen and a temperature sensor connected via a system bus. The processor is used to provide computing and control capabilities. The memory 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 operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the terminal device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a blood oxygen saturation detection method based on pulse signal calibration is implemented. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the temperature sensor is pre-set inside the terminal device to detect the operating temperature of the internal device.

[0167] Those skilled in the art will understand that Figure 4 The principle block diagram shown in the figure is only a block diagram of a partial structure related to the solution of the present invention, and does not constitute a limitation on the terminal device to which the solution of the present invention is applied. The specific terminal device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0168] In some embodiments, an embodiment of the present application provides a terminal device, comprising a processor and a memory, the memory being configured to store a computer program, the processor being configured to call and execute the computer program stored in the memory to perform the steps of the blood oxygen saturation detection method based on pulsating signal calibration provided in the first aspect of the embodiment of the present application. A fourth aspect of the embodiment of the present application provides a computer-readable storage medium, the computer-readable storage medium being configured to store a computer program that causes a computer to perform the steps of the blood oxygen saturation detection method based on pulsating signal calibration provided in the first aspect of the embodiment of the present application.

[0169] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the above-described method embodiments. Any reference to memory, storage, database, or other media used in the various embodiments provided herein may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct RAMbus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM).

[0170] The technical features of the above embodiments can be combined without changing the basic principles of this application. In order to make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0171] The above embodiments merely illustrate several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent application. It should be noted that a person skilled in the art could make various modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. Therefore, the scope of patent protection for the present application shall be determined by the appended claims.

Claims

1. A blood oxygen saturation detection method based on pulse signal calibration, characterized in that: include: Get the user's face video; extracting a plurality of cardiovascular signals from the facial video based on a time sequence; performing deviation prediction processing on a pulsatile signal and a non-pulsatile signal of a first target cardiovascular signal using a preset regression model to obtain a deviation value, wherein the first target cardiovascular signal is any one of the multiple cardiovascular signals; calculating the blood oxygen saturation of a second target cardiovascular signal based on the deviation value to obtain a target blood oxygen saturation of the user, wherein the second target cardiovascular signal is any one of all cardiovascular signals except the first target cardiovascular signal; The step of obtaining a user's facial video includes: setting a plurality of light beams of different wavelengths; and collecting the facial video when the plurality of light beams of different wavelengths illuminate the user's face in a time division multiplexing manner; The step of extracting multiple cardiovascular signals from the facial video based on the time sequence includes: dividing the facial video based on the time sequence and multiple different wavelength information to obtain multiple sub-videos, wherein each sub-video corresponds to a type of wavelength information; performing cardiovascular pulse signal extraction processing on each sub-video to obtain a cardiovascular signal corresponding to each sub-video; The steps of performing deviation prediction processing on the pulsating signal and the non-pulsating signal of the first target cardiovascular signal using a preset regression model to obtain a deviation value include: preprocessing both the pulsating signal and the non-pulsating signal of the first target cardiovascular signal to obtain a first preprocessed pulsating signal and a first preprocessed non-pulsating signal; performing amplitude feature extraction processing on the first preprocessed pulsating signal to obtain a first pulsating amplitude; performing amplitude feature extraction processing on the first preprocessed non-pulsating signal to obtain a first non-pulsating amplitude; and performing deviation prediction processing on the first pulsating amplitude and the first non-pulsating amplitude using the preset regression model to obtain the deviation value; The preset regression model is used to perform deviation prediction processing on the first pulsating amplitude and the first non-pulsating amplitude, and the step of obtaining the deviation value includes: performing amplitude feature extraction processing on the second target cardiovascular signal to obtain a second pulsating amplitude and a second non-pulsating amplitude; calculating a first pulsating ratio based on the first pulsating amplitude and the second pulsating amplitude; calculating a first non-pulsating ratio based on the first non-pulsating amplitude and the second non-pulsating amplitude; and using the preset regression model to perform deviation prediction processing on the first pulsating ratio and the first non-pulsating ratio to obtain the deviation value.

2. The blood oxygen saturation detection method based on pulse signal calibration according to claim 1, characterized in that: The step of performing deviation prediction processing on the first pulsation ratio and the first non-pulsation ratio using the preset regression model to obtain the deviation value includes: Using the pulse ratio regression model of the preset regression model to predict the first pulse ratio to obtain an initial blood oxygen saturation; Performing reverse reasoning on the initial blood oxygen saturation using the first non-pulsating ratio regression model of the preset regression model to obtain a standard non-pulsating ratio; A difference calculation process is performed on the standard non-pulsation ratio and the first non-pulsation ratio to obtain the deviation value.

3. The blood oxygen saturation detection method based on pulse signal calibration according to claim 1, characterized in that: The step of calculating the blood oxygen saturation of the second target cardiovascular signal based on the deviation value to obtain the target blood oxygen saturation of the user includes: calculating a second non-pulsatile ratio of a third target cardiovascular signal, wherein the third target cardiovascular signal is any one of all cardiovascular signals except the first target cardiovascular signal and the second target cardiovascular signal; summing the second non-pulsating ratio and the deviation value to obtain a calibrated non-pulsating ratio; The second non-pulsating ratio regression model of the preset regression model is used to perform prediction processing on the calibrated non-pulsating ratio to obtain the target blood oxygen saturation corresponding to the third target cardiovascular signal.

4. A blood oxygen saturation detection method and device based on pulse signal calibration, characterized in that: include: A facial video acquisition module is used to acquire the user's facial video; a signal extraction module, configured to extract a plurality of cardiovascular signals from the facial video based on a time sequence; a deviation prediction module, configured to perform deviation prediction processing on a pulsatile signal and a non-pulsatile signal of a first target cardiovascular signal using a preset regression model to obtain a deviation value, wherein the first target cardiovascular signal is any one of the plurality of cardiovascular signals; a target blood oxygen saturation calculation module, configured to calculate the blood oxygen saturation of a second target cardiovascular signal based on the deviation value to obtain the target blood oxygen saturation of the user, wherein the second target cardiovascular signal is any one of all cardiovascular signals except the first target cardiovascular signal; The facial video acquisition module is further configured to set a plurality of light beams of different wavelengths; and to capture the facial video when the plurality of light beams of different wavelengths are irradiated onto the user's face in a time division multiplexing manner; The signal extraction module is further configured to divide the facial video into multiple sub-videos based on the time sequence and the multiple different wavelength information to obtain a plurality of sub-videos, wherein each sub-video corresponds to a type of wavelength information; and perform cardiovascular pulse signal extraction processing on each sub-video to obtain a cardiovascular signal corresponding to each sub-video; The deviation prediction module is further configured to preprocess both the pulsating signal and the non-pulsating signal of the first target cardiovascular signal to obtain a first preprocessed pulsating signal and a first preprocessed non-pulsating signal; perform amplitude feature extraction processing on the first preprocessed pulsating signal to obtain a first pulsating amplitude; perform amplitude feature extraction processing on the first preprocessed non-pulsating signal to obtain a first non-pulsating amplitude; and perform deviation prediction processing on the first pulsating amplitude and the first non-pulsating amplitude using the preset regression model to obtain the deviation value; The deviation prediction module is also used to perform amplitude feature extraction processing on the second target cardiovascular signal to obtain a second pulsating amplitude and a second non-pulsating amplitude; calculate a first pulsating ratio based on the first pulsating amplitude and the second pulsating amplitude; calculate a first non-pulsating ratio based on the first non-pulsating amplitude and the second non-pulsating amplitude; and use the preset regression model to perform deviation prediction processing on the first pulsating ratio and the first non-pulsating ratio to obtain the deviation value.

5. A terminal device, characterized in that: include: A processor and a memory, the memory being used to store a computer program, the processor being used to call and run the computer program stored in the memory to execute the steps of the blood oxygen saturation detection method based on pulse signal calibration as described in any one of claims 1 to 3.

6. A computer-readable storage medium, characterized in that Used to store a computer program, wherein the computer program enables a computer to execute the steps of the blood oxygen saturation detection method based on pulse signal calibration as described in any one of claims 1 to 3.

Images