Sign sensing band signal correction method based on pressure signal and intelligent bed

Abstract

The invention belongs to the field of smart home, and discloses a sign sensing band signal correction method based on a pressure signal and a smart bed, and the method comprises the following steps: obtaining a first relationship between pressure and angle based on historical pressure data and historical angle data; acquiring a second relationship between the pressure and the original signal quality of the sign sensing belt based on the historical pressure data and the historical original signal quality data of the sign sensing belt; acquiring a third relationship between the angle and the original signal quality of the sign sensing band based on the historical angle data and the historical original signal quality data of the sign sensing band; constructing a signal quality correction model according to the first relation, the second relation and the third relation; according to the real-time pressure signal, the real-time angle signal and the signal quality correction model, physical sign sensing band signal correction is executed.

Description

Technical Field

[0001] This invention belongs to the field of smart home, and particularly relates to a method for correcting vital sign sensors based on pressure signals and a smart bed. Background Technology

[0002] A vital signs sensing band is a wearable or deployable flexible sensing device that integrates various sensors (such as piezoelectric sensors, capacitive sensors, strain gauges, etc.) to achieve real-time acquisition and monitoring of human physiological signals. Its core principle is to utilize the physical signals (such as pressure changes, bioelectric signals, mechanical vibrations, etc.) generated when the sensor comes into contact with the human body, and then convert them into quantifiable vital signs parameters through signal processing algorithms.

[0003] In the field of smart homes, vital sign sensor bands are usually used with adjustable electric bed frames or electric mattresses. However, in the current technology, when the user is not lying flat, the vital sign sensor band usually has poor signal quality, resulting in incorrect vital sign data output.

[0004] Existing technologies typically address poor sensor signal quality by addressing both hardware and software aspects, but both have certain shortcomings.

[0005] Hardware-level signal enhancement technologies typically include using low-noise piezoelectric films or capacitive sensors to reduce background noise, or using hydrogel electrodes or dry electrodes (such as silver-plated fabric) combined with elastic supports to ensure stable skin contact area, or increasing the number of vital signs sensors and performing fusion compensation through multimodal data. These methods involve increasing the number of sensors or optimizing the hardware, which not only has a higher cost but also affects the comfort experience during use. Software-level signal enhancement techniques include real-time filtering of motion artifacts through adaptive filtering, detection of sudden noise segments (such as signal distortion during body movement) using sliding windows, marking them as invalid data or triggering a resampling mechanism, or using CNN or LSTM networks to train models based on historical high-quality signals to perform end-to-end mapping and repair of real-time noise signals. These methods are highly dependent on the diversity of models and computing power, and prolong the process of outputting vital signs data, which can easily lead to lag.

[0006] In existing technologies, signal correction algorithms are essentially passive adjustments driven by signals. They rely on the statistical characteristics of the signal itself for processing and are difficult to cope with signal quality fluctuations caused by changes in the hardware installation environment (such as mattress pressure and bed frame angle), resulting in low detection accuracy and insufficient applicability in smart sleep scenarios. Summary of the Invention

[0007] This application addresses the problems existing in the prior art by proposing a method for correcting vital sign sensor band signals based on pressure signals and an intelligent bed. By adjusting the sampling rate of the vital sign sensor band in real time, the method enhances the adaptability and accuracy of vital sign output data in non-ideal scenarios (changes in user position / angle on the bed). The method is simple to implement and low in cost.

[0008] To achieve the above objectives, this application provides the following technical solution: a method for correcting vital sign sensor band signals based on pressure signals, comprising the following steps: obtaining a first relationship between pressure and angle based on historical pressure data and historical angle data; obtaining a second relationship between pressure and the original signal quality of the vital sign sensor band based on historical pressure data and historical original signal quality data of the vital sign sensor band; obtaining a third relationship between angle and the original signal quality of the vital sign sensor band based on historical angle data and historical original signal quality data of the vital sign sensor band; constructing a signal quality correction model based on the first, second, and third relationships; and performing vital sign sensor band signal correction based on real-time pressure signals, real-time angle signals, and the signal quality correction model.

[0009] Optionally, obtaining the first relationship between pressure and angle based on historical pressure data and historical angle data includes obtaining the pressure area and total pressure at different angles based on historical pressure data corresponding to historical angle data, and constructing the first relationship.

[0010] Optionally, obtaining a second relationship between pressure and the original signal quality of the vital signs sensor band based on historical pressure data and historical vital signs sensor band original signal quality data includes constructing a second relationship based on the total historical pressure, historical pressure area, and historical vital signs sensor band original signal quality data.

[0011] Optionally, the third relationship between the angle and the original signal quality of the vital signs sensor band can be obtained based on historical angle data and historical vital signs sensor band original signal quality data. This includes calculating the mean square error and standard deviation at the same angle based on the amplitude data in the historical vital signs sensor band original signal quality data, and constructing the third relationship.

[0012] Optionally, constructing a signal quality correction model based on the first, second, and third relationships includes obtaining pressure influence correction coefficients based on the first and second relationships; and constructing a signal quality correction model based on the third relationship and the pressure influence correction coefficients.

[0013] Optionally, it also includes: constructing a sensitivity correction relationship based on historical sampling rates and historical sensitivity; and performing vital sign sensor band signal correction based on real-time pressure signals, real-time angle signals, signal quality correction models, and sensitivity correction relationships.

[0014] Optionally, the vital sign sensor band signal correction is performed based on the real-time pressure signal, real-time angle signal, signal quality correction model, and sensitivity correction relationship. This includes obtaining the current signal quality based on the real-time pressure signal, real-time angle signal, and signal quality correction model; obtaining the sensitivity compensation value based on the current signal quality and a preset signal quality threshold; obtaining the sampling correction value based on the sensitivity compensation value and sensitivity correction relationship; and performing vital sign sensor band signal correction based on the sampling correction value.

[0015] Optionally, constructing a signal quality correction model based on the first, second, and third relationships also includes constructing an angle influence relationship based on the first and second relationships using the least squares method; constructing a pressure influence relationship based on the first and second relationships using the flat reference state and multi-angle contact state; constructing a first signal quality correction model based on the first, second, and third relationships; and constructing a second signal quality correction model based on the angle influence relationship and the third relationship.

[0016] Optionally, the vital sign sensor band signal correction is performed based on the real-time pressure signal, the real-time angle signal, and the signal quality correction model. This includes, if a real-time pressure signal exists, calling the first signal quality correction model to obtain the vital sign sensor band signal correction value based on the real-time pressure signal and the real-time angle signal; if no real-time pressure signal exists, calling the second signal quality correction model to obtain the vital sign sensor band signal correction value based on the real-time angle signal.

[0017] Another technical solution provided in this application is a smart bed, which includes the above-mentioned vital sign sensor band signal correction method based on pressure signals. The smart bed also includes: an electric bed frame for adjusting the angle; and a mattress. The electric bed frame or the mattress is provided with a sensor band and a control device. The control device is electrically connected to the sensor band and can acquire data, analyze and adjust the parameters of each device.

[0018] The beneficial effects of this application are as follows: 1. This application addresses the impact of angle changes on signal quality by altering the pressure (total pressure, pressure area) between the human body and the sensing belt. Therefore, it quantifies the influence weight of pressure on the signal, calculates the pressure correction coefficient, corrects the signal quality based on the correction coefficient, and finally compensates for sensitivity loss by adjusting the sampling rate. If the pressure signal cannot be obtained, the correction coefficient can be indirectly calculated using the angle of the electric bed.

[0019] 2. This application enhances the adaptability and accuracy of vital sign output data in non-ideal scenarios (changes in user position / angle on the bed) by adjusting the sampling rate of the vital sign sensor band in real time. It is simple to implement and low in cost. Attached Figure Description

[0020] Figure 1This is a schematic diagram of a vital sign sensor band signal correction method based on pressure signals according to one embodiment of this application.

[0021] Figure 2 For the purposes of this application Figure 1 The diagram shows a sensitivity correction relationship structure in one embodiment. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description of this application is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely one preferred embodiment of this application and are only used to explain this application. They do not limit the scope of protection of this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0023] Example 1: like Figure 1-2 As shown, a method for correcting vital sign sensor band signals based on pressure signals includes the following steps: The primary relationship between pressure and angle is obtained based on historical pressure data and historical angle data. A second relationship between pressure and the original signal quality of the vital signs sensor band was obtained based on historical pressure data and historical vital signs sensor band raw signal quality data. The third relationship between historical perspective data and historical vital sign sensor raw signal quality data is obtained based on historical perspective data and historical vital sign sensor raw signal quality data. Construct a signal quality correction model based on the first, second, and third relationships; A sensitivity correction relationship is constructed based on historical sampling rates and historical sensitivity. The vital signs sensor band signal is corrected based on real-time pressure signal, real-time angle signal, signal quality correction model, and sensitivity correction relationship.

[0024] In one specific embodiment, obtaining the first relationship between pressure and angle based on historical pressure data and historical angle data includes obtaining the pressure area and total pressure at different angles based on historical pressure data corresponding to historical angle data, and constructing the first relationship. Specifically, after the user lies down, the pressure area within the monitoring range of the vital signs sensor band is read by the pressure sensor in a supine, resting state. Sum of stress ; and collect data from multiple angles: randomly adjust J groups of angles, and record the pressure area and total pressure at different angles i. And randomly sampled data at different angles for a duration of [duration missing]. The original signal, which is the amplitude at different sampling points n. The number of sampling points is determined by the sampling rate. When the sampling rate N is 100Hz, it means there are 100 sampling points per second, corresponding to 100 sets of amplitude datasets. J≥5, and the angle range covers 0° to 60° (the common lying angle of the human body), with the interval between each set of angles not exceeding 15° to ensure uniform data distribution.

[0025] In one specific embodiment, obtaining a second relationship between pressure and the original signal quality of the vital signs sensor band based on historical pressure data and historical vital signs sensor band original signal quality data includes constructing a second relationship based on the total historical pressure, the historical pressure area, and the historical vital signs sensor band original signal quality data.

[0026] The third relationship between the angle and the original signal quality of the vital signs sensor band is obtained based on historical angle data and historical vital signs sensor band original signal quality data. This relationship is constructed by calculating the mean square error and standard deviation at the same angle based on the amplitude data in the historical vital signs sensor band original signal quality data.

[0027] The signal quality correction model is constructed based on the first, second, and third relationships, including obtaining the pressure influence correction coefficient based on the first and second relationships; and constructing the signal quality correction model based on the third relationship and the pressure influence correction coefficient.

[0028] Specifically, in calculating the mean square error and standard deviation at the same angle based on the amplitude data from the original signal quality data of the historical vital sign sensor band, the signal quality of the vital sign sensor band at different angles is calculated. As shown in Formula 1 below: (Formula) (1) in, The amplitude at each sampling point; Let S represent the average amplitude in the i-th data set; S is the pressure area at the i-th angle. This represents the standard deviation of the amplitude in the i-th data set. Therefore, by calculating the ratio of the mean square error to the standard deviation of the signal, the stability of the signal is quantified (the smaller the ratio, the higher the signal quality). The larger (the larger).

[0029] Specifically, the pressure effect correction factor As shown in Formula 2 below: (Formula) (2) in, This is the pressure weighting coefficient; The total pressure in the supine baseline state; denoted as the pressure area in the lying-flat baseline state; F represents the total pressure at the i-th angle; and S represents the pressure area at the i-th angle. This is a comprehensive correction coefficient characterizing the impact of the contact pressure between the chest cavity and the mattress on the quality of the original signal from the vital signs sensor band. When the total pressure increases or the pressure area decreases, the coupling between the chest cavity and the sensor becomes tighter, and the amplitude of the original signal is usually enhanced (the noise ratio decreases). A value >1 improves signal quality; conversely, a decrease in total pressure or an increase in pressure area (such as when the bed frame is raised, causing the chest cavity to be suspended) leads to signal quality degradation. <1, indicating a need to increase sensitivity).

[0030] Specifically, the signal quality correction model calculates the corrected signal quality based on the signal quality at different angles and the pressure effect correction coefficient. This value represents the required signal quality at different angles and is also the target of correction. As shown in Formula 3 below: (Formula) (3) Among them, signal quality Pressure influence coefficient .

[0031] Specifically, pressure weight The determination is shown in Formula 4 below: (Formula) (4) in, This indicates the signal quality when lying flat, and the signal quality at various angles. Pressure effect correction factor ; predict signal quality by calculation With actual signal quality The error is calculated by comparing all candidates in the interval 0-1. To minimize the error, choose the option that minimizes the error. Thus, the sum of pressures (F) and the pressure area (S) at each angle are determined. The weights in the equation.

[0032] In one specific embodiment, performing vital sign sensor band signal correction based on real-time pressure signal, real-time angle signal, signal quality correction model, and sensitivity correction relationship includes obtaining the current signal quality based on the real-time pressure signal, real-time angle signal, and signal quality correction model; obtaining the sensitivity compensation value based on the current signal quality and a preset signal quality threshold; obtaining the sampling correction value based on the sensitivity compensation value and sensitivity correction relationship; and performing vital sign sensor band signal correction using the sampling correction value.

[0033] Sampling rate adjustment (sensitivity compensation): When the user's position changes, it is necessary to compensate by increasing the sensitivity to maintain signal quality. The impact; for example When the angle is raised to 30 degrees, but To maintain a signal quality of 20dB, the sensitivity needs to be increased by 1.5dB. In this invention, the sensitivity is increased by adjusting the sampling rate of the vital signs sensor band. Increasing the sampling rate enhances the ability to capture weak signals. The adjusted sampling rate is... As shown in Formula 5 below: (Formula) (5) Where N is the element sampling rate; The signal quality at the i-th angle; This is the pressure influence correction coefficient for the i-th angle group; Signal quality in a flat reference state; signal quality after calibration. When below the benchmark By increasing the sampling rate Enhance sensitivity (the higher the sampling rate, the stronger the ability to capture weak signals), so that the signal quality can be restored to the reference level.

[0034] In a specific embodiment, constructing the signal quality correction model based on the first, second, and third relationships further includes constructing an angle influence relationship based on the first and second relationships using the least squares method; constructing a pressure influence relationship based on the first and second relationships using the supine reference state and multi-angle contact state; constructing a first signal quality correction model based on the first, second, and third relationships; and constructing a second signal quality correction model based on the angle influence relationship and the third relationship. Specifically, if a real-time pressure signal exists, the first signal quality correction model is invoked to obtain the corrected value of the vital sign sensor band signal based on the real-time pressure signal and the real-time angle signal; if no real-time pressure signal exists, the second signal quality correction model is invoked to obtain the corrected value of the vital sign sensor band signal based on the real-time angle signal.

[0035] In one specific embodiment, when a pressure signal cannot be obtained during actual use, the angle of the electric bed can be read. Calculate the angle influence coefficient and replace the formulas 3, 4, and 5. This method establishes a nonlinear mapping relationship between angle and pressure influence, as shown in Formula 6 below: (Formula) (6) in, Indicates the influence coefficient of angle; The angle of the electric bed can be calculated using formula 4; this is clear. The acquisition method: "By collecting at least 30 sets of different angles" Calculate the corresponding pressure data Fi, Si for each group. , and then For the target value, fit using the least squares method. In "To ensure the fitting error is ≤5%"; Scenario: "When the pressure sensor malfunctions or the signal is lost..." It can be used as a temporary alternative, but its accuracy is lower than that of other alternatives. (Error range 10%-15%). The constant term "1" indicates that the angle is 0° (resting state). =1, meaning there is no angle effect; This indicates the effect of angle increase on pressure, for example, when the angle increases from 0 to 60 degrees. From 0 to The cosine function decreases from 1 to 0, making From 0 to 1, ensure It increases monotonically with angle.

[0036] In a specific embodiment, obtaining the pressure influence correction coefficient based on the first and second relationships further includes obtaining the extreme pressure influence correction coefficient based on a quadratic term fitting of the first and second relationships. For users with uneven pressure distribution in the upper body (common extreme body types), a quadratic term can be introduced into Formula 2 to improve the fitting accuracy. Because the pressure distribution of users with extreme body types is special, the quadratic term can amplify the influence of outliers. For example, when the local pressure F is much greater than the reference value... hour, The growth rate of is faster than that of the first term, making the model more sensitive to high-pressure areas; similarly, when the contact area S is too small, It will also increase significantly, reflecting the effect of pressure concentration. As shown in Formula 7 below: (Formula) (7) in, This is the pressure weighting coefficient; Sref is the total pressure under the supine reference state; Sref is the pressure area under the supine reference state; F is the total pressure under the i-th angle; S is the pressure area under the i-th angle.

[0037] Example 2: This embodiment provides a smart bed, which includes the aforementioned method for correcting vital signs based on pressure signals using a sensor band. The smart bed also includes an electric bed frame and a mattress. The electric bed frame is angle-adjustable, and a sensor band is provided on the surface of the electric bed frame or inside the mattress. The sensor band can detect various physiological data, pressure signals, etc. A control device is connected to the above-mentioned equipment, enabling it to acquire data, analyze it, and adjust the parameters of each device.

[0038] In other embodiments, this application can be applied not only to smart beds, but also to devices with adjustable angles such as electric sofas.

[0039] The above-described specific embodiments are preferred embodiments of a vital sign sensor band signal correction method based on pressure signals and a smart bed according to this application. They are not intended to limit the specific scope of this application. The scope of this application includes but is not limited to the specific embodiments described above. All equivalent changes made in accordance with the shape and structure of this application are within the protection scope of this application.

Claims

1. A method of correcting a signal of a vital sign sensing band based on a pressure signal, characterized by, The method comprises the following steps: obtaining a first relationship between pressure and angle based on historical pressure data and historical angle data; obtaining a second relationship between pressure and original signal quality of the sign sensing belt based on historical pressure data and historical original signal quality data of the sign sensing belt; obtaining a third relationship between angle and original signal quality of the sign sensing belt based on historical angle data and historical original signal quality data of the sign sensing belt; constructing a signal quality correction model according to the first relationship, the second relationship and the third relationship; performing sign sensing belt signal correction according to real-time pressure signal, real-time angle signal and the signal quality correction model.

2. The method according to claim 1, wherein The first relationship between pressure and angle based on historical pressure data and historical angle data comprises: obtaining pressure area and pressure total under different angles based on historical pressure data corresponding to historical angle data, and constructing the first relationship.

3. The method of claim 1, wherein the pressure signal based vital sensor band signal correction method is characterized by, The second relationship between pressure and original signal quality of the sign sensing belt based on historical pressure data and historical original signal quality data of the sign sensing belt comprises: constructing the second relationship based on historical pressure total, historical pressure area and historical original signal quality data of the sign sensing belt.

4. The method of claim 1, wherein the pressure signal based vital sensor band signal correction method is characterized by, The third relationship between angle and original signal quality of the sign sensing belt based on historical angle data and historical original signal quality data of the sign sensing belt comprises: calculating mean square error and standard deviation under the same angle based on amplitude data in the historical original signal quality data of the sign sensing belt, and constructing the third relationship.

5. The method of claim 1, wherein the pressure signal based vital sensor band signal correction method is characterized by, The signal quality correction model constructed according to the first relationship, the second relationship and the third relationship comprises: obtaining a pressure influence correction coefficient based on the first relationship and the second relationship; constructing the signal quality correction model based on the third relationship and the pressure influence correction coefficient.

6. The method of claim 1, wherein the pressure signal based vital sensor band signal correction method is characterized by, Further comprising: constructing a sensitivity correction relationship based on historical sampling rate and historical sensitivity; performing sign sensing belt signal correction according to real-time pressure signal, real-time angle signal, the signal quality correction model and the sensitivity correction relationship.

7. The method of claim 6, wherein the pressure signal based vital band signal correction method is characterized by, The sign sensing belt signal correction performed according to real-time pressure signal, real-time angle signal, the signal quality correction model and the sensitivity correction relationship comprises: obtaining current signal quality according to real-time pressure signal, real-time angle signal and the signal quality correction model; obtaining a sensitivity to-be-compensated value according to the current signal quality and a preset signal quality threshold; obtaining a sampling correction value according to the sensitivity to-be-compensated value and the sensitivity correction relationship; performing sign sensing belt signal correction with the sampling correction value.

8. The method of claim 1, wherein the pressure signal based vital sensor band signal correction method is characterized by, The signal quality correction model constructed according to the first relationship, the second relationship and the third relationship further comprises: constructing an angle influence relationship according to the first relationship and the second relationship based on the least square method; constructing a pressure influence relationship according to the first relationship and the second relationship based on a lying down reference state and a multi-angle contact state; constructing a first signal quality correction model based on the first relationship, the second relationship and the third relationship; constructing a second signal quality correction model based on the angle influence relationship and the third relationship.

9. The method of claim 8, wherein the pressure signal based vital sign belt signal correction method further comprises: The sign sensing belt signal correction performed according to real-time pressure signal, real-time angle signal and the signal quality correction model comprises: If the real-time pressure signal exists, a first signal quality correction model is called to obtain a vital sign sensing belt signal correction value based on the real-time pressure signal and the real-time angle signal; If the real-time pressure signal does not exist, a second signal quality correction model is called to obtain a vital sign sensing belt signal correction value based on the real-time angle signal.

10. A smart bed, characterized by The smart bed comprises the vital sign sensing belt signal correction method based on the pressure signal according to any one of claims 1-9, and further comprises: a motorized bed frame for adjusting the angle; a mattress; wherein the motorized bed frame surface or the mattress interior is provided with a sensing belt and a control device, the control device is electrically connected to the sensing belt, and the control device can acquire data, analyze, and adjust parameters of each device.

Images