Intelligent human health sign measuring method and physical examination equipment

Abstract

The invention relates to the technical field of human health detection, in particular to an intelligent human health sign measuring method and physical examination equipment. The method comprises the following steps: acquiring a face video, acquiring a multi-wavelength optical signal and processing the signal to extract a pulse signal, further predicting physiological signals such as heart rate, heart rate variability, respiratory rate, oxyhemoglobin saturation and blood pressure, and intelligently predicting a human health degree result according to the weight and height of a human body. The method has the core advantage of non-contact, is convenient and efficient, and is suitable for the scenes of telemedicine, family health monitoring, ICU (intensive care unit) monitoring, factory worker health management and the like.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of human health detection, in particular to an intelligent human health sign measuring method and a physical examination device. BACKGROUND

[0002] With the improvement of living conditions, modern people pay more and more attention to their own and family health. The detection of health items such as heart rate, heart rate variability, blood oxygen saturation, blood pressure, etc. usually needs to be detected by blood testing, cuff measuring device, etc. The blood testing method is not suitable for daily detection, and the cuff device makes people feel uncomfortable, even invasive examination method. These contact detection methods cause some adults to be reluctant to use traditional health examination methods. How to provide a method and device to non-contact measure blood flow, heart rate, blood oxygen saturation and blood pressure is a problem to be solved by those skilled in the art. SUMMARY

[0003] In order to solve the above technical problems, the present application provides an intelligent human health sign measuring method, comprising the following steps: S1, obtaining height data and weight data of a human body, and obtaining face video data through a spectral camera; S2, analyzing the face video data by photoplethysmography or its derived non-contact technology to obtain heart rate data, blood oxygen saturation data and blood pressure data; S3, comprehensively evaluating the human health degree result according to the height data, weight data, heart rate data, blood oxygen saturation data and blood pressure data.

[0004] Preferably, the measurement of the heart rate data comprises the following steps: A1, analyzing the subtle changes of the face when the blood flows through the face video data to extract blood volume pulse (BVP) signals; A2, after filtering and denoising, calculating the heart rate data by time domain peak value detection or frequency domain FFT transformation.

[0005] Preferably, in step A1, the periodic beating of the heart will cause the arterial blood volume to change regularly with the heartbeat, and the absorption of blood to a specific wavelength of light will also fluctuate. When the heart contracts, the blood volume increases, and the light absorption increases. When the heart relaxes, the blood volume decreases, and the light absorption decreases. The spectral camera captures the subtle changes of the facial skin to extract the blood volume pulse (BVP) signals.

[0006] Preferably, in step A1, the measurement of the blood oxygen saturation data comprises the following steps: According to the obvious difference in the absorption characteristics of oxygenated hemoglobin and reduced hemoglobin in the blood in the red light (600-700 nm) and near-infrared light (800-1000 nm) wave bands, the BVP signal is collected through the red light channel (and the near-infrared light channel), and the blood oxygen saturation is derived by using the change of light absorption ratio at the two wavelengths and combining the Lambert-Beer law.

[0007] Preferably, in step A1, the measurement of the blood pressure data comprises the following steps: B1, collecting a pulse wave signal through a multi-spectral channel, B2, extracting the correlation characteristics of signals in different wave bands, and indirectly inverting the blood pressure value by combining a pre-constructed blood pressure estimation model.

[0008] Preferably, the pre-constructed blood pressure estimation model comprises a multiple linear regression or an AI model trained based on a large number of samples.

[0009] Preferably, a physical examination device is used to implement the intelligent human health sign measurement method described above, characterized in that it comprises a spectral camera module, a wide dynamic camera module, a radar height measurement module, a weighing module, and a processor module. The spectral camera module and the wide dynamic camera module are used to obtain face video data, the radar height measurement module and the weighing module are used to obtain height data and weight data, and the processor module is used to comprehensively evaluate the human health degree result according to the height data, the weight data, the heart rate data, the blood oxygen saturation data, and the blood pressure data.

[0010] As can be seen from the above, by obtaining face video, multi-wavelength light signal collection and signal processing, the pulse signal can be extracted, which is further used to predict physiological signals such as heart rate, heart rate variability, respiratory rate, blood oxygen saturation, and blood pressure. According to the weight and height of the human body, the intelligent prediction of the human health degree result is realized, and the health indicators of the human body are fused, calculated, and analyzed. The core advantage is non-contact, convenient, and efficient, and it is suitable for remote medical treatment, family health monitoring, ICU monitoring, factory worker health management, and other scenes. BRIEF DESCRIPTION OF DRAWINGS

[0011] 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 description of the embodiments of the present application or the prior art. Obviously, the drawings in the following description are only part of the embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0012] Figure 1 The flow chart of the intelligent human health sign measurement method of the embodiments of the present application; Figure 2 The structure diagram of the physical examination device is shown in the present application. DETAILED DESCRIPTION

[0013] The technical solutions in the embodiments of the present application will be clearly and completely described in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative labor fall within the protection scope of the present application. EMBODIMENT

[0014] To solve the above technical problems, the present embodiment provides a kind of intelligent human health sign determination method, as shown in Figure 1 The method comprises the following steps: S1, the height data of human body, weight data, through spectral camera obtains face video data; S2, through photoplethysmography or its derived non-contact technology, face video data is analyzed, and heart rate data, blood oxygen saturation data and blood pressure data are obtained; S3, according to height data, weight data, heart rate data, blood oxygen saturation data and blood pressure data, the health degree result of human body is obtained by comprehensive evaluation.

[0015] In the above scheme, the measurement of heart rate data comprises the following steps: A1, through face video data, the subtle change of blood flow face is analyzed, so as to extract blood volume pulse wave (BVP) signal; A2, after filtering and denoising, through time domain peak detection or frequency domain FFT transformation, heart rate data is calculated.

[0016] Wherein, in step A1, the periodic pulsation of heart will cause the arterial blood volume to change regularly with heartbeat, and the absorption amount of blood to specific wavelength light (such as green light and red light) will also fluctuate, when heart contracts, blood volume increases, light absorption enhances, when heart relaxes, blood volume decreases, light absorption weakens, spectral camera captures the subtle change of spectral color of facial skin blood, so as to extract blood volume pulse wave (BVP) signal, after filtering and denoising, through time domain peak detection and frequency domain FFT transformation, heart rate data is calculated.

[0017] In the filtering denoising process of step A2, an adaptive filtering algorithm is adopted, which can automatically adjust the filtering parameters according to the real-time characteristics of the signal, effectively remove the interference of environmental noise and motion artifacts on the BVP signal, and ensure the purity of the signal. Subsequently, the time domain peak detection method directly calculates the heart rate data by identifying the significant peak points in the BVP signal and combining the time interval of adjacent peak points; while the frequency domain FFT transform first converts the BVP signal from the time domain to the frequency domain, and then indirectly obtains the heart rate data by analyzing the peak position corresponding to the heart rate frequency in the frequency spectrum. The two methods are verified with each other, which improves the accuracy and reliability of heart rate measurement.

[0018] Further, in step S1, the measurement of blood oxygen saturation data includes the following steps: According to the obvious difference in absorption characteristics of oxyhemoglobin and reduced hemoglobin in the red light (600-700 nm) and near-infrared light (800-1000 nm) waveband in the blood, the BVP signal is collected through the red light channel and the near-infrared light channel, and the blood oxygen saturation data is derived by using the change of light absorption ratio at two wavelengths and combining the Lambert-Beer law. Specifically, when collecting the BVP signal, the red light channel emits red light in the 600-700 nm waveband, and the near-infrared light channel emits near-infrared light in the 800-1000 nm waveband. The two kinds of light are simultaneously irradiated to the facial skin, and the spectral camera captures the spectral color change of the facial skin blood after absorbing the two kinds of light, and then obtains the corresponding BVP signal. Then, the change of light absorption ratio at two wavelengths of red light and near-infrared light is accurately measured, and then the accurate blood oxygen saturation data is derived by algorithm formula according to the Lambert-Beer law. The signal difference of the multi-spectral channel needs to ensure the depth of light penetration and the contact stability with the skin.

[0019] Further, in step S1, the measurement of blood pressure data includes the following steps: B1, collecting pulse wave signals through multi-spectral channels, B2, extracting the correlation characteristics of signals of different wavebands, combining the pre-constructed blood pressure estimation model, and indirectly inverting the blood pressure value. The pre-constructed blood pressure estimation model includes a multiple linear regression or AI model trained based on a large number of samples. The correlation characteristics include feature point time difference and amplitude ratio.

[0020] Specifically, blood pressure is highly related to the propagation characteristics of the pulse wave (such as pulse transit time PTT) and the waveform characteristics (half-width, systolic area, etc.). In step B1, the multi-spectral channel emits light of different wavelengths, such as green light, red light, and near-infrared light, which irradiates the skin. The blood flow under the skin causes changes in absorption and reflection of light, and the spectral sensor captures these changes to collect pulse wave signals. In step B2, for the correlation features of different waveband signals, the feature point time difference refers to the time interval of the appearance of specific feature points in different waveband signals, and the amplitude ratio refers to the amplitude ratio of the feature points in different waveband signals. The pre-constructed blood pressure estimation model is obtained by training a large number of sample data of different blood pressure populations. These sample data contain the corresponding relationship between the correlation features of different waveband signals and the actual blood pressure values. Using multiple linear regression or AI models to learn and fit these relationships, when the correlation features of the newly collected different waveband signals are input, the corresponding blood pressure values can be indirectly inverted, without the need for cuff pressure, and can realize non-invasive continuous monitoring, greatly facilitating the user's daily health detection. In addition, based on the multiple linear regression or AI model trained based on a large number of samples, the respiratory rate, atrial fibrillation, emotions (stress, anxiety, depression, etc.), and human health condition classification can be accurately detected.

[0021] On the other hand, a physical examination device for implementing the above-mentioned intelligent human health sign measurement method includes a spectral camera module, a wide dynamic camera module, a radar height measurement module, a weighing module, and a processor module. The spectral camera module and the wide dynamic camera module are used to obtain face video data, the radar height measurement module and the weighing module are used to obtain height data and weight data, and the processor module is used to comprehensively evaluate the human health degree result based on the height data, the weight data, the heart rate data, the blood oxygen saturation data, and the blood pressure data.

[0022] Specifically, as Figure 2As shown, the above physical examination device comprises a main frame body 10, a touch display screen 20 embedded on the main frame body 10, a spectrum camera module 30 and a wide dynamic camera module 40 arranged on both sides of the touch display screen 20 on the main frame body 10, a radar height measurement module 50 arranged above the main frame body 10, a weighing module 60 connected below the main frame body 10, and a printing module 70 arranged below the touch display screen 20. After a human body stands on the weighing module 60 for weight measurement, the height of the human body is measured by the radar height measurement module 50. The face video information of the human body is obtained by the spectrum camera module 30 and the wide dynamic camera module 40, so that multiple physiological signals such as heart rate, blood oxygen saturation, blood pressure and the like of the human body are detected, and the detected results are reported and printed by the printing module 70. The user operation is simple and convenient. By collecting various detection modules on the main frame body 10, a multifunctional physical examination machine is realized, the human body detection efficiency is improved, and the equipment occupied space is reduced. The spectrum camera can indeed realize the measurement of blood flow, heart rate, blood oxygen saturation and blood pressure through multi-wavelength light signal acquisition and signal processing. The core advantage is non-contact, convenient and efficient, and is suitable for remote medical treatment, family health monitoring, ICU monitoring and the like.

[0023] Specifically, the main frame body 10 comprises a frame and a processor module, the processor module is in communication connection with the touch display screen 20, the spectrum camera module 30, the wide dynamic camera module 40, the radar height measurement module 50, the weighing module 60 and the printing module 70. For example, each module can be in communication connection with the processor module by wired or wireless mode. The spectrum camera module 30 can shoot and obtain a face video, the overexposure and underexposure in the picture are optimized by the wide dynamic camera module 40, so that a panoramic image with bright and dark parts can be obtained in the above environment, the definition of the face video is improved, and the face video is analyzed by the processor module. Generally, the pulse wave of the human body is obtained from the face video by PPG (photoplethysmography) technology, and multiple physiological signals such as heart rate, heart rate variability, blood oxygen saturation, blood pressure and the like are further obtained. The specific detection principle is not described here. The processor module combines the data of the radar height measurement module 50 and the weighing module 60 to preliminarily evaluate the health of the human body, and the result can be printed by the printing module 70 driven by the touch display screen 20 and the processor module. Further, a voice broadcast module 80 is also arranged on the main frame body 10, the voice broadcast module 80 is in communication connection with the processor module, and information broadcast or operation reminding is performed for the user.

[0024] Further, the top of the main rack body 10 is connected with a bent connecting block 51, and the radar height measuring module 50 is installed at the end of the bent connecting block 51. The existing height measuring instrument measures the height of the human body through the lifting of the pressing plate, and the lifting movement of the pressing plate is too complex for the structure of the physical examination machine, which is not conducive to the manufacture of the physical examination machine. Therefore, the radar height measuring module 50 measures the height by radar, and the infrared method can also be used to measure the height to improve the accuracy. The radar height measuring module 50 transmits the measured data to the processor module.

[0025] Further, since the weighing module 60 extends to one side of the main rack body 10, in order to facilitate the storage and protection of the weighing module 60 during the transfer of the whole machine, the weighing module 60 can be foldably connected to the main rack body 10. Specifically, the bottom of the main rack body 10 extends an installation block 11, and one side of the weighing module 60 is foldably connected to the installation block 11 through a hinge 61. The hinge 61 can realize the overturning of the weighing module 60, and the weighing module 60 can be folded to the side of the main rack body 10 during the transfer or carrying of the whole machine, so as to realize the storage of the weighing module 60 and avoid the collision between the weighing module 60 and the outside during the movement and carrying process.

[0026] Further, the weighing module 60 is connected to the main rack body 10 through a connecting piece after being folded. For example, the connecting piece can be a tape, a bolt, or a connecting plate. When the connecting plate is used, the connecting plate can be connected to the side of the weighing module 60 and the side of the main rack body 10 through screws, so that the weighing module 60 is fixed to the main rack body 10.

[0027] In summary, the present application obtains a face video, collects and processes a multi-wavelength light signal, extracts a pulse signal, further predicts physiological signals such as heart rate, heart rate variability, respiratory rate, blood oxygen saturation, blood pressure, and the like, and realizes intelligent prediction of the health degree of the human body according to the weight and height of the human body. The core advantage is non-contact, convenient and efficient, and is suitable for remote medical treatment, family health monitoring, ICU monitoring, factory worker health management and the like.

[0028] The above-described embodiments do not constitute a limitation on the protection scope of the technical solutions. Any modifications, equivalent replacements, and improvements made within the spirit and principles of the above-described embodiments shall fall within the protection scope of the technical solutions.

Claims

1. A method of intelligent human health vital sign measurement, characterized in that: The method comprises the following steps: S1, obtaining height data and weight data of a human body, and obtaining face video data through a spectral camera; S2, analyzing the face video data through photoplethysmography or a derived non-contact technology to obtain heart rate data, blood oxygen saturation data, and blood pressure data; S3, comprehensively evaluating the height data, weight data, heart rate data, blood oxygen saturation data, and blood pressure data to obtain a human health degree result.

2. The intelligent human health sign measuring method according to claim 1, characterized in that: The measurement of the heart rate data comprises the following steps: A1, analyzing the subtle changes in the face when blood flows to extract a blood volume pulse (BVP) signal from the face video data; A2, after filtering and noise reduction, calculating the heart rate data through time domain peak value detection and / or frequency domain FFT transformation.

3. The intelligent human health sign measuring method according to claim 2, characterized in that: In step S1, the periodic beating of the heart causes the arterial blood volume to change regularly with the heartbeat, and the amount of light absorbed by blood at a specific wavelength also fluctuates. When the heart contracts, the blood volume increases, and the light absorption increases. When the heart relaxes, the blood volume decreases, and the light absorption decreases. The spectral camera captures the subtle changes in the facial skin to extract the blood volume pulse (BVP) signal.

4. The intelligent human health sign measuring method according to claim 1, characterized in that: In step S1, the measurement of the blood oxygen saturation data comprises the following steps: According to the significant difference in the absorption characteristics of oxyhemoglobin and reduced hemoglobin in blood at red light (600-700 nm) and near-infrared light (800-1000 nm) bands, the BVP signal is collected through a red light channel (and a near-infrared light channel), the change in the light absorption ratio at the two wavelengths is used, and the Lambert-Beer law is used to derive the blood oxygen saturation.

5. The intelligent human health sign measuring method according to claim 1, characterized in that: In step S1, the measurement of the blood pressure data comprises the following steps: B1, collecting pulse wave signals through multiple spectral channels, B2, extracting the correlation features of different band signals, combining a pre-constructed blood pressure estimation model, and indirectly inverting the blood pressure value.

6. The intelligent human health sign measuring method according to claim 5, characterized in that: The pre-constructed blood pressure estimation model includes a multiple linear regression or AI model trained based on a large number of samples.

7. A physical examination apparatus for implementing the intelligent human health physical sign measurement method according to any one of claims 1-6, characterized in that: It comprises a spectral camera module, a wide dynamic camera module, a radar height measurement module, a weighing module, and a processor module. The spectral camera module and the wide dynamic camera module are used to obtain face video data, the radar height measurement module and the weighing module are used to obtain height data and weight data, and the processor module is used to comprehensively evaluate the height data, weight data, heart rate data, blood oxygen saturation data, and blood pressure data to obtain a human health degree result.

Images