Health state monitoring method and system

Abstract

The invention provides a health state monitoring method and system, and relates to the technical field of human health monitoring. Health state evaluation is performed by integrating the multi-dimensional physiological parameters, so that the comprehensiveness and accuracy of health monitoring are improved. The method comprises the following steps: collecting human body physiological parameters in real time based on a multi-parameter physiological sensor, wherein the human body physiological parameters comprise pulse, myoelectricity, blood pressure, body temperature, oxyhemoglobin saturation and heart rate variability; the collected human body physiological parameters are preprocessed; performing multi-source information fusion on the preprocessed human body physiological parameters based on an improved fuzzy D-S evidence theory; and constructing a health state model, and based on the health state model, judging a human health state according to a fusion result. The health state monitoring system is applied to the health state monitoring method.

Description

Technical Field

[0001] The present invention relates to the technical field of human health monitoring, and more specifically, to a health status monitoring method and system. Background Art

[0002] In the existing medical health monitoring system, traditional health monitoring methods often rely on the measurement of a single or a few physiological parameters, such as heart rate, blood pressure, blood sugar, etc. This monitoring method has limitations and is difficult to fully reflect the health status of individuals. In addition, due to the lack of an effective health risk assessment model, the ability to provide early warning and prevent diseases is limited, and effective diagnosis and treatment can often only be carried out after the disease occurs, which to a certain extent increases medical costs and reduces medical efficiency.

[0003] With the rapid development of the Internet of Things, big data, and artificial intelligence technologies, the field of medical health monitoring has ushered in new opportunities. By integrating IoT devices such as wearable devices and home medical devices, real-time monitoring and data collection of individual physiological parameters can be achieved. At the same time, using big data analysis and artificial intelligence algorithms, the collected physiological parameters can be deeply mined and analyzed, thereby achieving a comprehensive and dynamic assessment of the individual's health status. However, most of the health monitoring products or systems currently on the market are still at the simple data display and alarm level, lacking effective health risk assessment models and personalized health recommendations. Therefore, it is particularly important to develop a health monitoring method that can integrate multi-dimensional physiological parameters and use advanced algorithms for health risk assessment. Summary of the invention

[0004] The purpose of the present invention is to provide a health status monitoring method and system, which can improve the comprehensiveness and accuracy of health monitoring by comprehensively evaluating health status based on multi-dimensional physiological parameters.

[0005] In order to achieve the above object, the present invention provides a health status monitoring method, which comprises:

[0006] Step S1: collecting human physiological parameters in real time based on a multi-parameter physiological sensor, wherein the human physiological parameters include: pulse, electromyography, blood pressure, body temperature, blood oxygen saturation and heart rate variability;

[0007] Step S2: preprocessing the collected human physiological parameters;

[0008] Step S3: Based on the improved fuzzy DS evidence theory, multi-source information fusion is performed on the pre-processed human physiological parameters;

[0009] Step S4: construct a health status model, and judge the health status of the human body based on the health status model and the fusion result.

[0010] Preferably, the step S1 comprises:

[0011] The photoelectric pulse sensor uses infrared rays with a wavelength of 570nm or 870nm to detect changes in blood vessel transmittance and obtain pulse waveform signals.

[0012] Using surface electromyography sensors, the potential difference generated by muscle activity is detected to collect electromyography signals in real time;

[0013] Adopting the principle of blood pressure oscillometric method, the pressure change of blood vessel wall is detected by cuff pressurization, and blood pressure signal is collected in real time;

[0014] Use optical fiber radiation temperature sensor to collect human body temperature signals in real time;

[0015] Adopt photoelectric blood oxygen saturation sensor, which uses light of different wavelengths to penetrate the skin to detect blood oxygen content;

[0016] Heart rate variability is assessed by analyzing the time interval changes in the electrocardiogram or pulse signal.

[0017] Preferably, the step S2 comprises: using wavelet transform to filter and denoise the collected human physiological parameters, the formula is:

[0018] Among them, Z represents the wavelet coefficient, W i is the human physiological parameter after filtering and denoising, n represents the adjustment coefficient, λ i is the filtering threshold, i=1,2,3,4,5, representing pulse, electromyography, blood pressure, body temperature, blood oxygen saturation and heart rate variability respectively.

[0019] Preferably, the step S3: performing multi-source information fusion on the pre-processed human physiological parameters based on the improved fuzzy DS evidence theory comprises:

[0020] The pre-processed physiological parameters are fuzzified through membership functions and converted into fuzzy sets;

[0021] The fuzzy set of each physiological parameter is transformed into the corresponding basic trust assignment BBA;

[0022] The basic trust allocations BBA corresponding to the multiple physiological parameters are integrated to obtain the integrated basic trust allocation BBA;

[0023] The human health status is evaluated based on the fusion basic trust allocation (BBA).

[0024] Preferably, the step S4 comprises:

[0025] Step S4.1: construct a health status model based on deep learning, wherein the health status model uses the preprocessed and fused basic trust allocation BBA as input features and the human health status as output;

[0026] Step S4.2: Design and optimize the cross entropy loss function, the loss function formula is:

[0027]

[0028] Among them, L represents the total loss, N represents the total number of samples, m represents the total number of categories of health status, and y kj Indicates that the kth sample belongs to the jth health status category, It represents the probability that the kth sample predicted by the model belongs to the jth health status category.

[0029] Compared with the prior art, a health status monitoring method provided by the present invention has the following beneficial effects: The present invention provides a health status monitoring method, which includes: real-time collection of human physiological parameters based on multi-parameter physiological sensors, the human physiological parameters including: pulse, electromyography, blood pressure, body temperature, blood oxygen saturation and heart rate variability; preprocessing the collected human physiological parameters; multi-source information fusion of the preprocessed human physiological parameters based on the improved fuzzy DS evidence theory; constructing a health status model, and judging the health status of the human body based on the health status model according to the fusion results. The present invention conducts health status assessment by integrating multi-dimensional physiological parameters to improve the comprehensiveness and accuracy of health monitoring. By adopting the improved fuzzy DS evidence theory for multi-source information fusion, the information conflict and redundancy problems between different physiological parameters can be effectively solved.

[0030] The present invention also provides a health status monitoring system, the system comprising:

[0031] The acquisition module is used to acquire human physiological parameters in real time based on a multi-parameter physiological sensor, wherein the human physiological parameters include: pulse, electromyography, blood pressure, body temperature, blood oxygen saturation and heart rate variability;

[0032] A preprocessing module, used for preprocessing the collected human physiological parameters;

[0033] Information fusion module, used to fuse multi-source information of pre-processed human physiological parameters based on improved fuzzy DS evidence theory;

[0034] The model building module is used to build a health status model and judge the health status of the human body based on the health status model and the fusion results.

[0035] Preferably, in the acquisition module,

[0036] The photoelectric pulse sensor uses infrared rays with a wavelength of 570nm or 870nm to detect changes in blood vessel transmittance and obtain pulse waveform signals.

[0037] Using surface electromyography sensors, the potential difference generated by muscle activity is detected to collect electromyography signals in real time;

[0038] Adopting the principle of blood pressure oscillometric method, the pressure change of blood vessel wall is detected by cuff pressurization, and blood pressure signal is collected in real time;

[0039] Use optical fiber radiation temperature sensor to collect human body temperature signals in real time;

[0040] Adopt photoelectric blood oxygen saturation sensor, which uses light of different wavelengths to penetrate the skin to detect blood oxygen content;

[0041] Heart rate variability is assessed by analyzing the time interval changes in the electrocardiogram or pulse signal.

[0042] Preferably, in the preprocessing module, wavelet transform is used to filter and denoise the collected human physiological parameters, and the formula is:

[0043] Among them, Z represents the wavelet coefficient, W i is the human physiological parameter after filtering and denoising, n represents the adjustment coefficient, λ i is the filtering threshold, i=1,2,3,4,5, representing pulse, electromyography, blood pressure, body temperature, blood oxygen saturation and heart rate variability respectively.

[0044] Preferably, the information fusion module includes:

[0045] A fuzzy processing unit is used to perform fuzzy processing on the pre-processed physiological parameters through a membership function and convert them into fuzzy sets;

[0046] A basic trust allocation unit, used for converting the fuzzy set of each physiological parameter into a corresponding basic trust allocation BBA;

[0047] A trust allocation fusion unit, used for fusing basic trust allocations BBA corresponding to multiple physiological parameters to obtain a fused basic trust allocation BBA;

[0048] The evaluation unit is used to evaluate the health status of the human body based on the fused basic trust allocation BBA.

[0049] Preferably, the model building module comprises:

[0050] A model building unit, used to build a health status model based on deep learning, wherein the health status model uses the preprocessed and fused basic trust allocation BBA as an input feature and the human health status as an output;

[0051] The loss function unit is used to design and optimize the cross entropy loss function. The loss function formula is:

[0052]

[0053] Among them, L represents the total loss, N represents the total number of samples, m represents the total number of categories of health status, and y kj Indicates that the kth sample belongs to the jth health status category, It represents the probability that the kth sample predicted by the model belongs to the jth health status category.

[0054] Compared with the prior art, the beneficial effects of a health status monitoring system provided by the present invention are the same as the beneficial effects of a health status monitoring method described in the above technical solution, which will not be elaborated here.

[0055] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0057] Figure 1 A flow chart of a health status monitoring method provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0058] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0059] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0060] The embodiment of the present invention provides a health status monitoring method. Figure 1 A flow chart of a health status monitoring method provided by an embodiment of the present invention is shown as follows: Figure 1 As shown, the method includes:

[0061] Step S1: collecting human physiological parameters in real time based on a multi-parameter physiological sensor, wherein the human physiological parameters include: pulse, electromyography, blood pressure, body temperature, blood oxygen saturation and heart rate variability.

[0062] It should be noted that multiple information sources of different physiological parameter sensors are set in the wearable device to collect physiological indicators of multiple parts of the body. Specifically, a photoelectric pulse sensor is used to detect changes in blood vessel transmittance using infrared rays with a wavelength of 570nm or 870nm to obtain pulse waveform signals; a surface electromyography sensor is used to detect the potential difference generated by muscle activity and collect electromyography signals in real time; the blood pressure oscillometric method principle is used to detect changes in blood vessel wall pressure by applying pressure with a cuff and collect blood pressure signals in real time; an optical fiber radiation temperature sensor is used to collect human body temperature signals in real time; a photoelectric blood oxygen saturation sensor is used to detect blood oxygen content by using light of different wavelengths to penetrate the skin; and heart rate variability is evaluated by analyzing changes in time intervals in electrocardiograms or pulse signals.

[0063] Step S2: pre-processing the collected human physiological parameters.

[0064] It should be noted that wavelet transform is used to filter and denoise the collected human physiological parameters, and the formula is:

[0065] Among them, Z represents the wavelet coefficient, which represents the human physiological parameters after filtering and denoising, and W i is the human physiological parameter after filtering and denoising, n represents the adjustment coefficient, which is used to adjust the intensity and effect of filtering, and λ i is the filtering threshold, i=1,2,3,4,5, representing pulse, electromyography, blood pressure, body temperature, blood oxygen saturation and heart rate variability respectively.

[0066] Step S3: Based on the improved fuzzy DS evidence theory, multi-source information fusion is performed on the pre-processed human physiological parameters.

[0067] It should be noted that step S3 includes:

[0068] The pre-processed physiological parameters are fuzzified through membership functions and converted into fuzzy sets, and the precise physiological parameter values ​​are converted into fuzzy sets for subsequent information fusion. ;

[0069] The fuzzy set of each physiological parameter is converted into the corresponding basic trust allocation BBA. BBA represents each physiological parameter.

[0070] the confidence level of the parameter in the health status assessment;

[0071] The basic trust allocation BBA corresponding to multiple physiological parameters is integrated to obtain the integrated basic trust allocation BBA, and the accuracy of health status assessment is improved by integrating the information of multiple physiological parameters;

[0072] Based on the fused basic trust allocation BBA, the human health status is evaluated. By comparing the confidence of different health status categories, the most likely health status can be determined.

[0073] Step S4: construct a health status model, and judge the health status of the human body based on the health status model and the fusion result.

[0074] It should be noted that step S4 includes:

[0075] Step S4.1: Construct a health status model based on deep learning, wherein the health status model uses the preprocessed and fused basic trust allocation BBA as input features and the human health status as output; by training the model, the mapping relationship from physiological parameters to health status can be learned.

[0076] Step S4.2: Design and optimize the cross entropy loss function, the loss function formula is:

[0077]

[0078] Among them, L represents the total loss, N represents the total number of samples, m represents the total number of categories of health status, and y kj Indicates that the kth sample belongs to the jth health status category, It represents the probability that the kth sample predicted by the model belongs to the jth health status category. The cross entropy loss function is used to measure the difference between the health status predicted by the model and the actual health status. By optimizing the loss function, the prediction accuracy of the model can be improved, and by minimizing the loss function, the parameters of the model can be optimized and the performance of the model can be improved.

[0079] Compared with the prior art, the health status monitoring method provided by the embodiment of the present invention has the following beneficial effects:

[0080] 1. Comprehensively evaluate health status based on multi-dimensional physiological parameters to improve the comprehensiveness and accuracy of health monitoring.

[0081] 2. By adopting the improved fuzzy DS evidence theory to fuse multi-source information, the information conflict and redundancy problems between different physiological parameters can be effectively solved.

[0082] 3. The health status model built based on deep learning can automatically learn and adapt to the health characteristics of different users, improving the personalization level of health risk assessment.

[0083] 4. Provide comprehensive health status monitoring and personalized health advice to help with early warning and prevention of disease.

[0084] An embodiment of the present invention further provides a health status monitoring system, the system comprising:

[0085] The acquisition module is used to acquire human physiological parameters in real time based on a multi-parameter physiological sensor, wherein the human physiological parameters include: pulse, electromyography, blood pressure, body temperature, blood oxygen saturation and heart rate variability;

[0086] A preprocessing module, used for preprocessing the collected human physiological parameters;

[0087] Information fusion module, used to fuse multi-source information of pre-processed human physiological parameters based on improved fuzzy DS evidence theory;

[0088] The model building module is used to build a health status model and judge the health status of the human body based on the health status model and the fusion results.

[0089] Preferably, in the acquisition module,

[0090] The photoelectric pulse sensor uses infrared rays with a wavelength of 570nm or 870nm to detect changes in blood vessel transmittance and obtain pulse waveform signals.

[0091] Using surface electromyography sensors, the potential difference generated by muscle activity is detected to collect electromyography signals in real time;

[0092] Adopting the principle of blood pressure oscillometric method, the pressure change of blood vessel wall is detected by cuff pressurization, and blood pressure signal is collected in real time;

[0093] Use optical fiber radiation temperature sensor to collect human body temperature signals in real time;

[0094] Adopt photoelectric blood oxygen saturation sensor, which uses light of different wavelengths to penetrate the skin to detect blood oxygen content;

[0095] Heart rate variability is assessed by analyzing the time interval changes in the electrocardiogram or pulse signal.

[0096] Preferably, in the preprocessing module, wavelet transform is used to filter and denoise the collected human physiological parameters, and the formula is:

[0097] Among them, Z represents the wavelet coefficient, W i is the human physiological parameter after filtering and denoising, n represents the adjustment coefficient, λ i is the filtering threshold, i=1,2,3,4,5, representing pulse, electromyography, blood pressure, body temperature, blood oxygen saturation and heart rate variability respectively.

[0098] Preferably, the information fusion module includes:

[0099] A fuzzy processing unit is used to perform fuzzy processing on the pre-processed physiological parameters through a membership function and convert them into fuzzy sets;

[0100] A basic trust allocation unit, used for converting the fuzzy set of each physiological parameter into a corresponding basic trust allocation BBA;

[0101] A trust allocation fusion unit, used for fusing basic trust allocations BBA corresponding to multiple physiological parameters to obtain a fused basic trust allocation BBA;

[0102] The evaluation unit is used to evaluate the health status of the human body based on the fused basic trust allocation BBA.

[0103] Preferably, the model building module comprises:

[0104] A model building unit, used to build a health status model based on deep learning, wherein the health status model uses the preprocessed and fused basic trust allocation BBA as an input feature and the human health status as an output;

[0105] The loss function unit is used to design and optimize the cross entropy loss function. The loss function formula is:

[0106]

[0107] Among them, L represents the total loss, N represents the total number of samples, m represents the total number of categories of health status, and y kj Indicates that the kth sample belongs to the jth health status category, It represents the probability that the kth sample predicted by the model belongs to the jth health status category.

[0108] Compared with the prior art, the beneficial effects of a health status monitoring system provided by an embodiment of the present invention are the same as the beneficial effects of a health status monitoring method described in the above technical solution, which will not be elaborated here.

[0109] In addition, an embodiment of the present invention also provides an electronic device, including a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and run on the processor. The transceiver, the memory, and the processor are respectively connected via a bus. When the computer program is executed by the processor, each process of the above-mentioned health status monitoring method embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0110] In addition, an embodiment of the present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the various processes of the above-mentioned health status monitoring method embodiment are implemented and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0111] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technical solution that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A health status monitoring method, characterized in that: include: Step S1: collecting human physiological parameters in real time based on a multi-parameter physiological sensor, wherein the human physiological parameters include: pulse, electromyography, blood pressure, body temperature, blood oxygen saturation and heart rate variability; Step S2: preprocessing the collected human physiological parameters; Step S3: Based on the improved fuzzy DS evidence theory, multi-source information fusion is performed on the pre-processed human physiological parameters; Step S4: construct a health status model, and judge the health status of the human body based on the health status model and the fusion result.

2. A health status monitoring method according to claim 1, characterized in that: The step S1 comprises: The photoelectric pulse sensor uses infrared rays with a wavelength of 570nm or 870nm to detect changes in blood vessel transmittance and obtain pulse waveform signals. Using surface electromyography sensors, the potential difference generated by muscle activity is detected to collect electromyography signals in real time; Adopting the principle of blood pressure oscillometric method, the pressure change of blood vessel wall is detected by cuff pressurization, and blood pressure signal is collected in real time; Use optical fiber radiation temperature sensor to collect human body temperature signals in real time; Adopt photoelectric blood oxygen saturation sensor, which uses light of different wavelengths to penetrate the skin to detect blood oxygen content; Heart rate variability is assessed by analyzing the time interval changes in the electrocardiogram or pulse signal.

3. A health status monitoring method according to claim 1, characterized in that: The step S2 includes: using wavelet transform to filter and denoise the collected human physiological parameters, the formula is: Among them, Z represents the wavelet coefficient, W i is the human physiological parameter after filtering and denoising, n represents the adjustment coefficient, λ i is the filtering threshold, i=1,2,3,4,5, representing pulse, electromyography, blood pressure, body temperature, blood oxygen saturation and heart rate variability respectively.

4. A health status monitoring method according to claim 3, characterized in that: The step S3: performing multi-source information fusion on the pre-processed human physiological parameters based on the improved fuzzy DS evidence theory, comprises: The pre-processed physiological parameters are fuzzified through membership functions and converted into fuzzy sets; The fuzzy set of each physiological parameter is transformed into the corresponding basic trust assignment BBA; The basic trust allocations BBA corresponding to the multiple physiological parameters are integrated to obtain the integrated basic trust allocation BBA; The human health status is evaluated based on the fusion basic trust allocation (BBA).

5. A health status monitoring method according to claim 4, characterized in that: The step S4 comprises: Step S4.1: construct a health status model based on deep learning, wherein the health status model uses the preprocessed and fused basic trust allocation BBA as input features and the human health status as output; Step S4.2: Design and optimize the cross entropy loss function, the loss function formula is: Among them, L represents the total loss, N represents the total number of samples, m represents the total number of categories of health status, and y kj Indicates that the kth sample belongs to the jth health status category, It represents the probability that the kth sample predicted by the model belongs to the jth health status category.

6. A health status monitoring system, characterized in that: include: The acquisition module is used to acquire human physiological parameters in real time based on a multi-parameter physiological sensor, wherein the human physiological parameters include: pulse, electromyography, blood pressure, body temperature, blood oxygen saturation and heart rate variability; A preprocessing module, used for preprocessing the collected human physiological parameters; Information fusion module, used to fuse multi-source information of pre-processed human physiological parameters based on improved fuzzy DS evidence theory; The model building module is used to build a health status model and judge the health status of the human body based on the health status model and the fusion results.

7. A health status monitoring system according to claim 6, characterized in that: In the acquisition module, The photoelectric pulse sensor uses infrared rays with a wavelength of 570nm or 870nm to detect changes in blood vessel transmittance and obtain pulse waveform signals. Using surface electromyography sensors, the potential difference generated by muscle activity is detected to collect electromyography signals in real time; Adopting the principle of blood pressure oscillometric method, the pressure change of blood vessel wall is detected by cuff pressurization, and blood pressure signal is collected in real time; Use optical fiber radiation temperature sensor to collect human body temperature signals in real time; Adopt photoelectric blood oxygen saturation sensor, which uses light of different wavelengths to penetrate the skin to detect blood oxygen content; Heart rate variability is assessed by analyzing the time interval changes in the electrocardiogram or pulse signal.

8. A health status monitoring system according to claim 6, characterized in that: In the preprocessing module, wavelet transform is used to filter and denoise the collected human physiological parameters, and the formula is: Among them, Z represents the wavelet coefficient, W i is the human physiological parameter after filtering and denoising, n represents the adjustment coefficient, λ i is the filtering threshold, i=1,2,3,4,5, representing pulse, electromyography, blood pressure, body temperature, blood oxygen saturation and heart rate variability respectively.

9. A health status monitoring system according to claim 8, characterized in that: The information fusion module includes: A fuzzy processing unit is used to perform fuzzy processing on the pre-processed physiological parameters through a membership function and convert them into fuzzy sets; A basic trust allocation unit, used for converting the fuzzy set of each physiological parameter into a corresponding basic trust allocation BBA; A trust allocation fusion unit, used for fusing basic trust allocations BBA corresponding to multiple physiological parameters to obtain a fused basic trust allocation BBA; The evaluation unit is used to evaluate the health status of the human body based on the fused basic trust allocation BBA.

10. A health status monitoring system according to claim 9, characterized in that: The model building module includes: A model building unit, used to build a health status model based on deep learning, wherein the health status model uses the preprocessed and fused basic trust allocation BBA as an input feature and the human health status as an output; The loss function unit is used to design and optimize the cross entropy loss function. The loss function formula is: Among them, L represents the total loss, N represents the total number of samples, m represents the total number of categories of health status, and y kj Indicates that the kth sample belongs to the jth health status category, It represents the probability that the kth sample predicted by the model belongs to the jth health status category.

Images