Health monitoring system for intelligent closestool

Abstract

The invention provides a health monitoring system for an intelligent closestool. The health monitoring system comprises a user verification module, a current management module, a physiological parameter acquisition module, a cardiovascular and cerebrovascular disease prediction module, a urine sample acquisition module, a urinary system infection detection module, a comprehensive health scoring module, a cleaning and maintenance module and a data storage and updating module. The invention discloses an intelligent closestool. According to the health monitoring system for the intelligent closestool, through automatic detection of the infrared sensor and opening of the control valve, automatic collection of urine samples is achieved, human intervention is reduced, and the collection efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of smart home technology, and in particular to a health monitoring system for a smart toilet. Background Art

[0002] With the rapid development of smart home technology, people are paying more and more attention to health monitoring and management in daily life. As a common sanitary facility in the home, smart toilets not only provide a clean and comfortable experience, but also become an ideal platform for integrating health monitoring functions. In the field of medical health, early detection and diagnosis are the key to preventing and treating diseases. For common diseases such as cardiovascular and cerebrovascular diseases and urinary tract infections, if risk factors can be discovered in time through daily monitoring, it will greatly improve the treatment effect and reduce medical costs. However, some existing health monitoring systems for smart toilets may not be able to respond to urine flow in real time, realize real-time collection and continuous monitoring of urine samples, which limits the timely evaluation of disease changes, and urine testing lacks accurate quantification of results and accurate evaluation of disease levels; some existing health monitoring systems for smart toilets may not be able to provide continuous and real-time physical indicator data, which limits users' timely understanding and evaluation of changes in their own health status; physiological data from different sources and types may have different dimensions and numerical ranges, which may limit the generalization ability of machine learning models. Some existing health monitoring systems for smart toilets may not be able to provide continuous and long-term monitoring of disease changes, resulting in the inability to help patients understand the development of their disease in a timely manner. Summary of the invention

[0003] In order to solve the above-mentioned problem of the present invention, the present invention provides a health monitoring system for a smart toilet, comprising a user verification module, a current management module, a physiological parameter acquisition module, a cardiovascular and cerebrovascular disease prediction module, a urine sample acquisition module, a urinary system infection detection module, a comprehensive health scoring module, a cleaning and maintenance module and a data storage and update module; the user verification module is respectively connected to the above-mentioned other modules through a central controller, the current management module is connected to the physiological parameter acquisition module, the input end of the physiological parameter acquisition module is connected to the output end of the urine sample acquisition module, the input end of the cardiovascular and cerebrovascular disease prediction module is connected to the output end of the physiological parameter acquisition module, the input end of the urinary system infection detection module is connected to the output end of the physiological parameter acquisition module and the urine sample acquisition module, the input end of the comprehensive health scoring module is connected to the output end of the cardiovascular and cerebrovascular disease prediction module and the urinary system infection detection module, and the input end of the data storage and update module is connected to the input end of the cardiovascular and cerebrovascular disease prediction module, the urinary system infection detection module and the comprehensive health scoring module;

[0004] After the user identity authentication is successful, the user authentication module is used to obtain the user's facial image captured by the camera, extract key feature points in the facial image, apply a hash algorithm to the extracted key feature points to generate a feature vector, provide a stable and reliable data basis for identity authentication, compare the generated feature vector with the template feature vector stored in the database, calculate the Euclidean distance between the two feature vectors, convert the distance into a similarity score S, where S=1 / (1+D), D represents the Euclidean distance, and compare the similarity score S with a preset threshold T, providing a quantitative verification standard. When the similarity score S is greater than or equal to the preset threshold T, it indicates that the user identity authentication is successful, the toilet lid is controlled to open automatically, and contactless automatic operation is realized. A start signal is sent to the central controller, and the remaining modules of the system are started by the central controller, thereby enhancing the security of the smart toilet, preventing unauthorized use, and protecting user privacy.

[0005] After the user's identity verification is successful, the current management module is used to manage the electronic circuit of the bioelectrical impedance analyzer so that the low-frequency current can pass through the human body evenly, ensuring that the current intensity and frequency are within a safe range to avoid electric shock injuries to the human body. The conductive material in the toilet seat is used to transmit the current and achieve good electrical contact with the human body;

[0006] The physiological parameter acquisition module is used to start the physiological parameter data acquisition program to collect physiological parameter data from various sensors and bioelectrical impedance analyzers;

[0007] The cardiovascular and cerebrovascular disease prediction module is used to train a support vector machine model, predict the probability of cardiovascular and cerebrovascular diseases through the support vector machine model, generate a risk score report, and help users understand their own disease risks and take preventive measures;

[0008] The urine sample collection module is integrated with the infrared sensor. When the infrared sensor detects urine flow, it controls the urine sample detection device to open the valve and collect urine samples to the first detection area containing antibodies. The automatic detection of the infrared sensor and the control of the opening of the valve realize automatic collection and transmission of urine samples, reduce human intervention, improve collection efficiency, and reduce the risk of sample contamination and loss. The consistency and representativeness of the sample are ensured. In the first detection area, when the urine sample contains antigens, it will specifically bind to the antibodies pre-fixed in the first detection area to form an antibody complex. The antibody complex continues to be detected by the second detection area. In the second detection area, a color development reaction is triggered to form a color development area.

[0009] The urinary system infection detection module is used to quantify the color development area, evaluate the disease level and predict the disease change trend, which helps to track the disease change and provide a basis for clinical diagnosis;

[0010] The comprehensive health scoring module is used to integrate the outputs of the urinary system infection detection module and the cardiovascular and cerebrovascular disease prediction module to determine the comprehensive health status of the user and generate a comprehensive health score;

[0011] The cleaning and maintenance module is used to receive a cleaning instruction sent by the central controller each time the smart toilet is used, and start the cleaning device according to the cleaning instruction; the cleaning device cleans the biosensor array, the first detection area and the second detection area according to a preset cleaning program to maintain the hygiene and performance of the equipment;

[0012] The data storage and update module is used to store the latest disease assessment level, disease change trend, risk score report and comprehensive health score in the database, ensuring the long-term preservation and traceability of the data, and synchronously updating it to the user-side APP.

[0013] Further, the physiological parameter acquisition module includes a heart rate index data acquisition submodule, a urine biochemical index data acquisition submodule and a body index data acquisition submodule;

[0014] The heart rate index data acquisition submodule is used to acquire the user's heart rate index data measured by the radar sensor;

[0015] The urine biochemical index data acquisition submodule is integrated with the biosensor array, and acquires the biochemical index data in the detected urine sample from the biosensor array, wherein the biochemical index data includes protein content, pH value, electrolyte level, antigen concentration A, inflammatory factor concentration I and nucleic acid concentration V;

[0016] The body indicator data acquisition submodule is used to obtain the body's impedance data to electric current measured by the bioimpedance analyzer, extract impedance features from the impedance data, use the impedance data as the input of a pre-trained bioimpedance analysis model, automatically analyze the impedance data and predict the body indicator data. Through the bioimpedance analyzer, the user can complete the measurement of body indicator data while using the smart toilet. The body indicator data include blood pressure indicator data, arterial hardness indicator data, muscle mass indicator data, body water content data and body fat percentage data. The bioimpedance analyzer includes multiple electrodes and electronic circuits connected to the electrodes. The electrodes are in contact with the user's skin to transmit current and measure the impedance data of the current passing through the body.

[0017] Furthermore, the cardiovascular and cerebrovascular disease prediction module includes a model training submodule, a feature preprocessing submodule, a feature weight calculation submodule and a risk assessment submodule;

[0018] The model training submodule is used to construct a training data set including heart rate index data, body fat percentage, blood pressure index data and arterial hardness index data, which provides a rich data basis for the prediction of cardiovascular and cerebrovascular diseases. The training data set is used to train the support vector machine model, and the arterial hardness index feature H, body fat percentage feature F, blood pressure index feature P and heart rate index feature B in the training data set are extracted, which helps the model to learn and predict more accurately.

[0019] The feature preprocessing submodule is used to normalize the features to adapt to the support vector machine model;

[0020] The feature weight calculation submodule calculates the score S of each feature based on the loss function in the iterative training process of the model, and calculates the weight W of each feature based on the score S of each feature, where W=S*α, α is a preset adjustment coefficient;

[0021] The risk assessment submodule adjusts the degree of dependence of the trained support vector machine model on different features according to the weight W of each feature, so that the model can pay more attention to features with higher weights, thereby improving the accuracy of prediction. The training data set is used as the input of the support vector machine model, and the support vector machine model is applied to predict cardiovascular and cerebrovascular diseases to obtain the probability of illness. The probability of illness reflects the risk level of cardiovascular and cerebrovascular diseases. According to a preset risk threshold, the probability of illness is converted into a specific cardiovascular and cerebrovascular disease risk level. The risk level is divided into low risk, medium risk and high risk levels. The probability of illness and risk level are integrated into the risk scoring report to provide users with a comprehensive and easy-to-understand health assessment result.

[0022] Furthermore, the urinary system infection detection module includes an electrical signal analysis submodule, a disease level assessment submodule and a disease change trend prediction submodule;

[0023] The electric signal analysis submodule is used to quantify the color development area to obtain a color intensity value C, which provides a quantifiable indicator for disease assessment. The color intensity value C is compared with a preset color threshold value T1. When C is greater than or equal to T1, an electric signal E is output and the disease level assessment submodule is started, where E=k*(C T1), where k is a proportional coefficient. When C is less than T1, the output electric signal E is 0;

[0024] The disease level assessment submodule is used to multiply the antigen concentration A, the inflammatory factor concentration I and the nucleic acid concentration V by their corresponding weights W1A, W1I and W1V, respectively, to calculate the weighted disease severity index St, and provide a quantitative assessment of the severity of the disease, wherein St = W1A*A+W1I*I+W1V*V. When St is greater than or equal to the preset weighted severity index threshold Sth, the disease is judged to be a very serious first-level infection level. When St is less than Sth, the disease is judged to be a relatively serious second-level infection level. The infection level is judged to provide a basis for clinical treatment.

[0025] The disease change trend prediction submodule is used to record the sequence {St_1, St_2, ..., St_n} of the disease severity index St obtained by multiple consecutive measurements, to achieve continuous tracking of disease changes, to improve the continuity and accuracy of disease monitoring, and to calculate the change ΔSt of the disease severity index between two adjacent measurements, where ΔSt = St_(i+1)St_i. If ΔSt is greater than or equal to a preset change threshold ΔSth, it is judged that the disease has worsened. If ΔSt is less than ΔSth, it is judged that the disease has stabilized or improved. By analyzing the relationship between ΔSt and the preset change threshold ΔSth, the development trend of the disease is predicted to provide timely guidance for treatment.

[0026] Further, the comprehensive health score module includes a data fusion submodule and a health score generation submodule;

[0027] The data fusion submodule is responsible for receiving the disease assessment level, disease change trend and risk score report from the urinary system infection detection module and the cardiovascular and cerebrovascular disease prediction module, and fusing the received data to ensure that data from different sources can be comprehensively considered;

[0028] The health score generation submodule is used to use the fused data as the input of a pre-trained health score model, identify the correlation pattern between the scores of the health dimensions and the fused data, and generate a comprehensive health score. The clear and easy-to-understand comprehensive health score helps users better understand their health status.

[0029] Furthermore, a smart toilet includes a toilet cover, a camera, a radar sensor, a water tank, a toilet seat, a bioelectrical impedance analyzer, a barrel body, a water suction pipe, a catheter, a urine sample detection device, an infrared sensor and a valve, wherein the urine sample detection device includes a first detection area and a second detection area, the camera and the radar sensor are arranged on an outer surface of the water tank facing the toilet seat to facilitate capturing a user's facial image, and the camera is arranged at the middle upper end of the water tank, the radar sensor is located below the camera, the bioelectrical impedance analyzer is fixedly arranged on the inner upper surface of the barrel body, the toilet seat is arranged on one side of the water tank, and the bottom of the toilet seat is connected to the water suction pipe The urine sample detection device is fixedly arranged above the bottom inner surface of the barrel body and connected to the toilet seat through the catheter to realize efficient collection of urine samples. The infrared sensor is fixedly arranged above the urine sample detection device. The valve is arranged above the urine sample detection device and is located on one side edge of the infrared sensor. The biosensor array is fixedly arranged in the middle and upper part of the urine sample detection device and is located below the infrared sensor and the valve. The first detection area is connected to the second detection area to realize continuous detection and analysis of urine samples. The biosensor array is arranged on the side of the first detection area facing the toilet seat.

[0030] Furthermore, a smart toilet adopts the above-mentioned health monitoring system for a smart toilet.

[0031] The beneficial effects of the health monitoring system for a smart toilet of the present invention are as follows: through the automatic detection of the infrared sensor and the opening of the control valve, the automatic collection of urine samples is realized, which reduces human intervention and improves the collection efficiency. Automated collection reduces the risk of sample contamination and loss, and ensures the consistency and representativeness of the samples. The urinary system infection detection module can quantify the color development area and evaluate the severity of the disease according to the quantitative results, which is helpful to track the changes in the disease and provide a basis for clinical diagnosis. Through the bioelectrical impedance analyzer, the user can complete the measurement of body index data while using the smart toilet, which improves the convenience of monitoring. The feature weight calculation submodule calculates the feature score through the loss function and determines the feature weight accordingly, so that the model can identify and utilize more important features. By considering the feature weights, the risk assessment submodule enables the support vector machine model to pay more attention to important features when predicting, thereby improving the accuracy of the prediction. The disease change trend prediction submodule records the sequence of the disease severity index St to achieve continuous tracking of disease changes. By continuously monitoring the change in the disease severity index ΔSt, the continuity and accuracy of disease monitoring are improved. By analyzing the relationship between ΔSt and the preset change threshold ΔSth, the development trend of the disease is predicted, providing timely guidance for treatment. The fused data is used as the input of the pre-trained health scoring model to identify the correlation pattern between the scores of the health dimensions and the fused data, and generate a comprehensive health score. The clear and easy-to-understand comprehensive health score helps users better understand their health status. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 This is a system architecture diagram of a health monitoring system for a smart toilet according to the present invention;

[0033] Figure 2 This is a system architecture diagram of a physiological parameter acquisition module of a health monitoring system for a smart toilet of the present invention;

[0034] Figure 3 This is a system architecture diagram of a cardiovascular and cerebrovascular disease prediction module of a health monitoring system for a smart toilet of the present invention;

[0035] Figure 4 This is a system architecture diagram of a urinary system infection detection module of a health monitoring system for a smart toilet of the present invention;

[0036] Figure 5 This is a system architecture diagram of a comprehensive health scoring module of a health monitoring system for a smart toilet of the present invention;

[0037] Figure 6 This is a structural schematic diagram of a smart toilet of the present invention;

[0038] Figure 7This is a structural schematic diagram of a smart toilet in use state according to the present invention;

[0039] Figure 8 The present invention is a cross-sectional view of a smart toilet body.

[0040] Among them, 1-toilet lid, 2-camera, 3-radar sensor, 4-water tank, 5-toilet seat, 6-bioelectrical impedance analyzer, 61-electrode, 7-barrel body, 8-water suction pipe, 9-catheter, 10-urine sample detection device, 101-first detection area, 102-second detection area, 11-infrared sensor, 12-valve, 13-biosensor array, 14-user verification module, 141-central controller, 15-current management module, 151-electronic circuit, 16-physiological parameter acquisition module, 161-heart rate index data acquisition submodule, 162-urine biochemical index data acquisition submodule, 163-body Indicator data acquisition submodule, 17-cardiovascular and cerebrovascular disease prediction module, 171-model training submodule, 172-feature preprocessing submodule, 173-feature weight calculation submodule, 174-risk assessment submodule, 18-urine sample collection module, 19-urinary tract infection detection module, 191-electrical signal analysis submodule, 192-disease level assessment submodule, 193-disease change trend prediction submodule, 20-comprehensive health score module, 201-data fusion submodule, 202-health score generation submodule 202, 21-cleaning and maintenance module, 211-cleaning device, 22-data storage and update module. DETAILED DESCRIPTION

[0041] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] like Figure 1-8As shown, the present invention provides a health monitoring system for a smart toilet, including a user verification module 14, a current management module 15, a physiological parameter acquisition module 16, a cardiovascular and cerebrovascular disease prediction module 17, a urine sample acquisition module 18, a urinary system infection detection module 19, a comprehensive health scoring module 20, a cleaning and maintenance module 21 and a data storage and update module 22; the user verification module 14 is connected to the above-mentioned other modules respectively through a central controller 141, the current management module 15 is connected to the physiological parameter acquisition module 16, and the input end of the physiological parameter acquisition module 16 is connected to the urine sample acquisition module 18 The output end of the cardiovascular and cerebrovascular disease prediction module 17 is connected to the output end of the physiological parameter acquisition module 16, the input end of the urinary system infection detection module 19 is connected to the output end of the physiological parameter acquisition module 16 and the urine sample acquisition module 18, the input end of the comprehensive health scoring module 20 is connected to the output end of the cardiovascular and cerebrovascular disease prediction module 17 and the urinary system infection detection module 19, and the input end of the data storage and update module 22 is connected to the input end of the cardiovascular and cerebrovascular disease prediction module 17, the urinary system infection detection module 19 and the comprehensive health scoring module 20;

[0043] The user verification module 14 is used to obtain the facial image of the user captured by the camera 2, extract key feature points in the facial image, apply a hash algorithm to the extracted key feature points to generate a feature vector, provide a stable and reliable data basis for identity authentication, compare the generated feature vector with the template feature vector stored in the database, calculate the Euclidean distance between the two feature vectors, convert the distance into a similarity score S, where S=1 / (1+D), D represents the Euclidean distance, compare the similarity score S with a preset threshold T, and provide a quantitative verification standard. When the similarity score S is greater than or equal to the preset threshold T, it indicates that the user identity authentication is successful, the toilet lid 1 is controlled to open automatically, and contactless automatic operation is realized. A start signal is sent to the central controller 141, and the remaining modules of the system are started by the central controller 141, thereby enhancing the security of the smart toilet, preventing unauthorized use, and protecting user privacy.

[0044] After the user identity verification is successful, the current management module 15 is used to manage the electronic circuit 151 of the bioelectrical impedance analyzer 6, so that the low-frequency current can pass through the human body evenly, ensuring that the current intensity and frequency are within a safe range to avoid electric shock injuries to the human body. The conductive material in the toilet seat is used to transmit the current and achieve good electrical contact with the human body;

[0045] The physiological parameter acquisition module 16 is used to start the physiological parameter data acquisition program to collect physiological parameter data from various sensors and the bioelectrical impedance analyzer 6;

[0046] The cardiovascular and cerebrovascular disease prediction module 17 is used to train a support vector machine model, predict the probability of cardiovascular and cerebrovascular diseases through the support vector machine model, generate a risk score report, and help users understand their own disease risks and take preventive measures;

[0047] The urine sample collection module 18 is integrated with the infrared sensor 11. When the infrared sensor 11 detects urine flow, the urine sample detection device 10 is controlled to open the valve 12 to collect urine samples to the first detection area 101 containing antibodies. The automatic detection of the infrared sensor 11 and the control of the opening of the valve realize automatic collection and transmission of urine samples, reduce human intervention, improve collection efficiency, and reduce the risk of sample contamination and loss through automated collection, thereby ensuring the consistency and representativeness of the samples. In the first detection area 101, when the urine sample contains antigens, they will specifically bind to the antibodies pre-fixed in the first detection area 101 to form an antibody complex, and the antibody complex will continue to be detected by the second detection area 102. In the second detection area 102, a color development reaction is triggered to form a color development area.

[0048] The urinary system infection detection module 19 is used to quantify the color development area, evaluate the disease level and predict the disease change trend, which helps to track the disease change and provide a basis for clinical diagnosis;

[0049] The comprehensive health scoring module 20 is used to integrate the outputs of the urinary system infection detection module 19 and the cardiovascular and cerebrovascular disease prediction module 17 to determine the comprehensive health status of the user and generate a comprehensive health score;

[0050] The cleaning and maintenance module 21 is used to receive a cleaning instruction sent by the central controller 141 each time the smart toilet is used, and start the cleaning device 211 according to the cleaning instruction; the cleaning device 211 cleans the biosensor array 13, the first detection area 101 and the second detection area 102 according to a preset cleaning program to maintain the hygiene and performance of the equipment;

[0051] The data storage and update module 22 is used to store the latest disease assessment level, disease change trend, risk score report and comprehensive health score in the database, ensuring the long-term preservation and traceability of the data, and synchronously updating it to the user-side APP.

[0052] The physiological parameter acquisition module 16 includes a heart rate index data acquisition submodule 161, a urine biochemical index data acquisition submodule 162 and a body index data acquisition submodule 163.

[0053] The heart rate index data acquisition submodule 161 is used to acquire the user's heart rate index data measured by the radar sensor 3;

[0054] The urine biochemical index data acquisition submodule 162 is integrated with the biosensor array 13, and acquires the biochemical index data in the detected urine sample from the biosensor array 13, wherein the biochemical index data includes protein content, pH value, electrolyte level, antigen concentration A, inflammatory factor concentration I and nucleic acid concentration V;

[0055] The body indicator data acquisition submodule 163 is used to obtain the body's impedance data to electric current measured by the bioimpedance analyzer 6, extract impedance features from the impedance data, use the impedance data as the input of a pre-trained bioimpedance analysis model, automatically analyze the impedance data and predict the body indicator data. Through the bioimpedance analyzer, the user can complete the measurement of body indicator data while using the smart toilet. The body indicator data include blood pressure indicator data, arterial hardness indicator data, muscle mass indicator data, body water content data and body fat percentage data. The bioimpedance analyzer 6 includes multiple electrodes 61 and an electronic circuit 151 connected to the electrodes. The electrodes 61 contact the user's skin to transmit current and measure the impedance data of the current passing through the body.

[0056] The cardiovascular and cerebrovascular disease prediction module 17 includes a model training submodule 171, a feature preprocessing submodule 172, a feature weight calculation submodule 173 and a risk assessment submodule 174.

[0057] The model training submodule 171 is used to construct a training data set including heart rate index data, body fat percentage, blood pressure index data and arterial hardness index data, which provides a rich data basis for the prediction of cardiovascular and cerebrovascular diseases. The training data set is used to train the support vector machine model, and the arterial hardness index feature H, body fat percentage feature F, blood pressure index feature P and heart rate index feature B in the training data set are extracted, which helps the model to learn and predict more accurately.

[0058] The feature preprocessing submodule 172 is used to normalize the features to adapt to the support vector machine model;

[0059] The feature weight calculation submodule 173 calculates the score S of each feature based on the loss function in the model iterative training process, and calculates the weight W of each feature based on the score S of each feature, where W=S*α, α is a preset adjustment coefficient;

[0060] The risk assessment submodule 174 adjusts the degree of dependence of the trained support vector machine model on different features according to the weight W of each feature, so that the model can pay more attention to features with higher weights, thereby improving the accuracy of prediction. The training data set is used as the input of the support vector machine model, and the support vector machine model is applied to predict cardiovascular and cerebrovascular diseases to obtain the probability of illness. The probability of illness reflects the risk level of cardiovascular and cerebrovascular diseases. According to the preset risk threshold, the probability of illness is converted into a specific cardiovascular and cerebrovascular disease risk level. The risk level is divided into low risk, medium risk and high risk levels. The probability of illness and risk level are integrated into the risk scoring report to provide users with a comprehensive and easy-to-understand health assessment result.

[0061] The urinary system infection detection module 19 includes an electrical signal analysis submodule 191, a disease level assessment submodule 192, and a disease change trend prediction submodule 193.

[0062] The electric signal analysis submodule 191 is used to quantify the color development area to obtain the color intensity value C, which provides a quantifiable indicator for disease assessment. The color intensity value C is compared with the preset color threshold value T1. When C is greater than or equal to T1, the electric signal E is output and the disease level assessment submodule 192 is started, where E=k*(C T1), where k is the proportional coefficient. When C is less than T1, the output electric signal E is 0;

[0063] The disease level assessment submodule 192 is used to multiply the antigen concentration A, the inflammatory factor concentration I and the nucleic acid concentration V by their corresponding weights W1A, W1I and W1V, respectively, to calculate the weighted disease severity index St, and provide a quantitative assessment of the severity of the disease, wherein St = W1A*A+W1I*I+W1V*V. When St is greater than or equal to the preset weighted severity index threshold Sth, the disease is judged to be a very serious first-level infection level. When St is less than Sth, the disease is judged to be a relatively serious second-level infection level. The infection level is judged to provide a basis for clinical treatment.

[0064] The disease condition change trend prediction submodule 193 is used to record the sequence {St_1, St_2, ..., St_n} of the disease condition severity index St obtained by multiple consecutive measurements, to achieve continuous tracking of disease condition changes, to improve the continuity and accuracy of disease condition monitoring, and to calculate the change ΔSt of the disease condition severity index between two adjacent measurements, where ΔSt = St_(i+1)St_i. If ΔSt is greater than or equal to a preset change threshold ΔSth, the disease condition is judged to have worsened. If ΔSt is less than ΔSth, the disease condition is judged to have stabilized or improved. By analyzing the relationship between ΔSt and the preset change threshold ΔSth, the development trend of the disease condition is predicted to provide timely guidance for treatment.

[0065] The comprehensive health scoring module (20) comprises a data fusion submodule (201) and a health score generation submodule (202);

[0066] The data fusion submodule (201) is responsible for receiving the disease assessment level, disease change trend and risk score report from the urinary system infection detection module (19) and the cardiovascular and cerebrovascular disease prediction module (17), and fusing the received data to ensure that data from different sources can be comprehensively considered;

[0067] The health score generation submodule (202) is used to use the fused data as input to a pre-trained health score model, identify the association pattern between the scores of the health dimensions and the fused data, and generate a comprehensive health score. The clear and easy-to-understand comprehensive health score helps users better understand their health status.

[0068] A smart toilet comprises a toilet cover 1, a camera 2, a radar sensor 3, a water tank 4, a toilet seat 5, a bioelectrical impedance analyzer 6, a barrel body 7, a water suction pipe 8, a catheter 9, a urine sample detection device 10, an infrared sensor 11 and a valve 12, wherein the urine sample detection device 10 comprises a first detection area 101 and a second detection area 102, the camera 2 and the radar sensor 3 are arranged on an outer surface of the water tank 4 facing the toilet seat 5, so as to capture a user's facial image, and the camera 2 is arranged at the middle upper end of the water tank 4, the radar sensor 3 is located below the camera 2, the bioelectrical impedance analyzer 6 is fixedly arranged on the inner upper surface of the barrel body 7, the toilet seat 5 is arranged on one side of the water tank 4, and the bottom of the toilet seat 5 is connected to the water suction pipe 8 and the The catheter 9, the urine sample detection device 10 is fixedly arranged above the bottom inner surface of the barrel body 7, and is connected to the toilet seat 5 through the catheter 9 to achieve efficient collection of urine samples. The infrared sensor 11 is fixedly arranged above the urine sample detection device 10, and the valve 12 is arranged above the urine sample detection device 10, located on one side edge of the infrared sensor 11, and the biosensor array 13 is fixedly arranged in the middle and upper part of the urine sample detection device 10, located below the infrared sensor 11 and the valve 12. The first detection area 101 is connected to the second detection area 102 to achieve continuous detection and analysis of urine samples. The biosensor array 13 is arranged on the side of the first detection area 101 facing the toilet seat 5.

[0069] A smart toilet adopts the above-mentioned smart toilet health monitoring system.

[0070] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

[0071] The present invention and its embodiments are described above, which is not restrictive. What is shown in the accompanying drawings is only one of the embodiments of the present invention, and the actual content is not limited to this. In short, if ordinary technicians in this field are inspired by it and do not deviate from the purpose of the invention, they can creatively design structural methods and embodiments similar to the technical scheme, which should all fall within the scope of protection of the present invention.

Claims

1. A health monitoring system for a smart toilet, characterized in that: The invention comprises a user verification module (14), a current management module (15), a physiological parameter acquisition module (16), a cardiovascular and cerebrovascular disease prediction module (17), a urine sample acquisition module (18), a urinary system infection detection module (19), a comprehensive health scoring module (20), a cleaning and maintenance module (21) and a data storage and update module (22); the user verification module (14) is respectively connected to the above-mentioned other modules through a central controller (141), the current management module (15) is connected to the physiological parameter acquisition module (16), the input end of the physiological parameter acquisition module (16) is connected to the output end of the urine sample acquisition module (18), and the cardiovascular and cerebrovascular disease prediction module (17) is connected to the output end of the urine sample acquisition module (18). The input end of the cerebrovascular disease prediction module (17) is connected to the output end of the physiological parameter acquisition module (16), the input end of the urinary system infection detection module (19) is connected to the output ends of the physiological parameter acquisition module (16) and the urine sample acquisition module (18), the input end of the comprehensive health scoring module (20) is connected to the output ends of the cardiovascular and cerebrovascular disease prediction module (17) and the urinary system infection detection module (19), and the input end of the data storage and update module (22) is connected to the input ends of the cardiovascular and cerebrovascular disease prediction module (17), the urinary system infection detection module (19) and the comprehensive health scoring module (20); The user verification module (14) is used to obtain a facial image of a user captured by a camera (2), extract key feature points in the facial image, apply a hash algorithm to the extracted key feature points to generate a feature vector, compare the generated feature vector with a template feature vector stored in a database, calculate the Euclidean distance between the two feature vectors, convert the distance into a similarity score S, where S=1 / (1+D), D represents the Euclidean distance, compare the similarity score S with a preset threshold value T, and when the similarity score S is greater than or equal to the preset threshold value T, it indicates that the user identity verification is successful, control the toilet lid (1) to open automatically, and send a start signal to the central controller (141), so as to start the remaining modules of the system through the central controller (141); After the user identity verification is successful, the current management module (15) is used to manage the electronic circuit (151) of the bioelectrical impedance analyzer (6) so that the low-frequency current can pass through the human body evenly; The physiological parameter acquisition module (16) is used to start a physiological parameter data acquisition program to collect physiological parameter data from various sensors and a bioelectrical impedance analyzer (6); The cardiovascular and cerebrovascular disease prediction module (17) is used to train a support vector machine model, predict the probability of cardiovascular and cerebrovascular diseases through the support vector machine model, and generate a risk score report; The urine sample collection module (18) is integrated with an infrared sensor (11). When the infrared sensor (11) detects urine flow, it controls the urine sample detection device (10) to open the valve (12) to collect the urine sample to the first detection area (101) containing antibodies. In the first detection area (101), when the urine sample contains antigens, it will specifically bind to the antibodies pre-fixed in the first detection area (101) to form an antibody complex. The antibody complex is further detected by the second detection area (102). In the second detection area (102), a color development reaction is triggered to form a color development area. The urinary system infection detection module (19) is used to quantify the color development area, evaluate the disease level and predict the disease change trend; The comprehensive health scoring module (20) is used to integrate the outputs of the urinary system infection detection module (19) and the cardiovascular and cerebrovascular disease prediction module (17), judge the comprehensive health status of the user, and generate a comprehensive health score; The cleaning and maintenance module (21) is used to receive a cleaning instruction sent by the central controller (141) each time the smart toilet is used, and start the cleaning device (211) according to the cleaning instruction; the cleaning device (211) cleans the biosensor array (13), the first detection area (101) and the second detection area (102) according to a preset cleaning program; The data storage and update module (22) is used to store the latest disease assessment level, disease change trend, risk score report and comprehensive health score in the database, and synchronously update them to the user-side APP.

2. A health monitoring system for a smart toilet according to claim 1, characterized in that: The physiological parameter acquisition module (16) comprises a heart rate index data acquisition submodule (161), a urine biochemical index data acquisition submodule (162) and a body index data acquisition submodule (163); The heart rate index data acquisition submodule (161) is used to acquire the user's heart rate index data measured by the radar sensor (3); The urine biochemical index data acquisition submodule (162) is integrated with a biosensor array (13), and biochemical index data in the detected urine sample is acquired from the biosensor array (13), wherein the biochemical index data includes protein content, pH value, electrolyte level, antigen concentration A, inflammatory factor concentration I and nucleic acid concentration V; The body index data acquisition submodule (163) is used to acquire the body's impedance data to electric current measured by the bioimpedance analyzer (6), extract impedance features from the impedance data, use the impedance data as input of a pre-trained bioimpedance analysis model, automatically analyze the impedance data and predict body index data, wherein the body index data includes blood pressure index data, arterial hardness index data, muscle mass index data, body water content data and body fat percentage data. The bioimpedance analyzer (6) comprises a plurality of electrodes (61) and an electronic circuit (151) connected to the electrodes, and transmits electric current through the electrodes (61) in contact with the user's skin and measures the impedance data of the current passing through the body.

3. A health monitoring system for a smart toilet according to claim 2, characterized in that: The cardiovascular and cerebrovascular disease prediction module (17) includes a model training submodule (171), a feature preprocessing submodule (172), a feature weight calculation submodule (173) and a risk assessment submodule (174); The model training submodule (171) is used to construct a training data set including heart rate index data, body fat rate, blood pressure index data and arterial hardness index data, train a support vector machine model using the training data set, and extract arterial hardness index feature H, body fat rate feature F, blood pressure index feature P and heart rate index feature B from the training data set; The feature preprocessing submodule (172) is used to normalize the features to adapt to the support vector machine model; The feature weight calculation submodule (173) calculates the score S of each feature based on the loss function in the model iterative training process, and calculates the weight W of each feature based on the score S of each feature, wherein W=S*α, α is a preset adjustment coefficient; The risk assessment submodule (174) adjusts the degree of dependence of the trained support vector machine model on different features according to the weight W of each feature, takes the training data set as the input of the support vector machine model, applies the support vector machine model to predict cardiovascular and cerebrovascular diseases, and obtains the probability of illness. According to a preset risk threshold, the probability of illness is converted into a specific cardiovascular and cerebrovascular disease risk level. The risk level is divided into low risk, medium risk and high risk levels, and the probability of illness and risk level are integrated into a risk scoring report.

4. A health monitoring system for a smart toilet according to claim 3, characterized in that: The urinary system infection detection module (19) comprises an electrical signal analysis submodule (191), a disease level assessment submodule (192) and a disease change trend prediction submodule (193); The electric signal analysis submodule (191) is used to quantify the color development area to obtain a color intensity value C, compare the color intensity value C with a preset color threshold value T1, and when C is greater than or equal to T1, output an electric signal E and start the disease level assessment submodule (192), wherein E=k*(C T1), wherein k is a proportionality coefficient, and when C is less than T1, the output electric signal E is 0; The disease severity assessment submodule (192) is used to multiply the antigen concentration A, the inflammatory factor concentration I and the nucleic acid concentration V by their corresponding weights W1A, W1I and W1V, respectively, to calculate a weighted disease severity index St, wherein St = W1A*A+W1I*I+W1V*V. When St is greater than or equal to a preset weighted severity index threshold Sth, the disease is judged to be a very serious first-level infection level. When St is less than Sth, the disease is judged to be a relatively serious second-level infection level. The disease condition change trend prediction submodule (193) is used to record a sequence {St_1, St_2, ..., St_n} of disease condition severity index St obtained by multiple consecutive measurements, and calculate a change ΔSt of the disease condition severity index between two adjacent measurements, wherein ΔSt=St_(i+1)-St_i. If ΔSt is greater than or equal to a preset change threshold ΔSth, it is judged that the disease condition has worsened; if ΔSt is less than ΔSth, it is judged that the disease condition has stabilized or improved.

5. The health monitoring system for a smart toilet according to claim 4, characterized in that: The comprehensive health scoring module (20) comprises a data fusion submodule (201) and a health score generation submodule (202); The data fusion submodule (201) is responsible for receiving the disease assessment level, disease change trend and risk score report from the urinary system infection detection module (19) and the cardiovascular and cerebrovascular disease prediction module (17), and fusing the received data; The health score generation submodule (202) is used to use the fused data as input to a pre-trained health score model, identify the association pattern between the scores of the health dimensions and the fused data, and generate a comprehensive health score.

Images