Intelligent user monitoring system and method based on intelligent paper diaper
Through the intelligent diaper system, the urine data is analyzed, and the problems of trace elements detection lag and inconvenient health monitoring in infants and young children are solved, convenient and accurate health monitoring is achieved, and the risks of infants and young children's health are reduced.
Patent Information
- Application Number
- CN202411260287.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-10
- Publication Date
- 2024-12-13
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
In the prior art, trace elements in infants and young children lack detection lag and frequent hospital examinations are not good for the fragile immune system and lack convenient health monitoring methods.
The intelligent diaper system is adopted to obtain trace element data in urine from multiple predetermined time points, perform data regularization, multi-scale feature coding and feature emphasis, and use artificial intelligence analysis to judge the trace element content in the user's body to achieve intelligent monitoring.
It provides timely and accurate health feedback, reduces detection errors, reduces risks to infants and young children's health, and provides a convenient way to monitor health.
Smart Images
Figure CN119132652A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of intelligent monitoring, and more specifically, to a user intelligent monitoring system and method based on smart diapers. Background Art
[0002] The healthy growth of infants and young children cannot be separated from the supplement of trace elements. The lack of trace elements can lead to a series of symptoms in infants and young children, such as slow physical growth and development, inexplicable crying, and dislike of eating. Many mothers are often at a loss when faced with this situation. Regular child health care and testing methods, such as blood draws and urine tests, can accurately detect the content of trace elements in infants and young children, but this method often has a certain lag, and it is often detected after the lack of trace elements in infants and young children. At the same time, due to the fragile immune system of infants and young children, they cannot be taken to the hospital for examination frequently.
[0003] Therefore, a user intelligent monitoring solution based on smart diapers is needed. Summary of the invention
[0004] In view of the shortcomings of the prior art, the present application provides a user intelligent monitoring system and method based on smart diapers.
[0005] According to one aspect of the present application, a user intelligent monitoring system based on smart diapers is provided, which includes:
[0006] A user monitoring related data acquisition module is used to obtain trace element data in urine on the smart diaper at multiple predetermined time points within a predetermined time period, wherein the trace element data includes nitrite, urobilinogen, protein, occult blood, ketone bodies, bilirubin, glucose, ascorbic acid, urine calcium, and trace albumin;
[0007] A user monitoring related data regularization module, used for regularizing the trace element data in the urine on the smart diaper at the plurality of predetermined time points to obtain a urine trace element time series input matrix;
[0008] A multi-scale feature encoding module for user monitoring related data, used for performing multi-scale feature encoding on the urine trace element time series input matrix to obtain a multi-scale urine trace element time series feature matrix;
[0009] A user monitoring multi-scale feature emphasis module is used to perform feature emphasis on the multi-scale urine trace element time series feature matrix to obtain a urine trace element time series correlation feature emphasis matrix;
[0010] The module for generating the trace element monitoring results in the user's body is used to obtain the trace element monitoring results in the user wearing the smart diaper based on the urine trace element time series correlation feature emphasis matrix.
[0011] According to another aspect of the present application, a user intelligent monitoring method based on smart diapers is provided, which includes:
[0012] Acquire the trace element data in the urine of the smart diaper at multiple predetermined time points within a predetermined time period, wherein the trace element data includes nitrite, urobilinogen, protein, occult blood, ketone bodies, bilirubin, glucose, ascorbic acid, urine calcium, and trace albumin;
[0013] Regularizing the data of trace elements in urine on the smart diapers at the plurality of predetermined time points to obtain a time series input matrix of trace elements in urine;
[0014] Performing multi-scale feature encoding on the urine trace element time series input matrix to obtain a multi-scale urine trace element time series feature matrix;
[0015] Performing feature emphasis on the multi-scale urine trace element time series feature matrix to obtain a urine trace element time series correlation feature emphasis matrix;
[0016] Based on the urine trace element time series correlation feature emphasis matrix, the trace element monitoring results of the user wearing the smart diaper are obtained.
[0017] This application has significant technical effects due to the adoption of the above technical solutions:
[0018] The user intelligent monitoring system and method based on smart diapers provided in this application adopts data analysis technology based on artificial intelligence, and performs time series feature analysis on the trace element data in the urine of smart diapers at multiple predetermined time points within a predetermined time period, thereby judging whether the trace element content in the body of the user wearing the smart diapers is normal. In this way, intelligent monitoring of trace elements in the user's body is realized, which is conducive to timely obtaining feedback on the user's health status and provides a more convenient user health monitoring method. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.
[0020] Figure 1 The figure is a flow chart of a user intelligent monitoring method based on smart diapers according to an embodiment of the present application.
[0021] Figure 2Schematic diagram of data flow of a user intelligent monitoring method based on smart diapers according to an embodiment of the present application.
[0022] Figure 3 The present invention is a flowchart of performing multi-scale feature encoding on the urine trace element time series input matrix to obtain a multi-scale urine trace element time series feature matrix in the user intelligent monitoring method based on smart diapers according to an embodiment of the present application.
[0023] Figure 4 The present invention is a flowchart of obtaining the monitoring results of trace elements in the body of a user wearing a smart diaper based on the urine trace element time series correlation feature emphasis matrix in the user intelligent monitoring method based on smart diapers according to an embodiment of the present application.
[0024] Figure 5 4 is a system block diagram of a user intelligent monitoring system based on smart diapers according to an embodiment of the present application. DETAILED DESCRIPTION
[0025] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.
[0026] In today's society, people pay great attention to the healthy growth of infants and young children. Infancy is a crucial stage in life, and the healthy growth of infants and young children cannot be separated from the supplementation of various trace elements. Once the trace elements are deficient, infants and young children may show a series of worrying symptoms. For example, physical growth and development are obviously slow, and the gap with peers gradually widens; they often cry for no reason, which makes mothers feel distressed and at a loss; they also show a lack of interest in food and do not like to eat, which not only affects the nutritional intake, but also makes parents extremely anxious.
[0027] Many mothers often feel helpless when faced with these situations. At present, regular child health care and testing methods mainly include blood draws, urine tests, etc., which can more accurately detect the content of trace elements in infants and young children. However, it cannot be ignored that this detection method often has a certain lag, and it can usually only be detected after infants and young children have symptoms of trace element deficiency. At the same time, considering the fragile immune system of infants and young children, they cannot be taken to the hospital for examination frequently to avoid increasing the risk of infection. Therefore, a user intelligent monitoring solution based on smart diapers is expected.
[0028] In recent years, deep learning and neural networks have been widely used in computer vision, natural language processing, text signal processing and other fields. In addition, deep learning and neural networks have also demonstrated a level close to or even beyond that of humans in image classification, object detection, semantic segmentation, text translation and other fields. The development of deep learning and neural networks provides new solutions and solutions for user intelligent monitoring based on smart diapers.
[0029] Figure 1 The figure is a flow chart of a user intelligent monitoring method based on smart diapers according to an embodiment of the present application. Figure 2 Schematic diagram of data flow of the user intelligent monitoring method based on smart diapers according to an embodiment of the present application. Figure 1 and Figure 2 As shown, according to the embodiment of the present application, the user intelligent monitoring method based on smart diapers includes: S110, obtaining trace element data in urine on smart diapers at multiple predetermined time points within a predetermined time period, wherein the trace element data include nitrite, urobilinogen, protein, occult blood, ketone bodies, bilirubin, glucose, ascorbic acid, urine calcium, and trace albumin; S120, performing data regularization on the trace element data in urine on the smart diapers at the multiple predetermined time points to obtain a urine trace element time series input matrix; S130, performing multi-scale feature encoding on the urine trace element time series input matrix to obtain a multi-scale urine trace element time series feature matrix; S140, performing feature emphasis on the multi-scale urine trace element time series feature matrix to obtain a urine trace element time series correlation feature emphasis matrix; S150, based on the urine trace element time series correlation feature emphasis matrix, obtaining the trace element monitoring result in the body of the user wearing the smart diaper.
[0030] In step S110, the trace element data in the urine of the smart diaper at multiple predetermined time points within a predetermined time period is obtained, and the trace element data includes nitrite, urobilinogen, protein, occult blood, ketone bodies, bilirubin, glucose, ascorbic acid, urine calcium, and trace albumin. It should be understood that the content of trace elements in the human body changes over time, and they are affected by many factors, such as dietary intake, physiological state, disease status, etc. Therefore, it is difficult to fully reflect the true status of trace elements in infants and young children by relying solely on data at a single time point. By obtaining data at multiple predetermined time points, the changing trend and fluctuation of trace element content can be more accurately captured, thereby providing a more comprehensive health assessment. And in practical applications, due to the influence of various factors (such as measurement error, sample contamination, etc.), the data at a single time point may have large errors and interference. By obtaining data at multiple time points and performing statistical analysis, the impact of these errors and interferences on the final results can be effectively reduced, thereby improving the accuracy and reliability of trace element monitoring in the body of users wearing smart diapers. In particular, in a specific implementation of the embodiment of the present application, data on various trace elements in urine can be obtained through a urine detection device built into the smart diaper.
[0031] In step S120, the data of trace elements in urine on the smart diapers at the plurality of predetermined time points are regularized to obtain a urine trace element time series input matrix. Specifically, in an embodiment of the present application, the data of trace elements in urine on the smart diapers at the plurality of predetermined time points are regularized to obtain a urine trace element time series input matrix, including: arranging the trace element data in urine on the smart diapers at the plurality of predetermined time points according to the time dimension to obtain the urine trace element time series input matrix. It should be understood that the content of trace elements in the human body changes dynamically over time. Arranging the urine trace element data according to the time dimension can clearly observe the changing trend and fluctuation of the trace element content, which is helpful to analyze whether the content of trace elements in the body of the user wearing the smart diaper is normal. Each row in the arranged urine trace element time series input matrix data is the urine trace element data at a time point, and each column is a specific urine trace element monitoring indicator data.
[0032] In step S130, multi-scale feature encoding is performed on the urine trace element time series input matrix to obtain a multi-scale urine trace element time series feature matrix. Specifically, Figure 3 The present invention is a flowchart of performing multi-scale feature encoding on the urine trace element time series input matrix to obtain a multi-scale urine trace element time series feature matrix in the user intelligent monitoring method based on smart diapers according to an embodiment of the present application. Figure 3As shown, the multi-scale feature encoding of the urine trace element time series input matrix to obtain a multi-scale urine trace element time series feature matrix includes: S131, passing the urine trace element time series input matrix through a first urine trace element data time series feature analyzer to obtain a first-scale urine trace element time series feature matrix; S132, passing the urine trace element time series input matrix through a second urine trace element data time series feature analyzer to obtain a second-scale urine trace element time series feature matrix; S133, fusing the first-scale urine trace element time series feature matrix and the second-scale urine trace element time series feature matrix to obtain the multi-scale urine trace element time series feature matrix.
[0033] In order to be able to capture the content change of a certain trace element in a specific time period and the correlation between different trace elements and other information, it is necessary to pass the urine trace element time series input matrix through the first urine trace element data time series feature analyzer to obtain the first scale urine trace element time series feature matrix. In this application, the first urine trace element data time series feature analyzer is a convolutional neural network model using a first scale two-dimensional convolution kernel. It should be understood that the convolutional neural network can capture the local features in the urine trace element time series data through the sliding and convolution operations of the convolution kernel on the input matrix. And because the input matrix data is time-series, the convolutional neural network can also take into account the correlation of the data in the time dimension while extracting local features. Through multi-layer convolution and pooling operations, higher-level time series features can be gradually extracted, and these features can reflect the trend and pattern of trace element content changing over time. It is also worth mentioning that through multi-layer convolution and the nonlinear transformation of the activation function, the convolutional neural network can learn more complex and abstract feature representations. These features not only have stronger expressive power, but can also better adapt to data changes of different users and different time periods.
[0034] At the same time, the urine trace element time series input matrix is passed through the second urine trace element data time series feature analyzer to obtain the second scale urine trace element time series feature matrix. The second urine trace element data time series feature analyzer in the present application is a convolutional neural network model using a second scale two-dimensional convolution kernel, and the first scale is different from the second scale. It should be understood that the scale of the convolution kernel determines the area size of the input data covered by each convolution operation. Larger convolution kernels can capture a wider range of spatial (or temporal) features, while smaller convolution kernels are better at capturing details and local features. The second scale (usually larger than the first scale) convolution kernel can capture wider and more abstract features, which helps to enhance the expressive power of information in the urine trace element time series input matrix.
[0035] The first-scale urine trace element time series feature matrix is usually extracted using a smaller convolution kernel, which can capture the details and local features in the urine trace element time series data. These features are helpful for analyzing the changes, fluctuations and local anomalies of trace element content in a short period of time; the second-scale urine trace element time series feature matrix is extracted using a larger convolution kernel, which can capture broader and more abstract features, reflecting the trends and patterns over a longer period of time. These features help to understand the long-term changes in trace element content and the correlation between different elements; since the features of these two scales each provide different information, in order to obtain a more comprehensive and rich feature representation, the technical solution of this application fuses the first-scale urine trace element time series feature matrix and the second-scale urine trace element time series feature matrix to obtain the multi-scale urine trace element time series feature matrix. The multi-scale urine trace element time series feature matrix obtained by fusion contains not only detailed information, but also global information, which can better describe the characteristics of urine trace element time series data.
[0036] In an embodiment of the present application, a possible implementation method of fusing the first-scale urine trace element time series feature matrix and the second-scale urine trace element time series feature matrix to obtain the multi-scale urine trace element time series feature matrix is as follows: fusing the first-scale urine trace element time series feature matrix and the second-scale urine trace element time series feature matrix to obtain the multi-scale urine trace element time series feature matrix; wherein the cascade formula is:
[0037] M=Concat[M1,M2]
[0038] Among them, M1 represents the first-scale urine trace element time series feature matrix, M2 represents the second-scale urine trace element time series feature matrix, M represents the multi-scale urine trace element time series feature matrix, and Concat[·] represents the cascade function.
[0039] In step S140, the multi-scale urine trace element time series feature matrix is feature emphasized to obtain a urine trace element time series correlation feature emphasis matrix. Specifically, in an embodiment of the present application, the multi-scale urine trace element time series feature matrix is feature emphasized to obtain a urine trace element time series correlation feature emphasis matrix, including: passing the multi-scale urine trace element time series feature matrix through a urine trace element time series correlation feature emphasizer to obtain the urine trace element time series correlation feature emphasis matrix. In the present application, the urine trace element time series correlation feature emphasizer is a convolutional neural network model using a spatial attention mechanism. It should be understood that there is redundant information in the multi-scale urine trace element time series feature matrix obtained by fusion, and not all feature information in the feature matrix is equally important for judging whether the content of trace elements in the body of the user wearing the smart diaper is normal. In order to emphasize and highlight the more critical features for detecting abnormal trace elements in the body, a convolutional neural network model with a spatial attention mechanism is needed to process the multi-scale urine trace element time series feature matrix. The spatial attention mechanism can automatically learn the importance of different features in the spatial dimension and generate corresponding attention weights. These weights can reflect which features are more critical in temporal association, thereby emphasizing key features. Through the processing of the urine trace element temporal association feature emphasizer, the expression ability of the urine trace element temporal association feature can be further improved, thereby improving the accuracy of trace element abnormality identification.
[0040] In step S150, based on the urine trace element time series correlation feature emphasis matrix, the trace element monitoring results of the user wearing the smart diaper are obtained. Specifically, Figure 4 This is a flow chart of obtaining the monitoring results of trace elements in the body of a user wearing a smart diaper based on the urine trace element time series correlation feature emphasis matrix in the user intelligent monitoring method based on the smart diaper according to the embodiment of the present application. Figure 4 As shown, the monitoring result of trace elements in the body of the user wearing the smart diaper is obtained based on the urine trace element time series correlation feature emphasis matrix, including: S151, expanding the urine trace element time series correlation feature emphasis matrix to obtain a urine trace element time series correlation feature emphasis vector; S152, performing local feature offset correction based on feedback propagation representation on the urine trace element time series correlation feature emphasis vector to obtain an optimized urine trace element time series correlation feature emphasis vector; S153, passing the optimized urine trace element time series correlation feature emphasis vector through a urine analysis result generator based on a classifier to obtain a classification result, and the classification result is used to indicate whether the content of trace elements in the body of the user wearing the smart diaper is normal.
[0041] Considering that the urine trace element temporal correlation feature emphasis matrix has a high dimension, in order to simplify the data processing process and improve the calculation efficiency, it is necessary to expand the urine trace element temporal correlation feature emphasis matrix in the technical solution of the present application to obtain the urine trace element temporal correlation feature emphasis vector. It should be understood that in computer memory, the access of one-dimensional vectors is usually more efficient than that of matrices. Through the expansion operation, the data dimension is reduced, which is conducive to speeding up the data processing speed, thereby realizing more efficient real-time monitoring of urine.
[0042] In particular, during the measurement of urine trace elements, the detection equipment on the smart diaper will be affected by environmental factors, resulting in noise in the collected urine trace element data, and the individual's physiological state (such as diet, exercise, drugs, etc.) will affect the concentration of trace elements in urine. In addition, in some cases, the measurement data at a specific time point will be relatively small, which will lead to insufficient representativeness of local features, thereby introducing uncertainty. The influence of the above factors will lead to the presence of local feature disturbances in the urine trace element time series correlation feature emphasis vector obtained by encoding the collected urine trace element monitoring data. In order to reduce the possibility of classifiers reducing misclassification, thereby improving the accuracy of judging whether the urine trace element content is normal or not, it is necessary to improve the robustness of the classifier to local feature input disturbances. Therefore, in the technical solution of the present application, it is necessary to perform a local feature offset correction based on the feedback propagation representation on the urine trace element time series correlation feature emphasis vector to obtain an optimized urine trace element time series correlation feature emphasis vector.
[0043] Specifically, in an embodiment of the present application, the urine trace element time series association feature emphasis vector is subjected to a local feature offset correction based on a feedback propagation representation to obtain an optimized urine trace element time series association feature emphasis vector, including: determining a first weight matrix and a second weight matrix of the classifier before and after each iterative update; matrix multiplying the first weight matrix and the second weight matrix with the urine trace element time series association feature emphasis vector respectively to obtain a first urine trace element time series association feature vector and a second urine trace element time series association feature vector; calculating the positional difference between the first urine trace element time series association feature vector and the second urine trace element time series association feature vector to obtain a urine trace element time series association shift information representation vector; calculating the first weight matrix and the second weight matrix; and obtaining a urine trace element time series association feature vector. The F norm of the position mean vector between the first urine trace element time series association feature vector and the second urine trace element time series association feature vector is used as a shift compensation scaling factor; the urine trace element time series association shift information representation vector is linearly scaled by the shift compensation scaling factor to obtain a scaled urine trace element time series association shift information representation vector, and the scaled urine trace element time series association shift information representation vector is input into a Sigmoid activation function to obtain a urine trace element time series association backward propagation shift compensation representation vector; the position point multiplication of the urine trace element time series association backward propagation shift compensation representation vector and the urine trace element time series association feature emphasis vector is calculated to obtain the optimized urine trace element time series association feature emphasis vector.
[0044] In an embodiment of the present application, another expression of performing local feature offset correction based on the feedback propagation representation on the urine trace element time series association feature emphasis vector to obtain an optimized urine trace element time series association feature emphasis vector can be: processing the urine trace element time series association feature emphasis vector with the following formula to obtain the optimized urine trace element time series association feature emphasis vector; wherein the formula is:
[0045]
[0046] Wherein, M1 is the first weight matrix, M2 is the second weight matrix, V is the urine trace element time series correlation feature emphasis vector, represents matrix multiplication, Indicates adding by position point, ||·|| F represents the F norm of the feature vector, S represents the shift compensation scaling factor, ⊙ represents the point multiplication by position, represents point-by-point subtraction, sigmoid represents an activation function, and V' represents the optimized urine trace element temporal association feature emphasis vector.
[0047] In order to enhance the robustness of the classifier to local disturbances in the input feature vector, the urine trace element time series association feature emphasis vector is subjected to displacement compensation based on back propagation representation. The key to the displacement compensation based on back propagation representation of the urine trace element time series association feature emphasis vector is to capture the information of the weight matrix changing with the training process, and to use this information to generate the urine trace element time series association displacement information representation vector, and then generate the urine trace element time series association back propagation displacement compensation representation vector by scaling and nonlinear transformation (Sigmoid function). In this way, the urine trace element time series association feature emphasis vector is supported by the distribution difference of different classification scenarios in the training process of the classifier to describe the local displacement of the feature, so that it can not only quantify the displacement of the feature vector caused by the change of the weight matrix, but also generate a compensation vector according to the displacement information, the role of which is to weight and adjust each element in the feature vector according to the degree of influence of the disturbance, thereby reducing the influence of the disturbance on the classification result. In this way, the robustness of the classifier to local disturbances in the input data can be significantly improved, the stability and generalization ability of the classifier can be enhanced, and by reducing the impact of local disturbances on the classification results, the classifier can maintain a high classification accuracy when facing noisy or variant data.
[0048] Specifically, although the optimized urine trace element time series correlation feature emphasis vector contains rich feature information, this information is still the intermediate result after the original data has been processed in multiple steps. In order to convert these complex feature information into easy-to-understand classification results, such as "normal" or "abnormal", it is necessary to input the optimized urine trace element time series correlation feature emphasis vector into the classifier-based urine analysis result generator. As a machine learning model, the classifier can analyze and judge according to the input data and map it to different categories. That is, the normality of the trace element content in the body of the user wearing the smart diaper is judged based on the optimized urine trace element time series correlation feature emphasis vector information. In this way, by providing classification results in a timely and accurate manner, users can obtain timely feedback on their own health without having to go to the hospital for frequent testing, thereby improving the user experience.
[0049] In an embodiment of the present application, the optimized urine trace element time series association feature emphasis vector is passed through a urine analysis result generator based on a classifier to obtain a classification result, and the classification result is used to indicate whether the trace element content in the body of a user wearing a smart diaper is normal. One possible implementation method may be: using the fully connected layer of the classifier to perform fully connected encoding on the optimized urine trace element time series association feature emphasis vector to obtain a urine trace element time series association fully connected encoded feature vector; inputting the urine trace element time series association fully connected encoded feature vector into the Softmax classification function of the classifier to obtain the probability values belonging to each classification label in the optimized urine trace element time series association feature emphasis vector, and the classification labels include those used to indicate that the trace element content in the body of the user wearing the smart diaper is normal and those used to indicate that the trace element content in the body of the user wearing the smart diaper is abnormal; and determining the classification label corresponding to the largest of the probability values as the classification result.
[0050] In summary, the user intelligent monitoring method based on smart diapers according to the embodiment of the present application is explained, which adopts artificial intelligence-based data analysis technology, and performs time series feature analysis on the trace element data in the urine of the smart diapers at multiple predetermined time points within a predetermined time period, thereby judging whether the content of trace elements in the body of the user wearing the smart diapers is normal. In this way, intelligent monitoring of trace elements in the user's body is realized, which is conducive to timely obtaining feedback on the user's health status and provides a more convenient user health monitoring method.
[0051] Figure 5 FIG. 1 is a system block diagram of a user intelligent monitoring system based on smart diapers according to an embodiment of the present application. Figure 5As shown, according to the embodiment of the present application, the user intelligent monitoring system 100 based on the smart diaper includes: a user monitoring related data acquisition module 110, which is used to obtain the trace element data in the urine of the smart diaper at multiple predetermined time points within a predetermined time period, and the trace element data includes nitrite, urobilinogen, protein, occult blood, ketone bodies, bilirubin, glucose, ascorbic acid, urine calcium, and trace albumin; a user monitoring related data regularization module 120, which is used to regularize the trace element data in the urine of the smart diaper at the multiple predetermined time points to obtain the urine trace element time series output input matrix; a multi-scale feature encoding module 130 for user monitoring related data, used to perform multi-scale feature encoding on the urine trace element time series input matrix to obtain a multi-scale urine trace element time series feature matrix; a user monitoring multi-scale feature emphasis module 140, used to perform feature emphasis on the multi-scale urine trace element time series feature matrix to obtain a urine trace element time series correlation feature emphasis matrix; a user body trace element monitoring result generation module 150, used to obtain the trace element monitoring result of the user wearing the smart diaper based on the urine trace element time series correlation feature emphasis matrix.
[0052] Here, those skilled in the art can understand that the specific functions and operations of the various units and modules in the above-mentioned user intelligent monitoring system 100 based on smart diapers have been described in the above reference. Figures 1 to 4 The user intelligent monitoring method based on smart diapers has been introduced in detail, and therefore, its repeated description will be omitted.
[0053] In summary, the user intelligent monitoring system 100 based on the smart diaper according to the embodiment of the present application is explained, which adopts the data analysis technology based on artificial intelligence, and performs time series feature analysis on the trace element data in the urine of the smart diaper at multiple predetermined time points within a predetermined time period, thereby judging whether the content of trace elements in the body of the user wearing the smart diaper is normal. In this way, the intelligent monitoring of trace elements in the user's body is realized, which is conducive to timely obtaining feedback on the user's health status and provides a more convenient user health monitoring method.
[0054] As described above, the user intelligent monitoring system 100 based on smart diapers according to the embodiment of the present application can be implemented in various wireless terminals, such as a server for user intelligent monitoring based on smart diapers. In one example, the user intelligent monitoring system 100 based on smart diapers according to the embodiment of the present application can be integrated into the wireless terminal as a software module and / or a hardware module. For example, the user intelligent monitoring system 100 based on smart diapers can be a software module in the operating system of the wireless terminal, or can be an application developed for the wireless terminal; of course, the user intelligent monitoring system 100 based on smart diapers can also be one of the many hardware modules of the wireless terminal.
[0055] Alternatively, in another example, the smart diaper-based user intelligent monitoring system 100 and the wireless terminal may also be separate devices, and the smart diaper-based user intelligent monitoring system 100 may be connected to the wireless terminal via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.
[0056] The above is only a preferred embodiment of the present application and does not constitute any form of limitation to the present application. Although the present application has been disclosed as a preferred embodiment as above, it is not intended to limit the present application. Any technician familiar with the profession can make some changes or modifications to equivalent embodiments of the technical contents disclosed above without departing from the scope of the technical solution of the present application. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present application without departing from the content of the technical solution of the present application still falls within the scope of the technical solution of the present application.
Claims
1. A user intelligent monitoring method based on smart diapers, characterized in that: include: Acquire the trace element data in the urine of the smart diaper at multiple predetermined time points within a predetermined time period, wherein the trace element data includes nitrite, urobilinogen, protein, occult blood, ketone bodies, bilirubin, glucose, ascorbic acid, urine calcium, and trace albumin; Regularizing the data of trace elements in urine on the smart diapers at the plurality of predetermined time points to obtain a time series input matrix of trace elements in urine; Performing multi-scale feature encoding on the urine trace element time series input matrix to obtain a multi-scale urine trace element time series feature matrix; Performing feature emphasis on the multi-scale urine trace element time series feature matrix to obtain a urine trace element time series correlation feature emphasis matrix; Based on the urine trace element time series correlation feature emphasis matrix, the trace element monitoring results of the user wearing the smart diaper are obtained.
2. The user intelligent monitoring method based on smart diapers according to claim 1 is characterized in that: The trace element data in the urine on the smart diaper at the multiple predetermined time points are data regularized to obtain a urine trace element time series input matrix, including: arranging the trace element data in the urine on the smart diaper at the multiple predetermined time points according to the time dimension to obtain the urine trace element time series input matrix.
3. The user intelligent monitoring method based on smart diapers according to claim 2 is characterized in that: The urine trace element time series input matrix is subjected to multi-scale feature encoding to obtain a multi-scale urine trace element time series feature matrix, including: Passing the urine trace element time series input matrix through a first urine trace element data time series feature analyzer to obtain a first scale urine trace element time series feature matrix; Passing the urine trace element time series input matrix through a second urine trace element data time series feature analyzer to obtain a second scale urine trace element time series feature matrix; The first-scale urine trace element time series feature matrix and the second-scale urine trace element time series feature matrix are fused to obtain the multi-scale urine trace element time series feature matrix.
4. The user intelligent monitoring method based on smart diapers according to claim 3 is characterized in that: Feature emphasis is performed on the multi-scale urine trace element time series feature matrix to obtain a urine trace element time series correlation feature emphasis matrix, including: passing the multi-scale urine trace element time series feature matrix through a urine trace element time series correlation feature emphasizer to obtain the urine trace element time series correlation feature emphasis matrix.
5. The user intelligent monitoring method based on smart diapers according to claim 4 is characterized in that: The first urine trace element data temporal feature analyzer uses a convolutional neural network model with a first-scale two-dimensional convolution kernel, and the second urine trace element data temporal feature analyzer uses a convolutional neural network model with a second-scale two-dimensional convolution kernel. The first scale is different from the second scale, and the urine trace element temporal correlation feature emphasizer is a convolutional neural network model using a spatial attention mechanism.
6. The user intelligent monitoring method based on smart diapers according to claim 5, characterized in that: Based on the urine trace element time series correlation feature emphasis matrix, the trace element monitoring results of the user wearing the smart diaper are obtained, including: Expanding the urine trace element time series correlation feature emphasis matrix to obtain a urine trace element time series correlation feature emphasis vector; Performing local feature offset correction based on feedback propagation representation on the urine trace element time series correlation feature emphasis vector to obtain an optimized urine trace element time series correlation feature emphasis vector; The optimized urine trace element time series correlation feature emphasis vector is passed through a classifier-based urine analysis result generator to obtain a classification result, and the classification result is used to indicate whether the trace element content in the body of a user wearing the smart diaper is normal.
7. The user intelligent monitoring method based on smart diapers according to claim 6, characterized in that: The urine trace element time series correlation feature emphasis vector is subjected to local feature offset correction based on the feedback propagation representation to obtain an optimized urine trace element time series correlation feature emphasis vector, including: Determining a first weight matrix and a second weight matrix of the classifier before and after each iterative update; Performing matrix multiplication of the first weight matrix and the second weight matrix with the urine trace element time series association feature emphasis vector respectively to obtain a first urine trace element time series association feature vector and a second urine trace element time series association feature vector; Calculating the position difference between the first urine trace element time series association feature vector and the second urine trace element time series association feature vector to obtain a urine trace element time series association shift information representation vector; Calculating the F-norm of the position mean vector between the first urine trace element time series correlation feature vector and the second urine trace element time series correlation feature vector as a shift compensation scaling factor; Linearly scaling the urine trace element time series associated shift information representation vector with the shift compensation scaling factor to obtain a scaled urine trace element time series associated shift information representation vector, and inputting the scaled urine trace element time series associated shift information representation vector into a Sigmoid activation function to obtain a urine trace element time series associated backward propagation shift compensation representation vector; The optimized urine trace element time series association feature emphasis vector is obtained by calculating the position point multiplication of the urine trace element time series association backpropagation shift compensation representation vector and the urine trace element time series association feature emphasis vector.
8. A user intelligent monitoring system based on smart diapers, characterized in that: include: A user monitoring related data acquisition module is used to obtain trace element data in urine on the smart diaper at multiple predetermined time points within a predetermined time period, wherein the trace element data includes nitrite, urobilinogen, protein, occult blood, ketone bodies, bilirubin, glucose, ascorbic acid, urine calcium, and trace albumin; A user monitoring related data regularization module, used for regularizing the trace element data in the urine on the smart diaper at the plurality of predetermined time points to obtain a urine trace element time series input matrix; A multi-scale feature encoding module for user monitoring related data, used for performing multi-scale feature encoding on the urine trace element time series input matrix to obtain a multi-scale urine trace element time series feature matrix; A user monitoring multi-scale feature emphasis module is used to perform feature emphasis on the multi-scale urine trace element time series feature matrix to obtain a urine trace element time series correlation feature emphasis matrix; The module for generating the trace element monitoring results in the user's body is used to obtain the trace element monitoring results in the user wearing the smart diaper based on the urine trace element time series correlation feature emphasis matrix.
9. The user intelligent monitoring system based on smart diapers according to claim 8, characterized in that: The user monitoring related data regularization module is used to: arrange the trace element data in the urine on the smart diapers at the multiple predetermined time points according to the time dimension to obtain the urine trace element time series input matrix.
10. The user intelligent monitoring system based on smart diapers according to claim 9, characterized in that: The user's body trace element monitoring result generation module includes: A correlation emphasis feature matrix dimension reduction unit, used for expanding the urine trace element time series correlation feature emphasis matrix to obtain a urine trace element time series correlation feature emphasis vector; A correlation feature vector optimization unit, used for performing a local feature offset correction based on a feedback propagation representation on the urine trace element time series correlation feature emphasis vector to obtain an optimized urine trace element time series correlation feature emphasis vector; The optimized correlation feature parsing unit is used to pass the optimized urine trace element time series correlation feature emphasis vector through a classifier-based urine analysis result generator to obtain a classification result, and the classification result is used to indicate whether the trace element content in the body of the user wearing the smart diaper is normal.