Intelligent nutrition management method for endocrine patients

Abstract

The invention discloses an intelligent nutrition management method for an endocrine patient, and relates to the technical field of intelligent health management, and the method comprises the steps: transmitting collected physiological parameters to a mobile application program of mobile application equipment through employing a wireless transmission technology; collecting psychological data of the user, and analyzing the collected physiological parameters and psychological data by adopting a deep learning algorithm; and converting the transmitted brain wave mode by using a machine learning algorithm of the mobile application equipment to complete intelligent nutrition management on the health of the user. Through the multi-modal data fusion and brain wave transmission technology, the collection and analysis precision of physiological parameters and psychological data is remarkably improved, comprehensive real-time monitoring and intelligent management of the health state of the user are achieved, compared with the prior art, the emotional state and the physiological state of the user can be more accurately recognized, and the user experience is improved. Personalized health suggestions and early warning are provided, and the health management level and life quality of the user are effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent health management, and in particular to an intelligent nutrition management method for endocrine patients. Background Art

[0002] In today's society, with the accelerated pace of life and changes in environmental factors, more and more people are facing various health problems, especially diseases related to the endocrine system. Disorders of the endocrine system not only affect the physical health of individuals, but may also cause psychological problems, such as mood swings, anxiety and depression. These problems are often related to the dysfunction of specific areas of the brain, such as the amygdala. In addition, the incidence of brain diseases such as epilepsy is also increasing year by year. These diseases have a serious impact on the quality of life of patients. Therefore, the development of a technical solution that can effectively manage endocrine health while monitoring and improving brain function has become an urgent need of all sectors of society.

[0003] At present, although the existing health management technologies can provide a certain degree of support, such as assisting the treatment of endocrine diseases through dietary advice, there are still major deficiencies in accuracy and personalized services. First, most existing health management systems lack consideration of the patient's specific brain structure and functional status, and are unable to provide personalized nutritional interventions based on the specific conditions of different patients. Secondly, for patients with brain diseases such as epilepsy, existing technologies are difficult to achieve real-time monitoring and early warning of the condition, which greatly limits its application value in emergency situations. Furthermore, current technical solutions often neglect the maintenance of mental health and fail to fully consider the impact of factors such as mood swings on endocrine balance. In order to solve the above problems, the present invention adopts an intelligent nutritional management method for endocrine patients to provide a more comprehensive and personalized health management solution. Summary of the invention

[0004] In view of the problems existing in the existing intelligent nutrition management methods for endocrine patients, the present invention is proposed.

[0005] Therefore, the purpose of the present invention is to provide an intelligent nutrition management method for endocrine patients. The present invention aims to solve the problem of inaccurate collection and analysis of physiological parameters and psychological data in the prior art by adopting multimodal data fusion and brain wave transmission technology.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: In a first aspect, an embodiment of the present invention provides an intelligent nutrition management method for endocrine patients, which includes collecting physiological parameters of a user through a mobile application device, and sending the collected physiological parameters to a mobile application of the mobile application device using wireless transmission technology; Introducing emotion recognition modules based on mobile applications to collect users’ psychological data, and using deep learning algorithms to analyze the collected physiological parameters and psychological data; Through the system radio wave transmission function of the mobile application device, the analyzed physiological parameters and psychological data are transmitted to the user's brain through brain wave patterns, and the transmitted brain wave patterns are converted using the machine learning algorithm of the mobile application device, providing a data display interface for the mobile application device to complete intelligent nutritional management of the user's health.

[0007] As a preferred solution of the intelligent nutrition management method for endocrine patients described in the present invention, wherein: the collecting of the physiological parameters of the user includes installing a monitoring device inside the mobile application device, the monitoring device includes a heart rate monitoring device, a blood pressure monitoring device and a blood sugar monitoring device, the heart rate monitoring device includes using photoplethysmography to filter the noise generated by the user's behavior state, the behavior state includes a static state and a motion state; According to the automatic sampling frequency of the heart rate monitoring device, if the frequency is 0, the user is in a static state; if the frequency is not 0, the user is in a moving state, and the noise is filtered through the multi-wavelength optical sensor based on the different behavior states of the user; The filtering of the noise includes collecting light signals using a multi-wavelength optical sensor and preprocessing the collected light signals, wherein the preprocessing includes extracting the heart rate component monitored by the heart rate monitoring device by analyzing the time and frequency domain characteristics of the light signals, and removing the noise component related to the movement; The blood pressure monitoring device includes a pressure sensor installed inside the mobile application device, and the pressure sensor includes a dynamic calibration function to compare with the blood pressure meter and automatically correct the measurement result; The blood glucose monitoring device includes a micro-needle technology installed inside the mobile application device, which penetrates the skin surface through the micro-needle to collect the glucose concentration in the user's interstitial fluid.

[0008] As a preferred solution of the intelligent nutrition management method for endocrine patients of the present invention, wherein: the sending of the collected physiological parameters to the mobile application of the mobile application device includes using an environmental compensation algorithm to correct the monitoring error caused by the change of the external environment; When changes in the external environment are detected, the environmental compensation algorithm automatically adjusts the reference baseline of the optical signal to complete the calibration of the heart rate monitoring; When changes in the external environment are detected, the environmental compensation algorithm automatically adjusts the dynamic calibration function of the pressure sensor to complete the calibration of blood pressure monitoring; When changes in the external environment are detected, the environmental compensation algorithm automatically adjusts the position of the microneedle penetrating the skin surface to complete the calibration of blood pressure monitoring; The specific formula for automatic adjustment of the reference baseline environmental compensation algorithm is: in, Indicates the optical signal strength, Indicates the ambient temperature, Indicates the ambient humidity. Indicates the reference ambient temperature, Indicates the reference ambient humidity, represents the calibration coefficient for temperature, represents the calibration factor for humidity, and represents the polynomial order, Indicates the reference baseline of the optical signal intensity after calibration; The dynamic calibration adjustment formula of the blood pressure monitoring device is: in, Indicates the pressure sensor reading, Indicates the ambient temperature, Indicates the ambient humidity. Indicates the reference ambient temperature, Indicates the reference ambient humidity, represents the calibration coefficient for temperature, represents the calibration factor for humidity, and represents the polynomial order, Indicates the calibrated pressure sensor reading; The adjustment formula for the position where the micro needle penetrates the skin surface is: in, Indicates the depth of penetration of the micro needles into the skin surface. represents the reference penetration depth, Indicates the ambient temperature, Indicates the ambient humidity. Indicates the reference ambient temperature, Indicates the reference ambient humidity, represents the calibration coefficient for temperature, represents the calibration factor for humidity, and Represents the polynomial order.

[0009] As a preferred solution of the intelligent nutrition management method for endocrine patients described in the present invention, wherein: the introduction of the emotion recognition module includes acquiring user language by using an acoustic feature extraction method, and performing emotion classification on the acquired user language by using a pre-trained deep learning model, wherein the pre-trained deep learning model includes standardization of user language, data cleaning, and missing value filling; The standardization includes standardizing the extracted acoustic features using a Z-score standardization format, providing a visual interface to display a feature distribution graph of the user's language before and after standardization, and the user selects a standardization method through a mobile application to view the feature distribution graphs before and after standardization; The data cleaning includes using a noise reduction algorithm to remove background noise from the user's speech, and the mobile application automatically adjusts the noise reduction level to adapt to different external environment changes; The missing value filling includes removing missing values ​​in the user language background noise according to the mobile application's automatic detection noise reduction algorithm, and the mobile application sets the filling value according to the mean or median of the missing value.

[0010] As a preferred solution of the intelligent nutrition management method for endocrine patients described in the present invention, wherein: the use of a noise reduction algorithm to remove the background noise of the user's language includes decomposing the background noise into wavelet coefficients of different scales, introducing a threshold to process the wavelet coefficients, and completing the removal of the background noise; The calculation formula of the wavelet coefficient is: in, represents the speech signal, represents the wavelet basis function, represents the scale parameter, represents the translation parameter, Indicates at location and scale The wavelet coefficients at represents the time variable; The noise reduction level is determined according to the signal characteristics of the wavelet coefficients, and the calculation formula of the threshold is: in, represents a function, Indicates the maximum value, Indicates the threshold value; The background noise decomposition includes primary noise reduction, secondary noise reduction and tertiary noise reduction; When the background noise level When the background noise of the user's speech is removed, the first level noise reduction is selected; When the background noise level When the background noise of the user's speech is removed, the secondary noise reduction is selected; When the background noise level When the background noise of the user's speech is removed, the third level of noise reduction is selected.

[0011] As a preferred solution of the intelligent nutrition management method for endocrine patients described in the present invention, wherein: the analysis of the collected physiological parameters and psychological data includes multimodal data fusion of the collected physiological parameters and the denoised psychological data, and the multimodal data fusion includes labeling the pre-trained deep learning model; The annotation includes optimizing a pre-trained deep learning model through cross-validation of physiological parameters and denoised psychological data; The cross-validation method includes dividing the collected physiological parameters and denoised psychological data into K subsets, using K-1 subsets as training sets and using subsets other than K-1 as validation sets; The optimization of the pre-trained deep learning model includes repeating the pre-training of K subsets K times to complete multiple labeling of the pre-trained deep learning model.

[0012] As a preferred solution of the intelligent nutritional management method for endocrine patients described in the present invention, the transmission to the user's brain through brain wave patterns includes designing an electric wave transmission protocol, generating brain wave patterns using a pre-trained deep learning model, and converting the physiological parameters after multimodal data fusion and the psychological data after denoising into brain wave signals. The data display interface includes completing health risk warnings based on the evaluation results of the brain wave signals, recording the user's health change trends, and completing intelligent nutritional management of the user's health.

[0013] In a second aspect, an embodiment of the present invention provides an intelligent nutrition management system for endocrine patients, which includes: a collection module, which collects physiological parameters of a user through a mobile application device, and sends the collected physiological parameters to a mobile application of the mobile application device using wireless transmission technology; The analysis module introduces an emotion recognition module based on the mobile application, collects the user's psychological data, and uses a deep learning algorithm to analyze the collected physiological parameters and psychological data; The management module transmits the analyzed physiological parameters and psychological data to the user's brain through brain wave patterns through the system electric wave transmission function of the mobile application device, transforms the transmitted brain wave patterns using the machine learning algorithm of the mobile application device, provides a data display interface of the mobile application device, and completes intelligent nutritional management of the user's health.

[0014] In a third aspect, an embodiment of the present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the processor executes the computer program, any step of the above-mentioned intelligent nutrition management method for endocrine patients is implemented.

[0015] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the above-mentioned intelligent nutrition management method for endocrine patients is implemented.

[0016] The beneficial effects of the present invention are as follows: the present invention significantly improves the collection and analysis accuracy of physiological parameters and psychological data through multimodal data fusion and brain wave transmission technology, and realizes comprehensive real-time monitoring and intelligent management of the user's health status. Compared with the existing technology, the present invention can more accurately identify the user's emotional state and physiological condition, provide personalized health advice and early warnings, and effectively improve the user's health management level and quality of life. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them: Figure 1 A specific flow chart of an intelligent nutrition management method for endocrine patients provided by one embodiment of the present invention.

[0018] Figure 2 A schematic diagram of a management method for an intelligent nutrition management method for endocrine patients provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0020] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0021] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0022] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0023] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0024] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0025] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, provides an intelligent nutrition management method for endocrine patients, comprising: S1: The user's physiological parameters are collected through a mobile application device, and the collected physiological parameters are sent to a mobile application of the mobile application device using wireless transmission technology.

[0026] Wherein, collecting the physiological parameters of the user includes installing a monitoring device inside the mobile application device, the monitoring device includes a heart rate monitoring device, a blood pressure monitoring device and a blood sugar monitoring device, the heart rate monitoring device includes using a photoplethysmography method to filter the noise generated by the user's behavior state, and the behavior state includes a static state and a motion state; According to the automatic sampling frequency of the heart rate monitoring device, if the frequency is 0, the user is in a static state; if the frequency is not 0, the user is in a moving state, and the noise is filtered through the multi-wavelength optical sensor based on the different behavior states of the user; Filtering the noise includes collecting light signals using a multi-wavelength optical sensor, and preprocessing the collected light signals, wherein the preprocessing includes extracting the heart rate component monitored by the heart rate monitoring device by analyzing the time and frequency domain characteristics of the light signals, and removing the noise component related to the movement; The blood pressure monitoring device includes a pressure sensor installed inside the mobile application device, and the pressure sensor includes a dynamic calibration function to compare with the blood pressure meter and automatically correct the measurement result; The blood glucose monitoring device involves installing micro-needle technology inside a mobile application device, which penetrates the surface of the skin and collects glucose concentration in the user's interstitial fluid.

[0027] Furthermore, a variety of monitoring devices are installed inside the mobile application device, including a heart rate monitoring device, a blood pressure monitoring device and a blood sugar monitoring device. The heart rate monitoring device adopts photoplethysmography and collects optical signals through a multi-wavelength optical sensor, which can effectively filter out the noise generated by the user's behavior state. The heart rate monitoring device determines the user's behavior state according to the automatic sampling frequency; if the sampling frequency is 0 Hz, it indicates that the user is in a stationary state; if the sampling frequency is not 0 Hz, it indicates that the user is in motion. In the stationary state, a single wavelength optical sensor is used for monitoring; in the motion state, a multi-wavelength optical sensor is used to extract the heart rate component by analyzing the time and frequency domain characteristics of the optical signal, and at the same time remove the noise component related to the motion, and the signal-to-noise ratio can be improved to more than 30 dB.

[0028] The blood pressure monitoring device installs a pressure sensor inside the mobile application device, uses the dynamic calibration function to compare with the standard blood pressure monitor, and automatically corrects the measurement results. The dynamic calibration function ensures the accuracy of the measurement results by adjusting the sensitivity and offset of the sensor in real time. The measurement error can be controlled within ±3 mmHg.

[0029] The blood glucose monitoring device installs micro-needle technology inside the mobile application device, which penetrates the surface of the skin through the micro-needle to collect the glucose concentration in the user's interstitial fluid. The diameter of the micro-needle is about 0.2 mm and the penetration depth is about 1 mm. It can accurately collect blood glucose data without causing obvious pain. The sampling frequency of the micro-needle is once every 5 minutes to ensure the continuity and reliability of the data.

[0030] The collected physiological parameters are sent in real time to the mobile application of the mobile application device through wireless transmission technology. The transmission rate of wireless transmission technology is 1 Mbps, which ensures the efficiency and stability of data transmission. In addition, encryption technology is used during data transmission to protect user privacy and data security.

[0031] S1.1: sending the collected physiological parameters to a mobile application of a mobile application device including using an environmental compensation algorithm to correct monitoring errors caused by changes in the external environment; When changes in the external environment are detected, the environmental compensation algorithm automatically adjusts the reference baseline of the optical signal to complete the calibration of the heart rate monitoring; When changes in the external environment are detected, the environmental compensation algorithm automatically adjusts the dynamic calibration function of the pressure sensor to complete the calibration of blood pressure monitoring; When changes in the external environment are detected, the environmental compensation algorithm automatically adjusts the position of the microneedle penetrating the skin surface to complete the calibration of blood pressure monitoring; The specific formula for automatic adjustment of the environmental compensation algorithm based on the reference baseline is: in, Indicates the optical signal strength, Indicates the ambient temperature, Indicates the ambient humidity. Indicates the reference ambient temperature, Indicates the reference ambient humidity, represents the calibration coefficient for temperature, represents the calibration factor for humidity, and represents the polynomial order, Indicates the reference baseline of the optical signal intensity after calibration; The dynamic calibration adjustment formula for the blood pressure monitoring device is: in, Indicates the pressure sensor reading, Indicates the ambient temperature, Indicates the ambient humidity. Indicates the reference ambient temperature, Indicates the reference ambient humidity, represents the calibration coefficient for temperature, represents the calibration factor for humidity, and represents the polynomial order, Indicates the calibrated pressure sensor reading; The formula for adjusting the position where the micro needles penetrate the surface of the skin is: in, Indicates the depth of penetration of the micro needles into the skin surface. represents the reference penetration depth, Indicates the ambient temperature, Indicates the ambient humidity. Indicates the reference ambient temperature, Indicates the reference ambient humidity, represents the calibration coefficient for temperature, represents the calibration factor for humidity, and Represents the polynomial order.

[0032] Furthermore, when a change in the external environment is detected, the environmental compensation algorithm automatically adjusts the reference baseline of the optical signal to complete the calibration of the heart rate monitoring. For example, when the ambient temperature changes from 20°C to 25°C and the humidity changes from 50% to 60%, the algorithm automatically adjusts the reference baseline of the optical signal according to the preset temperature calibration coefficient of 0.02 and the humidity calibration coefficient of 0.01 to ensure the accuracy of the heart rate monitoring. When a change in the external environment is detected, the environmental compensation algorithm automatically adjusts the dynamic calibration function of the pressure sensor to complete the calibration of the blood pressure monitoring. For example, when the ambient temperature changes from 20°C to 25°C and the humidity changes from 50% to 60%, the algorithm automatically adjusts the reading of the pressure sensor according to the preset temperature calibration coefficient of 0.03 and the humidity calibration coefficient of 0.02 to ensure the accuracy of the blood pressure monitoring. The measurement error after calibration can be controlled within ±2 mmHg, when the external environment changes are detected, the environmental compensation algorithm automatically adjusts the position of the micro needle penetrating the skin surface to complete the calibration of blood glucose monitoring. For example, when the ambient temperature changes from 20°C to 25°C and the humidity changes from 50% to 60%, the algorithm will automatically adjust the penetration depth of the micro needle according to the preset temperature calibration coefficient of 0.01 and humidity calibration coefficient of 0.01 to ensure the accuracy of blood glucose monitoring. The penetration depth error after calibration can be controlled within ±0.05 mm. The environmental compensation algorithm calibration data is shown in Table 1 below: Table 1 Environmental compensation algorithm calibration data table Parameter Type Value before environmental change Value after environment change Temperature calibration coefficient Humidity calibration factor Error range after calibration Heart rate monitoring 20°C, 50% 25°C, 60% 0.02 0.01 - Blood pressure monitoring 20°C, 50% 25°C, 60% 0.03 0.02 ±2 mmHg Blood glucose monitoring 20°C, 50% 25°C, 60% 0.01 0.01 ±0.05 mm Table 1 shows that in the intelligent nutrition management method, the environmental compensation algorithm is used to correct the physiological parameter monitoring error caused by the external environment change. When the ambient temperature changes from 20°C to 25°C and the humidity changes from 50% to 60%, the heart rate monitoring automatically adjusts the reference baseline of the light signal through the temperature calibration coefficient of 0.02 and the humidity calibration coefficient of 0.01 to ensure the accuracy of the heart rate monitoring; the blood pressure monitoring automatically adjusts the reading of the pressure sensor through the temperature calibration coefficient of 0.03 and the humidity calibration coefficient of 0.02 to ensure the accuracy of the blood pressure monitoring, and the measurement error after calibration is controlled within ±2 mmHg; the blood sugar monitoring automatically adjusts the penetration depth of the micro needle through the temperature calibration coefficient of 0.01 and the humidity calibration coefficient of 0.01 to ensure the accuracy of the blood sugar monitoring, and the penetration depth error after calibration is controlled within ±0.05 mm. Through these environmental compensation measures, the present invention can effectively correct the monitoring errors caused by the external environment changes and ensure the accuracy and reliability of the physiological parameters.

[0033] S2: Introduce emotion recognition module based on mobile application to collect users’ psychological data, and use deep learning algorithm to analyze the collected physiological parameters and psychological data.

[0034] The introduction of the emotion recognition module includes obtaining user language by using an acoustic feature extraction method, and performing emotion classification on the obtained user language using a pre-trained deep learning model. The pre-trained deep learning model includes standardization of user language, data cleaning, and missing value filling. Standardization includes standardizing the extracted acoustic features using the Z-score standardization format, providing a visual interface to display the feature distribution graph before and after standardization of the user's language, and the user selects the standardization method through the mobile application to view the feature distribution graph before and after standardization; Data cleaning includes using noise reduction algorithms to remove background noise from user speech. At this time, the mobile application automatically adjusts the noise reduction level to adapt to different external environment changes; Missing value filling includes removing missing values ​​from the user's language background noise based on the mobile application's automatic detection noise reduction algorithm, and the mobile application sets the filling value based on the mean or median of the missing value.

[0035] Furthermore, in the intelligent nutrition management method, an emotion recognition module is introduced based on a mobile application to collect the user's psychological data, and a deep learning algorithm is used to analyze the collected physiological parameters and psychological data. Specifically, the emotion recognition module obtains the user's language through an acoustic feature extraction method, and uses a pre-trained deep learning model to perform emotion classification on the user's language. The pre-trained deep learning model includes standardization of the user's language, data cleaning, and missing value filling. The standardization process uses a Z-score standardization format to process the extracted acoustic features. The mobile application provides a visualization interface to display the feature distribution diagram of the user's language before and after standardization. The user can select a standardization method through the application and view the feature distribution diagram before and after standardization. For example, the feature value range after standardization is controlled between -3 and 3. The data cleaning process uses a noise reduction algorithm to remove the background noise of the user's language. The mobile application automatically adjusts the noise reduction level according to the ambient noise level. For example, when low-level noise reduction is used in a quiet environment, the noise level is set to 0.5 when the noise level is lower than 30 dB; when high-level noise reduction is used in a noisy environment, the noise level is higher than 70 dB. dB, the noise reduction level is set to 0.9. The missing value filling process automatically detects the noise reduction algorithm through the mobile application to remove the missing values ​​in the user's language background noise, and sets the filling value according to the mean or median of the missing value. For example, when the missing value ratio exceeds 10%, the median filling is used preferentially to ensure the integrity and accuracy of the data. Through these steps, the present invention can efficiently and accurately collect and analyze the user's physiological parameters and psychological data, and provide personalized health management and emotional support for the user.

[0036] S2.1: Using a noise reduction algorithm to remove the background noise of the user's speech includes decomposing the background noise into wavelet coefficients of different scales, introducing a threshold to process the wavelet coefficients, and completing the removal of the background noise; The calculation formula of wavelet coefficients is: in, represents the speech signal, represents the wavelet basis function, represents the scale parameter, represents the translation parameter, Indicates at location and scale The wavelet coefficients at represents the time variable; The noise reduction level is determined according to the signal characteristics of the wavelet coefficients, and the specific calculation formula of the threshold is: in, represents a function, Indicates the maximum value, Indicates the threshold value; The background noise decomposition includes primary noise reduction, secondary noise reduction and tertiary noise reduction; When the background noise level When the background noise of the user's speech is removed, the first level noise reduction is selected; When the background noise level When the background noise of the user's speech is removed, the secondary noise reduction is selected; When the background noise level When the background noise of the user's speech is removed, the third level of noise reduction is selected.

[0037] Furthermore, in the intelligent nutrition management method, the background noise of the user's language is removed by using a noise reduction algorithm, including decomposing the background noise into wavelet coefficients of different scales, and introducing a threshold to process the wavelet coefficients to complete the removal of the background noise. Specifically, the wavelet coefficients are calculated by wavelet transform, wherein the speech signal is decomposed into wavelet coefficients under different scales and translation parameters, the noise reduction level is determined according to the signal characteristics of the wavelet coefficients, the threshold is determined by calculating the maximum value of the wavelet coefficients, and the background noise decomposition includes primary noise reduction, secondary noise reduction and tertiary noise reduction, which are specifically classified as follows: when the background noise level is lower than 30 dB, primary noise reduction is selected, and the noise reduction threshold is set to 10; when the background noise level is between 30 dB and 60 dB, secondary noise reduction is selected, and the noise reduction threshold is set to 20; when the background noise level is higher than 60 dB, tertiary noise reduction is selected, and the noise reduction threshold is set to 30. The present invention can select a suitable noise reduction level according to different environmental noise levels, effectively remove background noise, and improve the clarity and quality of the speech signal.

[0038] S2.2: Analyzing the collected physiological parameters and psychological data includes multimodal data fusion of the collected physiological parameters and the denoised psychological data, and the multimodal data fusion includes annotating the pre-trained deep learning model; The annotation includes optimizing the pre-trained deep learning model through cross-validation of physiological parameters and denoised psychological data; The cross-validation method involves dividing the collected physiological parameters and denoised psychological data into K subsets, using K-1 subsets as training sets and the subsets other than K-1 as validation sets; Optimizing the pre-trained deep learning model includes repeating the pre-training on the K subsets K times to complete multiple labeling of the pre-trained deep learning model.

[0039] In the intelligent nutrition management method, the collected physiological parameters and psychological data are analyzed, including multimodal data fusion of the collected physiological parameters and denoised psychological data, and the multimodal data fusion is labeled through a pre-trained deep learning model. Specifically, the labeling process uses a cross-validation method to optimize the pre-trained deep learning model, and the collected physiological parameters and denoised psychological data are divided into 10 subsets, 9 subsets are used as training sets each time, and the remaining 1 subset is used as a validation set. The pre-training is repeated 10 times to complete multiple labeling of the pre-trained deep learning model. For example, each subset contains 1,000 data records, and there are a total of 10,000 data records. Through 10 cross-validations, it is ensured that each subset is used as a validation set once, thereby improving the generalization ability and accuracy of the model. Through these steps, the present invention can effectively integrate multimodal data, optimize the performance of deep learning models, and provide users with more accurate health management and emotional support.

[0040] S3: Through the system radio wave transmission function of the mobile application device, the analyzed physiological parameters and psychological data are transmitted to the user's brain through the brain wave pattern. The transmitted brain wave pattern is transformed using the machine learning algorithm of the mobile application device, and a data display interface of the mobile application device is provided to complete the intelligent nutrition management of the user's health.

[0041] Among them, transmitting to the user's brain through brainwave patterns includes designing an electric wave transmission protocol, using a pre-trained deep learning model to generate brainwave patterns, converting the physiological parameters after multimodal data fusion and the psychological data after denoising into brainwave signals, and the data display interface includes completing health risk warnings based on the evaluation results of brainwave signals, recording the user's health change trends, and completing intelligent nutritional management of the user's health.

[0042] In a preferred embodiment, an intelligent nutrition management system for endocrine patients includes a collection module, which collects the user's physiological parameters through a mobile application device, and sends the collected physiological parameters to a mobile application of the mobile application device using wireless transmission technology; an analysis module, which introduces an emotion recognition module based on the mobile application, collects the user's psychological data, and uses a deep learning algorithm to analyze the collected physiological parameters and psychological data; a management module, which transmits the analyzed physiological parameters and psychological data to the user's brain through brain wave patterns through the system radio wave transmission function of the mobile application device, converts the transmitted brain wave patterns using the machine learning algorithm of the mobile application device, provides a data display interface for the mobile application device, and completes intelligent nutrition management of the user's health.

[0043] The above-mentioned unit modules may be embedded in or independent of a processor in a computer device in the form of hardware, or may be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to the above-mentioned modules.

[0044] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0045] In summary, the present invention significantly improves the collection and analysis accuracy of physiological parameters and psychological data through multimodal data fusion and brainwave transmission technology, and realizes comprehensive real-time monitoring and intelligent management of the user's health status. Compared with the existing technology, the present invention can more accurately identify the user's emotional state and physiological condition, provide personalized health advice and warnings, and effectively improve the user's health management level and quality of life.

[0046] Example 2, reference Figure 1 and Figure 2 , which is the second embodiment of the present invention, provides an intelligent nutrition management method for endocrine patients. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through simulation experiments.

[0047] During the experiment of the intelligent nutrition management method, the user's physiological parameters, including heart rate, blood pressure and blood sugar, were first collected through a mobile application device, and then sent to the mobile application of the mobile application device using wireless transmission technology. Specifically, the heart rate monitoring device adopted photoplethysmography, and collected optical signals through multi-wavelength optical sensors when the user was stationary and in motion, and the signal-to-noise ratio was increased to more than 35 dB; the blood pressure monitoring device used a dynamic calibration function to ensure that the measurement error was controlled within ±2.5 mmHg; the blood sugar monitoring device penetrated the surface of the skin through a micro-needle, and the sampling frequency was once every 4 minutes to ensure the continuity and reliability of the data.

[0048] An emotion recognition module is introduced based on a mobile application to collect users' psychological data. A deep learning algorithm is used to analyze the collected physiological parameters and psychological data. The emotion recognition module obtains user language through an acoustic feature extraction method, and uses a pre-trained deep learning model to perform emotion classification on the user language. The pre-trained deep learning model includes standardization of user language, data cleaning, and missing value filling. The standardization process uses the Z-score standardization format, and the feature value range is controlled between -2.5 and 2.5; the data cleaning process uses a noise reduction algorithm to remove background noise, and automatically adjusts the noise reduction level according to the ambient noise level. For example, when the noise level is 40 dB, the noise reduction level is set to 0.6; the missing value filling process automatically detects the noise reduction algorithm through a mobile application to remove missing values ​​in the background noise of the user's language, and sets the filling value according to the mean or median of the missing value. When the proportion of missing values ​​exceeds 12%, median filling is used preferentially.

[0049] The collected physiological parameters and denoised psychological data were subjected to multimodal data fusion and annotated by a pre-trained deep learning model. The annotation process used a cross-validation method to optimize the pre-trained deep learning model. The collected physiological parameters and denoised psychological data were divided into 8 subsets, 7 subsets were used as training sets each time, and the remaining 1 subset was used as a validation set. The pre-training was repeated 8 times to complete the multiple annotation of the pre-trained deep learning model. Each subset contained 1200 data records, and there were 9600 data records in total. Through 8 cross-validations, it was ensured that each subset was used as a validation set once. The average accuracy of the model reached 92%, and the F1 score reached 0.91. The experimental data of the present invention are shown in Table 2 below: Table 2 Experimental data table of the present invention Parameter Type Specific data describe Heart rate monitoring The signal-to-noise ratio is increased to more than 35 dB Signal-to-noise ratio of multi-wavelength optical sensors in static and moving states Blood pressure monitoring The measurement error is controlled within ±2.5 mmHg Dynamic calibration function ensures measurement error Blood glucose monitoring The sampling frequency is once every 4 minutes The sampling frequency of microneedles penetrating the skin surface Emotion Recognition The eigenvalue range is controlled between -2.5 and 2.5 Z-score standardized format Noise Reduction Algorithm When the noise level is 40 dB, the noise reduction level is set to 0.6 Automatically adjusts noise cancellation level based on ambient noise level Missing value filling When the missing value ratio exceeds 12%, the median is preferred. Automatically detect and fill missing values Multimodal data fusion 8 subsets, each with 1200 data records Data division for cross validation Cross Validation Average accuracy 92%, F1 score 0.91 The results of 8-fold cross validation Table 2 includes the physiological parameters and psychological data of users collected by mobile application devices during the experiment of the intelligent nutrition management method. The multimodal data fusion adopts an 8-fold cross-validation optimization model, with 1,200 data in each subset, an average model accuracy of 92%, and an F1 score of 0.91. It should be noted that the above embodiments are only used to illustrate the technical solution of the present invention and are not limited thereto. Although the present invention is described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention. The comparison between the present invention and the prior art is shown in Table 3 below: Table 3 Comparison table between the present invention and the prior art Compare Projects Prior art The present invention Advantages Physiological parameter monitoring Uses simple sensors with low signal-to-noise ratio Using multi-wavelength optical sensors, the signal-to-noise ratio is significantly improved Higher signal-to-noise ratio, higher data quality Psychological data collection The emotion recognition method is single and the background noise processing effect is limited Introducing deep learning models and multi-level noise reduction algorithms More accurate emotion recognition, better noise processing Multimodal data fusion Lack of effective data fusion methods Multimodal data fusion using pre-trained deep learning models More efficient multimodal data integration to improve analysis accuracy Environmental compensation algorithm No or simple environmental compensation Automatically adjusted environmental compensation algorithm More accurate environmental adaptation, reducing monitoring errors Data Transfer Low transmission rate and average security High-speed wireless transmission, data encryption Faster data transfer, higher data security Model Optimization Lack of systematic model optimization methods K-fold cross validation method is used for model optimization Higher model generalization ability and stability user interface The user interface has a single function and poor interactivity Provides rich visual interface and interactive functions Better user experience, higher user satisfaction Health Tips Generic health advice lacks personalization Provide personalized health advice and warnings More personalized health management, more effective health intervention Data display The data display format is single and the amount of information is limited Comprehensive data display interface to record health trends More comprehensive data display, more detailed health records System Integration Each module is independent and has low integration Highly integrated system, each module works together More efficient system operation, better overall performance Table 3 includes the significant improvement in the accuracy of collection and analysis of physiological parameters and psychological data through advanced multimodal data fusion, environmental compensation algorithm, deep learning model optimization and highly integrated system design. At the same time, personalized health advice and rich user interface are provided to enhance the user experience and health management level. Compared with the existing technology, the present invention has obvious advantages in data quality, analysis accuracy, user interaction and system integration.

Claims

1. An intelligent nutrition management method for endocrine patients, characterized in that: include: The user's physiological parameters are collected through the mobile application device, and the collected physiological parameters are sent to the mobile application of the mobile application device using wireless transmission technology; Introducing emotion recognition modules based on mobile applications to collect users’ psychological data, and using deep learning algorithms to analyze the collected physiological parameters and psychological data; Through the system radio wave transmission function of the mobile application device, the analyzed physiological parameters and psychological data are transmitted to the user's brain through brain wave patterns, and the transmitted brain wave patterns are converted using the machine learning algorithm of the mobile application device, providing a data display interface for the mobile application device to complete intelligent nutritional management of the user's health.

2. The intelligent nutrition management method for endocrine patients according to claim 1, characterized in that: The collecting of the physiological parameters of the user includes installing a monitoring device inside the mobile application device, the monitoring device includes a heart rate monitoring device, a blood pressure monitoring device and a blood sugar monitoring device, the heart rate monitoring device includes using photoplethysmography to filter the noise generated by the user's behavior state, the behavior state includes a static state and a motion state; According to the automatic sampling frequency of the heart rate monitoring device, if the frequency is 0, the user is in a static state; if the frequency is not 0, the user is in a moving state, and the noise is filtered through the multi-wavelength optical sensor based on the different behavior states of the user; The filtering of the noise includes collecting light signals using a multi-wavelength optical sensor and preprocessing the collected light signals, wherein the preprocessing includes extracting the heart rate component monitored by the heart rate monitoring device by analyzing the time and frequency domain characteristics of the light signals, and removing the noise component related to the movement; The blood pressure monitoring device includes a pressure sensor installed inside the mobile application device, and the pressure sensor includes a dynamic calibration function to compare with the blood pressure meter and automatically correct the measurement result; The blood glucose monitoring device includes a micro-needle technology installed inside the mobile application device, which penetrates the skin surface through the micro-needle to collect the glucose concentration in the user's interstitial fluid.

3. The intelligent nutrition management method for endocrine patients according to claim 2, characterized in that: The sending of the collected physiological parameters to the mobile application of the mobile application device includes using an environmental compensation algorithm to correct monitoring errors caused by changes in the external environment; When changes in the external environment are detected, the environmental compensation algorithm automatically adjusts the reference baseline of the optical signal to complete the calibration of the heart rate monitoring; When changes in the external environment are detected, the environmental compensation algorithm automatically adjusts the dynamic calibration function of the pressure sensor to complete the calibration of blood pressure monitoring; When changes in the external environment are detected, the environmental compensation algorithm automatically adjusts the position of the microneedle penetrating the skin surface to complete the calibration of blood pressure monitoring; The automatic adjustment formula of the environmental compensation algorithm of the reference baseline is: in, Indicates the optical signal strength, Indicates the ambient temperature, Indicates the ambient humidity. Indicates the reference ambient temperature, Indicates the reference ambient humidity, represents the calibration coefficient for temperature, represents the calibration factor for humidity, and represents the polynomial order, Indicates the reference baseline of the optical signal intensity after calibration; The dynamic calibration adjustment formula of the blood pressure monitoring device is: in, Indicates the pressure sensor reading, Indicates the ambient temperature, Indicates the ambient humidity. Indicates the reference ambient temperature, Indicates the reference ambient humidity, represents the calibration coefficient for temperature, represents the calibration factor for humidity, and represents the polynomial order, Indicates the calibrated pressure sensor reading; The adjustment formula for the position where the micro needle penetrates the skin surface is: in, Indicates the depth of penetration of the micro needles into the skin surface. represents the reference penetration depth, Indicates the ambient temperature, Indicates the ambient humidity. Indicates the reference ambient temperature, Indicates the reference ambient humidity, represents the calibration coefficient for temperature, represents the calibration factor for humidity, and Represents the polynomial order.

4. The intelligent nutrition management method for endocrine patients according to claim 3, characterized in that: The introduction of the emotion recognition module includes acquiring user language by using an acoustic feature extraction method, and performing emotion classification on the acquired user language by using a pre-trained deep learning model, wherein the pre-trained deep learning model includes standardization of the user language, data cleaning, and missing value filling; The standardization includes standardizing the extracted acoustic features using a Z-score standardization format, providing a visual interface to display a feature distribution graph of the user's language before and after standardization, and the user selects a standardization method through a mobile application to view the feature distribution graphs before and after standardization; The data cleaning includes using a noise reduction algorithm to remove background noise from the user's speech, and the mobile application automatically adjusts the noise reduction level to adapt to different external environment changes; The missing value filling includes removing missing values ​​in the user language background noise according to the mobile application's automatic detection noise reduction algorithm, and the mobile application sets the filling value according to the mean or median of the missing value.

5. The intelligent nutrition management method for endocrine patients according to claim 4, characterized in that: The method of removing the background noise of the user's speech by using a noise reduction algorithm includes decomposing the background noise into wavelet coefficients of different scales, introducing a threshold to process the wavelet coefficients, and completing the removal of the background noise; The calculation formula of the wavelet coefficient is: in, represents the speech signal, represents the wavelet basis function, represents the scale parameter, represents the translation parameter, Indicates at location and scale The wavelet coefficients at represents the time variable; The noise reduction level is determined according to the signal characteristics of the wavelet coefficients, and the calculation formula of the threshold is: in, represents a function, Indicates the maximum value, Indicates the threshold value; The background noise decomposition includes primary noise reduction, secondary noise reduction and tertiary noise reduction; When the background noise level When the background noise of the user's speech is removed, the first level noise reduction is selected; When the background noise level When the background noise of the user's speech is removed, the secondary noise reduction is selected; When the background noise level When the background noise of the user's speech is removed, the third level of noise reduction is selected.

6. The intelligent nutrition management method for endocrine patients according to claim 5, characterized in that: The analyzing of the collected physiological parameters and psychological data includes multimodal data fusion of the collected physiological parameters and the denoised psychological data, and the multimodal data fusion includes labeling the pre-trained deep learning model; The annotation includes optimizing a pre-trained deep learning model through cross-validation of physiological parameters and denoised psychological data; The cross-validation method includes dividing the collected physiological parameters and denoised psychological data into K subsets, using K-1 subsets as training sets and using subsets other than K-1 as validation sets; The optimization of the pre-trained deep learning model includes repeating the pre-training of K subsets K times to complete multiple labeling of the pre-trained deep learning model.

7. The intelligent nutrition management method for endocrine patients according to claim 6, characterized in that: The transmission to the user's brain through brainwave patterns includes designing an electric wave transmission protocol, generating brainwave patterns using a pre-trained deep learning model, converting the physiological parameters after multimodal data fusion and the psychological data after denoising into brainwave signals, and the data display interface includes completing health risk warnings based on the evaluation results of the brainwave signals, recording the user's health change trends, and completing intelligent nutritional management of the user's health.

8. An intelligent nutrition management system for endocrine patients, based on the intelligent nutrition management method for endocrine patients according to any one of claims 1 to 7, characterized in that: include: A collection module collects the user's physiological parameters through a mobile application device and sends the collected physiological parameters to a mobile application of the mobile application device using wireless transmission technology; The analysis module introduces an emotion recognition module based on the mobile application, collects the user's psychological data, and uses a deep learning algorithm to analyze the collected physiological parameters and psychological data; The management module transmits the analyzed physiological parameters and psychological data to the user's brain through brain wave patterns through the system electric wave transmission function of the mobile application device, transforms the transmitted brain wave patterns using the machine learning algorithm of the mobile application device, provides a data display interface of the mobile application device, and completes intelligent nutritional management of the user's health.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the intelligent nutrition management method for endocrine patients described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the intelligent nutrition management method for endocrine patients described in any one of claims 1 to 7 are implemented.

Images