An intelligent auxiliary system and method for dermatological rehabilitation nursing

Abstract

This invention discloses an intelligent auxiliary system and method for dermatological rehabilitation and nursing. The invention relates to the field of dermatological rehabilitation and nursing technology, and includes a symptom recognition module for acquiring image data of the patient's skin through image acquisition equipment and multispectral imaging technology; it is also used to identify skin disease symptoms. This intelligent auxiliary system and method for dermatological rehabilitation and nursing, through image preprocessing and enhancement unit processing, improves the quality and realism of images, providing a reliable foundation for accurate identification of skin disease symptoms. By combining machine learning algorithms and a skin disease symptom database for comparison and identification, it can generate symptom data more accurately. It can identify the potential impact of the patient's lifestyle habits on skin symptoms at multiple levels, providing a scientific basis for developing personalized rehabilitation recommendations. It can dynamically adjust rehabilitation recommendations based on the patient's lifestyle data, improving the pertinence and effectiveness of the rehabilitation recommendations.

Description

Technical Field

[0001] The present invention relates to the technical field of dermatology rehabilitation nursing, and in particular to an intelligent auxiliary system and method for dermatology rehabilitation nursing. Background Art

[0002] With the improvement of people's living standards and changes in the environment, the incidence of skin diseases is rising, and the demand for dermatological diagnosis and treatment is also increasing. Traditional dermatological rehabilitation care mainly relies on the experience of doctors and the self-management of patients, and often has the following problems: On the one hand, when diagnosing and treating skin diseases, doctors mainly rely on naked eye observation and patient descriptions. It is difficult to comprehensively and accurately identify and evaluate the symptoms and changes of skin diseases. Moreover, it is difficult for doctors to understand the impact of patients' habits and behaviors in daily life on skin diseases, and it is difficult to give personalized rehabilitation advice. On the other hand, patients often lack professional guidance and supervision during the rehabilitation care process, and it is difficult to adhere to correct living habits and treatment plans. For example, patients may not know which diets, work and rest schedules and contacts will affect the recovery of skin diseases, and may forget to take medicines, apply ointments or perform care on time, affecting the treatment effect and patient satisfaction. Summary of the invention

[0003] In order to solve the above technical problems, the present invention is implemented through the following technical solutions: an intelligent auxiliary system and method for dermatological rehabilitation nursing, including a symptom recognition module, which is used to obtain image data of the patient's skin through an image acquisition device and multi-spectral imaging technology; it is also used to identify skin disease symptoms, including the type of skin disease, the shape, color, distribution range of the rash, and the degree of eczema, and generate symptom data; wherein the image data is shot at multiple angles and under different light to ensure that the system comprehensively collects skin disease symptoms;

[0004] The habit collection module is connected to the symptom recognition module, and collects the patient's living habit data based on the patient's living habit questionnaire and sensor equipment. The living habit data includes the corresponding data of diet, work and rest rules and contact objects;

[0005] The collection of lifestyle data involves real-time monitoring of patients’ sleep quality, activity levels, and dietary records through wearable devices, as well as collecting patients’ self-reported daily conditions through mobile phone applications. The lifestyle data is then improved based on the daily conditions to ensure the integrity and accuracy of the data.

[0006] The data processing module is used to correlate symptom data with lifestyle data and identify the potential impact of patients’ lifestyle habits on skin symptoms;

[0007] It is also used to fill in missing values ​​in lifestyle data, remove outliers, and normalize various lifestyle data according to standardized algorithms for subsequent data analysis;

[0008] The rehabilitation suggestion module generates rehabilitation suggestions for skin diseases based on the potential impacts analyzed by the data processing module, pushes the rehabilitation suggestions to patients through mobile phone applications, and sets a reminder mechanism to ensure that patients implement the rehabilitation suggestions on time.

[0009] Preferably, the symptom recognition module further includes an image preprocessing unit and an image enhancement unit:

[0010] The image preprocessing unit is used to process the image data of the patient's skin based on denoising, color balancing and edge detection technology to obtain real image data to improve the accuracy of recognition;

[0011] The image enhancement unit is used to enhance the real image data through image detail compensation technology when the image resolution is low or the light conditions are poor, to obtain multispectral image data and ensure the accuracy of symptom recognition;

[0012] The symptom recognition module also compares and identifies multispectral image data with existing skin disease big data based on machine learning algorithms and a skin disease symptom database to obtain symptom data and achieve high-precision diagnosis of different types of skin diseases.

[0013] Preferably, the habit collection module includes a wearable device and a mobile phone application:

[0014] Wearable devices are used to monitor the patient's body temperature, heart rate, sleep status, daily exercise and diet information in real time, obtain the patient's living habit data, and upload the living habit data to the system database;

[0015] The mobile phone application is used to regularly push lifestyle questionnaires to patients. Patients fill out the questionnaires through voice input to obtain the patients' self-reported daily conditions, including eating habits, work pressure, skin care product usage and allergic reactions. The lifestyle data obtained through wearable devices are supplemented based on the patients' self-reported daily conditions to obtain complete lifestyle data.

[0016] The application also has a daily reminder function to urge patients to complete the health log on time and ensure the comprehensiveness of lifestyle data.

[0017] Preferably, the process of supplementing the life habit data obtained through the wearable device based on the patient's self-reported daily conditions to obtain complete life habit data is as follows:

[0018] By comparing and analyzing the patient's self-reported daily conditions collected by the mobile phone application with the life habit data uploaded by the wearable device, it can be checked whether the dietary information recorded by the wearable device is consistent with the patient's dietary habits described in the questionnaire, and whether the sleep conditions monitored by the wearable device are consistent with the patient's self-reported impact of work pressure on sleep; when the patient provides information in the questionnaire that is not monitored by the wearable device, the information is marked as pending registration information, and the pending registration information is supplemented to the life habit data; then the supplemented patient's self-reported daily conditions and the life habit data of the wearable device are integrated to form complete life habit data.

[0019] Preferably, the data processing module includes:

[0020] Impact Identification Unit: Using algorithms based on time series analysis and multivariate regression, we conduct multi-level correlation analysis between symptom data and lifestyle data to identify the potential impact of patients’ lifestyles on skin symptoms, including the impact of patients’ diet, daily routine, and contact materials on the frequency and severity of skin disease attacks;

[0021] Correction reminder unit: Generates reminder information based on abnormal values ​​in lifestyle data, and sends the reminder information in the form of pop-up information to the wearable device of the habit collection module, prompting the patient to reconfirm or correct the lifestyle data, where the pop-up information includes confirmation and correction.

[0022] Preferably, the process of identifying the potential impact of the patient's lifestyle on skin symptoms is:

[0023] Step 1: Time series analysis:

[0024] The symptom data and lifestyle data are divided into different time periods in chronological order, using days, weeks and months as units, to form symptom time series data sets and life time series data sets. The symptom time series data sets are then analyzed to observe the frequency and severity of skin disease attacks and obtain the symptom change trend. At the same time, the time series of diet, work and rest, and contact objects in the life time series data sets are analyzed to obtain the law of life changes.

[0025] Obtaining the law of life changes: using the comprehensive correlation coefficient to calculate the correlation coefficient between the time series of symptom data and life habit data, detecting the correlation between them, and obtaining the comprehensive coefficient, that is, the result of time series analysis. The comprehensive coefficient includes the Pearson coefficient and the Spearman coefficient. There are two cases when using the comprehensive correlation coefficient to calculate the period;

[0026] One method: When the symptom data and the lifestyle data are in a linear relationship, the Pearson correlation coefficient is used to calculate the correlation coefficient between the time series of the symptom data and the lifestyle data, and the correlation between them is detected to obtain the Pearson coefficient;

[0027] Another method is to use the Spearman correlation coefficient to calculate the correlation coefficient between the time series of symptom data and lifestyle data when the relationship between the symptom data and lifestyle data is nonlinear, and to detect the correlation between them to obtain the Spearman coefficient;

[0028] Among them, the Pearson correlation coefficient calculates the correlation coefficient between the time series of symptom data and lifestyle data, detects the correlation between them, and obtains the Pearson coefficient; the Pearson coefficient is between -1 and 1;

[0029] The calculation formula of Pearson correlation coefficient is: ;

[0030] in, and There are two variables: the observed values ​​of lifestyle data and the observed values ​​of symptom data; and There are two variables: the sample mean of lifestyle data and the sample mean of symptom data; in addition, the Spearman coefficient ranges from -1 to +1, where:

[0031] The calculation formula of Spearman correlation coefficient is: ;

[0032] Where: d i is the difference between the levels of two variables, where the two variables refer to lifestyle data and symptom data respectively; n is the number of data pairs based on lifestyle data and symptom data in the two variables;

[0033] The Spearman correlation coefficient is calculated as follows: assign a rank to each observation of each variable; calculate the difference in ranks d for each pair of observations i ; Square all the level differences and then sum them; Substitute the result into the above formula to calculate r s ;

[0034] Step 2: Algorithm analysis of multivariate regression:

[0035] The various factors in the lifestyle data were used as independent variables, and the skin symptom data were used as dependent variables; when the dependent variable was a continuous variable and had a linear relationship with the independent variable, a linear regression model was selected based on the characteristics and data type of the skin symptoms, and the model parameters were calculated by solving the normal equations; when the observations were independent of each other, a polynomial regression model was selected based on the characteristics and data type of the skin symptoms, and the model parameters were calculated by the least squares method;

[0036] The coefficient of determination R² is used to evaluate the degree of fit of the model to the data. The value of R² is between 0 and 1. The closer R² is to 1, the better the model fits the data. When R² is close to 1, it means that the degree of fit is good and the prediction ability is strong, indicating that the model can effectively identify the potential impact of lifestyle habits on skin symptoms. According to the parameter estimation results of the regression model, the direction and degree of the impact of various lifestyle factors on skin symptoms are explained to obtain the regression analysis results, that is, the results of multivariate regression analysis.

[0037] The results of the regression analysis are as follows:

[0038] If the regression coefficient of the lifestyle factor is positive, it means that the factor is positively correlated with skin symptoms, that is, the increase of this factor will lead to the aggravation of skin symptoms;

[0039] On the contrary, if the regression coefficient of the lifestyle factor is negative, it indicates that the factor is negatively correlated with skin symptoms, that is, the increase of this factor will alleviate skin symptoms.

[0040] Preferably, the rehabilitation suggestion module generates rehabilitation suggestions for skin diseases based on the potential impacts generated by the data processing module and in combination with the preferred practice solutions in the skin disease rehabilitation database:

[0041] Rehabilitation suggestions include lifestyle adjustments, diet plans, medication recommendations, skin care product recommendations, exercise recommendations, and the rehabilitation suggestions are adjusted based on the patient's lifestyle data;

[0042] The rehabilitation advice module includes a reminder unit, which is used to push rehabilitation advice to patients through a mobile phone application on a regular basis based on a preset reminder mechanism to ensure that patients comply with the treatment plan in their daily lives.

[0043] Preferably, the process of regularly pushing rehabilitation suggestions to patients through a mobile phone application based on a preset reminder mechanism is as follows:

[0044] Determine the reminder time interval based on the patient's condition and rehabilitation needs. For patients who need to strictly adhere to a specific diet plan, you can set reminders before the three meals a day; for patients who need to take medication on time, you can set reminders every few hours; select multiple reminder methods based on the patient's usage habits and preferences to remind the patient, including sound prompts, vibrations, and pop-up notifications to ensure that the reminder can effectively attract the patient's attention; and adjust the reminder mechanism based on the patient's situation and needs to obtain the adjusted preset reminder mechanism. For patients with poor sleep quality, reminders can be set during the quieter time periods during the day to avoid affecting the patient's rest;

[0045] Then, the reminder unit and the mobile application are interfaced to ensure that the rehabilitation suggestions can be accurately pushed to the mobile application; when the rehabilitation suggestions are generated based on the reminder unit, they are transmitted to the mobile application via the network communication protocol; after the mobile application receives the rehabilitation suggestions, it stores them in the local database so as to remind when needed;

[0046] Set a timer in the mobile phone application to trigger a reminder according to a preset reminder time interval; wherein the timer is implemented based on the scheduled task function provided by the mobile phone operating system;

[0047] When the timer is triggered, the mobile phone application checks whether the triggering conditions of the reminder are currently met; determines whether the current time is within the preset reminder time range, whether the patient is in a specific state (such as resting or busy), etc.; if the triggering conditions are met, a reminder is given; otherwise, wait for the next time the timer is triggered; according to the preset reminder mechanism, the mobile phone application performs the reminder operation; if set to sound reminder, a specific reminder sound is played; if set to vibration, the vibration function of the mobile phone is triggered; if set to pop-up notification, a pop-up window of rehabilitation suggestions is displayed on the mobile phone screen; while reminding, the mobile phone application displays it in the form of pictures so that the patient can clearly understand the rehabilitation measures that need to be taken and the specific content of the rehabilitation suggestions is displayed.

[0048] Preferably, the intelligent auxiliary system for dermatological rehabilitation care also includes a remote diagnosis module, which is used to summarize the patient's symptom data, normalized living habit data and rehabilitation suggestions into preliminary analysis results, and transmit them to the dermatologist's diagnosis platform through a wireless network. The doctor remotely checks the patient's condition based on the preliminary analysis results of the auxiliary system and provides further diagnosis and treatment opinions. At the same time, the auxiliary system synchronously updates the patient's treatment record in real time.

[0049] An intelligent auxiliary method for dermatological rehabilitation nursing, comprising the following steps:

[0050] Step 1: Symptom identification and preprocessing: Use multispectral imaging technology and image acquisition equipment to obtain image data of the patient's skin, then use the image preprocessing unit to perform denoising, color balancing and edge detection, and use the image enhancement unit to compensate for the details of the low-resolution image. Then, based on the machine learning algorithm and the skin disease symptom database, generate symptom data;

[0051] Step 2: Collection and integration of lifestyle data: Monitor and record patients’ lifestyle habits through wearable devices, use mobile phone applications to collect patients’ lifestyle data through voice input, compare and analyze the self-reported data collected by mobile phone applications with the lifestyle data of wearable devices, and supplement the missing lifestyle data related information;

[0052] Step 3: Data analysis and impact identification: Correlation analysis is performed between symptom data and lifestyle data, and the data is analyzed using time series analysis and multivariate regression algorithms to obtain analysis results;

[0053] Step 4: Generation and push of rehabilitation suggestions: Based on the analysis results, rehabilitation suggestions are generated and pushed regularly through the reminder unit of the mobile application;

[0054] Step 5: Remote diagnosis support: Symptom data, lifestyle data, and rehabilitation suggestions are aggregated and transmitted via wireless networks, allowing doctors to view patient data remotely.

[0055] The present invention provides an intelligent auxiliary system and method for dermatological rehabilitation nursing, which has the following beneficial effects:

[0056] 1. The intelligent auxiliary system and method for dermatological rehabilitation nursing use image acquisition equipment and multispectral imaging technology to obtain image data of the patient's skin from multiple angles and under different light conditions, and after image preprocessing and processing by the enhancement unit, the quality and authenticity of the image are improved, providing a reliable basis for accurately identifying skin disease symptoms. Combining machine learning algorithms and skin disease symptom databases for comparison and identification, symptom data can be generated more accurately, and wearable devices can be used to monitor patients' sleep quality, activity level, diet records and other living habit data in real time. At the same time, mobile phone applications are used to collect patients' self-reported daily conditions, including eating habits, work pressure, skin care product usage and allergic reactions, etc., to achieve a comprehensive understanding of patients' living habits, and through comparative analysis and supplementary mechanisms, complete living habit data is obtained, improving the accuracy and completeness of the data.

[0057] 2. The intelligent auxiliary system and method for dermatological rehabilitation care, through the data processing module, correlates the symptom data with the lifestyle data, and adopts the time series analysis and multivariate regression algorithms. It can identify the potential impact of the patient's lifestyle on skin symptoms at multiple levels, including the impact of diet, work and rest, and contact objects on the frequency and severity of skin diseases, providing a scientific basis for formulating personalized rehabilitation recommendations; through the trimming reminder unit, it can timely detect abnormal values ​​in the lifestyle data and send them to the wearable device in the form of a pop-up message, prompting the patient to reconfirm or correct, thus ensuring the accuracy and reliability of the data.

[0058] 3. The intelligent auxiliary system and method for dermatological rehabilitation nursing, based on the potential impact analyzed by the data processing module, combines the preferred practice plans in the dermatology rehabilitation database to generate rehabilitation suggestions including lifestyle adjustment, diet plan, drug use suggestions, skin care product recommendations, exercise suggestions, etc., and can dynamically adjust the rehabilitation suggestions based on the patient's lifestyle data, thereby improving the pertinence and effectiveness of the rehabilitation suggestions. Through the timed reminder mechanism of the mobile phone application, the rehabilitation suggestions are pushed to the patients in a timely manner to ensure that the patients can comply with the treatment plan in their daily lives. The reminder methods are diverse and can be selected according to the patient's usage habits and preferences, thereby improving the effectiveness of the reminders. At the same time, the rehabilitation suggestions are displayed in the form of pictures, etc., making it easier for patients to understand and accept. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 This is a flowchart of an intelligent auxiliary system for dermatology rehabilitation nursing of the present invention;

[0060] Figure 2 The present invention is a flow chart of an intelligent auxiliary method for dermatological rehabilitation nursing. DETAILED DESCRIPTION

[0061] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. The embodiments of the present invention are provided for the purpose of illustration and description, and are not intended to be exhaustive or to limit the present invention to the disclosed forms. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to better illustrate the principles and practical applications of the present invention, and to enable those of ordinary skill in the art to understand the present invention and thereby design various embodiments with various modifications suitable for specific uses.

[0062] like Figure 1 and Figure 2 As shown, the present invention provides a technical solution: an intelligent auxiliary system and method for dermatological rehabilitation nursing, including a symptom recognition module, which is used to obtain image data of the patient's skin through an image acquisition device and a multi-spectral imaging technology; it is also used to identify skin disease symptoms, including the type of skin disease, the shape, color, distribution range of the rash, and the degree of eczema, and generate symptom data; wherein the image data is shot at multiple angles and under different light to ensure that the system comprehensively collects skin disease symptoms;

[0063] The habit collection module is connected to the symptom recognition module, and collects the patient's living habit data based on the patient's living habit questionnaire and sensor equipment. The living habit data includes the corresponding data of diet, work and rest rules and contact objects;

[0064] The collection of lifestyle data involves real-time monitoring of patients’ sleep quality, activity levels, and dietary records through wearable devices, as well as collecting patients’ self-reported daily conditions through mobile phone applications. The lifestyle data is then improved based on the daily conditions to ensure the integrity and accuracy of the data.

[0065] The data processing module is used to correlate symptom data with lifestyle data and identify the potential impact of patients’ lifestyle habits on skin symptoms;

[0066] It is also used to fill in missing values ​​in lifestyle data, remove outliers, and normalize various lifestyle data according to standardized algorithms for subsequent data analysis;

[0067] The rehabilitation suggestion module generates rehabilitation suggestions for skin diseases based on the potential impacts analyzed by the data processing module, pushes the rehabilitation suggestions to patients through mobile phone applications, and sets up a reminder mechanism to ensure that patients follow the rehabilitation suggestions on time. Doctors can remotely view patients' symptom data and lifestyle data through the auxiliary system, and give corresponding diagnosis and treatment suggestions based on the preliminary rehabilitation suggestions provided by the rehabilitation suggestion module.

[0068] The symptom recognition module also includes an image preprocessing unit and an image enhancement unit:

[0069] The image preprocessing unit is used to process the image data of the patient's skin based on denoising, color balancing and edge detection technology to obtain real image data to improve the accuracy of recognition;

[0070] The image enhancement unit is used to enhance the real image data through image detail compensation technology when the image resolution is low or the light conditions are poor, to obtain multispectral image data and ensure the accuracy of symptom recognition;

[0071] The symptom recognition module also compares and identifies multispectral image data with existing skin disease big data based on machine learning algorithms and a skin disease symptom database to obtain symptom data and achieve high-precision diagnosis of different types of skin diseases.

[0072] The habit collection module includes wearable devices and mobile applications:

[0073] Wearable devices are used to monitor the patient's body temperature, heart rate, sleep status, daily exercise and diet information in real time, obtain the patient's living habit data, and upload the living habit data to the system database; built-in sensors can detect changes in the patient's body temperature, record heart rate fluctuations, monitor sleep duration and depth, count the number of steps in daily activities, and identify the time and type of food intake;

[0074] The mobile phone application is used to regularly push lifestyle questionnaires to patients. Patients fill out the questionnaires through voice input to obtain the patients' self-reported daily conditions, including eating habits, work pressure, skin care product usage and allergic reactions. The lifestyle data obtained through wearable devices are supplemented based on the patients' self-reported daily conditions to obtain complete lifestyle data. The questionnaire content covers eating habits, work pressure, skin care product usage and allergic reactions. Patients can be asked whether they prefer spicy food in their daily diet, how stressed they are at work, what skin care products they use and whether they have allergic reactions to specific substances. Patients can easily describe their daily conditions with voice, and the application converts the voice into text and records it in the questionnaire.

[0075] The application also has a daily reminder function to urge patients to complete the health log on time and ensure the comprehensiveness of lifestyle data.

[0076] The process of supplementing the life habit data obtained through wearable devices based on the patient's self-reported daily conditions to obtain complete life habit data is as follows:

[0077] Comparing and analyzing the patient's self-reported daily conditions collected by the mobile phone application with the life habit data uploaded by the wearable device can check whether the dietary information recorded by the wearable device is consistent with the patient's dietary habits described in the questionnaire, and whether the sleep status monitored by the wearable device is consistent with the patient's self-reported impact of work pressure on sleep; when the patient provides information in the questionnaire that is not monitored by the wearable device, the information is marked as pending registration information and supplemented to the life habit data. For example, if the patient mentioned in the questionnaire that he had recently used a new skin care product, but the wearable device could not monitor this information related to the skin care product, the auxiliary system would supplement the use of the skin care product in the form of skin care product information to the life habit data. According to the data; the supplemented patient's self-reported daily conditions and the living habits data of wearable devices are then integrated to form complete living habits data; during the integration, taking work stress as an example: the work stress situation described by the patient in the questionnaire is combined with the heart rate changes monitored by wearable devices, and the impact of work stress on the patient's physical condition is comprehensively considered; in this way, the living habits data are supplemented and improved in terms of work stress conditions, and the remaining data are supplemented in the same way; through the above process, the living habits data obtained by wearable devices can be supplemented based on the patient's self-reported daily conditions, so as to obtain more complete and accurate living habits data, and provide more comprehensive information support for the intelligent assistance system of dermatological rehabilitation care.

[0078] The data processing module includes:

[0079] Impact Identification Unit: Using algorithms based on time series analysis and multivariate regression, we conduct multi-level correlation analysis between symptom data and lifestyle data to identify the potential impact of patients’ lifestyles on skin symptoms, including the impact of patients’ diet, daily routine, and contact materials on the frequency and severity of skin disease attacks;

[0080] Correction reminder unit: Generates reminder information based on abnormal values ​​in lifestyle data, and sends the reminder information in the form of pop-up information to the wearable device of the habit collection module, prompting the patient to reconfirm or correct the lifestyle data, where the pop-up information includes confirmation and correction.

[0081] The process for identifying the potential impact of a patient's lifestyle habits on their skin symptoms is:

[0082] Step 1: Time series analysis:

[0083] The symptom data and lifestyle data are divided into different time periods in chronological order, using days, weeks and months as units, to form symptom time series data sets and life time series data sets. The symptom time series data sets are then analyzed to observe the frequency and severity of skin disease attacks and obtain the symptom change trend. At the same time, the time series of diet, work and rest, and contact objects in the life time series data sets are analyzed to obtain the law of life changes.

[0084] Among them, time series analysis can be used to observe the changing trend of symptoms over time, as well as the changes in lifestyle data at different time points, so as to find out the time correlation; and multivariate regression can simultaneously consider the impact of multiple lifestyle factors on skin symptoms and determine the importance of each factor;

[0085] Obtaining the law of life changes: using the comprehensive correlation coefficient to calculate the correlation coefficient between the time series of symptom data and life habit data, detecting the correlation between them, and obtaining the comprehensive coefficient, that is, the result of time series analysis. The comprehensive coefficient includes the Pearson coefficient and the Spearman coefficient. There are two cases when using the comprehensive correlation coefficient to calculate the period;

[0086] One method: When the symptom data and the lifestyle data are in a linear relationship, the Pearson correlation coefficient is used to calculate the correlation coefficient between the time series of the symptom data and the lifestyle data, and the correlation between them is detected to obtain the Pearson coefficient;

[0087] Another method is to use the Spearman correlation coefficient to calculate the correlation coefficient between the time series of symptom data and lifestyle data when the relationship between the symptom data and lifestyle data is nonlinear, and to detect the correlation between them to obtain the Spearman coefficient;

[0088] Among them, the Pearson correlation coefficient calculates the correlation coefficient between the time series of symptom data and lifestyle data, detects the correlation between them, and obtains the Pearson coefficient. When the Pearson coefficient is high, it indicates that the lifestyle factor has a potential impact on skin symptoms;

[0089] The Pearson coefficient is between -1 and 1, where a Pearson coefficient close to 1 indicates a strong positive correlation between the two variables, a Pearson coefficient close to -1 indicates a strong negative correlation between the two variables, and a Pearson coefficient close to 0 indicates no linear correlation between the two variables.

[0090] The calculation formula of Pearson correlation coefficient is: ;

[0091] in, and There are two variables: the observed values ​​of lifestyle data and the observed values ​​of symptom data; and There are two variables: the sample mean of the lifestyle data and the sample mean of the symptom data;

[0092] In addition, the range of the Spearman coefficient is from -1 to +1, where: the Spearman coefficient close to +1 indicates a positive monotonic relationship, as one variable: lifestyle data increases, the other variable: symptom data also increases; the Spearman coefficient close to -1 indicates a negative monotonic relationship, as one variable: lifestyle data increases, the other variable: symptom data decreases; the Spearman coefficient close to 0 indicates that there is no monotonic relationship between lifestyle data and symptom data;

[0093] The calculation formula of Spearman correlation coefficient is: ;

[0094] Where: d i is the difference between the levels of two variables, where the two variables refer to lifestyle data and symptom data respectively; n is the number of data pairs based on lifestyle data and symptom data in the two variables, that is, each data pair contains an observation value of the lifestyle data variable and an observation value of the symptom data variable; the data pairs are collected at the same time and are used to evaluate the relationship between the two variables: lifestyle data and symptom data;

[0095] The Spearman correlation coefficient is calculated as follows: assign a rank to each observation of each variable; calculate the difference in ranks d for each pair of observations i ; Square all the level differences and then sum them; Substitute the result into the above formula to calculate r s ;

[0096] The calculation of the Spearman correlation coefficient can avoid the influence of outliers. It is not only suitable for measuring the strength of nonlinear relationships, but also for non-normally distributed data.

[0097] Step 2: Algorithm analysis of multivariate regression:

[0098] The various factors in the lifestyle data were used as independent variables, and the skin symptom data were used as dependent variables; when the dependent variable was a continuous variable and had a linear relationship with the independent variable, a linear regression model was selected based on the characteristics and data type of the skin symptoms, and the model parameters were calculated by solving the normal equations; when the observations were independent of each other, a polynomial regression model was selected based on the characteristics and data type of the skin symptoms, and the model parameters were calculated by the least squares method;

[0099] The coefficient of determination R² is used to evaluate the degree of fit of the model to the data. The value of R² is between 0 and 1. The closer R² is to 1, the better the model fits the data. When R² is close to 1, it means that the degree of fit is good and the prediction ability is strong, indicating that the model can effectively identify the potential impact of lifestyle habits on skin symptoms. According to the parameter estimation results of the regression model, the direction and degree of the impact of various lifestyle factors on skin symptoms are explained to obtain the regression analysis results, that is, the results of multivariate regression analysis.

[0100] The results of the regression analysis are as follows:

[0101] If the regression coefficient of the lifestyle factor is positive, it means that the factor is positively correlated with skin symptoms, that is, the increase of this factor will lead to the aggravation of skin symptoms;

[0102] On the contrary, if the regression coefficient of the lifestyle factor is negative, it means that the factor is negatively correlated with skin symptoms, that is, the increase of the factor will alleviate skin symptoms;

[0103] The results of time series analysis and multivariate regression analysis were combined to determine the potential impact of the patient's lifestyle habits on skin symptoms; if both analysis methods showed that a certain lifestyle factor was strongly correlated with skin symptoms, it could be determined that the factor had a potential impact on skin symptoms; for example, if both time series analysis and multivariate regression analysis showed that the patient's intake of spicy food in the diet was positively correlated with the frequency of skin disease attacks, it could be determined that the intake of spicy food had a potential impact on the patient's skin symptoms and might increase the frequency and severity of skin diseases.

[0104] The rehabilitation suggestion module generates rehabilitation suggestions for skin diseases based on the potential impacts generated by the data processing module and the preferred practice plans in the skin disease rehabilitation database:

[0105] Rehabilitation suggestions include lifestyle adjustments, diet plans, medication recommendations, skin care product recommendations, exercise recommendations, and the rehabilitation suggestions are adjusted based on the patient's lifestyle data;

[0106] The rehabilitation advice module includes a reminder unit, which is used to push rehabilitation advice to patients through a mobile phone application on a regular basis based on a preset reminder mechanism to ensure that patients comply with the treatment plan in their daily lives.

[0107] Based on the preset reminder mechanism, the process of pushing rehabilitation suggestions to patients through mobile phone applications at regular intervals is as follows:

[0108] Determine the reminder time interval based on the patient's condition and rehabilitation needs. For patients who need to strictly adhere to a specific diet plan, you can set reminders before the three meals a day; for patients who need to take medication on time, you can set reminders every few hours; select multiple reminder methods based on the patient's usage habits and preferences to remind the patient, including sound prompts, vibrations, and pop-up notifications to ensure that the reminder can effectively attract the patient's attention; and adjust the reminder mechanism based on the patient's situation and needs to obtain the adjusted preset reminder mechanism. For patients with poor sleep quality, reminders can be set during the quieter time periods during the day to avoid affecting the patient's rest;

[0109] Then, the reminder unit and the mobile application are interfaced to ensure that the rehabilitation suggestions can be accurately pushed to the mobile application; when the rehabilitation suggestions are generated based on the reminder unit, they are transmitted to the mobile application via the network communication protocol; after the mobile application receives the rehabilitation suggestions, it stores them in the local database so as to remind when needed;

[0110] Set a timer in the mobile phone application to trigger a reminder according to a preset reminder time interval; wherein the timer is implemented based on the scheduled task function provided by the mobile phone operating system;

[0111] When the timer is triggered, the mobile phone application checks whether the triggering conditions of the reminder are currently met; determines whether the current time is within the preset reminder time range, whether the patient is in a specific state (such as resting or busy), etc.; if the triggering conditions are met, a reminder is issued; otherwise, wait for the next time the timer is triggered; according to the preset reminder mechanism, the mobile phone application performs the reminder operation; if it is set to sound reminder, a specific reminder sound is played; if it is set to vibration, the vibration function of the mobile phone is triggered; if it is set to pop-up notification, a pop-up window of rehabilitation suggestions is displayed on the mobile phone screen; while reminding, the mobile phone application displays it in the form of pictures so that the patient can clearly understand the rehabilitation measures that need to be taken and the specific content of the rehabilitation suggestions are displayed;

[0112] Through the above process, the reminder unit can push rehabilitation suggestions to patients through mobile phone applications at regular intervals through a preset reminder mechanism, helping patients better comply with the treatment plan and promote recovery from skin diseases.

[0113] The intelligent auxiliary system for dermatological rehabilitation care also includes a remote diagnosis module, which is used to summarize the patient's symptom data, normalized living habit data, and rehabilitation suggestions into preliminary analysis results, and transmit them to the dermatologist's diagnosis platform through a wireless network. The doctor can remotely view the patient's condition based on the preliminary analysis results of the auxiliary system; the remote diagnosis module summarizes the patient's symptom data, normalized living habit data, and rehabilitation suggestions into preliminary analysis results, and transmits them to the dermatologist's diagnosis platform through a wireless network. The doctor can remotely view the patient's condition and provide the patient with more professional diagnosis and treatment suggestions. This not only facilitates patients and improves the utilization efficiency of medical resources, but also helps doctors better understand the patient's condition and rehabilitation status and adjust treatment plans in a timely manner.

[0114] Based on the application of the above technical solutions, the auxiliary system can achieve comprehensive and accurate symptom recognition and data collection, in-depth data analysis and potential impact identification, personalized rehabilitation suggestions and reminder mechanisms, and remote diagnosis functions, providing all-round intelligent assistance for the rehabilitation care of dermatology patients.

[0115] When using an intelligent auxiliary system for dermatological rehabilitation nursing, firstly, symptom recognition and preprocessing are performed: multispectral imaging technology and image acquisition equipment are used to obtain image data of the patient's skin, and then denoising, color balancing and edge detection are performed through the image preprocessing unit, and details of the low-resolution image are compensated in the image enhancement unit, and symptom data is generated based on the machine learning algorithm and the comparison with the skin disease symptom database; then, the life habit data is collected and integrated: the patient's life habits are monitored and recorded through wearable devices, and the patient's life habit data is collected through voice input using a mobile phone application, and the self-reported data collected by the mobile phone application is compared and analyzed with the life habit data of the wearable device, and the missing life habit data related information is supplemented;

[0116] Then, through data analysis and impact identification: the symptom data and lifestyle data are correlated and analyzed, and the data is analyzed using time series analysis and multivariate regression algorithms to obtain analysis results; then rehabilitation suggestions are generated and pushed: based on the analysis results, rehabilitation suggestions are generated and pushed regularly through the reminder unit of the mobile phone application; finally, remote diagnosis support is provided: the symptom data, lifestyle habit data and rehabilitation suggestions are summarized and transmitted via the wireless network, so that doctors can view the patient's data remotely.

[0117] Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field and related fields without creative work should fall within the scope of protection of the present invention. The structures, devices and operating methods not specifically described and explained in the present invention are implemented according to the conventional means in the field unless otherwise specified and limited.

Claims

1. An intelligent auxiliary system for dermatological rehabilitation nursing, characterized in that: include: A symptom recognition module, used to obtain image data of the patient's skin through an image acquisition device and multispectral imaging technology; It is also used to identify symptoms of skin diseases and generate symptom data; wherein the image data is shot at multiple angles and under different lighting conditions; The habit collection module is connected to the symptom recognition module. Based on the patient's living habit questionnaire and sensor equipment, the patient's living habit data is collected. The living habit data includes corresponding data on diet, work and rest schedule, and contact objects. The living habit data is collected through wearable devices to monitor the patient's sleep quality, activity level, and diet records in real time, and the patient's self-reported daily conditions are collected through mobile phone applications, and the living habit data is improved based on the daily conditions. The data processing module is used to correlate symptom data with lifestyle data and identify the potential impact of patients' lifestyles on skin symptoms; it is also used to fill in missing values ​​in lifestyle data, remove outliers, and normalize various lifestyle data according to a standardized algorithm; The rehabilitation suggestion module generates rehabilitation suggestions for skin diseases based on the potential impact analyzed by the data processing module, pushes the rehabilitation suggestions to patients through mobile phone applications, and sets reminder mechanisms; The symptom recognition module also includes an image preprocessing unit and an image enhancement unit: The image preprocessing unit is used to process the image data of the patient's skin based on denoising, color balancing and edge detection technology to obtain real image data; The image enhancement unit is used to enhance the real image data by using the image detail compensation technology when the image resolution is low, so as to obtain multispectral image data; The symptom recognition module also compares and identifies multispectral image data with existing skin disease big data based on machine learning algorithms and skin disease symptom databases to obtain symptom data; The habit collection module includes wearable devices and mobile phone applications: Wearable devices are used to monitor the patient's body temperature, heart rate, sleep status, daily exercise and diet information in real time, obtain the patient's living habit data, and upload the living habit data to the system database; The mobile phone application is used to push a lifestyle questionnaire to the patient. The patient fills in the questionnaire through voice input to obtain the patient's self-reported daily conditions, including eating habits, work pressure, skin care product usage and allergic reactions. The lifestyle data obtained through the wearable device is supplemented based on the patient's self-reported daily conditions to obtain complete lifestyle data; The app also has a daily reminder function to urge patients to complete the health log on time; The process of supplementing the life habit data obtained through the wearable device based on the patient's self-reported daily conditions to obtain complete life habit data is as follows: Compare and analyze the patient's self-reported daily conditions collected by the mobile phone application with the life habit data uploaded by the wearable device; when the patient provides information in the questionnaire that is not monitored by the wearable device, mark the information as pending registration information and add the pending registration information to the life habit data; then integrate the supplemented patient's self-reported daily conditions with the life habit data of the wearable device to form a complete life habit data; The data processing module comprises: Impact Identification Unit: Using algorithms based on time series analysis and multivariate regression, we conduct multi-level correlation analysis between symptom data and lifestyle data to identify the potential impact of patients’ lifestyles on skin symptoms, including the impact of patients’ diet, daily routine, and contact materials on the frequency and severity of skin disease attacks; Correction reminder unit: generates reminder information based on abnormal values ​​in the living habit data, and sends the reminder information in the form of a pop-up box information to the wearable device of the habit collection module, prompting the patient to reconfirm or correct the living habit data, where the pop-up box information includes confirmation and correction; The process of identifying the potential impact of the patient's lifestyle on skin symptoms is: Step 1: Time series analysis: The symptom data and lifestyle data are divided into different time periods in chronological order, using days, weeks and months as units, to form symptom time series data sets and life time series data sets. The symptom time series data sets are then analyzed to observe the frequency and severity of skin disease attacks and obtain the symptom change trend. At the same time, the time series of diet, work and rest, and contact objects in the life time series data sets are analyzed to obtain the law of life changes. Obtaining the law of life changes: using the comprehensive correlation coefficient to calculate the correlation coefficient between the time series of symptom data and life habit data, and obtaining the comprehensive coefficient, that is, the result of time series analysis. The comprehensive coefficient includes the Pearson coefficient and the Spearman coefficient. There are two cases when using the comprehensive correlation coefficient to calculate the period; One: When the symptom data and lifestyle data are in a linear relationship, the Pearson correlation coefficient is used to calculate the correlation coefficient between the time series of symptom data and lifestyle data to obtain the Pearson coefficient; Another method is to use the Spearman correlation coefficient to calculate the correlation coefficient between the time series of symptom data and lifestyle data when the relationship between symptom data and lifestyle data is nonlinear, and obtain the Spearman coefficient; Step 2: Algorithm analysis of multivariate regression: The various factors in the lifestyle data were used as independent variables, and the skin symptom data were used as dependent variables; when the dependent variable was a continuous variable and had a linear relationship with the independent variable, a linear regression model was selected based on the characteristics and data type of the skin symptoms, and the model parameters were calculated by solving the normal equations; when the observations were independent of each other, a polynomial regression model was selected based on the characteristics and data type of the skin symptoms, and the model parameters were calculated by the least squares method; The coefficient of determination R² is used to evaluate the degree of fit of the model to the data. The value of R² is between 0 and 1. When R² is close to 1, it indicates a good degree of fit and strong predictive ability, indicating that the model can effectively identify the potential impact of lifestyle habits on skin symptoms. Based on the parameter estimation results of the regression model, the direction and degree of the impact of various lifestyle factors on skin symptoms are explained to obtain the regression analysis results, that is, the results of multivariate regression analysis. The results of the regression analysis are as follows: If the regression coefficient of the lifestyle factor is positive, it means that the factor is positively correlated with skin symptoms, that is, the increase of this factor will lead to the aggravation of skin symptoms; On the contrary, if the regression coefficient of the lifestyle factor is negative, it means that the factor is negatively correlated with skin symptoms, that is, the increase of the factor will alleviate skin symptoms; The rehabilitation suggestion module generates rehabilitation suggestions for skin diseases based on the potential impact generated by the data processing module and in combination with the preferred practice solutions in the skin disease rehabilitation database: Rehabilitation suggestions include lifestyle adjustments, diet plans, medication recommendations, skin care product recommendations, exercise recommendations, and the rehabilitation suggestions are adjusted based on the patient's lifestyle data; The rehabilitation suggestion module includes a reminder unit, which is used to push rehabilitation suggestions to the patient through a mobile phone application at a fixed time based on a preset reminder mechanism; The process of regularly pushing rehabilitation suggestions to patients through mobile phone applications based on the preset reminder mechanism is as follows: Determine the reminder time interval according to the patient's condition and rehabilitation needs; select multiple reminder methods to remind the patient according to the patient's usage habits and preferences, including sound prompts, vibrations, and pop-up notifications; and adjust the reminder mechanism based on the patient's condition and needs to obtain the adjusted preset reminder mechanism; Then, an interface is developed between the reminder unit and the mobile phone application; when a rehabilitation suggestion is generated based on the reminder unit, it is transmitted to the mobile phone application via a network communication protocol; after the mobile phone application receives the rehabilitation suggestion, the rehabilitation suggestion is stored in a local database; Set a timer in the mobile phone application to trigger a reminder according to a preset reminder time interval; wherein the timer is implemented based on the scheduled task function provided by the mobile phone operating system; When the timer is triggered, the mobile application checks whether the triggering conditions of the reminder are currently met; if the triggering conditions are met, a reminder is issued; otherwise, the reminder waits for the next time the timer is triggered; according to the preset reminder mechanism, the mobile application executes the reminder operation; while giving the reminder, the mobile application displays the specific content of the rehabilitation suggestions in the form of pictures; The intelligent auxiliary system for dermatological rehabilitation care also includes a remote diagnosis module, which is used to summarize the patient's symptom data, normalized living habit data and rehabilitation suggestions into preliminary analysis results, and transmit them to the dermatologist's diagnosis platform via a wireless network. The doctor remotely checks the patient's condition based on the preliminary analysis results of the auxiliary system.

2. An intelligent assistance method for dermatological rehabilitation nursing according to an intelligent assistance system for dermatological rehabilitation nursing according to claim 1, characterized in that: The following steps are involved: Step 1: Symptom identification and preprocessing: Use multispectral imaging technology and image acquisition equipment to obtain image data of the patient's skin, then use the image preprocessing unit to perform denoising, color balancing and edge detection, and use the image enhancement unit to compensate for the details of the low-resolution image. Then, based on the machine learning algorithm and the skin disease symptom database, generate symptom data; Step 2: Collection and integration of lifestyle data: Monitor and record patients’ lifestyle habits through wearable devices, use mobile phone applications to collect patients’ lifestyle data through voice input, compare and analyze the self-reported data collected by mobile phone applications with the lifestyle data of wearable devices, and supplement the missing lifestyle data related information; Step 3: Data analysis and impact identification: Correlation analysis is performed between symptom data and lifestyle data, and the data is analyzed using time series analysis and multivariate regression algorithms to obtain analysis results; Step 4: Generation and push of rehabilitation suggestions: Based on the analysis results, rehabilitation suggestions are generated and pushed regularly through the reminder unit of the mobile application; Step 5: Remote diagnosis support: Symptom data, lifestyle data, and rehabilitation suggestions are aggregated and transmitted via wireless networks, allowing doctors to view patient data remotely.

Images