Intelligent follow-up visit management system for skin disease patients

Abstract

The invention discloses an intelligent follow-up visit management system for skin disease patients, and relates to the technical field of Internet. During operation of the system, multi-source data is acquired from mobile equipment, medical equipment and environment acquisition equipment worn by the patients, a comprehensive original data set D covering illness conditions and environment factors is formed, summarizing calculation is performed, and a follow-up visit result is obtained; according to the method, the dynamic follow-up scoring index DP is generated, so that the follow-up frequency adjustment identifier SF is generated, accurate regulation and control of the follow-up frequency are realized, and the defect that a fixed follow-up period cannot respond to the change of an illness state in real time is overcome. The evaluation feedback and adjustment module calculates a follow-up visit frequency adjustment identifier SF for the second time based on follow-up visit data fed back by the patient, and substitutes the follow-up visit frequency adjustment identifier SF into the data processing and feature extraction module for iterative optimization; the system can continuously adapt to dynamic changes of illness conditions, and precision and individuation of follow-up visit plans are achieved.

Description

Technical Field

[0001] The present invention relates to the field of Internet technology, and in particular to an intelligent follow-up management system for patients with skin diseases. Background Art

[0002] In the overall field of modern medical management, follow-up management is a crucial branch that runs through the entire process from initial diagnosis and treatment to patient recovery. With the advancement of medical technology, more and more disease management has begun to adopt intelligent follow-up systems to support personalized diagnosis and treatment needs. Among this type of follow-up management, follow-up management of patients with skin diseases is particularly important due to the large fluctuations in their condition, complex and diverse symptoms, and easy recurrence.

[0003] In the existing methods of follow-up management of skin diseases, although regular follow-up can help patients control their condition to a certain extent, the traditional fixed follow-up cycle is usually difficult to meet the actual needs of patients with skin diseases. Because the condition of patients with skin diseases is greatly affected by external factors, such as climate, environment and seasonal changes, a fixed follow-up cycle may result in fluctuations in the condition not being captured and treated in a timely manner, increasing the risk of recurrence and worsening of the condition. In addition, fixed-cycle follow-up also makes resource allocation unreasonable: during the stable period of the disease, patients do not need frequent follow-up, but during the acute fluctuation period, the disease may be delayed due to insufficient follow-up. Therefore, a dynamic follow-up frequency adjustment mechanism based on changes in the condition is particularly critical, which can allocate follow-up resources more scientifically, improve follow-up efficiency, and reduce unnecessary follow-up burdens. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides an intelligent follow-up management system for patients with skin diseases, which solves the problems mentioned in the background technology.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: an intelligent follow-up management system for skin disease patients, including a patient data collection module, a data processing and feature extraction module, a dynamic frequency evaluation module, a follow-up suggestion generation module and an evaluation feedback and adjustment module;

[0006] The patient data collection module collects data from multiple sources through mobile devices, medical devices and environmental collection devices worn by patients to form an original data set D;

[0007] The data processing and feature extraction module performs data processing on the original data set D, including denoising and format standardization, and performs feature extraction on the processed original data set D to generate a feature vector F;

[0008] The dynamic frequency evaluation module aggregates several feature vectors in the feature vector F, generates a dynamic follow-up scoring index DP, matches it with a preset dynamic follow-up threshold DPth, obtains the patient follow-up evaluation and regulation evaluation result, and marks the follow-up frequency adjustment flag SF;

[0009] The follow-up suggestion generation module modifies the follow-up time period according to the patient's follow-up frequency adjustment flag SF, and generates a suggestion notification and push plan;

[0010] The evaluation feedback and adjustment module triggers the follow-up data feedback mechanism according to the patient's recommended notification and push plan, and brings the feedback information into the dynamic frequency evaluation module for secondary calculation to obtain the follow-up frequency adjustment flag SF, and simultaneously brings the follow-up frequency adjustment flag SF into the data processing and feature extraction module for iterative optimization.

[0011] Preferably, the patient data acquisition module includes a patient status unit and an environmental data unit;

[0012] The patient status unit receives the patient's lesion image data and symptom data in real time through the mobile device and medical device worn by the patient, and forms a patient status subset Dp={IMG, P, I, H};

[0013] Among them, the patient's lesion image data includes image data IMG, which is obtained by taking pictures with a mobile device and collecting image data of the lesion area regularly with medical equipment; the symptom data includes pain state P, itching state I and redness and swelling state H, which is obtained by having the patient fill out a questionnaire and a scoring form in real time on a mobile device;

[0014] The environmental data unit obtains the meteorological data of the patient area through the positioning function integrated in the mobile device and the medical device, including humidity S, temperature T, air quality QZ and ultraviolet intensity UV, to form a patient area meteorological data subset Dh = {S, T, QZ, UV}, and merges it with the patient status subset Dp to form an original data set D = {IMG, P, I, H, Dp, Dhc}.

[0015] Preferably, the data processing and feature extraction module includes a data preprocessing unit and a feature extraction unit;

[0016] The data preprocessing unit performs data processing on the original data set D, including data denoising and format standardization;

[0017] Among them, the data denoising processing includes using Gaussian filtering and mean filtering to perform denoising on the original data set D; the format standardization processing includes using normalization processing to perform format standardization on the original data set D after denoising, and obtaining the standard data set Dc = {IMGc, Pc, Ic, Hc, Sc, Tc, QZc, UVc}, where IMGc represents standardized image data, Pc represents standardized pain state, Ic represents standardized itching state, Hc represents standardized redness and swelling state, Sc represents standardized humidity, Tc represents standardized temperature, QZc represents standardized air quality, and UVc represents standardized ultraviolet intensity.

[0018] Preferably, the feature extraction unit performs feature extraction on the standard data set Dc after format standardization processing, including extracting color features, extended features and texture features of the image data IMG, obtaining the color feature vector Col, the extended feature vector Exp and the texture feature vector Tex, and generating the feature vector F = {Pc, Ic, Hc, Sc, Tc, QZc, UVc, Col, Exp, Tex}.

[0019] Preferably, the color feature vector Col is obtained by analyzing the image data IMG using the HSV color space. Specifically, the color feature vector Col = {μH, μS, μV, σH 2 ,σS 2 ,σV 2}, μH, μS and μV represent the mean hue, saturation and brightness of the image data IMG respectively, σH 2 ,σS 2 and σV 2 They represent hue variance, saturation variance, and brightness variance respectively;

[0020] The extended feature vector Exp determines the contour of the lesion in the image data IMG by using the Canny edge detection algorithm, calculates the area A and the perimeter L of the lesion in the image data IMG, and then calculates the compactness factor C by the area A and the perimeter C, specifically, the extended feature vector Exp={C};

[0021] The texture feature vector Tex extracts texture feature information of the lesion area in the image data IMG by using the gray level co-occurrence matrix GLCM method to obtain contrast Con and energy Ene. Specifically, the texture feature vector Tex={Con, Ene}.

[0022] Preferably, the dynamic frequency assessment module includes a feature summary unit and a dynamic assessment unit;

[0023] The feature summarization unit summarizes several feature vectors in the feature vector F, obtains the symptom feature summary factor Sym, the environmental meteorological summary factor Hjm, the color feature vector summary factor ColH and the physical feature vector summary factor TexH, and performs fitting calculation to generate a dynamic follow-up scoring index DP;

[0024] The disease symptom summary factor Sym is obtained by summarizing the standardized pain state Pc, the standardized itching state Ic and the standardized redness and swelling state Hc in the feature vector F;

[0025] The environmental meteorological summary factor Hjm is obtained by summarizing the standardized humidity Sc, the standardized temperature Tc, the standardized air quality QZc and the standardized ultraviolet intensity UVc in the characteristic vector F;

[0026] The color feature vector summary factor ColH is obtained by summarizing the color feature vectors Col in the feature vector F;

[0027] The texture feature vector summary factor TexH is obtained by summarizing the texture feature vectors Tex in the feature vector F.

[0028] Preferably, the dynamic follow-up scoring index DP is obtained by the following calculation formula:

[0029]

[0030] Wherein, sin represents the sine function.

[0031] Preferably, the dynamic evaluation unit matches the dynamic follow-up scoring index DP with a preset dynamic follow-up threshold DPth, obtains the patient follow-up evaluation regulation evaluation result, and marks the follow-up frequency adjustment flag SF;

[0032] The patient follow-up assessment and control assessment results are obtained by the following matching method:

[0033] When the dynamic follow-up scoring index DP>dynamic follow-up threshold DPth, the patient follow-up evaluation regulation result is obtained as the need to increase the regulation result, and the follow-up frequency adjustment flag SF=1 is generated;

[0034] When the dynamic follow-up scoring index DP is less than the dynamic follow-up threshold DPth, the patient follow-up evaluation regulation result is obtained as a non-regulation result, and a follow-up frequency adjustment flag SF=0 is generated.

[0035] When the dynamic follow-up scoring index DP is less than the dynamic follow-up threshold DPth, the patient follow-up evaluation regulation result is obtained to be the need to reduce the regulation result, and a follow-up frequency adjustment flag SF=-1 is generated.

[0036] Preferably, the follow-up suggestion generating module includes a follow-up time adjusting unit and a suggestion notification generating unit;

[0037] The follow-up time adjustment unit modifies the follow-up time period according to the patient's follow-up frequency adjustment flag SF, wherein the modified follow-up time period is specifically calculated by the follow-up frequency adjustment flag SF and the adjustment coefficient k to obtain a modified follow-up time period Tnew, and proportionally adjusts the current visit time period by applying the modified follow-up time period Tnew;

[0038] The suggestion notification generating unit fills in a preset suggestion notification template according to the patient's follow-up frequency adjustment flag SF and the revised follow-up time period Tnew, and generates a suggestion notification and a push plan;

[0039] The suggested notification and push scheme are obtained by the following generation method:

[0040] When the follow-up frequency adjustment flag SF≠0, a prompt notification plan for the patient is generated, including prompting the patient to adjust the follow-up frequency to the revised follow-up time period Tnew, and the plan is pushed to relevant medical staff and patients.

[0041] Preferably, the evaluation feedback and adjustment module triggers the follow-up data feedback mechanism according to the patient's recommended notification and push plan, and brings the feedback information into the dynamic frequency evaluation module for secondary calculation to obtain the follow-up frequency adjustment mark SF, and brings the follow-up frequency adjustment mark SF into the data processing and feature extraction module for iterative optimization, including adjusting the acquisition of the iterative adjustment color feature vector Col;

[0042] The follow-up data feedback mechanism is specifically triggered by the follow-up frequency adjustment flag SF. When the follow-up frequency adjustment flag SF≠0, the follow-up data feedback mechanism is adjusted to twice the frequency to feed back the patient's original data set D, which is marked as feedback information.

[0043] The present invention provides an intelligent follow-up management system for patients with skin diseases, which has the following beneficial effects:

[0044] (1) When the system is running, it obtains multi-source data from mobile devices, medical devices, and environmental collection devices worn by patients to form a comprehensive original data set D covering the condition and environmental factors. It performs summary calculations to generate a dynamic follow-up score index DP, thereby generating a follow-up frequency adjustment mark SF, achieving precise control of the follow-up frequency and making up for the defect that the fixed follow-up cycle fails to respond to changes in the condition in real time. It also generates personalized recommendation notifications and push plans so that patients and doctors can know the follow-up adjustment plan in real time. The evaluation feedback and adjustment module recalculates the follow-up frequency adjustment mark SF based on the follow-up data fed back by patients, and substitutes it into the data processing and feature extraction module for iterative optimization, so that the system can continuously adapt to the dynamic changes of the condition, realize the precision and personalization of the follow-up plan, and thoroughly improve the shortcomings of traditional follow-up management that cannot be dynamically updated and optimized.

[0045] (2) Through the comprehensive extraction and quantification of the color feature vector Col, the extended feature vector Exp, and the texture feature vector Tex, the system can accurately reflect the color consistency, shape regularity, and surface texture characteristics of the lesion area from the lesion image. The dynamic follow-up scoring index DP is fitted and generated, so that the system can quantify the degree of fluctuation of the patient's condition. By comparing the dynamic follow-up scoring index DP with the preset dynamic follow-up threshold DPth, the system can generate a follow-up frequency adjustment mark SF to dynamically adjust the follow-up frequency, thereby avoiding the problem of too frequent or delayed follow-up, ensuring that the follow-up plan can closely match the patient's condition changes, and realizing the real-time and efficient condition monitoring.

[0046] (3) The follow-up time adjustment unit dynamically generates a revised follow-up time period Tnew according to the follow-up frequency adjustment mark SF and the adjustment coefficient k, ensuring that the patient's follow-up period is consistent with the real-time changes in the condition, avoiding the problem of incompatibility of follow-up frequency in traditional follow-up plans. It is recommended that the notification generation unit further generates accurate follow-up adjustment notifications based on the follow-up frequency adjustment mark SF and the revised follow-up time period Tnew and pushes them to patients and medical staff to ensure the instant transmission of information. In addition, the evaluation feedback and adjustment module controls the feedback frequency of the follow-up data feedback mechanism through the follow-up frequency adjustment mark SF, achieving double frequency feedback on the patient's original data set D, thereby enhancing the density of condition monitoring and the timeliness of data. Adjusting the filter window size according to the follow-up frequency can smooth the data to improve the analysis stability when the condition fluctuates greatly, and retain more details when the condition fluctuates less, ensuring the sensitivity of data processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 The present invention is a schematic diagram of a block diagram of an intelligent follow-up management system for skin disease patients. DETAILED DESCRIPTION

[0048] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0049] Example 1

[0050] The present invention provides an intelligent follow-up management system for patients with skin diseases. Figure 1 , including patient data collection module, data processing and feature extraction module, dynamic frequency assessment module, follow-up suggestion generation module and assessment feedback and adjustment module;

[0051] The patient data collection module collects data from multiple sources through mobile devices, medical devices and environmental collection devices worn by patients to form an original data set D;

[0052] The data processing and feature extraction module performs data processing on the original data set D, including denoising and format standardization, and performs feature extraction on the processed original data set D to generate a feature vector F;

[0053] The dynamic frequency evaluation module aggregates several feature vectors in the feature vector F, generates a dynamic follow-up scoring index DP, matches it with a preset dynamic follow-up threshold DPth, obtains the patient follow-up evaluation and regulation evaluation result, and marks the follow-up frequency adjustment flag SF;

[0054] The follow-up suggestion generation module modifies the follow-up time period according to the patient's follow-up frequency adjustment flag SF, and generates a suggestion notification and push plan;

[0055] The evaluation feedback and adjustment module triggers the follow-up data feedback mechanism according to the patient's recommended notification and push plan, and brings the feedback information into the dynamic frequency evaluation module for secondary calculation to obtain the follow-up frequency adjustment flag SF, and simultaneously brings the follow-up frequency adjustment flag SF into the data processing and feature extraction module for iterative optimization.

[0056] In this embodiment, by acquiring multi-source data from mobile devices, medical devices and environmental collection devices worn by patients, a comprehensive original data set D covering the condition and environmental factors is formed, and the feature vector F is extracted to ensure the accuracy and consistency of the analysis data, overcome the disadvantages of inconsistent manual screening data, perform summary calculations, generate a dynamic follow-up score index DP, and match it with the preset dynamic follow-up threshold DPth, thereby generating a follow-up frequency adjustment mark SF, achieving precise control of the follow-up frequency, and making up for the defect that the fixed follow-up cycle fails to respond to changes in the condition in real time. And generate personalized suggestion notifications and push solutions, so that patients and doctors can know the follow-up adjustment plan in real time, solving the pain point of untimely information communication. In addition, the evaluation feedback and adjustment module recalculates the follow-up frequency adjustment flag SF based on the follow-up data fed back by patients, and substitutes it into the data processing and feature extraction module for iterative optimization, so that the system can continuously adapt to the dynamic changes of the disease and realize the precision and personalization of the follow-up plan, which thoroughly improves the shortcomings of traditional follow-up management that cannot be dynamically updated and optimized. In summary, this system has established a closed-loop feedback mechanism through multi-module collaboration, so that the follow-up frequency can adapt to the dynamic changes of the disease, realize the intelligent adjustment of the follow-up plan, improve the efficiency and accuracy of follow-up, and improve the patient's condition monitoring effect.

[0057] Example 2

[0058] This embodiment is explained in Example 1, please refer to Figure 1 ,Specifically: the patient data acquisition module includes a patient status unit and an environmental data unit;

[0059] The patient status unit receives the patient's lesion image data and symptom data in real time through the mobile device and medical device worn by the patient, and forms a patient status subset Dp={IMG, P, I, H};

[0060] Among them, the patient's lesion image data includes image data IMG, which is obtained by taking pictures with a mobile device and collecting image data of the lesion area regularly with medical equipment; the symptom data includes pain state P, itching state I and redness and swelling state H, which is obtained by having the patient fill out a questionnaire and a scoring form in real time on a mobile device;

[0061] The environmental data unit obtains the meteorological data of the patient area through the positioning function integrated in the mobile device and the medical device, including humidity S, temperature T, air quality QZ and ultraviolet intensity UV, to form a patient area meteorological data subset Dh = {S, T, QZ, UV}, and merges it with the patient status subset Dp to form an original data set D = {IMG, P, I, H, Dp, Dhc}.

[0062] The data processing and feature extraction module includes a data preprocessing unit and a feature extraction unit;

[0063] The data preprocessing unit performs data processing on the original data set D, including data denoising and format standardization;

[0064] Among them, the data denoising processing includes using Gaussian filtering and mean filtering to perform denoising on the original data set D; the format standardization processing includes using normalization processing to perform format standardization on the original data set D after denoising, and obtaining the standard data set Dc = {IMGc, Pc, Ic, Hc, Sc, Tc, QZc, UVc}, where IMGc represents standardized image data, Pc represents standardized pain state, Ic represents standardized itching state, Hc represents standardized redness and swelling state, Sc represents standardized humidity, Tc represents standardized temperature, QZc represents standardized air quality, and UVc represents standardized ultraviolet intensity.

[0065] The feature extraction unit performs feature extraction on the standard data set Dc after format standardization processing, including extracting color features, extended features and texture features of the image data IMG, obtaining color feature vector Col, extended feature vector Exp and texture feature vector Tex, and generating a feature vector F = {Pc, Ic, Hc, Sc, Tc, QZc, UVc, Col, Exp, Tex}.

[0066] Example 3

[0067] This embodiment is explained in Example 2. Please refer to Figure 1 Specifically: the color feature vector Col is obtained by analyzing the image data IMG using the HSV color space. Specifically, the color feature vector Col = {μH, μS, μV, σH 2 ,σS 2 ,σV 2}, μH, μS and μV represent the mean hue, saturation and brightness of the image data IMG respectively, σH 2 ,σS 2 and σV 2 They represent hue variance, saturation variance, and brightness variance respectively. The color feature vector Col reflects the change amplitude of the color of the lesion area, which can be used to evaluate the consistency and change trend of the lesion color.

[0068] The extended feature vector Exp determines the contour of the lesion in the image data IMG by using the Canny edge detection algorithm, calculates the area A and the perimeter L of the lesion in the image data IMG, and then calculates the compactness factor C by the area A and the perimeter C. Specifically, the extended feature vector Exp={C}, wherein the compactness factor C is used to reflect the regularity of the lesion shape. The closer the value is to 1, the more round the lesion area is, and the larger the value is, the more irregular the lesion expansion is.

[0069] The compactness factor C is given by The calculation formula is obtained, where π is the constant of circumference;

[0070] The texture feature vector Tex extracts texture feature information of the lesion area in the image data IMG by using the gray level co-occurrence matrix GLCM method to obtain contrast Con and energy Ene, specifically, the texture feature vector Tex = {Con, Ene};

[0071] The contrast ratio is determined by The calculation formula is obtained, where N represents the gray level, GLCM(i,j) represents the value of the i-th row and j-th column in the gray level co-occurrence matrix, which specifically represents the frequency of occurrence of pixel gray pairs in the lesion area. When the contrast Con increases, it means that the surface texture of the lesion is rougher;

[0072] The energy Ene is The calculation formula is used, where the higher the energy Ene, the smoother the surface of the lesion area.

[0073] The dynamic frequency assessment module includes a feature summary unit and a dynamic assessment unit;

[0074] The feature summarization unit summarizes several feature vectors in the feature vector F, obtains the symptom feature summary factor Sym, the environmental meteorological summary factor Hjm, the color feature vector summary factor ColH and the physical feature vector summary factor TexH, and performs fitting calculation to generate a dynamic follow-up scoring index DP;

[0075] The disease symptom summary factor Sym is obtained by summarizing the standardized pain state Pc, the standardized itching state Ic and the standardized redness and swelling state Hc in the feature vector F;

[0076] The symptom characteristic summary factor Sym is obtained by the following calculation formula:

[0077]

[0078] In the formula, ln represents the natural logarithm function;

[0079] The environmental meteorological summary factor Hjm is obtained by summarizing the standardized humidity Sc, the standardized temperature Tc, the standardized air quality QZc and the standardized ultraviolet intensity UVc in the characteristic vector F;

[0080] The environmental meteorological summary factor Hjm is Obtain calculation formula;

[0081] The color feature vector summary factor ColH is obtained by summarizing the color feature vectors Col in the feature vector F;

[0082] The color feature vector summary factor ColH is obtained by Obtain calculation formula;

[0083] The texture feature vector summary factor TexH is obtained by summarizing the texture feature vectors Tex in the feature vector F;

[0084] The texture feature vector summary factor TexH is obtained by Obtain the calculation formula.

[0085] The dynamic follow-up scoring index DP is obtained by the following calculation formula:

[0086]

[0087] In the formula, sin represents the sine function, Sym*Him represents the combined influence of subjective symptoms and environment, and the interaction between the disease characteristic summary factor Sym and the environmental meteorological summary factor Hjm reflects the possibility that the disease condition fluctuates with the environment. It represents the ratio of the square of color feature to texture feature, which amplifies the impact of color fluctuation on the condition. (1+sin(Sym+TexH)) means taking the sine value of Sym+TexH to ​​generate a fluctuation term, making the condition change more consistent with the natural fluctuation characteristics.

[0088] The dynamic evaluation unit matches the dynamic follow-up scoring index DP with a preset dynamic follow-up threshold DPth, obtains the patient follow-up evaluation regulation evaluation result, and marks the follow-up frequency adjustment flag SF;

[0089] The patient follow-up assessment and control assessment results are obtained by the following matching method:

[0090] When the dynamic follow-up scoring index DP>dynamic follow-up threshold DPth, the patient follow-up evaluation regulation result is obtained as the need to increase the regulation result, and the follow-up frequency adjustment flag SF=1 is generated;

[0091] When the dynamic follow-up scoring index DP is less than the dynamic follow-up threshold DPth, the patient follow-up evaluation regulation result is obtained as a non-regulation result, and a follow-up frequency adjustment flag SF=0 is generated.

[0092] When the dynamic follow-up scoring index DP is less than the dynamic follow-up threshold DPth, the patient follow-up evaluation regulation result is obtained to be the need to reduce the regulation result, and a follow-up frequency adjustment flag SF=-1 is generated.

[0093] In this embodiment, by comprehensive extraction and quantification of color feature vector Col, extended feature vector Exp and texture feature vector Tex, the system can accurately reflect the color consistency, shape regularity and surface texture characteristics of the lesion area from the lesion image. And fit to generate a dynamic follow-up scoring index DP, so that the system can quantify the degree of fluctuation of the patient's condition. By comparing the dynamic follow-up scoring index DP with the preset dynamic follow-up threshold DPth, the system can generate a follow-up frequency adjustment mark SF to dynamically adjust the follow-up frequency, thereby avoiding the problem of too frequent or delayed follow-up, ensuring that the follow-up plan can closely match the patient's condition changes, and realizing the real-time and high efficiency of condition monitoring. This scheme can not only reflect the fluctuation trend of the patient's condition, but also generate automated control prompts during the follow-up process, making the follow-up frequency more accurate and reliable, reducing the intervention of manual decision-making, and effectively reducing the burden on patients. In summary, the system provides intelligent and full-process follow-up support for patients and doctors, realizes personalized and dynamic condition management, thereby greatly improving the efficiency and scientificity of follow-up services.

[0094] Example 4

[0095] This embodiment is explained in Example 3, please refer to Figure 1 ,Specifically: the follow-up suggestion generation module includes a follow-up time adjustment unit and a suggestion notification generation unit;

[0096] The follow-up time adjustment unit modifies the follow-up time period according to the patient's follow-up frequency adjustment flag SF, wherein the modified follow-up time period is specifically calculated by the follow-up frequency adjustment flag SF and the adjustment coefficient k to obtain a modified follow-up time period Tnew, and proportionally adjusts the current visit time period by applying the modified follow-up time period Tnew;

[0097] The modified follow-up time period Tnew is Obtain calculation formula;

[0098] The suggestion notification generating unit fills in a preset suggestion notification template according to the patient's follow-up frequency adjustment flag SF and the revised follow-up time period Tnew, and generates a suggestion notification and a push plan;

[0099] The suggested notification and push scheme are obtained by the following generation method:

[0100] When the follow-up frequency adjustment flag SF≠0, a prompt notification plan for the patient is generated, including prompting the patient to adjust the follow-up frequency to the revised follow-up time period Tnew, and the plan is pushed to relevant medical staff and patients.

[0101] The evaluation feedback and adjustment module triggers the follow-up data feedback mechanism according to the patient's recommended notification and push plan, and brings the feedback information into the dynamic frequency evaluation module for secondary calculation to obtain the follow-up frequency adjustment mark SF, and brings the follow-up frequency adjustment mark SF into the data processing and feature extraction module for iterative optimization, including adjusting the acquisition of the iterative adjustment color feature vector Col, specifically, when the follow-up frequency adjustment mark SF=1, the filter window is increased to smooth the data with large fluctuations, and when the follow-up frequency adjustment mark SF=-1, the window is reduced to retain more detailed information;

[0102] The follow-up data feedback mechanism is specifically triggered by the follow-up frequency adjustment flag SF. When the follow-up frequency adjustment flag SF≠0, the follow-up data feedback mechanism is adjusted to twice the frequency to feed back the patient's original data set D, which is marked as feedback information.

[0103] In this embodiment, the follow-up time adjustment unit dynamically generates a revised follow-up time period Tnew according to the follow-up frequency adjustment mark SF and the adjustment coefficient k, ensuring that the patient's follow-up period is consistent with the real-time changes in the condition, avoiding the problem of incompatibility of the follow-up frequency in the traditional follow-up scheme. It is recommended that the notification generation unit further generates an accurate follow-up adjustment notification based on the follow-up frequency adjustment mark SF and the revised follow-up time period Tnew and pushes it to the patient and medical staff to ensure the instant transmission of information. In addition, the evaluation feedback and adjustment module controls the feedback frequency of the follow-up data feedback mechanism through the follow-up frequency adjustment mark SF, achieving double frequency feedback of the patient's original data set D, thereby enhancing the density of condition monitoring and the timeliness of the data. Adjusting the filter window size according to the follow-up frequency can smooth the data to improve the analysis stability when the condition fluctuates greatly, and retain more details when the condition fluctuates less, ensuring the sensitivity of data processing.

[0104] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent follow-up management system for patients with skin diseases, characterized by: It includes patient data collection module, data processing and feature extraction module, dynamic frequency assessment module, follow-up suggestion generation module and assessment feedback and adjustment module; The patient data collection module collects data from multiple sources through mobile devices, medical devices and environmental collection devices worn by patients to form an original data set D; The data processing and feature extraction module performs data processing on the original data set D, including denoising and format standardization, and performs feature extraction on the processed original data set D to generate a feature vector F; The dynamic frequency evaluation module aggregates several feature vectors in the feature vector F, generates a dynamic follow-up scoring index DP, matches it with a preset dynamic follow-up threshold DPth, obtains the patient follow-up evaluation and regulation evaluation result, and marks the follow-up frequency adjustment flag SF; The follow-up suggestion generation module modifies the follow-up time period according to the patient's follow-up frequency adjustment flag SF, and generates a suggestion notification and push plan; The evaluation feedback and adjustment module triggers the follow-up data feedback mechanism according to the patient's recommended notification and push plan, and brings the feedback information into the dynamic frequency evaluation module for secondary calculation to obtain the follow-up frequency adjustment flag SF, and simultaneously brings the follow-up frequency adjustment flag SF into the data processing and feature extraction module for iterative optimization.

2. The intelligent follow-up management system for skin disease patients according to claim 1, characterized in that: The patient data acquisition module includes a patient status unit and an environmental data unit; The patient status unit receives the patient's lesion image data and symptom data in real time through the mobile device and medical device worn by the patient, and forms a patient status subset Dp={IMG, P, I, H}; Among them, the patient's lesion image data includes image data IMG, which is obtained by taking pictures with a mobile device and collecting image data of the lesion area regularly with medical equipment; the symptom data includes pain state P, itching state I and redness and swelling state H, which is obtained by having the patient fill out a questionnaire and a scoring form in real time on a mobile device; The environmental data unit obtains the meteorological data of the patient area through the positioning function integrated in the mobile device and the medical device, including humidity S, temperature T, air quality QZ and ultraviolet intensity UV, to form a patient area meteorological data subset Dh = {S, T, QZ, UV}, and merges it with the patient status subset Dp to form an original data set D = {IMG, P, I, H, Dp, Dhc}.

3. The intelligent follow-up management system for skin disease patients according to claim 2, characterized in that: The data processing and feature extraction module includes a data preprocessing unit and a feature extraction unit; The data preprocessing unit performs data processing on the original data set D, including data denoising and format standardization; Among them, the data denoising processing includes using Gaussian filtering and mean filtering to perform denoising on the original data set D; the format standardization processing includes using normalization processing to perform format standardization on the original data set D after denoising, and obtaining the standard data set Dc = {IMGc, Pc, Ic, Hc, Sc, Tc, QZc, UVc}, where IMGc represents standardized image data, Pc represents standardized pain state, Ic represents standardized itching state, Hc represents standardized redness and swelling state, Sc represents standardized humidity, Tc represents standardized temperature, QZc represents standardized air quality, and UVc represents standardized ultraviolet intensity.

4. The intelligent follow-up management system for skin disease patients according to claim 3, characterized in that: The feature extraction unit performs feature extraction on the standard data set Dc after format standardization processing, including extracting color features, extended features and texture features of the image data IMG, obtaining color feature vector Col, extended feature vector Exp and texture feature vector Tex, and generating a feature vector F = {Pc, Ic, Hc, Sc, Tc, QZc, UVc, Col, Exp, Tex}.

5. The intelligent follow-up management system for skin disease patients according to claim 4, characterized in that: The color feature vector Col is obtained by analyzing the image data IMG using the HSV color space. Specifically, the color feature vector Col = {μH, μS, μV, σH 2 ,σS 2 ,σV 2 }, μH, μS and μV represent the mean hue, saturation and brightness of the image data IMG respectively, σH 2 ,σS 2 and σV 2 They represent hue variance, saturation variance, and brightness variance respectively; The extended feature vector Exp determines the contour of the lesion in the image data IMG by using the Canny edge detection algorithm, calculates the area A and the perimeter L of the lesion in the image data IMG, and then calculates the compactness factor C by the area A and the perimeter C, specifically, the extended feature vector Exp={C}; The texture feature vector Tex extracts texture feature information of the lesion area in the image data IMG by using the gray level co-occurrence matrix GLCM method to obtain contrast Con and energy Ene. Specifically, the texture feature vector Tex={Con, Ene}.

6. The intelligent follow-up management system for skin disease patients according to claim 5, characterized in that: The dynamic frequency assessment module includes a feature summary unit and a dynamic assessment unit; The feature summarization unit summarizes several feature vectors in the feature vector F, obtains the symptom feature summary factor Sym, the environmental meteorological summary factor Hjm, the color feature vector summary factor ColH and the physical feature vector summary factor TexH, and performs fitting calculation to generate a dynamic follow-up scoring index DP; The disease symptom summary factor Sym is obtained by summarizing the standardized pain state Pc, the standardized itching state Ic and the standardized redness and swelling state Hc in the feature vector F; The environmental meteorological summary factor Hjm is obtained by summarizing the standardized humidity Sc, the standardized temperature Tc, the standardized air quality QZc and the standardized ultraviolet intensity UVc in the characteristic vector F; The color feature vector summary factor ColH is obtained by summarizing the color feature vectors Col in the feature vector F; The texture feature vector summary factor TexH is obtained by summarizing the texture feature vectors Tex in the feature vector F.

7. The intelligent follow-up management system for skin disease patients according to claim 6, characterized in that: The dynamic follow-up scoring index DP is obtained by the following calculation formula: Wherein, sin represents the sine function.

8. The intelligent follow-up management system for skin disease patients according to claim 6, characterized in that: The dynamic evaluation unit matches the dynamic follow-up scoring index DP with a preset dynamic follow-up threshold DPth, obtains the patient follow-up evaluation regulation evaluation result, and marks the follow-up frequency adjustment flag SF; The patient follow-up assessment and control assessment results are obtained by the following matching method: When the dynamic follow-up scoring index DP>dynamic follow-up threshold DPth, the patient follow-up evaluation regulation result is obtained as the need to increase the regulation result, and the follow-up frequency adjustment flag SF=1 is generated; When the dynamic follow-up scoring index DP is less than the dynamic follow-up threshold DPth, the patient follow-up evaluation regulation result is obtained as a non-regulation result, and a follow-up frequency adjustment flag SF=0 is generated. When the dynamic follow-up scoring index DP is less than the dynamic follow-up threshold DPth, the patient follow-up evaluation regulation result is obtained to be the need to reduce the regulation result, and a follow-up frequency adjustment flag SF=-1 is generated.

9. The intelligent follow-up management system for skin disease patients according to claim 8, characterized in that: The follow-up suggestion generation module includes a follow-up time adjustment unit and a suggestion notification generation unit; The follow-up time adjustment unit modifies the follow-up time period according to the patient's follow-up frequency adjustment flag SF, wherein the modified follow-up time period is specifically calculated by the follow-up frequency adjustment flag SF and the adjustment coefficient k to obtain a modified follow-up time period Tnew, and proportionally adjusts the current visit time period by applying the modified follow-up time period Tnew; The suggestion notification generating unit fills in a preset suggestion notification template according to the patient's follow-up frequency adjustment flag SF and the revised follow-up time period Tnew, and generates a suggestion notification and a push plan; The suggested notification and push scheme are obtained by the following generation method: When the follow-up frequency adjustment flag SF≠0, a prompt notification plan for the patient is generated, including prompting the patient to adjust the follow-up frequency to the revised follow-up time period Tnew, and the plan is pushed to relevant medical staff and patients.

10. The intelligent follow-up management system for skin disease patients according to claim 9, characterized in that: The evaluation feedback and adjustment module triggers the follow-up data feedback mechanism according to the patient's recommended notification and push plan, and brings the feedback information into the dynamic frequency evaluation module for secondary calculation to obtain the follow-up frequency adjustment mark SF, and brings the follow-up frequency adjustment mark SF into the data processing and feature extraction module for iterative optimization, including adjusting the acquisition of the iterative adjustment color feature vector Col; The follow-up data feedback mechanism is specifically triggered by the follow-up frequency adjustment flag SF. When the follow-up frequency adjustment flag SF≠0, the follow-up data feedback mechanism is adjusted to twice the frequency to feed back the patient's original data set D, which is marked as feedback information.

Images