Cancer survivor chronic pain management system based on patient reporting outcomes
Through real-time data acquisition and multi-dimensional analysis, the chronic pain management system for cancer survivors uses personalized characteristics and rules engines to provide accurate pain management suggestions, solving the problem that existing systems cannot capture pain fluctuations in time and lack of analysis depth, and achieving Efficient and accurate pain management effects.
Patent Information
- Application Number
- CN202411074952.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-07
- Publication Date
- 2024-11-19
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing chronic pain management system for cancer survivors lacks real-time data update capabilities, cannot capture pain fluctuations in time, and the depth and breadth of analysis are insufficient, and it cannot provide comprehensive and personalized decision-making support, and cannot help patients identify pain patterns and trends. It provides scientific advice on chronic pain management.
设计了一种基于患者报告结果的癌症幸存者慢性疼痛管理系统,通过实时数据采集和多维度分析,利用个性化特征和规则引擎提供精准的药物、非药物和生活管理策略管理建议,并根据反馈 动态调整,包括数据采集模块、数据分析引擎、个性化建议生成器和用户交互模块,通过疼痛波动分析公式、疼痛潜力分析函数和情绪-疼痛相互作用公式进行评估和分析,自动分组并设定疼痛 Manage priorities and policies.
Through multi-dimensional data analysis and personalized suggestions generation, it provides efficient, accurate and flexible chronic pain management solutions to ensure real-time and effective data management, help patients dynamically adjust pain management strategies, and improve the accuracy of pain management and patient quality of life.
Smart Images

Figure CN118983047A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of medical information management, and more particularly to a chronic pain management system for cancer survivors based on patient-reported outcomes. Background Art
[0002] After surviving cancer, cancer survivors often face the challenge of chronic pain due to the side effects of drugs and treatments, which seriously affects the quality of life of patients. Effective chronic pain management is crucial to the long-term health and well-being of cancer survivors. In this process, patient-reported outcomes (PROs) are an important self-assessment tool that records patients' subjective feelings about pain, quality of life, emotional state, etc., providing valuable insights for patients' chronic pain management and helping patients better record their daily experiences and symptom change data. Existing pain management usually relies on PROs data submitted by patients on a regular basis, lacks real-time data update capabilities, relies on patient self-reports at fixed time intervals, and cannot capture patients' pain fluctuations in a timely manner. At the same time, when facing chronic pain, cancer patients need to adjust their self-management strategies according to their own pain fluctuations. However, the existing system lacks sufficient analytical depth and breadth, and cannot provide patients with comprehensive and personalized decision support, nor can it help patients identify pain patterns and trends and provide them with scientific chronic pain management recommendations. Summary of the invention
[0003] In order to overcome the above-mentioned defects of the prior art, the present invention provides a chronic pain management system for cancer survivors based on patient-reported results. It accurately evaluates the chronic pain of cancer survivors and its impact on the quality of life through real-time data collection and multi-dimensional analysis, and uses personalized features and rule engines to provide precise management recommendations on drugs, non-drugs and life management strategies, and dynamically adjusts according to feedback.
[0004] To achieve the above object, the present invention provides the following technical solutions:
[0005] A chronic pain management system for cancer survivors based on patient-reported results includes a data acquisition module, a data analysis engine, a personalized suggestion generator, a user interaction module and a data storage module. The data acquisition module is connected to the data analysis engine, the data engine is connected to the personalized suggestion generator, and the user interaction module and the data storage module are both connected to the data acquisition module, the data analysis engine and the personalized suggestion generator. The data analysis engine obtains pre-processed multi-source data from PROs report results of cancer survivors, extracts key features from the original data and generates new features by applying mathematical changes, evaluates the severity of chronic pain of cancer survivors based on a pain fluctuation analysis formula, analyzes the long-term impact of chronic pain on the quality of life of cancer survivors based on a pain potential analysis function, and quantifies the dynamic impact of emotional changes on the perception of chronic pain of cancer survivors based on an emotion-pain interaction formula. The personalized suggestion generator performs personalized feature analysis based on the data obtained by the data analysis engine, realizes automatic grouping of chronic pain of cancer survivors based on a rule engine, sets different pain management priorities and strategies based on analysis results, automatically generates chronic pain drug management data, non-drug management data and life management data, and dynamically adjusts the priority and content of suggestions based on patient feedback and continuous data updates.
[0006] As a further scheme of the present invention, the data acquisition module obtains pain intensity data through the visual analog scale and the digital rating scale in the patient-reported outcomes (PROs), obtains the emotional score data of cancer survivors through the quality of life questionnaire through the PHQ-9, GAD-7, and BDI scales, obtains the patient's health status assessment in multiple life dimensions, obtains the patient's basic biological data, past medical history, current diagnosis, treatment plan, and medication records through background information and clinical data, monitors the patient's heart rate, number of steps, and activity level in real time through wearable devices, performs data preprocessing and data cleaning, and transmits the data to the data storage module for storage through encryption.
[0007] As a further solution of the present invention, in the data engine, the pain fluctuation analysis formula analyzes the impact of emotional fluctuations on chronic pain by integrating the change rate and acceleration of the intensity of chronic pain of cancer survivors and the patient's emotional fluctuation coefficient, and identifies the severity of chronic pain of cancer survivors. The pain fluctuation analysis formula is:
[0008]
[0009] Where: K PTI is the pain fluctuation analysis value, P is the pain intensity score, t is the calculation time point, is the rate of change in pain intensity score, is the acceleration of pain intensity, and It was obtained by time series analysis method, α was the density function of pain intensity score, which was obtained by analyzing the data of visual analog scale and digital rating scale, and β was the mood fluctuation coefficient, which was obtained by analyzing the data of mood state assessment scale.
[0010] As a further solution of the present invention, in the pain fluctuation analysis formula of the data analysis engine, the process of obtaining the density function of the pain intensity score includes:
[0011] Step 11, statistically analyzing the collected visual analog scale and digital rating scale data to obtain the mean value and standard deviation of the ratings;
[0012] Step 12, density function construction: Based on the statistical analysis results, the density function of the pain intensity score is constructed by the kernel density estimation method to obtain a smooth density curve.
[0013] Step 13, parameter calibration and verification: calibrate the parameters of the density function by comparing with historical data, verify the density function using data at different time points, and further optimize the adjustment parameters of the density function.
[0014] Step 14, result output: storing the calculated density function of the pain intensity score and related statistics in a data storage module;
[0015] Step 15, data visualization: draw a distribution curve of the scores and visualize the distribution of the scores.
[0016] As a further solution of the present invention, in the pain fluctuation analysis formula of the data analysis engine, the process of obtaining the emotion fluctuation coefficient includes:
[0017] Step 21, data standardization: Obtain the patient's regular PHQ-9, GAD-7, and BDI scale data, integrate the data of different scales by weighted average method to report their emotional state, and standardize the comprehensive score;
[0018] Step 22, mood fluctuation analysis: calculate the standard deviation of the patient's mood scores at different time points, analyze the frequency of changes in the mood state score by counting the number of changes exceeding a preset threshold, and calculate the mood fluctuation coefficient using regression analysis based on the amplitude and frequency of the mood fluctuation;
[0019] Step 23, parameter verification and optimization: Use historical data and cross-sample data to verify the accuracy and consistency of the sentiment fluctuation coefficient, and adjust the calculation parameter settings based on the verification results;
[0020] Step 24, result output: store the emotion fluctuation coefficient and related analysis data in the data storage module, and display the emotion fluctuation analysis results through charts and visualization tools.
[0021] As a further solution of the present invention, in the data analysis engine, the pain potential analysis function evaluates the long-term impact of chronic pain on the quality of life by combining the quality of life change rate obtained from the quality of life questionnaire data and the personalized coefficient of the cancer survivors with the current chronic pain change rate. The formula of the pain potential analysis function is:
[0022]
[0023] Where: K PPF is the density function value of the pain intensity score, Q is the quality of life score, γ is the personalized coefficient of cancer survivors, which is obtained through machine learning model analysis, and t0 is the initial moment of the analysis.
[0024] As a further solution of the present invention, in the pain potential analysis function formula analyzed by the data engine, the process of obtaining the personalized coefficient of cancer survivors includes:
[0025] Step 31, data feature extraction: obtain age, gender, weight, and height from the basic biological data of cancer survivors, extract keywords from past medical history, current diagnosis, treatment plan, and medication records, obtain heart rate, number of steps, and activity level from real-time monitoring data, and obtain key features in the collected data based on principal component analysis technology;
[0026] Step 32, machine learning model training: select a support vector machine model, train the model using historical data, and use a cross-validation method to evaluate model performance and optimize model parameters;
[0027] Step 33, personalized coefficient prediction: the key features obtained in step 31 are used as the input of the model, and the machine learning model trained in step 32 is used to output the personalized coefficient of each cancer survivor. The model and personalized coefficient are dynamically adjusted through continuous data input and feedback. The accuracy and stability of the model are verified using an independent test data set. The model parameters are adjusted according to the verification results, and the model is retrained regularly.
[0028] Step 34, result output and application: store the personalized coefficient of each patient in the data storage module.
[0029] As a further solution of the present invention, in the data analysis engine, the emotion-pain interaction formula quantifies the dynamic impact of emotion changes on the perception of chronic pain by cancer survivors through the emotion impact coefficient and the emotion change rate. The emotion-pain interaction formula is:
[0030]
[0031] Where: K EPIMis the emotion-pain interaction analysis value, E is the current emotion score, σ is the emotion influence coefficient, is the emotion-pain interaction coefficient, σ and Obtained through machine learning model analysis.
[0032] As a further solution of the present invention, in the emotion-pain interaction formula of the data analysis engine, the process of obtaining the emotion impact coefficient includes:
[0033] Step 41, data standardization: collecting data from PHQ-9, GAD-7, and BDI to obtain emotional score data of cancer survivors, and standardizing the scores of different emotional scales;
[0034] Step 42, emotion change analysis: calculating the change rate of emotion scores of cancer survivors at different time points based on the difference method;
[0035] Step 43, prediction of emotion impact coefficient: select the random forest model, input the emotion change rate, basic biological data, medical history, and current treatment plan, use historical data to train the model, try to use the cross-validation method to evaluate the model performance during training, use the trained model to predict new data, and obtain the emotion impact coefficient;
[0036] Step 44, parameter verification and optimization: using an independent verification data set to test the emotion impact coefficient, and adjusting the model parameters according to the verification results to improve the model structure and regularly update the model based on the relationship between the predicted emotion impact coefficient and the actual emotion and pain perception data;
[0037] Step 45, result output: the predicted emotion impact coefficient and related analysis data are stored in the data storage module, and the emotion impact coefficient is combined with the emotion change rate and other related variables in the emotion-pain interaction formula to quantify the dynamic impact of emotion changes on pain perception.
[0038] As a further solution of the present invention, in the emotion-pain interaction formula of the data analysis engine, the process of obtaining the emotion-pain interaction coefficient includes:
[0039] Step 51, data standardization: standardizing the patient's emotional state data and pain intensity data;
[0040] Step 52, interaction effect analysis: using a multivariate regression analysis model to analyze the interaction effect between emotion and pain intensity, with a particular focus on how emotion changes regulate changes in pain perception, using historical data to train the multivariate regression analysis model to predict the emotion-pain interaction coefficient, with input features being the standardized current emotion score, pain intensity score, emotion change rate, and pain change rate;
[0041] Step 53, interaction coefficient prediction: Use the trained model to analyze new emotion and pain data, predict the emotion-pain interaction coefficient, and dynamically adjust and update the model based on new data input and patient feedback;
[0042] Step 54, parameter validation: Use an independent validation data set to test the emotion-pain interaction coefficient, compare the pain response predicted by the model with the actual data, analyze the performance of the model in different patient groups, adjust the model parameters and structure according to the validation results, and retrain the model regularly;
[0043] Step 55, result output and application: The emotion-pain interaction coefficient and related analysis data are stored in the data storage module, and the emotion-pain interaction coefficient is combined with the emotion influence coefficient and other related variables in the emotion-pain interaction formula to quantify the dynamic impact of emotion changes on pain perception.
[0044] As a further scheme of the present invention, a personalized suggestion generator receives pain fluctuation analysis values, density function values of pain intensity scores, emotion-pain interaction analysis values and personalization coefficients, density functions of pain intensity scores, emotion impact coefficients, and emotion-pain interaction coefficients obtained by a data analysis engine, integrates the data, extracts key features of cancer survivors from these data, combines the extracted key features using a weighted summation method, forms a comprehensive score for assessing the risk level of cancer survivors, and divides the risk level into high risk, medium risk, and low risk according to a pre-set risk level classification standard, and generates suggestions based on a rule engine.
[0045] As a further solution of the present invention, in the personalized suggestion generator, the preset rules are:
[0046] High-risk cancer survivors need to adjust the dosage of existing analgesics, recommend more auxiliary plans, and increase the level of exercise in daily life; medium-risk cancer survivors maintain the current plan, conduct regular assessments, and maintain the current level of exercise; low-risk cancer survivors gradually reduce the amount and frequency of analgesics and maintain the current level of exercise.
[0047] As a further solution of the present invention, the personalized suggestion generator collects feedback data after the patient implements the suggestions, dynamically adjusts the suggestion generation content and priority based on the feedback information and the input of new data, and regularly evaluates and updates the rule base.
[0048] As a further solution of the present invention, the user interaction module transmits the information generated by the personalized suggestion generator to the patient and medical staff for display, and readjusts the personalized suggestions based on the interaction of the cancer survivor and the interaction of the medical staff, and supports the issuance of adjustment suggestion information to the real-time monitoring device.
[0049] Compared with the prior art, the technical effects of the present invention are as follows: the collected multi-dimensional data is deeply analyzed through the data analysis engine, key features are extracted and new features are generated, providing a basis for data-driven decision support; the personalized suggestion generator uses the rule engine to automatically group according to the analysis results, set different data management priorities and strategies, and generate personalized management data; through continuous data updates and patient feedback, the content and priority of management suggestions are dynamically adjusted to ensure the real-time and effectiveness of data management; in terms of data collection, analysis, personalized suggestion generation and dynamic optimization, an efficient, accurate and flexible data management solution is provided. BRIEF DESCRIPTION OF THE DRAWINGS
[0050] Figure 1 It is a system block diagram of the present invention. DETAILED DESCRIPTION
[0051] The following will be combined with the drawings in the present invention to clearly and completely describe the technical solutions in the present invention. Obviously, the described technical solutions are only part of the present invention but not all of it. Based on the content of the present invention, all other technical solutions obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.
[0052] like Figure 1 As shown, a chronic pain management system for cancer survivors based on patient-reported results proposed by the present invention includes a data acquisition module, a data analysis engine, a personalized suggestion generator, a user interaction module and a data storage module. The data acquisition module is connected to the data analysis engine, the data engine is connected to the personalized suggestion generator, and the user interaction module and the data storage module are both connected to the data acquisition module, the data analysis engine and the personalized suggestion generator. The data analysis engine obtains pre-processed multi-source data from the PROs report results of cancer survivors, extracts key features from the original data and generates new features by applying mathematical changes, evaluates the severity of chronic pain in cancer survivors based on the pain fluctuation analysis formula, analyzes the long-term impact of chronic pain on the quality of life of cancer survivors based on the pain potential analysis function, quantifies the dynamic impact of emotional changes on the perception of chronic pain in cancer survivors based on the emotion-pain interaction formula, and the personalized suggestion generator performs personalized feature analysis based on the data obtained by the data analysis engine. Automatic grouping of chronic pain in cancer survivors is achieved based on the rule engine. Different pain management priorities and strategies are set based on the analysis results, and chronic pain drug management data, non-drug management data and life management data are automatically generated. The priority and content of the suggestions are dynamically adjusted according to patient feedback and continuous data updates.
[0053] The present invention uses a data analysis engine to perform in-depth analysis on the collected multi-dimensional data, extract key features and generate new features, providing a basis for data-driven decision support. The personalized suggestion generator uses a rule engine to automatically group according to the analysis results, set different data management priorities and strategies, and generate personalized management data. Through continuous data updates and patient feedback, the content and priority of management suggestions are dynamically adjusted to ensure the real-time and effectiveness of data management. In terms of data collection, analysis, personalized suggestion generation and dynamic optimization, an efficient, accurate and flexible data management solution is provided.
[0054] It should be noted that the data collection module obtains pain intensity data through the visual analog scale and digital rating scale in patient-reported outcomes (PROs), obtains emotional score data of cancer survivors through quality of life questionnaires through PHQ-9, GAD-7, and BDI scales, obtains health status assessments of patients in multiple life dimensions, obtains patients' basic biological data, past medical history, current diagnosis, treatment plan, and medication records through background information and clinical data, monitors patients' heart rate, number of steps, and activity level in real time through wearable devices, performs data preprocessing and data cleaning, and transmits the data to the data storage module for storage through encryption.
[0055] It should be noted that in the data engine, the pain fluctuation analysis formula analyzes the impact of emotional fluctuations on chronic pain by combining the rate of change and acceleration of chronic pain intensity of cancer survivors and the patient's emotional fluctuation coefficient, and identifies the severity of chronic pain in cancer survivors. The pain fluctuation analysis formula is:
[0056]
[0057] Where: K PTI is the pain fluctuation analysis value, P is the pain intensity score, t is the calculation time point, is the rate of change in pain intensity score, is the acceleration of pain intensity, and It was obtained by time series analysis method, α was the density function of pain intensity score, which was obtained by analyzing the data of visual analog scale and digital rating scale, and β was the mood fluctuation coefficient, which was obtained by analyzing the data of mood state assessment scale.
[0058] This formula not only takes into account the instantaneous change of pain intensity (rate of change), but also includes the trend of pain change (acceleration), thus providing a comprehensive assessment of pain fluctuations. By introducing the emotion fluctuation coefficient (β), the formula can quantify the impact of emotion changes on pain perception, providing a scientific basis for understanding and managing the relationship between patients' emotions and pain. The application of the formula provides precise numerical data, which can be further used to generate personalized management recommendations and improve the accuracy and effectiveness of decision support. PTI is the output of the formula, which indicates the overall strength and severity of pain fluctuations. Higher values indicate more severe pain fluctuations. It is the rate of change of pain intensity score over time, indicating the speed of change of pain intensity per unit time. By analyzing the patient's pain score data in time series and calculating the difference between consecutive time points, it reflects the changing trend of instantaneous pain and is an important indicator for evaluating pain volatility. is the rate of change of the pain intensity score change rate, that is, the acceleration of the pain score change. Further time series analysis of the change rate data is performed to calculate its change rate, revealing the trend and acceleration of pain changes, and providing information for understanding the aggravation or relief of pain fluctuations. α is the emotion fluctuation coefficient, which reflects the influence of emotion changes on pain perception. By analyzing the data of the emotion state assessment scale (PHQ-9, GAD-7), the influence of emotion state changes on pain intensity is calculated, and the regulatory effect of emotion fluctuations on pain perception is quantified, providing a quantitative basis for analyzing the relationship between emotion and pain. By combining the density function of the pain score and the emotion fluctuation coefficient, the formula can comprehensively evaluate the degree of pain fluctuation and the potential impact of emotion on pain, helping to understand the complex pain-emotion interaction relationship. The formula includes the time dimension of pain changes. By analyzing the pain change rate and acceleration, it can capture the instantaneous changes and long-term trends of pain, providing a comprehensive volatility analysis.
[0059] It should be further specified that in the pain fluctuation analysis formula of the data analysis engine, the process of obtaining the density function of the pain intensity score includes:
[0060] Step 11, statistically analyzing the collected visual analog scale and digital rating scale data to obtain the mean value and standard deviation of the ratings;
[0061] Step 12, density function construction: Based on the statistical analysis results, the density function of the pain intensity score is constructed by the kernel density estimation method to obtain a smooth density curve.
[0062] Step 13, parameter calibration and verification: calibrate the parameters of the density function by comparing with historical data, verify the density function using data at different time points, and further optimize the adjustment parameters of the density function.
[0063] Step 14, result output: storing the calculated density function of the pain intensity score and related statistics in a data storage module;
[0064] Step 15, data visualization: draw a distribution curve of the scores and visualize the distribution of the scores.
[0065] It should be further specified that in the pain fluctuation analysis formula of the data analysis engine, the process of obtaining the emotion fluctuation coefficient includes:
[0066] Step 21, data standardization: Obtain the patient's regular PHQ-9, GAD-7, and BDI scale data, integrate the data of different scales by weighted average method to report their emotional state, and standardize the comprehensive score;
[0067] Step 22, mood fluctuation analysis: calculate the standard deviation of the patient's mood scores at different time points, analyze the frequency of changes in the mood state score by counting the number of changes exceeding a preset threshold, and calculate the mood fluctuation coefficient using regression analysis based on the amplitude and frequency of the mood fluctuation;
[0068] Step 23, parameter verification and optimization: Use historical data and cross-sample data to verify the accuracy and consistency of the sentiment fluctuation coefficient, and adjust the calculation parameter settings based on the verification results;
[0069] Step 24, result output: store the emotion fluctuation coefficient and related analysis data in the data storage module, and display the emotion fluctuation analysis results through charts and visualization tools.
[0070] It should be noted that in the data analysis engine, the pain potential analysis function uses the quality of life change rate obtained from the quality of life questionnaire data and the personalized coefficient of cancer survivors, combined with the current chronic pain change rate, to evaluate the long-term impact of chronic pain on quality of life. The formula of the pain potential analysis function is:
[0071]
[0072] Where: K PPF is the density function value of the pain intensity score, Q is the quality of life score, γ is the personalized coefficient of cancer survivors, which is obtained through machine learning model analysis, and t0 is the initial moment of the analysis.
[0073] Chronic pain not only affects the patient's immediate physical sensation, but also has a profound impact on the long-term quality of life. This formula aims to comprehensively evaluate the long-term impact of chronic pain on quality of life and help understand the dynamic relationship between pain and quality of life. Each cancer survivor has different backgrounds, medical histories, psychological states and other factors, which leads to different effects of pain on each person. By introducing a personalized coefficient, the formula can take into account individual differences and provide a more accurate assessment. In an integral form, the formula can capture the cumulative effects of changes in quality of life and pain in the time dimension, taking into account the temporal continuity of chronic pain and its effects, and providing a long-term observation perspective. The rate of change in quality of life reflects the dynamic changes in the patient's quality of life. By integrating it, the formula can evaluate the overall trend of changes in quality of life and provide a comprehensive long-term assessment. The personalized coefficient makes the analysis more precise and personalized. Through a machine learning model, the personalized coefficient reflects the intensity of the impact of individual characteristics on pain perception, thereby quantifying the impact of a patient's specific situation on quality of life. The integral form allows the evaluation of the cumulative effects of changes in quality of life and pain, providing data support for long-term management and prediction. This helps to develop long-term management strategies, not just short-term relief plans. The formula provides quantitative data that can be directly used in decision support systems. This data helps the healthcare team and patients understand the long-term impact of pain and develop personalized pain management device parameters and intervention strategies based on this data.
[0074] It should be further explained that in the pain potential analysis function formula analyzed by the data engine, the process of obtaining the personalized coefficient of cancer survivors includes:
[0075] Step 31, data feature extraction: obtain age, gender, weight, and height from the basic biological data of cancer survivors, extract keywords from past medical history, current diagnosis, treatment plan, and medication records, obtain heart rate, number of steps, and activity level from real-time monitoring data, and obtain key features in the collected data based on principal component analysis technology;
[0076] Step 32, machine learning model training: select a support vector machine model, train the model using historical data, and use a cross-validation method to evaluate model performance and optimize model parameters;
[0077] Step 33, personalized coefficient prediction: the key features obtained in step 31 are used as the input of the model, and the machine learning model trained in step 32 is used to output the personalized coefficient of each cancer survivor. The model and personalized coefficient are dynamically adjusted through continuous data input and feedback. The accuracy and stability of the model are verified using an independent test data set. The model parameters are adjusted according to the verification results, and the model is retrained regularly.
[0078] Step 34, result output and application: store the personalized coefficient of each patient in the data storage module.
[0079] It should be noted that in the data analysis engine, the emotion-pain interaction formula quantifies the dynamic impact of emotion changes on cancer survivors' perception of chronic pain through the emotion impact coefficient and the emotion change rate. The emotion-pain interaction formula is:
[0080]
[0081] Where: K EPIM is the emotion-pain interaction analysis value, E is the current emotion score, σ is the emotion influence coefficient, is the emotion-pain interaction coefficient, σ and Obtained through machine learning model analysis.
[0082] The impact of emotional changes on pain perception is an important factor in cancer survivors. Emotional fluctuations can exacerbate or relieve pain perception. Through the emotion-pain interaction formula, the system can quantify the dynamic impact of emotional changes on pain and provide an in-depth understanding of the complex relationship between emotion and pain. Each patient may react differently to emotional changes. The introduction of the emotion impact coefficient and the emotion-pain interaction coefficient can capture the individual's specific response to emotional changes and provide a more accurate basis for personalized pain management. By analyzing the emotion change rate and emotion score, the formula can not only evaluate the immediate impact of current emotions on pain, but also predict the future trend of emotional changes on pain perception. This dynamic assessment is of great significance for formulating intervention measures in advance. The emotion impact coefficient and emotion change rate parts of the formula provide a quantitative assessment of the impact of emotional fluctuations on pain perception, helping the medical team understand how changes in emotional state regulate the intensity of pain perception. By introducing the emotion-pain interaction coefficient and the product term of the emotion score and pain intensity, the formula can capture the complex interactive relationship between emotion and pain, which helps to identify the special impact of emotional state on pain. The parameters in the formula (σ and ) is obtained through machine learning model analysis, reflecting the individual patient's emotional and pain characteristics, allowing the system to provide more accurate pain management suggestions for each patient, especially in the combination of emotional management and pain relief strategies. The data provided by the emotion-pain interaction formula can support clinical decision-making and provide a reliable data basis for the decision support system. EPIM Indicates the comprehensive impact of emotional changes on pain perception. The higher the value, the more significant the impact of emotions on pain, which helps quantify the overall impact of emotional fluctuations on pain perception and provides data support for personalized management. σ (emotional impact coefficient) reflects the coefficient of the regulatory effect of emotional changes on pain perception. It is obtained by analyzing historical data and individual patient characteristics and using machine learning model training to quantify the sensitivity of emotional fluctuations to pain changes. It is the rate of change of emotion score per unit time. The rate of change of emotion score is calculated through the regular data of emotion assessment scale (PHQ-9, GAD-7) to reflect the dynamic changes of emotion state. (Emotion-Pain Interaction Coefficient) reflects the coefficient of the interaction between emotional state and pain perception. It is obtained by analyzing the relationship between emotional score and pain intensity through machine learning model, capturing the interactive effect of emotion and pain perception, and quantifying the complex impact of emotional state on pain perception. E (Current Emotion Score) reflects the patient's current emotional state. It is regularly obtained through emotional state assessment tools (PHQ-9, GAD-7) to provide quantitative data of immediate emotional state. P (Current Pain Intensity Score) is usually obtained through a visual analog scale (VAS) or a numerical rating scale (NRS). It is the pain intensity data regularly self-reported by patients, providing quantitative data of immediate pain perception.
[0083] It should be further explained that in the emotion-pain interaction formula of the data analysis engine, the process of obtaining the emotion impact coefficient includes:
[0084] Step 41, data standardization: collecting data from PHQ-9, GAD-7, and BDI to obtain emotional score data of cancer survivors, and standardizing the scores of different emotional scales;
[0085] Step 42, emotion change analysis: calculating the change rate of emotion scores of cancer survivors at different time points based on the difference method;
[0086] Step 43, prediction of emotion impact coefficient: select the random forest model, input the emotion change rate, basic biological data, medical history, and current treatment plan, use historical data to train the model, try to use the cross-validation method to evaluate the model performance during training, use the trained model to predict new data, and obtain the emotion impact coefficient;
[0087] Step 44, parameter verification and optimization: using an independent verification data set to test the emotion impact coefficient, and adjusting the model parameters according to the verification results to improve the model structure and regularly update the model based on the relationship between the predicted emotion impact coefficient and the actual emotion and pain perception data;
[0088] Step 45, result output: the predicted emotion impact coefficient and related analysis data are stored in the data storage module, and the emotion impact coefficient is combined with the emotion change rate and other related variables in the emotion-pain interaction formula to quantify the dynamic impact of emotion changes on pain perception.
[0089] It should be noted that in the emotion-pain interaction formula of the data analysis engine, the process of obtaining the emotion-pain interaction coefficient includes:
[0090] Step 51, data standardization: standardizing the patient's emotional state data and pain intensity data;
[0091] Step 52, interaction effect analysis: using a multivariate regression analysis model to analyze the interaction effect between emotion and pain intensity, with a particular focus on how emotion changes regulate changes in pain perception, using historical data to train the multivariate regression analysis model to predict the emotion-pain interaction coefficient, with input features being the standardized current emotion score, pain intensity score, emotion change rate, and pain change rate;
[0092] Step 53, interaction coefficient prediction: Use the trained model to analyze new emotion and pain data, predict the emotion-pain interaction coefficient, and dynamically adjust and update the model based on new data input and patient feedback;
[0093] Step 54, parameter validation: Use an independent validation data set to test the emotion-pain interaction coefficient, compare the pain response predicted by the model with the actual data, analyze the performance of the model in different patient groups, adjust the model parameters and structure according to the validation results, and retrain the model regularly;
[0094] Step 55, result output and application: The emotion-pain interaction coefficient and related analysis data are stored in the data storage module, and the emotion-pain interaction coefficient is combined with the emotion influence coefficient and other related variables in the emotion-pain interaction formula to quantify the dynamic impact of emotion changes on pain perception.
[0095] It should be noted that the personalized recommendation generator receives the pain fluctuation analysis value, the density function value of the pain intensity score, the emotion-pain interaction analysis value and the personalization coefficient, the density function of the pain intensity score, the emotion impact coefficient, and the emotion-pain interaction coefficient obtained by the data analysis engine, integrates the data, extracts the key features of cancer survivors from these data, and uses the weighted summation method to combine the extracted key features to form a comprehensive score for evaluating the risk level of cancer survivors. According to the pre-set risk level classification criteria, the score is divided into high risk, medium risk and low risk, and recommendations are generated based on the rule engine.
[0096] It should be noted that in the personalized suggestion generator, the preset rules are:
[0097] High-risk cancer survivors need to adjust the dosage of existing analgesics, recommend more auxiliary plans, and increase the level of exercise in daily life; medium-risk cancer survivors maintain the current plan, conduct regular assessments, and maintain the current level of exercise; low-risk cancer survivors gradually reduce the amount and frequency of analgesics and maintain the current level of exercise.
[0098] It should be noted that the personalized suggestion generator collects feedback data from patients after they implement the suggestions, dynamically adjusts the suggestion generation content and priority based on the feedback information and the input of new data, and regularly evaluates and updates the rule base.
[0099] It should be noted that the user interaction module transmits the information generated by the personalized suggestion generator to patients and medical staff for display, and readjusts the personalized suggestions based on the interactions of cancer survivors and medical staff, and supports the issuance of adjustment suggestion information to real-time monitoring devices.
[0100] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
[0101] Finally: The above description is only an implementation mode of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.
Claims
1. A chronic pain management system for cancer survivors based on patient-reported outcomes, characterized in that: It includes a data acquisition module, a data analysis engine, a personalized suggestion generator, a user interaction module and a data storage module. The data acquisition module is connected to the data analysis engine, the data engine is connected to the personalized suggestion generator, and the user interaction module and the data storage module are both connected to the data acquisition module, the data analysis engine and the personalized suggestion generator. The data analysis engine obtains pre-processed multi-source data from the PROs report results of cancer survivors, extracts key features from the original data and generates new features by applying mathematical changes, evaluates the severity of chronic pain in cancer survivors based on the pain fluctuation analysis formula, analyzes the long-term impact of chronic pain on the quality of life of cancer survivors based on the pain potential analysis function, and quantifies the dynamic impact of emotional changes on the chronic pain perception of cancer survivors based on the emotion-pain interaction formula. The personalized suggestion generator performs personalized feature analysis based on the data obtained by the data analysis engine, realizes automatic grouping of chronic pain in cancer survivors based on the rule engine, sets different pain management priorities and strategies based on the analysis results, automatically generates chronic pain drug management data, non-drug management data and life management data, and dynamically adjusts the priority and content of suggestions based on patient feedback and continuous data updates.
2. A chronic pain management system for cancer survivors based on patient-reported outcomes according to claim 1, characterized in that: The data collection module obtains pain intensity data through the visual analog scale and digital rating scale in patient-reported outcomes (PROs), obtains emotional score data of cancer survivors through the PHQ-9, GAD-7, and BDI scales, and obtains the health status assessment of patients in multiple life dimensions. It obtains the patient's basic biological data, past medical history, current diagnosis, treatment plan, and medication records through background information and clinical data, monitors the patient's heart rate, number of steps, and activity level in real time through wearable devices, performs data preprocessing and data cleaning, and transmits the data to the data storage module for storage through encryption.
3. A chronic pain management system for cancer survivors based on patient reported outcomes according to claim 1, characterized in that: In the data engine, the pain fluctuation analysis formula analyzes the impact of emotional fluctuations on chronic pain by combining the rate of change and acceleration of chronic pain intensity of cancer survivors and the patient's emotional fluctuation coefficient, and identifies the severity of chronic pain in cancer survivors. The pain fluctuation analysis formula is: Where: K PTI is the pain fluctuation analysis value, P is the pain intensity score, t is the calculation time point, is the rate of change in pain intensity score, is the acceleration of pain intensity, and It was obtained by time series analysis method, α was the density function of pain intensity score, which was obtained by analyzing the data of visual analog scale and digital rating scale, and β was the mood fluctuation coefficient, which was obtained by analyzing the data of mood state assessment scale.
4. A chronic pain management system for cancer survivors based on patient reported outcomes according to claim 3, characterized in that: In the pain fluctuation analysis formula of the data analysis engine, the process of obtaining the density function of the pain intensity score includes: Step 11, statistically analyzing the collected visual analog scale and digital rating scale data to obtain the mean value and standard deviation of the ratings; Step 12, density function construction: Based on the statistical analysis results, the density function of the pain intensity score is constructed by the kernel density estimation method to obtain a smooth density curve. Step 13, parameter calibration and verification: calibrate the parameters of the density function by comparing with historical data, verify the density function using data at different time points, and further optimize the adjustment parameters of the density function. Step 14, result output: storing the calculated density function of the pain intensity score and related statistics in a data storage module; Step 15, data visualization: draw a distribution curve of the scores and visualize the distribution of the scores.
5. A chronic pain management system for cancer survivors based on patient reported outcomes according to claim 3, characterized in that: In the pain fluctuation analysis formula of the data analysis engine, the process of obtaining the emotion fluctuation coefficient includes: Step 21, data standardization: Obtain the patient's regular PHQ-9, GAD-7, and BDI scale data, integrate the data of different scales by weighted average method to report their emotional state, and standardize the comprehensive score; Step 22, mood fluctuation analysis: calculate the standard deviation of the patient's mood scores at different time points, analyze the frequency of changes in the mood state score by counting the number of changes exceeding a preset threshold, and calculate the mood fluctuation coefficient using regression analysis based on the amplitude and frequency of the mood fluctuation; Step 23, parameter verification and optimization: Use historical data and cross-sample data to verify the accuracy and consistency of the sentiment fluctuation coefficient, and adjust the calculation parameter settings based on the verification results; Step 24, result output: store the emotion fluctuation coefficient and related analysis data in the data storage module, and display the emotion fluctuation analysis results through charts and visualization tools.
6. A chronic pain management system for cancer survivors based on patient reported outcomes according to claim 3, characterized in that: In the data analysis engine, the pain potential analysis function uses the quality of life change rate obtained from the quality of life questionnaire data and the personalized coefficient of cancer survivors, combined with the current chronic pain change rate, to evaluate the long-term impact of chronic pain on quality of life. The formula for the pain potential analysis function is: Where: K PPF is the density function value of the pain intensity score, Q is the quality of life score, γ is the personalized coefficient of cancer survivors, which is obtained through machine learning model analysis, and t0 is the initial moment of the analysis.
7. A chronic pain management system for cancer survivors based on patient reported outcomes according to claim 6, characterized in that: In the pain potential analysis function formula analyzed by the data engine, the process of obtaining the personalized coefficient of cancer survivors includes: Step 31, data feature extraction: obtain age, gender, weight, and height from the basic biological data of cancer survivors, extract keywords from past medical history, current diagnosis, treatment plan, and medication records, obtain heart rate, number of steps, and activity level from real-time monitoring data, and obtain key features in the collected data based on principal component analysis technology; Step 32, machine learning model training: select a support vector machine model, train the model using historical data, and use a cross-validation method to evaluate model performance and optimize model parameters; Step 33, personalized coefficient prediction: the key features obtained in step 31 are used as the input of the model, and the machine learning model trained in step 32 is used to output the personalized coefficient of each cancer survivor. The model and personalized coefficient are dynamically adjusted through continuous data input and feedback. The accuracy and stability of the model are verified using an independent test data set. The model parameters are adjusted according to the verification results, and the model is retrained regularly. Step 34, result output and application: store the personalized coefficient of each patient in the data storage module.
8. The chronic pain management system for cancer survivors based on patient-reported outcomes according to claim 3, characterized in that: In the data analysis engine, the emotion-pain interaction formula quantifies the dynamic impact of emotion changes on cancer survivors' perception of chronic pain through the emotion impact coefficient and emotion change rate. The emotion-pain interaction formula is: Where: K EPIM is the emotion-pain interaction analysis value, E is the current emotion score, σ is the emotion influence coefficient, is the emotion-pain interaction coefficient, σ and Obtained through machine learning model analysis.
9. A chronic pain management system for cancer survivors based on patient reported outcomes according to claim 8, characterized in that: In the emotion-pain interaction formula of the data analysis engine, the process of obtaining the emotion impact coefficient includes: Step 41, data standardization: collecting data from PHQ-9, GAD-7, and BDI to obtain emotional score data of cancer survivors, and standardizing the scores of different emotional scales; Step 42, emotion change analysis: calculating the change rate of emotion scores of cancer survivors at different time points based on the difference method; Step 43, prediction of emotion impact coefficient: select the random forest model, input the emotion change rate, basic biological data, medical history, and current treatment plan, use historical data to train the model, try to use the cross-validation method to evaluate the model performance during training, use the trained model to predict new data, and obtain the emotion impact coefficient; Step 44, parameter verification and optimization: using an independent verification data set to test the emotion impact coefficient, and adjusting the model parameters according to the verification results to improve the model structure and regularly update the model based on the relationship between the predicted emotion impact coefficient and the actual emotion and pain perception data; Step 45, result output: the predicted emotion impact coefficient and related analysis data are stored in the data storage module, and the emotion impact coefficient is combined with the emotion change rate and other related variables in the emotion-pain interaction formula to quantify the dynamic impact of emotion changes on pain perception.
10. The chronic pain management system for cancer survivors based on patient reported outcomes according to claim 8, characterized in that: In the emotion-pain interaction formula of the data analysis engine, the process of obtaining the emotion-pain interaction coefficient includes: Step 51, data standardization: standardizing the patient's emotional state data and pain intensity data; Step 52, interaction effect analysis: using a multivariate regression analysis model to analyze the interaction effect between emotion and pain intensity, with a particular focus on how emotion changes regulate changes in pain perception, using historical data to train the multivariate regression analysis model to predict the emotion-pain interaction coefficient, with input features being the standardized current emotion score, pain intensity score, emotion change rate, and pain change rate; Step 53, interaction coefficient prediction: using the trained model to analyze new emotion and pain data, predict the emotion-pain interaction coefficient, and dynamically adjust and update the model based on new data input and patient feedback; Step 54, parameter validation: Use an independent validation data set to test the emotion-pain interaction coefficient, compare the pain response predicted by the model with the actual data, analyze the performance of the model in different patient groups, adjust the model parameters and structure according to the validation results, and retrain the model regularly; Step 55, result output and application: The emotion-pain interaction coefficient and related analysis data are stored in the data storage module, and the emotion-pain interaction coefficient is combined with the emotion influence coefficient and other related variables in the emotion-pain interaction formula to quantify the dynamic impact of emotion changes on pain perception.
Citation Information
Cited By
Follow-up visit consultation system for chronic pain patients
CN120708943A
Intelligent cancer pain follow-up visit system and method based on large model
CN120766902A
Chronic pain relieving method based on virtual reality and time interference nerve regulation
CN121687488A
