Post-radiotherapy patient nursing auxiliary decision-making system based on artificial intelligence

Abstract

The invention discloses a post-radiotherapy patient nursing auxiliary decision-making system based on artificial intelligence, and relates to the field of medical rehabilitation nursing, and the system comprises a management and control module which serves as a core management end of each function module and carries out the editing and sending of a control instruction; the data archive library is used for storing all the collected user personal data and analysis data thereof; the data acquisition module is used for acquiring basic physique data, illness state related data and radiotherapy plans of the patient through a preset period; individual data of patients are analyzed through a prediction model, personalized nursing suggestions are generated based on medical history, health indexes and radiotherapy plans, personalized nursing schemes are provided for each patient according to curative effect analysis of similar patients, customization is carried out according to unique illness states, physiques and treatment plans of the patients, professional knowledge and data in the medical field are combined, and the patient health management system is established. Potential abnormal changes are output, and patients and medical staff are helped to make timely decisions.

Description

Technical Field

[0001] The present invention relates to the technical field of medical rehabilitation nursing, and in particular to an artificial intelligence-based decision-making support system for nursing patients after radiotherapy. Background Art

[0002] With the increase in the number of patients after radiotherapy, the medical system is facing the problem of insufficient nursing resources. The care of patients after radiotherapy requires close cooperation between radiologists, oncologists, nurses and other related professionals. At present, medical imaging, radiotherapy technology and computer science are developing rapidly. The technical level of radiotherapy and related fields has been significantly improved. The application of artificial intelligence technology in data processing, pattern recognition and predictive analysis has greatly improved the accuracy of medical decision-making. Individual differences play an important role in cancer treatment. Patients have different genetic backgrounds, tumor types, and responses to treatment. Therefore, personalized treatment and care have become an important trend. Patients after radiotherapy often face multiple health problems, such as the management of radiotherapy side effects, the prevention of complications, and mental health issues. Existing nursing decision-making systems often adopt universal standards and lack personalized adjustments for individual patients, which can easily lead to patients not being fully considered and meticulously managed during the nursing process, affecting the patient's recovery effect. It is difficult to effectively integrate multi-dimensional data from different sources. The lack of a comprehensive analysis system may lead to reduced accuracy and effectiveness of decision support. It is often impossible to self-learn and improve through the collected data. When faced with new patient data and clinical conditions, it is difficult to adapt to and optimize the decision support model. Summary of the invention

[0003] In view of the above-mentioned shortcomings of the prior art, the present invention provides an artificial intelligence-based post-radiotherapy patient care decision-making support system, which can effectively solve the problems of the prior art.

[0004] To achieve the above objectives, the present invention is implemented through the following technical solutions.

[0005] The present invention discloses an artificial intelligence-based post-radiotherapy patient nursing decision-making auxiliary system, comprising: The control module, as the core management end of each functional module, edits and sends control instructions; Data archive, used to store all collected user personal data and its analysis data; A data collection module is used to collect the patient's basic physical data, disease-related data and radiotherapy plan through a preset cycle; The plan analysis module is used to convert the dose, cycle and target site in the patient's radiotherapy plan into numerical values ​​that can be used by the model, read the radiotherapy history data, and analyze the factors that affect the patient's body in the radiotherapy plan; The model building module is used to build a prediction model based on a machine learning algorithm, input the patient's current status data, historical treatment data, and future radiotherapy plan into the prediction model, and output the patient's physical condition and abnormal disease characteristics after the next cycle of radiotherapy; The anomaly identification module is used to identify the types of anomaly characteristics output by the model building module and classify them; Nursing strategy library, used to store standard nursing strategies for different abnormal characteristics. Nursing strategies are stored in a structured manner. Intelligent matching module, used to search the nursing strategy library one by one according to the abnormal characteristics output by the current prediction model, and match several nursing strategies suitable for the current patient based on collaborative filtering of similarity; A plan generation unit is used to integrate several matched nursing strategies to generate a complete nursing plan for the current cycle; The interactive module is used to provide a patient interactive interface, personal identity verification and exclusive care plan review.

[0006] Furthermore, the basic physical data collected by the data collection module include: blood pressure, heart rate, blood oxygen, weight and body temperature; the disease-related data include: tumor marker levels, immune indicators, blood routine results and affected area status; the radiotherapy plan includes: radiotherapy dose, frequency and location.

[0007] Furthermore, the data acquisition module is interactively connected with the filtering module via a wireless network, and the filtering module is used to filter redundant and useless data collected by the data acquisition module, and normalize the filtered data.

[0008] Furthermore, the calculation formula of the prediction model constructed by the model construction module is: ; In the formula, represents the target variable for prediction, indicating an indicator of the patient's physical condition or disease status after the next cycle of radiotherapy; represents the intercept of the model, reflecting the basic health status effect; Represents the number of indicators; Represents the independent variable The coefficient of , which indicates the degree of influence of each independent variable on the predicted value; Represents the independent variable vector, indicating various factors that affect the predicted target Y; Represents a nonlinear function, which is used to capture the complex relationship between the patient's constitution and disease status and the input features; Represents the error term, which indicates the random interference factors that the forecasting model fails to capture.

[0009] Furthermore, the nonlinear function The calculation formula is: ; In the formula, Represents the weight matrix of the Lth layer, with a size of , Represents the number of neurons in the current layer, represents the number of neurons in the previous layer, represents the output of the previous layer, Represents the bias term of the Lth layer, which is , Represents the input feature vector.

[0010] Furthermore, the As the output of the previous layer, it is the value after the activation function, expressed as: ; In the formula, g represents the activation function, which is used to introduce nonlinearity. Representative The weight matrix of the layer, Representative The output of the layer, Representative The bias term of the layer.

[0011] Furthermore, the standard nursing strategies stored in the nursing strategy library include: Therapeutic interventions: medication recommendations and treatments for specific abnormalities; Nutritional support: dietary advice for weakened immunity; Psychological care: Psychological counseling programs to relieve patients' anxiety or depression.

[0012] Furthermore, the solution generation unit has submodules deployed at the lower level, and the submodules include: a reading module, an adjustment module and an update module, the reading module and the adjustment module are interactively connected via a wireless network, and the adjustment module and the update module are interactively connected via a wireless network, wherein: The reading module is used to identify all nursing strategies and summarize them by implementation period; An adjustment module, used for receiving parameters for manually customizing the nursing strategy; The updating module is used to receive the adjustment commands sent by the regulating module and update the nursing strategy in real time.

[0013] Furthermore, the interactive module is interactively connected with the feedback module via a wireless network, and the feedback module is used to collect feedback data from patients after they implement the nursing plan, evaluate the completion of the current nursing plan and patient satisfaction, and serve as a reference for adjusting the nursing plan.

[0014] Furthermore, the management and control module is interactively connected to the data archive and the data acquisition module through a wireless network, the data acquisition module is interactively connected to the plan analysis module through a wireless network, the plan analysis module is interactively connected to the model construction module through a wireless network, the model construction module is interactively connected to the abnormality identification module through a wireless network, the intelligent matching module is interactively connected to the abnormality identification module, the nursing strategy library and the plan generation unit through a wireless network, and the plan generation unit is interactively connected to the interaction module through a wireless network.

[0015] Compared with the prior art, the technical solution provided by the present invention has the following beneficial effects: Through predictive models, we analyze patients' individual data and generate personalized care recommendations based on their medical history, health indicators, and radiotherapy plans. Based on the efficacy analysis of similar patients, we provide each patient with a personalized care plan that is tailored to the patient's unique condition, constitution, and treatment plan. Combined with expertise and data in the medical field, we output potential abnormal changes to help patients and medical staff make timely decisions.

[0016] Through intelligent data analysis, the efficiency and accuracy of nursing decisions can be improved and human errors can be reduced. The system automatically analyzes and integrates information to provide decision-making support for medical staff. Through automated and intelligent processes, the workload of medical staff can be reduced and work efficiency can be improved.

[0017] By continuously collecting new data for self-learning and optimization, we can improve future performance, promptly use newly collected data to fine-tune the model, ensure the accuracy and timeliness of the predictive model, and continuously optimize the algorithm and data processing process to adapt to changing clinical needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0019] Figure 1 A schematic diagram of the framework of an artificial intelligence-based decision-making support system for nursing post-radiotherapy patients in the present invention; Figure 2It is a schematic diagram of the framework of the solution generation unit in the present invention.

[0020] The numbers in the figure represent, respectively, 1. Management and control module; 2. Data archive; 3. Data acquisition module; 4. Plan analysis module; 5. Model building module; 6. Abnormal identification module; 7. Nursing strategy library; 8. Intelligent matching module; 9. Plan generation unit; 91. Reading module; 92. Adjustment module; 93. Update module; 10. Interaction module; 11. Feedback module; 12. Filtering module. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution 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 part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] The present invention will be further described below in conjunction with the embodiments.

[0023] Embodiment 1: This embodiment is an artificial intelligence-based post-radiotherapy patient care decision-making support system, such as Figure 1 As shown, including: Control module 1, as the core management end of each functional module, edits and sends control instructions; Data archive 2, used to store all collected user personal data and its analysis data; safely store all user personal data and its analysis results; The data collection module 3 is used to collect the patient's basic physical data, disease-related data and radiotherapy plan through a preset cycle; the collected basic physical data include: blood pressure, heart rate, blood oxygen, weight and body temperature, the disease-related data include: tumor marker levels, immune indicators, blood routine results and affected area status, and the radiotherapy plan includes: radiotherapy dose, frequency and location; the data collection module 3 is interactively connected with the filtering module 12 through a wireless network, and the filtering module 12 is used to filter redundant and useless data collected by the data collection module 3, and normalize the filtered data. Regularly and automatically collect the patient's physical data, disease-related data and radiotherapy plan, which provides basic data for subsequent analysis. Periodic data collection helps to monitor the patient's condition in real time, so as to achieve early detection and early intervention; Plan analysis module 4 is used to convert the dose, cycle and target site in the patient's radiotherapy plan into numerical values ​​that can be used by the model, read the radiotherapy history data, and analyze the factors that affect the patient's body in the radiotherapy plan, such as the dose accumulation effect and the sensitivity of the tissue around the target; convert the information of the radiotherapy plan into standardized data that can be used by the prediction model, and conduct in-depth analysis in combination with historical data, which helps to identify possible physical reactions during radiotherapy and enhance the targetedness of personalized treatment; Model building module 5 is used to build a prediction model based on a machine learning algorithm, input the patient's current status data, historical treatment data and future radiotherapy plan into the prediction model, and output the patient's physical condition and abnormal characteristics of the disease after the next cycle of radiotherapy, so as to provide a scientific basis for follow-up care; The abnormality identification module 6 is used to identify the type of abnormal characteristics output by the model building module 5 and classify them; the abnormal conditions output by the prediction model are classified and graded, so that the hospital and the nursing team can take corresponding nursing measures according to different abnormal characteristics, thereby improving the pertinence and effectiveness of nursing; Nursing strategy library 7 is used to store standard nursing strategies for different abnormal characteristics. The nursing strategies are stored in a structured manner. The standard nursing strategies include: Therapeutic interventions: medication recommendations and treatments for specific abnormalities; Nutritional support: dietary advice for weakened immunity; Psychological care: Psychological counseling programs to relieve patients’ anxiety or depression; The nursing strategies for different abnormalities are stored in a structured manner, including treatment, nutrition and psychological care strategies, making the nursing strategies more systematic and professional through standardized management methods; The intelligent matching module 8 is used to retrieve the nursing strategy library 7 one by one according to the abnormal characteristics output by the current prediction model, and match a number of nursing strategies suitable for the current patient based on similarity collaborative filtering; by matching the current prediction data with the nursing strategy library 7 one by one, based on similarity collaborative filtering technology, the nursing strategy most suitable for the patient can be selected. This flexible matching can significantly improve nursing efficiency and accuracy and enhance patient experience; The plan generation unit 9 is used to integrate the matched nursing strategies to generate a complete nursing plan for the current cycle; multiple matched nursing strategies are integrated into a complete nursing plan to ensure that the needs of the patient are fully covered and the treatment effect is optimized; The interactive module 10 is used to provide a patient interactive interface, personal identity verification and exclusive nursing plan review; the interactive module 10 is interactively connected with the feedback module 11 through a wireless network, and the feedback module 11 is used to collect feedback data after the patient implements the nursing plan, evaluate the completion of the current nursing plan and patient satisfaction, and serve as a reference for adjusting the nursing plan; like Figure 1 As shown, the control module 1 is interactively connected with the data archive 2 and the data acquisition module 3 through a wireless network, the data acquisition module 3 is interactively connected with the plan analysis module 4 through a wireless network, the plan analysis module 4 is interactively connected with the model construction module 5 through a wireless network, the model construction module 5 is interactively connected with the abnormality recognition module 6 through a wireless network, the intelligent matching module 8 is interactively connected with the abnormality recognition module 6, the nursing strategy library 7 and the solution generation unit 9 through a wireless network, and the solution generation unit 9 is interactively connected with the interaction module 10 through a wireless network; In the specific implementation of this embodiment, the global function module is controlled by the control module 1, the personal medical records of all patients are stored by the data archive 2, the personal data and medical data of the user are collected by the data collection module 3, and the personal data and medical data are filtered by the filtering module 12; The patient's current radiotherapy plan is analyzed through the plan analysis module 4, and a prediction model is built through the model building module 5. By inputting the user's current personal physical data, disease data and radiotherapy plan, the patient's future physical abnormalities are output. Based on these abnormalities, the nursing strategy library 7 is indexed through the intelligent matching module 8 to obtain the nursing strategy for the abnormal data. The nursing strategy is read through the reading module 91. The patient or professional personnel such as a doctor adjusts the strategy through the adjustment module 92. The nursing strategy is updated in real time through the update module 93. The user views the nursing plan composed of several nursing strategies through the interaction module 10, and the patient gives feedback on the implementation effect of the nursing plan through the feedback module 11; Analyze the patient's individual data through predictive models, generate personalized nursing recommendations based on medical history, health indicators and radiotherapy plans, and provide each patient with personalized nursing plans based on the efficacy analysis of similar patients. Tailor-made nursing plans are based on the patient's unique condition, constitution and treatment plan. Combined with professional knowledge and data in the medical field, potential abnormal changes are output to help patients and medical staff make timely decisions. Through intelligent data analysis, the efficiency and accuracy of nursing decisions are improved, human errors are reduced, and the system automatically analyzes and integrates information to provide decision-making support for medical staff. Through automated and intelligent processes, the workload of medical staff is reduced and work efficiency is improved; By continuously collecting new data for self-learning and optimization, we can improve future performance, promptly use newly collected data to fine-tune the model, ensure the accuracy and timeliness of the predictive model, and continuously optimize the algorithm and data processing process to adapt to changing clinical needs.

[0024] Compared with the existing technology, the present invention reduces the time and errors of manual intervention and improves the efficiency of data processing through automated data collection and processing. Through the use of machine learning models, it can conduct in-depth analysis of individual patients' responses and needs, thereby improving the accuracy of nursing plans. The system can match the most appropriate nursing strategy according to the different data characteristics of each patient, thereby achieving more personalized nursing plans.

[0025] Real-time data collection and analysis capabilities provide a basis for medical staff to respond quickly, enabling them to detect abnormal situations in a timely manner and take necessary measures. The structured storage and standardized management of nursing strategies make the implementation of nursing plans more orderly and effective when managing large-scale patients. Through these advantages, the system can significantly improve the quality of care for patients after radiotherapy and provide patients with more comprehensive and higher-quality medical support.

[0026] Embodiment 2: In other aspects, this embodiment also provides a calculation formula of a prediction model, specifically: ; In the formula, represents the target variable for prediction, indicating an indicator of the patient's physical condition or disease status after the next cycle of radiotherapy; Represents the intercept of the model, which means that when all independent variables When the values ​​of are all zero, baseline level, which reflects the basic health status influence, Represents the number of indicators; Represents the independent variable The coefficient indicates the influence of each independent variable on the predicted value. express For each additional unit, The expected amount of change can include basic patient data, such as age, gender, weight, etc., and can also include information related to the radiotherapy plan, such as radiotherapy dose, procedure time, etc. represents the independent variable vector, which indicates the various factors that affect the predicted target Y, including various clinical and lifestyle parameters, such as the patient's age, body mass index, radiotherapy dose, expected number of radiotherapy days, and past medical history. Represents a nonlinear function, which is used to capture the complex relationship between the patient's constitution and disease status and the input features; represents the error term, which indicates the random interference factors that the forecast model fails to capture. It is an independent and identically distributed random variable, and its mean is usually assumed to be zero; This prediction model formula combines the characteristics of linear and nonlinear parts, which can more effectively process complex patient data and radiotherapy plans, and provide more accurate predictions of physical constitution and disease status; Nonlinear functions The calculation formula is: ; In the formula, Represents the weight matrix of the Lth layer, with a size of , Represents the number of neurons in the current layer, Represents the number of neurons in the previous layer. The weight matrix is ​​used for linear transformation to map the input of the current layer to the output of this layer. represents the output of the previous layer, Represents the bias term of the Lth layer, which is , the bias term helps adjust the output of the predicted value to make it more flexible, represents the input feature vector, which represents the set of features required for calculation and can be a combination of multiple clinical and lifestyle parameters; As the output of the previous layer, it is the value after the activation function, expressed as: ; In the formula, g represents the activation function, which is used to introduce nonlinearity so that the model can learn complex feature relationships. It acts on the output of a linear transformation to determine whether a neuron is activated. Representative The weight matrix of the layer, Representative The output of the layer, Representative The bias term of the layer; By using the nonlinear function f(Z), the prediction model can capture the complex relationship between input features and enhance the prediction ability. In practical applications, through the combination of multiple levels and activation functions, it can more effectively learn the potential patterns of patients' physical conditions and disease changes.

[0027] Embodiment 3: In this embodiment, as Figure 2 As shown, the solution generation unit 9 has submodules deployed at the lower level, including: a reading module 91, an adjustment module 92 and an update module 93, the reading module 91 and the adjustment module 92 are interactively connected via a wireless network, and the adjustment module 92 and the update module 93 are interactively connected via a wireless network, wherein: A reading module 91 is used to identify all nursing strategies and summarize them by implementation period; The reading module 91 is responsible for identifying all available nursing strategies in the system and summarizing them according to the implementation period, so that the nursing team can quickly view and obtain nursing strategy information for different treatment stages, which is convenient for the overall planning and adjustment of patient treatment. Through systematic summary, the nursing team can quickly select the most appropriate strategy, improve nursing work efficiency, and reduce errors caused by information asymmetry; An adjustment module 92, used for receiving parameters for manually customizing the nursing strategy; The adjustment module 92 enables nursing staff to flexibly adjust the parameters of the nursing strategy according to clinical needs. In the actual nursing process, individual differences of patients and the uncertainty of their response to treatment often require timely adjustment of the nursing strategy. The implementation of the adjustment module 92 provides the nursing team with a higher degree of freedom, enabling them to respond quickly and make effective strategy adjustments based on the patient's physical condition and response; An updating module 93 is used to receive the adjustment command sent by the adjusting module 92 and update the nursing strategy in real time; The update module 93 is responsible for receiving the adjustment commands from the adjustment module 92 and updating the nursing strategy in real time. As the patient's condition changes, the nursing team can immediately detect and track and gradually adjust the nursing strategy to ensure that the implemented nursing measures are effective. Real-time updating of the nursing strategy enhances the timeliness and adaptability of nursing and ensures that the nursing process is consistent with the patient's actual needs.

[0028] Compared with existing technologies: Compared with fixed nursing plans, the present invention allows nursing staff to make independent adjustments based on the patient's real-time condition, thereby improving the adaptability of nursing plans. The system updates nursing strategies in real time, effectively avoiding nursing omissions caused by information lags, ensuring that patients can obtain the most appropriate nursing support. The nursing strategy identification and summary functions of reading module 91 ensure that a comprehensive nursing strategy view is provided regularly, so that nursing staff can efficiently select appropriate strategies, thereby efficiently integrating and implementing nursing measures at different stages. Through the design of manually customized adjustment parameters, the interactive experience between nursing staff and the system is improved, making it more in line with the habits of actual clinical operations.

[0029] The present invention controls the global function module through the control module 1, stores the personal medical records of all patients through the data archive 2, collects the user's personal data and medical data through the data collection module 3, and filters the personal data and medical data through the filtering module 12; analyzes the patient's current radiotherapy plan through the plan analysis module 4, builds a prediction model through the model building module 5, outputs the patient's future physical abnormalities by inputting the user's current personal physical data, medical data and radiotherapy plan, and based on these abnormalities, indexes are performed in the nursing strategy library 7 through the intelligent matching module 8 to obtain nursing strategies for abnormal data, reads the nursing strategies through the reading module 91, and the patients or professionals such as doctors adjust the strategies through the adjustment module 92, and update the nursing strategies in real time through the updating module 93. The user views the nursing plan composed of several nursing strategies through the interactive module 10, and the patient feedbacks the implementation effect of the nursing plan through the feedback module 11; Through real-time data collection and analysis, we provide evidence-based medical decisions, use machine learning algorithms and similarity matching to customize nursing plans, improve patient satisfaction, predict abnormal characteristics that may occur in patients in advance, strengthen monitoring and management of high-risk patients, and improve the standardization and effectiveness of nursing through a structured nursing strategy library7, and reduce arbitrariness and errors.

[0030] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An artificial intelligence-based decision-making support system for post-radiotherapy patient care, characterized in that: include: The control module, as the core management end of each functional module, edits and sends control instructions; Data archive, used to store all collected user personal data and its analysis data; A data collection module is used to collect the patient's basic physical data, disease-related data and radiotherapy plan through a preset cycle; The plan analysis module is used to convert the dose, cycle and target site in the patient's radiotherapy plan into numerical values ​​that can be used by the model, read the radiotherapy history data, and analyze the factors that affect the patient's body in the radiotherapy plan; The model building module is used to build a prediction model based on a machine learning algorithm, input the patient's current status data, historical treatment data, and future radiotherapy plan into the prediction model, and output the patient's physical condition and abnormal disease characteristics after the next cycle of radiotherapy; The anomaly identification module is used to identify the types of anomaly characteristics output by the model building module and classify them; Nursing strategy library, used to store standard nursing strategies for different abnormal characteristics. Nursing strategies are stored in a structured manner. Intelligent matching module, used to search the nursing strategy library one by one according to the abnormal characteristics output by the current prediction model, and match several nursing strategies suitable for the current patient based on collaborative filtering of similarity; A plan generation unit is used to integrate several matched nursing strategies to generate a complete nursing plan for the current cycle; The interactive module is used to provide a patient interactive interface, personal identity verification and exclusive care plan review.

2. According to claim 1, an artificial intelligence-based decision-making support system for post-radiotherapy patient care is characterized in that: The basic physical data collected by the data collection module include: blood pressure, heart rate, blood oxygen, weight and body temperature; the disease-related data include: tumor marker levels, immune indicators, blood routine results and the status of the affected area; the radiotherapy plan includes: radiotherapy dose, frequency and location.

3. The artificial intelligence-based decision-making support system for post-radiotherapy patient care according to claim 1 is characterized in that: The data acquisition module is interactively connected with the filtering module via a wireless network. The filtering module is used to filter redundant and useless data collected by the data acquisition module and normalize the filtered data.

4. The artificial intelligence-based decision-making support system for post-radiotherapy patient care according to claim 1, characterized in that: The calculation formula of the prediction model constructed by the model construction module is: ； In the formula, represents the target variable for prediction, indicating an indicator of the patient's physical condition or disease status after the next cycle of radiotherapy; represents the intercept of the model, reflecting the basic health status effect; Represents the number of indicators; Represents the independent variable The coefficient of , which indicates the influence of each independent variable on the predicted value; Represents the independent variable vector, indicating various factors that affect the predicted target Y; Represents a nonlinear function, which is used to capture the complex relationship between the patient's physical condition and disease status and the input features; Represents the error term, which indicates the random interference factors that the forecasting model fails to capture.

5. The artificial intelligence-based decision-making support system for post-radiotherapy patient care according to claim 4, characterized in that: The nonlinear function The calculation formula is: ； In the formula, Represents the weight matrix of the Lth layer, with a size of , Represents the number of neurons in the current layer, represents the number of neurons in the previous layer, represents the output of the previous layer, Represents the bias term of the Lth layer, which is , Represents the input feature vector.

6. The artificial intelligence-based decision-making support system for post-radiotherapy patient care according to claim 5, characterized in that: Said As the output of the previous layer, it is the value after the activation function, expressed as: ； In the formula, g represents the activation function, which is used to introduce nonlinearity. Representative The weight matrix of the layer, Representative The output of the layer, Representative The bias term of the layer.

7. The artificial intelligence-based decision-making support system for post-radiotherapy patient care according to claim 1, characterized in that: The standard nursing strategies stored in the nursing strategy library include: Therapeutic interventions: medication recommendations and treatments for specific abnormalities; Nutritional support: dietary advice for weakened immunity; Psychological care: Psychological counseling programs to relieve patients' anxiety or depression.

8. The artificial intelligence-based decision-making support system for post-radiotherapy patient care according to claim 1, characterized in that: The scheme generation unit has submodules deployed at the lower level, and the submodules include: a reading module, an adjustment module and an update module, the reading module and the adjustment module are interactively connected via a wireless network, and the adjustment module and the update module are interactively connected via a wireless network, wherein: The reading module is used to identify all nursing strategies and summarize them by implementation period; An adjustment module, used for receiving parameters for manually customizing the nursing strategy; The updating module is used to receive the adjustment commands sent by the regulating module and update the nursing strategy in real time.

9. The artificial intelligence-based decision-making support system for post-radiotherapy patient care according to claim 1, characterized in that: The interactive module is interactively connected with the feedback module via a wireless network. The feedback module is used to collect feedback data after the patient executes the nursing plan, evaluate the completion of the current nursing plan and the patient's satisfaction, and serve as a reference for adjusting the nursing plan.

10. The artificial intelligence-based decision-making support system for post-radiotherapy patient care according to claim 1, characterized in that: The management and control module is interactively connected to the data archive and the data acquisition module through a wireless network, the data acquisition module is interactively connected to the plan analysis module through a wireless network, the plan analysis module is interactively connected to the model construction module through a wireless network, the model construction module is interactively connected to the abnormality identification module through a wireless network, the intelligent matching module is interactively connected to the abnormality identification module, the nursing strategy library and the plan generation unit through a wireless network, and the plan generation unit is interactively connected to the interaction module through a wireless network.

Images