Psychosomatic medicine mixed psychological intervention mode collaborative management system and method

Abstract

The application particularly relates to the technical field of big data integration, and discloses a psychosomatic medical mixed psychological intervention mode collaborative management system and method, which comprises a multi-source data acquisition module, an intelligent evaluation and diagnosis module, a personalized scheme generation module, a multi-disciplinary collaborative execution module, a whole-cycle monitoring feedback module and a system iteration optimization module. Through the construction of a patient multidimensional data set, the intelligent evaluation and diagnosis module automatically calculates the psychosomatic correlation degree and psychosomatic comorbidity risk total score through a psychosomatic state evaluation model, generates a precise patient portrait, and based on the intelligent evaluation and diagnosis result, automatically generates a personalized intervention scheme through rule matching and algorithm optimization. Through the whole-cycle monitoring feedback module, according to the patient intervention progress, the monitoring focus and frequency are set in stages, and a multidimensional therapeutic effect evaluation system is constructed to calculate the therapeutic effect evaluation total score, realize whole-cycle closed-loop monitoring and evaluation, greatly improve the information sharing and collaborative efficiency, and significantly improve the personalized intervention precision.

Description

Technical Field

[0001] This invention relates to the field of big data integration technology, and more specifically, to a collaborative management system and method for a hybrid psychological intervention model in psychosomatic medicine. Background Technology

[0002] The psychosomatic medicine hybrid psychological intervention model is a composite intervention system in the field of psychosomatic medicine that combines traditional offline intervention, modern digital intervention, physiological regulation and psychological intervention. With the high incidence of psychosomatic comorbidities, the transformation of medical models and the development trend of mature digital technology, the collaborative management of the psychosomatic medicine hybrid psychological intervention model has been promoted.

[0003] The existing psychosomatic medicine hybrid psychological intervention model management method is mainly interdisciplinary collaborative diagnosis and treatment management. Through team building and division of labor, multi-dimensional assessment mechanism and joint diagnosis and treatment process, it integrates multidisciplinary resources and technical means to form a systematic intervention system.

[0004] However, it still has some shortcomings in actual use. First, the efficiency of information sharing and collaboration is low. The existing psychosomatic medicine hybrid psychological intervention model management method mainly relies on manual transmission of medical records and holding offline meetings for information exchange. Manual transmission is prone to problems such as information omission and delay, resulting in information not being synchronized in a timely manner, which affects the timeliness and accuracy of collaborative diagnosis and treatment.

[0005] Second, the precision of personalized intervention is insufficient. The existing psychosomatic medicine hybrid psychological intervention model management method mainly relies on the doctor's clinical experience for judgment. Doctors have difficulty processing a large amount of patient data comprehensively and quickly, and cannot accurately capture the potential correlation between data. This results in limited sensitivity to changes in patients' conditions and accuracy of judgment, and the intervention plan is difficult to accurately match the individual differences and dynamic changes in patients' conditions.

[0006] Third, the full-cycle monitoring and evaluation is not comprehensive. The existing psychosomatic medicine hybrid psychological intervention model management methods are limited by technical conditions, making it difficult to achieve continuous and comprehensive monitoring of patients after discharge. Furthermore, the monitoring data is not obtained in a timely and comprehensive manner, and there is a lack of a systematic long-term evaluation mechanism, making it difficult to comprehensively and objectively reflect the long-term effects of the intervention. Summary of the Invention

[0007] In view of this, embodiments of the present invention provide a collaborative management system and method for a psychosomatic medicine hybrid psychological intervention model. By constructing a full-process information sharing and real-time collaboration mechanism, a model evaluation and dynamic adjustment mechanism, and a full-stage monitoring, multi-dimensional evaluation and feedback mechanism, the present invention effectively solves the problems of low information sharing and collaboration efficiency, insufficient accuracy of personalized intervention, and incomplete full-cycle monitoring and evaluation mentioned in the background art.

[0008] To achieve the above objectives, the present invention provides the following technical solution: a collaborative management system for a psychosomatic medicine hybrid psychological intervention model, comprising a multi-source data acquisition module, an intelligent assessment and diagnosis module, a personalized plan generation module, a multidisciplinary collaborative execution module, a full-cycle monitoring and feedback module, and a system iteration and optimization module.

[0009] Multi-source data acquisition module: Collects multi-dimensional patient data, including medical data, psychological data, and life scenario data, constructs a multi-dimensional patient dataset, and transmits it to the intelligent assessment and diagnosis module;

[0010] Intelligent assessment and diagnosis module: Analyzes multi-dimensional patient datasets, calculates the dynamic psychosomatic correlation of physiological and psychological indicators based on psychosomatic medicine-specific logic, and outputs the psychosomatic comorbidity risk level by combining a logistic regression model with medical history weights.

[0011] Personalized treatment plan generation module: Dynamically generates and adjusts personalized intervention plans based on intelligent assessment and diagnosis results, and transmits them to the multidisciplinary collaborative execution module;

[0012] Multidisciplinary collaborative execution module: Construct an online and offline integrated multidisciplinary collaboration platform to collaboratively execute personalized intervention plans across multiple disciplines and transmit plan execution data to the full-cycle monitoring and feedback module;

[0013] Full-cycle monitoring and feedback module: Real-time monitoring of patient data throughout the entire treatment cycle, obtaining efficacy evaluation results, establishing feedback triggering mechanisms, setting differentiated monitoring frequencies based on psychosomatic comorbidity risk levels, and pushing efficacy evaluation results and feedback information to the personalized plan generation module and system iteration optimization module;

[0014] System Iterative Optimization Module: Based on multi-dimensional patient datasets, efficacy evaluation results, and feedback information, the model is continuously iterated and optimized, and the iterated model is synchronized to the preceding module.

[0015] The collaborative management approach of the psychosomatic medicine hybrid psychological intervention model includes the following steps:

[0016] S1: Collect multi-dimensional patient data through the multi-source data acquisition module, including medical data, psychological data, and life scenario data, to construct a multi-dimensional patient dataset;

[0017] S2: Analyze multi-dimensional patient datasets, calculate the dynamic psychosomatic correlation of physiological and psychological indicators based on psychosomatic medicine-specific logic, and output the psychosomatic comorbidity risk level;

[0018] S3: The personalized intervention plan is dynamically generated and adjusted based on the intelligent assessment and diagnosis results through the personalized plan generation module;

[0019] S4: Build an online and offline integrated multidisciplinary collaboration platform to enable multidisciplinary collaborative execution of personalized intervention plans and obtain plan execution data;

[0020] S5: Real-time monitoring of patient data throughout the entire treatment cycle, obtaining efficacy evaluation results, establishing feedback triggering mechanisms and monitoring frequencies, and feeding back efficacy evaluation results and feedback information to the personalized plan generation module;

[0021] S6: The system iterative optimization module continuously optimizes the model based on multi-dimensional patient datasets, efficacy evaluation results, and feedback information.

[0022] The technical effects and advantages of this invention are as follows:

[0023] 1. This invention constructs a full-process information sharing and real-time collaboration mechanism. Multi-dimensional patient data collected by the multi-source data acquisition module is transmitted in real time to the intelligent assessment and diagnosis module. The intervention plan is formulated by the personalized plan generation module and executed collaboratively by the multidisciplinary collaboration platform. The entire process is without human intervention, which greatly improves the efficiency of information sharing and collaboration.

[0024] 2. This invention constructs a model evaluation and dynamic adjustment mechanism, builds a multi-dimensional patient dataset, and uses an intelligent assessment and diagnosis module to automatically calculate the psychosomatic correlation degree and total psychosomatic comorbidity risk score through a psychosomatic state assessment model, generating an accurate patient profile. Based on the intelligent assessment and diagnosis results, it automatically generates personalized intervention plans through rule matching and algorithm optimization, dynamically adapting to individual differences and significantly improving the accuracy of personalized intervention.

[0025] 3. This invention constructs a full-stage monitoring, multi-dimensional assessment, and feedback mechanism. Through the full-cycle monitoring and feedback module, the monitoring focus and frequency are set in stages according to the patient's intervention progress. A multi-dimensional efficacy assessment system is constructed to calculate the total efficacy assessment score, thereby realizing full-cycle closed-loop monitoring and assessment, covering all stages of diagnosis and treatment, and comprehensively reflecting the long-term effects of the intervention. Attached Figure Description

[0026] Figure 1 This is a schematic diagram of the overall structure of the present invention.

[0027] Figure 2 This is a schematic diagram of the method steps of the present invention.

[0028] Figure 3 This is a schematic diagram of the steps for determining the risk level of psychosomatic comorbidity according to the present invention. Detailed Implementation

[0029] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0030] As attached Figure 1 The collaborative management system of the psychosomatic medicine hybrid psychological intervention model shown includes a multi-source data acquisition module, an intelligent assessment and diagnosis module, a personalized plan generation module, a multidisciplinary collaborative execution module, a full-cycle monitoring and feedback module, and a system iteration and optimization module.

[0031] It should be further explained that the output of the multi-source data acquisition module is connected to the input of the intelligent assessment and diagnosis module, the output of the intelligent assessment and diagnosis module is connected to the input of the personalized solution generation module, the output of the personalized solution generation module is connected to the input of the multidisciplinary collaborative execution module, the output of the multidisciplinary collaborative execution module is connected to the input of the personalized solution generation module and the full-cycle monitoring and feedback module, and the output of the full-cycle monitoring and feedback module is connected to the input of the system iterative optimization module.

[0032] The specific embodiments of the present invention include the following:

[0033] Multi-source data acquisition module: Collects multi-dimensional patient data, including medical data, psychological data, and life scenario data, constructs a multi-dimensional patient dataset, and transmits it to the intelligent assessment and diagnosis module;

[0034] Furthermore, medical data is obtained by connecting to the hospital's clinical system, including basic patient information, laboratory test data, and imaging test data; psychological data includes subjective psychological data and objective psychological state data; and life scenario data includes exercise data, sleep data, and basic physiological data.

[0035] In this embodiment, it should be specifically noted that the hospital clinical system refers to the hospital's existing information system. The patient's basic information is obtained by connecting to the HIS hospital information system, including the patient's identity information, past medical history, and current inpatient / outpatient treatment information. The laboratory test data is obtained by connecting to the LIS laboratory information system, including physiological and biochemical indicators and stress-inflammatory indicators. The physiological and biochemical indicators include complete blood count, liver and kidney function, electrolytes, and blood glucose / glycated hemoglobin. The stress-inflammatory indicators include cortisol level and C-reactive protein. The imaging examination data is obtained by connecting to the PACS image archiving system, including cardiac ultrasound, brain MRI, and chest CT.

[0036] It should be specifically noted that psychological data is acquired through a combination of subjective scale assessment and behavioral monitoring. Subjective psychological scales are obtained through subjective scale assessment, including emotion screening scales, somatic symptom scales, and functional impact scales. Emotion screening scales include the Self-Rating Depression Scale, Self-Rating Anxiety Scale, PHQ-9, and GAD-7. Somatic symptom scales include SCL-90 and SSS. Functional impact scales include SF-36 and WSAS. Objective psychological state data is acquired through non-contact technology and professional equipment, including voice emotion data, facial expression data, and biofeedback data. Voice emotion data includes speech rate and voice emotion intensity. Facial expression data refers to the patient's facial micro-expressions, which are matched with an emotion-expression database using image recognition technology to quantify the intensity of emotions such as pleasure, sadness, anger, and fear. Biofeedback data includes skin conductivity and SDNN. SDNN refers to the standard deviation of the RR interval of all sinus heartbeats, reflecting autonomic nervous function. The SDNN of depressed patients is often <50ms.

[0037] It should be noted that the life scenario data is obtained through wearable devices. The exercise data includes daily steps, exercise duration and sedentary time, the sleep data includes total sleep duration and sleep structure, and the basic physiological data includes real-time heart rate and blood pressure.

[0038] It should be further explained that the construction of the patient multidimensional dataset requires data cleaning, data standardization, and data association labeling of the patient's multidimensional data to form a structured dataset with a unique patient identifier, data type, and collection time.

[0039] Intelligent assessment and diagnosis module: Analyzes multi-dimensional patient datasets, calculates the dynamic psychosomatic correlation of physiological and psychological indicators based on psychosomatic medicine-specific logic, and outputs the psychosomatic comorbidity risk level by combining a logistic regression model with medical history weights.

[0040] In this embodiment, it is important to note that the construction of the psychosomatic state assessment model requires selecting standardized scores Si of core features strongly correlated with psychosomatic comorbidity from the patient's multidimensional dataset as input variables. These features include medical, psychological, and lifestyle characteristics. The weights Wi of each feature are determined by combining random forest feature importance ranking with clinical expert scoring to ensure that the impact of key indicators on the assessment results aligns with clinical understanding. The psychosomatic state assessment model is constructed using a fusion algorithm of logistic regression and gradient boosting tree. A 5-year clinical case database from the psychosomatic department of a tertiary hospital is used, divided into training and test sets at an 8:2 ratio. The model parameters are optimized through 5-fold cross-validation.

[0041] Further explanation is needed regarding the medical dimension features, which include cortisol levels, HRV standardized values, physical illness severity, and inflammatory markers; the psychological dimension features, which include PHQ-9 score, GAD-7 score, SCL-90 score, skin conductivity, and emotional intensity of speech; and the lifestyle dimension features, which include the number of sleep interruptions and the duration of insufficient exercise. The HRV standardized value is the result after standardization by SDNN indicators.

[0042] Furthermore, the calculation of the psychosomatic correlation degree requires obtaining the core feature data from the patient's multi-dimensional dataset, including medical dimension features, psychological dimension features, and lifestyle dimension features. The correlation strength between the medical dimension features and the psychological dimension features is calculated using the Pearson correlation coefficient to obtain the psychosomatic correlation degree r, and the psychosomatic correlation matrix is ​​output.

[0043] In this embodiment, it should be specifically noted that the calculation of the psychosomatic correlation degree requires the use of the following formula:

[0044] ,

[0045] The psychosomatic correlation coefficient r is calculated, where n is the sample size of the patient's data in the past 7 days, x is the standardized score of medical dimension features, such as the standardized value of HRV, and y is the standardized score of psychological dimension features, such as the GAD-7 score. The value of r ranges from [-1, 1]. In the psychosomatic correlation matrix, |r| ≥ 0.7 indicates a strong correlation, 0.5 ≤ |r| < 0.7 indicates a moderate correlation, and |r| < 0.5 indicates a weak correlation. For example, HRV and GAD-7 are strongly negatively correlated, with r = -0.89, indicating that the higher the patient's anxiety level, the weaker the autonomic nervous system function, which is consistent with the clinical mechanism that anxiety activates the sympathetic nervous system and inhibits the autonomic nervous system.

[0046] Furthermore, the output of the psychosomatic comorbidity risk level requires calculating the total psychosomatic comorbidity risk score. This calculation necessitates importing the standardized scores Si of the core features from the patient's multi-dimensional dataset into the psychosomatic state assessment model, obtaining the feature weights Wi corresponding to each core feature, and then using the formula:

[0047] ,

[0048] The total risk score R of psychosomatic comorbidity is calculated, where Si is the standardized score of the i-th core feature, Wi is the feature weight corresponding to the i-th core feature, and l1 is the number of core features. The risk level of psychosomatic comorbidity is output according to the preset evaluation index. The risk level of psychosomatic comorbidity includes the first risk level, the second risk level, the third risk level and the fourth risk level, and the risk level is automatically iterated once every 48 hours as the dataset is updated.

[0049] In this embodiment, it should be specifically noted that the feature weights Wi corresponding to each core feature are obtained by combining expert opinions and the importance of algorithm features. Among them, the weights of HRV standardized value, cortisol level, physical disease severity, and inflammation index in the medical dimension feature are 0.12, 0.10, 0.08, and 0.05, respectively. Among the psychological dimension features, PHQ-9 score, GAD-7 score, SCL-90 score, skin conductivity, and voice emotion intensity are 0.02, respectively. Among the life dimension features, the weight of sleep interruption frequency is 0.1, and the weight of insufficient exercise duration is 0.15. The value range of R is [0, 100].

[0050] It needs to be further explained that, as Figure 3 As shown, the steps for determining the risk level of psychosomatic comorbidity are as follows:

[0051] A1: When R≤30, it is in the first risk level, indicating that the patient's psychosomatic comorbidity risk is mild, manifested as mild psychological symptoms, mild abnormalities in physical indicators, and no obvious stress events;

[0052] A2: When 30 < R ≤ 60, it is in the second risk level, which means that the patient's psychosomatic comorbidity risk is moderate, manifested as moderate psychological symptoms, moderate abnormality of physical indicators, and the presence of general stress events;

[0053] A3: When 60 < R ≤ 85, it is in the third risk level, which means that the patient's psychosomatic comorbidity risk is severe, manifested as severe psychological symptoms, severe abnormal physical indicators and high stress events.

[0054] A4: When R > 85, it is in the fourth risk level, indicating that the patient's psychosomatic comorbidity risk is extremely high, manifested as extremely severe psychological symptoms and acute onset of physical illness.

[0055] It should be further explained that the patient profile includes the patient's unique identifier, basic identity information, severity of physical illness, degree of psychological state, degree of somatization symptoms, number of sleep interruptions, and duration of insufficient exercise. For example, patient P2025001, male, 45 years old, diagnosed with coronary heart disease 2 years ago, has a psychological state of moderate anxiety, mild depression, and moderate somatization symptoms. In terms of daily life, he has had 4 sleep interruptions per night in the past week and exercises less than 30 minutes per day.

[0056] Personalized treatment plan generation module: Dynamically generates and adjusts personalized intervention plans based on intelligent assessment and diagnosis results, and transmits them to the multidisciplinary collaborative execution module;

[0057] Furthermore, the generation of personalized intervention plans requires the establishment of a mapping rule base for psychosomatic comorbidity risk levels and intervention directions based on intelligent assessment and diagnosis results. Intervention directions include medical intervention, psychological intervention, and lifestyle intervention. Patient cases with a similarity of >85% to the current patient's total psychosomatic comorbidity risk score and patient profile are screened using a collaborative filtering algorithm. An initial intervention plan is generated by combining the mapping rule base and similar patient cases. The initial intervention plan is then optimized through a multi-objective optimization model to obtain a personalized intervention plan.

[0058] In this embodiment, it should be specifically noted that the total risk score of psychosomatic comorbidity and the similarity of patient profile need to be calculated and weighted summed based on the total risk score of psychosomatic comorbidity and patient profile. The multi-objective optimization model takes maximizing intervention effect, maximizing patient compliance and minimizing safety risk as training objectives, optimizes parameters in combination with individual patient constraints, and is verified by clinical experts. An expert team composed of psychotherapists conducts the final review of the algorithm-generated plan to ensure that the personalized intervention plan conforms to clinical practice. The personalized intervention plan is updated in real time through the efficacy evaluation results and feedback information fed back by the subsequent full-cycle monitoring and feedback module to optimize the personalized intervention plan.

[0059] Multidisciplinary collaborative execution module: Construct an online and offline integrated multidisciplinary collaboration platform to collaboratively execute personalized intervention plans across multiple disciplines and transmit plan execution data to the full-cycle monitoring and feedback module;

[0060] Furthermore, a multidisciplinary collaboration platform refers to the integration of personnel from cardiology, psychology, and rehabilitation departments to collaboratively implement personalized intervention plans, track the entire process of personalized intervention plans, record implementation data, including training records and treatment logs, and automatically allocate offline and online resources based on personalized intervention plans.

[0061] In this embodiment, it is important to specify the roles of the multidisciplinary collaboration platform, which includes clinicians, psychotherapists, community nurses, patients, and their families. Clinicians are responsible for the treatment of physical illnesses, communicate weekly with psychotherapists about changes in the patient's mental state, determine whether fluctuations in physical symptoms are related to psychological factors, and adjust the medical intervention plan accordingly. Psychotherapists are responsible for the psychological intervention content and provide feedback to clinicians on the effectiveness of the intervention. Community nurses are responsible for daily monitoring and follow-up tasks, and when abnormal physiological indicators or deterioration in the patient's mental state are detected, they immediately synchronize with clinicians and psychotherapists to initiate emergency treatment. Patients actively complete home intervention tasks, provide real-time feedback on their physical and psychological feelings, and family members assist in supervising the patient's implementation of the plan, recording the patient's home performance, and assisting in seeking help in case of emergencies.

[0062] It should be further clarified that offline resources include, but are not limited to, rehabilitation equipment and psychological counseling rooms, while online resources include, but are not limited to, remote consultation with experts and digital training tools, to ensure the continuity of intervention. Multidisciplinary collaboration platforms should allow medical staff to adjust the treatment plan online.

[0063] Full-cycle monitoring and feedback module: Real-time monitoring of patient data throughout the entire treatment cycle, obtaining efficacy evaluation results, establishing feedback triggering mechanisms, setting differentiated monitoring frequencies based on psychosomatic comorbidity risk levels, and pushing efficacy evaluation results and feedback information to the personalized plan generation module and system iteration optimization module;

[0064] Furthermore, the data for the entire treatment cycle includes data from the treatment period, rehabilitation period, and follow-up period. Monitoring priorities and frequencies are set for each period. Data for the entire treatment cycle is collected based on the monitoring priorities and frequencies. The data for the entire treatment cycle includes physical health data, mental health data, and behavioral performance data. To obtain the efficacy evaluation results, an efficacy evaluation model needs to be constructed. The data for the entire treatment cycle is imported into the efficacy evaluation model to calculate the total efficacy evaluation score and obtain the efficacy evaluation results.

[0065] In this embodiment, it should be specifically noted that the monitoring focus of the treatment period data is on the immediate effect and safety of the intervention plan, with a high monitoring frequency: physiological indicators 2-3 times daily, psychological scales once a week, and adverse drug reaction records once a day. The monitoring focus of the recovery period data is on the consolidation of the intervention effect and the patient's self-management ability, with a medium monitoring frequency: physiological indicators once a day, psychological scales once every 2 weeks, and home intervention implementation statistics once a week. The monitoring focus of the follow-up period data is on the long-term maintenance of the intervention effect and the risk of relapse, with a low monitoring frequency: physiological indicators once every 2 weeks, psychological scales once a month, and relapse risk assessment once every 3 months. The monitoring frequency will be temporarily increased when abnormal situations occur.

[0066] It should be specifically noted that the physiological health data is medical data collected by the multi-source data acquisition module, the mental health data is psychological data collected by the multi-source data acquisition module, and the behavioral execution data includes psychological training completion rate, medication adherence, exercise task completion, dietary structure, and daily routine.

[0067] It should be further explained that the calculation of the total efficacy evaluation score needs to be based on the preset evaluation criteria in the efficacy evaluation model to determine the contribution score of each indicator in the data of the entire treatment cycle. Combined with clinical guidelines and expert opinions, the weight corresponding to each indicator is determined. The total efficacy evaluation score B is calculated by weighting the contribution score of each indicator and the corresponding weight. Where B≥85 indicates excellent efficacy, 70≤B<85 indicates good efficacy, 50≤B<70 indicates average efficacy, and B<50 indicates unsatisfactory efficacy. The efficacy evaluation result is obtained from this.

[0068] Furthermore, the feedback triggering mechanism needs to be based on the efficacy evaluation results, which include substandard, average, good, and excellent. When the efficacy is substandard, the triggering mechanism optimizes the feedback; when the efficacy is good, the triggering mechanism maintains the feedback; and when the efficacy is excellent, the triggering mechanism strengthens the feedback.

[0069] The differentiated monitoring frequency is 2 times per day for the fourth risk level, 1 time per day for the third risk level, 1 time per day for the second risk level, and 1 time per day for the first risk level.

[0070] In this embodiment, it should be specifically explained that the core objective of the program optimization feedback is to adjust the intervention program to improve the effect. For example, if the patient's total efficacy score is 48 points and the main reason is that the exercise completion rate is not up to standard, the program optimization feedback is triggered, and it is recommended to adjust the exercise plan and increase exercise monitoring. The core objective of the program maintenance feedback is to maintain the current program and consolidate the efficacy. For example, if the patient's total efficacy score is 78 points, the program maintenance feedback is triggered, and it is recommended to continue to implement the current program and appropriately reduce the monitoring frequency. The core objective of the program reinforcement feedback is to strengthen preventive measures and reduce the probability of relapse. For example, if the patient's total efficacy score is 88 points, the program reinforcement feedback is triggered, and it is recommended to increase the psychological training for preventing relapse once a month and follow up once every two weeks.

[0071] System Iterative Optimization Module: Based on multi-dimensional patient datasets, efficacy evaluation results, and feedback information, the model is continuously iterated and optimized, and the iterated model is synchronized to the preceding module.

[0072] Furthermore, the model includes a psychosomatic state assessment model and a therapeutic effect assessment model. The optimization of the psychosomatic state assessment model requires recalculating the weights of each core feature based on the therapeutic effect assessment results and feedback information using a gradient boosting tree. The optimization of the therapeutic effect assessment model requires the introduction of dynamic features.

[0073] In this embodiment, it should be specifically noted that the optimization of the psychosomatic state assessment model also requires the introduction of an attention mechanism, allowing the model to automatically focus on features that have a greater impact on the diagnostic results. The iterative effect target is to improve the accuracy of risk level determination to ≥95%, reduce the bias rate of psychosomatic correlation analysis to <5%, and improve the assessment accuracy of special populations to ≥93%. The dynamic features introduced into the efficacy assessment model are the rate of change in adherence over the past two weeks and the fluctuation range of physiological indicators, to improve the model's sensitivity to short-term changes.

[0074] like Figure 2 As shown, this embodiment provides a collaborative management method for a psychosomatic medicine hybrid psychological intervention model, including the following steps:

[0075] S1: Collect multi-dimensional patient data through the multi-source data acquisition module, including medical data, psychological data, and life scenario data, to construct a multi-dimensional patient dataset;

[0076] S2: Analyze multi-dimensional patient datasets, calculate the dynamic psychosomatic correlation of physiological and psychological indicators based on psychosomatic medicine-specific logic, and output the psychosomatic comorbidity risk level;

[0077] S3: The personalized intervention plan is dynamically generated and adjusted based on the intelligent assessment and diagnosis results through the personalized plan generation module;

[0078] S4: Build an online and offline integrated multidisciplinary collaboration platform to enable multidisciplinary collaborative execution of personalized intervention plans and obtain plan execution data;

[0079] S5: Real-time monitoring of patient data throughout the entire treatment cycle, obtaining efficacy evaluation results, establishing feedback triggering mechanisms and monitoring frequencies, and feeding back efficacy evaluation results and feedback information to the personalized plan generation module;

[0080] S6: The system iterative optimization module continuously optimizes the model based on multi-dimensional patient datasets, efficacy evaluation results, and feedback information.

[0081] Secondly: The accompanying drawings of the embodiments disclosed in this invention only involve the structures involved in the embodiments disclosed in this invention. Other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of this invention can be combined with each other.

[0082] In conclusion, the above description is only a preferred embodiment 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 within the protection scope of the present invention.

Claims

1. A synergistic management system for psychosomatic medicine mixed psychological intervention model, characterized in that, The system comprises a multi-source data acquisition module, an intelligent evaluation and diagnosis module, a personalized scheme generation module, a multidisciplinary collaborative execution module, a whole-cycle monitoring feedback module, and a system iterative optimization module. The multi-source data acquisition module acquires multi-dimensional data of a patient, including medical data, psychological data, and life scene data, constructs a multi-dimensional data set of the patient, and delivers the data set to the intelligent evaluation and diagnosis module. The intelligent evaluation and diagnosis module analyzes the multi-dimensional data set of the patient, calculates a dynamic psychosomatic correlation degree of physiological-psychological indicators based on a psychosomatic medicine exclusive logic, and outputs a psychosomatic comorbidity risk level in combination with a logic regression model incorporating a disease history weight. The personalized scheme generation module dynamically generates and adjusts a personalized intervention scheme based on the intelligent evaluation and diagnosis result, and delivers the scheme to the multidisciplinary collaborative execution module. The multidisciplinary collaborative execution module constructs an online and offline integrated multidisciplinary collaboration platform, and collaboratively executes the personalized intervention scheme, and delivers scheme execution data to the whole-cycle monitoring feedback module. The whole-cycle monitoring feedback module monitors real-time data of a patient in a whole-cycle diagnosis and treatment, obtains a therapeutic effect evaluation result, formulates a feedback triggering mechanism, sets a differential monitoring frequency according to a psychosomatic comorbidity risk level, and pushes the therapeutic effect evaluation result and feedback information to the personalized scheme generation module and the system iterative optimization module. The system iterative optimization module continuously iteratively optimizes a model based on a multi-dimensional data set of a patient, a therapeutic effect evaluation result, and feedback information, and synchronizes the iteratively optimized model to the previous modules.

2. The psychosomatic medicine mixed psychological intervention mode collaborative management system according to claim 1, characterized in that: The medical data is acquired by interfacing with a hospital clinical system, including patient basic information, laboratory examination data, and imaging examination data, the psychological data includes subjective psychological table data and objective psychological state data, and the life scene data includes exercise data, sleep data, and basic physiological data.

3. The psychosomatic medicine mixed psychological intervention mode collaborative management system according to claim 1, characterized in that: The calculation of the psychosomatic correlation degree requires obtaining core feature data in the multi-dimensional data set of the patient, including medical dimension features, psychological dimension features, and life dimension features, and calculating the correlation strength of the medical dimension features and the psychological dimension features by using a Pearson correlation coefficient to obtain a psychosomatic correlation matrix.

4. The psychosomatic medicine mixed psychological intervention mode synergistic management system according to claim 1, characterized in that: The output of the psychosomatic comorbidity risk level requires calculating a psychosomatic comorbidity risk total score, which requires importing core feature standardized scores Si in the multi-dimensional data set of the patient into a psychosomatic state evaluation model, obtaining feature weights Wi corresponding to each core feature, and calculating the psychosomatic comorbidity risk total score R by the formula: R = ∑(Si*Wi), wherein Si is the standardized score of the i-th core feature, Wi is the feature weight corresponding to the i-th core feature, and 11 is the number of core features, and outputting a psychosomatic comorbidity risk level according to a preset evaluation index, wherein the risk level comprises a first risk level, a second risk level, a third risk level, and a fourth risk level, and the risk level is automatically iterated every 48 hours as the data set is updated. ， ​ 5. The psychosomatic medicine mixed psychological intervention mode synergistic management system according to claim 1, characterized in that: The generation of the personalized intervention scheme needs to establish a mapping rule library of psychosomatic comorbidity risk level and intervention direction based on the intelligent evaluation diagnosis result, the intervention direction includes medical intervention, psychological intervention and life intervention, the patient cases with a psychosomatic comorbidity risk total score similar to the current patient and a patient portrait similarity of > 85% are screened through a collaborative filtering algorithm, an initial intervention scheme is generated combining the mapping rule library and the similar patient cases, and the personalized intervention scheme is obtained by optimizing the initial intervention scheme through a multi-objective optimization model.

6. The psychosomatic medicine mixed psychological intervention mode synergistic management system according to claim 1, characterized in that: The multidisciplinary collaboration platform refers to the multidisciplinary personnel integrating the departments of cardiology, psychology and rehabilitation, which collaboratively execute the personalized intervention scheme, track the personalized intervention scheme throughout the process, record the scheme execution data including training records and diagnosis and treatment logs, and automatically dispatch offline resources and online resources according to the personalized intervention scheme. 7.The psychosomatic medicine mixed psychological intervention mode synergistic management system according to claim 1, characterized in that: The diagnosis and treatment whole cycle data includes diagnosis and treatment period data, rehabilitation period data and follow-up period data, the monitoring focus and monitoring frequency of each period are set, the diagnosis and treatment whole cycle data is collected based on the monitoring focus and monitoring frequency, and the diagnosis and treatment whole cycle data includes physiological health data, psychological health data and behavior execution data; the efficacy evaluation result is obtained by constructing an efficacy evaluation model, importing the diagnosis and treatment whole cycle data into the efficacy evaluation model to calculate the total score of the efficacy evaluation model, and obtaining the efficacy evaluation result. 8.The psychosomatic medicine mixed psychological intervention mode synergistic management system according to claim 4, characterized in that: The feedback trigger mechanism needs to be developed based on the efficacy evaluation result, and the efficacy evaluation result includes substandard, general, good and excellent; when the efficacy effect is substandard, the scheme optimization feedback is triggered; when the efficacy effect is good, the scheme maintenance feedback is triggered; and when the efficacy effect is excellent, the scheme strengthening feedback is triggered. The differential monitoring frequency is 2 times per day for the fourth risk level, 1 time per day for the third risk level, 1 time every 3 days for the second risk level, and 1 time every 5 days for the first risk level. 9.The psychosomatic medicine mixed psychological intervention mode synergistic management system according to claim 1, characterized in that: The model includes a psychosomatic state evaluation model and an efficacy evaluation model; the optimization of the psychosomatic state evaluation model needs to be based on the efficacy evaluation result and feedback information, and the gradient boosting tree is used to recalculate the weight of each core feature; and the optimization of the efficacy evaluation model needs to introduce dynamic features.

10. A synergistic management method of psychosomatic medicine mixed psychological intervention model, characterized in that, The collaborative management system applied to the psychosomatic medicine mixed psychological intervention mode of any one of claims 1-9 comprises the following steps: S1: collecting patient multidimensional data through a multi-source data acquisition module, including medical data, psychological data and life scene data, and constructing a patient multidimensional data set; S2: analyzing the patient multidimensional data set, calculating the dynamic psychosomatic correlation degree of physiological-psychological indicators based on psychosomatic medicine exclusive logic, and outputting the psychosomatic comorbidity risk level; S3: dynamically generating and adjusting the personalized intervention scheme based on the intelligent evaluation diagnosis result through the personalized scheme generation module; S4: constructing an online and offline integrated multidisciplinary collaboration platform, and collaboratively executing the personalized intervention scheme to obtain scheme execution data; S5: real-time monitoring of patient diagnosis and treatment whole cycle data, obtaining efficacy evaluation results, developing feedback trigger mechanisms and monitoring frequencies, and feeding back the efficacy evaluation results and feedback information to the personalized scheme generation module; S6: continuously iteratively optimizing the model by the system iterative optimization module based on the multidimensional data set of the patient, the efficacy evaluation result and the feedback information.

Images