Medical risk assessment system and method based on multi-source data analysis

Abstract

The invention discloses a medical risk assessment system and method based on multi-source data analysis, and relates to the technical field of intelligent medical analysis. The system acquires physiological data and psychological data of a patient, and performs correlation analysis on the physiological data and the psychological data to obtain an influence coefficient YX of a psychological state of the patient on physiological health; a patient comprehensive medical risk score CMRI (t) is obtained, a medical risk prediction model is constructed by using a random forest algorithm, the patient comprehensive medical risk score at a future time point is predicted in combination with historical medical data, and an early warning mechanism is implemented in combination with a set medical risk change threshold value JKFY, so that the health condition of the patient can be prevented from further deteriorating. According to the method, the hospitalization rate and the occurrence frequency of first-aid events are reduced, so that the medical cost and resource waste are reduced, the treatment compliance of the patient is improved, the specific health requirement of the patient can be effectively met, and the overall health result of the patient is improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent medical analysis technology, and in particular to a medical risk assessment system and method based on multi-source data analysis. Background Art

[0002] In existing medical analysis, more and more studies have shown that mental health has an extremely important impact on overall physical health, especially in the management of chronic diseases, postoperative recovery, and high-risk patients. Changes in mental state often aggravate the condition or affect treatment outcomes. Therefore, medical risk assessment based on mental state has extremely high practical value in modern medical and health management.

[0003] At present, although medical technology continues to advance and the means of monitoring patients' physiological health are gradually improving, the monitoring of psychological state and the analysis of its correlation with physiological health are still insufficient. The specific problem is that the existing medical system mostly focuses on the collection and evaluation of physical health data, such as physiological indicators such as blood pressure, blood sugar, and heart rate, while ignoring the direct or indirect impact of psychological factors on physical health. In addition, although there are some standardized scales and testing methods for mental health problems in medicine, such as anxiety scales and depression scales, these assessment tools often lack dynamic correlation with real-time physiological data, and it is difficult to fully reflect the impact of changes in patients' psychological state on physical risks during actual treatment. This fragmented assessment method can easily lead to a lag in disease management and delay the timing of intervention for potential high-risk patients. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides a medical risk assessment system and method based on multi-source data analysis, which solves the problems in the above-mentioned background technology.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a medical risk assessment system based on multi-source data analysis, including a multi-source data acquisition module, a multi-source data analysis module, a comprehensive assessment module, a medical risk prediction module and an early warning module;

[0006] The multi-source data acquisition module is used to collect physiological data and psychological data of patients, and pre-process the collected physiological data and psychological data of patients to construct a multi-source data set of patients;

[0007] The multi-source data analysis module is used to analyze the patient's physiological data and psychological data in the patient's multi-source data set, obtain the mental health risk score XL and the physiological health risk score SL, and perform correlation analysis on the mental health risk score XL and the physiological health risk score SL to obtain the influence coefficient YX of the patient's psychological state on physiological health;

[0008] The comprehensive assessment module is used to obtain the patient's comprehensive medical risk score CMRI(t) based on the mental health risk score XL, the physical health risk score SL and the influence coefficient of psychology on physical health YX;

[0009] The medical risk prediction module is used to construct a patient medical risk prediction model, predict the patient's future medical risk changes, obtain the patient's comprehensive medical risk score CMRI change curve at a future time point, and obtain the patient's medical risk change rate ΔCMRI(t+j) based on the patient's comprehensive medical risk score CMRI change curve at a future time point;

[0010] The early warning module is used to preset the patient medical risk change threshold JKFY, compare and analyze the patient medical risk change rate ΔCMRI(t+j) and the patient medical risk change threshold JKFY, evaluate the patient's medical risk, and formulate a personalized health management plan.

[0011] Preferably, the multi-source data acquisition module includes a physiological data acquisition unit, a psychological data acquisition unit and a data preprocessing unit;

[0012] The physiological data acquisition unit is used to collect physiological data of K patients N times a day according to a standardized procedure, including collecting heart rate variability HRV, blood oxygen saturation SpO2 and hemoglobin Hb data of the patients;

[0013] The patient's heart rate variability HRV data is obtained by monitoring the electrical activity of the heart using an electrocardiogram sensor;

[0014] The blood oxygen saturation SpO2 data is obtained by using photoplethysmography;

[0015] The hemoglobin Hb data is obtained by analyzing the patient's blood using a fully automatic blood analyzer;

[0016] The psychological data collection unit is used to collect psychological data from patients using a psychological health assessment tool and a wearable emotion monitoring device, including collecting the patient's psychological stress index PSS, anxiety index HARS and depression index HDRS;

[0017] The patient's psychological stress index PSS is obtained by evaluating the standardized PSS-14 questionnaire;

[0018] The anxiety index HARS is obtained by using the Hamilton Anxiety Rating Scale;

[0019] The depression index HDRS is obtained through face-to-face interviews with mental health professionals;

[0020] The data preprocessing unit is used to preprocess the collected patient physiological data and psychological data, detect and remove abnormal values ​​and noise levels, and perform dimensionless processing on the patient's physiological data and psychological data to obtain standardized patient physiological data and psychological data and construct a patient data set.

[0021] Preferably, the multi-source data analysis module includes a physiological data analysis unit, a psychological data analysis unit and a correlation analysis unit;

[0022] The physiological data analysis unit is used to obtain the patient's physiological health risk score SL according to the patient's physiological data in the patient data set. The patient's physiological health risk score SL is obtained in the following manner:

[0023]

[0024] In the formula, S HRV，i is the heart rate variability index obtained for the patient at the i-th time, S SpO2，i is the blood oxygen saturation index obtained by the patient for the i-th time, S Hb，i is the hemoglobin index of the patient obtained for the i-th time, i = {1, 2, ..., N}, N is the number of data collections in one day, ω1, ω2 and ω3 are weight coefficients;

[0025] The heart rate variability index S obtained for the patient for the i-th time HRV，i , blood oxygen saturation index S SpO2，i and hemoglobin index S Hb，i The method of obtaining is:

[0026]

[0027] In the formula, X HRV，i is the heart rate variability data obtained for the i-th time after data standardization, HRV LL It is the lower limit of the medical standard range of heart rate variability, HRV UL is the upper limit of the medical standard range of heart rate variability, X SpO2，i is the blood oxygen saturation data obtained for the ith time after data standardization, SpO2 LL It is the lower limit of the medical standard range of blood oxygen saturation, SpO2 UL is the upper limit of the medical standard range of blood oxygen saturation, X Hb，i is the hemoglobin data obtained for the ith time after data standardization, Hb LL It is the lower limit of the medical standard range of hemoglobin, Hb UL It is the upper limit of the medical standard range of heart rate variability;

[0028] The psychological data analysis unit is used to obtain the patient's psychological health risk score XL according to the patient's psychological data in the patient data set. The patient's psychological health risk score XL is obtained in the following manner:

[0029]

[0030] In the formula, PSS is the patient's psychological stress index PSS, anxiety index HARS and depression index HDRS, and are all weight coefficients.

[0031] Preferably, the correlation analysis unit is used to perform correlation analysis on the patient's physical health risk score SL and the patient's mental health risk score XL to obtain the influence coefficient YX of the patient's mental state on physical health. The influence coefficient YX of the patient's mental state on physical health is obtained in the following manner:

[0032]

[0033] Where COV(XL, SL) is the covariance between the patient's mental health risk score XL and the patient's physical health risk score SL, σ XL，K is the standard deviation of the mental health risk score XL of K patients, σ SL，K is the standard deviation of the physiological health risk score SL of K patients.

[0034] Preferably, the comprehensive assessment module is used to obtain the patient's comprehensive medical risk score CMRI(t) based on the patient's physical health risk score SL, the patient's mental health risk score XL and the patient's psychological state's influence coefficient on physical health YX. The patient's comprehensive medical risk score CMRI(t) is obtained in the following manner:

[0035] CMRI(t)=α·SL+β·XL·(1+YX);

[0036] Where α and β are the weight coefficients of the patient's physiological health risk score SL and the patient's mental health risk score XL, respectively.

[0037] Preferably, the medical risk prediction module includes a model building unit and a risk prediction unit;

[0038] The model building unit is used to build a medical risk prediction model based on a random forest algorithm, collect historical medical data of patients, divide the historical medical data of patients into a test set and a validation set, obtain historical physiological health risk scores and mental health risk scores based on the test set data, input the historical physiological health risk scores and mental health risk scores as feature data into the medical risk prediction model to train the medical risk prediction model, output the patient's comprehensive medical risk score CMRI, use the training set data to evaluate the medical risk prediction model, calculate the predicted mean square error MSE, and optimize the model parameters;

[0039] The above patient historical medical data is obtained from the patient's electronic health record system.

[0040] Preferably, the risk prediction unit is used to use the patient medical data collected in real time, input the trained medical risk prediction model, and obtain the patient's comprehensive medical risk score CMRI(t+j) at a future time point;

[0041] A patient comprehensive medical risk score curve graph is drawn based on the patient comprehensive medical risk score CMRI(t+j) obtained at the future time point, where the x-axis represents time and the y-axis represents the CMRI value. Based on the patient comprehensive medical risk score curve graph and the current patient comprehensive medical risk score CMRI(t), the patient medical risk change rate ΔCMRI(t+j) is obtained. The patient medical risk change rate ΔCMRI(t+j) is obtained as follows:

[0042] ΔCMRI(t+j)=CMRI(t+j)-CMRI(t);

[0043] Where CMRI(t+j) is the patient's comprehensive medical risk score at a future time point, and t+j represents the predicted future time j.

[0044] Preferably, the early warning module includes a patient risk early warning unit and a personalized medical plan formulation unit;

[0045] The patient risk warning unit is used to preset the patient medical risk change threshold JKFY, compare and analyze the patient medical risk change rate ΔCMRI(t+j) and the patient medical risk change threshold JKFY, and evaluate the patient medical risk. The specific evaluation content is as follows:

[0046] If the patient's medical risk change rate ΔCMRI(t+j) is greater than or equal to the patient's medical risk change threshold JKFY, that is, ΔCMRI(t+j)≥JKFY, the early warning mechanism is triggered, and the patient's medical risk report is continuously generated to notify the medical staff until the medical staff responds;

[0047] If the patient's medical risk change rate ΔCMRI(t+j) is less than the patient's medical risk change threshold JKFY, that is, ΔCMRI(t+j)<JKFY, the patient's medical data is recorded, and a patient medical risk report is generated and stored in the patient's electronic health record system.

[0048] Preferably, the personalized medical plan formulation unit is used to, when the patient medical risk change rate ΔCMRI(t+j) is greater than or equal to the patient medical risk change threshold JKFY, trigger the early warning mechanism, notify the patient's attending physician and psychologist based on the generated patient medical risk report, formulate a personalized treatment plan, including adjusting the patient's drug treatment plan and formulating psychological intervention measures, and increase the patient's health monitoring frequency;

[0049] The above-mentioned psychological interventions include psychological counseling and relaxation training for patients;

[0050] Record the events and monitoring data that trigger the alarm mechanism in detail, establish a feedback mechanism, regularly review the handling effects of warning events, evaluate the effectiveness of mental health interventions, and adjust relevant treatment strategies.

[0051] A medical risk assessment method based on multi-source data analysis includes the following steps:

[0052] S1. Collecting physiological data and psychological data of patients, and preprocessing the collected physiological data and psychological data of patients to construct a multi-source data set of patients;

[0053] S2. Analyze the physiological data and psychological data of the patient in the multi-source data set of the patient to obtain the mental health risk score XL and the physiological health risk score SL, and perform correlation analysis on the mental health risk score XL and the physiological health risk score SL to obtain the influence coefficient YX of the patient's psychological state on physiological health;

[0054] S3. Obtain the patient's comprehensive medical risk score CMRI(t) based on the mental health risk score XL, the physical health risk score SL and the coefficient of psychological influence on physical health YX;

[0055] S4. Construct a patient medical risk prediction model to predict the patient's future medical risk changes, obtain the patient's comprehensive medical risk score CMRI change curve at a future time point, and obtain the patient's medical risk change rate ΔCMRI(t+j) based on the patient's comprehensive medical risk score CMRI change curve at a future time point;

[0056] S5. Preset the patient medical risk change threshold JKFY, compare and analyze the patient medical risk change rate ΔCMRI(t+j) and the patient medical risk change threshold JKFY, evaluate the patient's medical risk, and develop a personalized medical plan.

[0057] The present invention provides a medical risk assessment system and method based on multi-source data analysis, which has the following beneficial effects:

[0058] (1) By integrating multi-source data, including physical and mental health information, a comprehensive medical risk assessment system is established. This comprehensive data analysis can not only identify the patient's physical health risks, but also reveal the impact of psychological state on physical health. Through correlation analysis, doctors can clearly understand the relationship between mental health and physical health, thereby providing patients with a more comprehensive assessment basis. This comprehensive health management can help patients obtain more accurate diagnosis and treatment, improve their overall health level, and strengthen their active participation in health management.

[0059] (2) Through real-time data monitoring and the construction of dynamic prediction models, this solution can promptly identify changes in patients' health risks and set thresholds for medical risk changes. Once the patient's medical risk change rate exceeds the preset threshold, the system can automatically trigger the early warning mechanism and promptly notify medical staff to intervene. This timely response mechanism significantly improves the ability to prevent potential health risks, can identify and deal with possible health problems at an early stage, and reduce the risk of patients' health deterioration. At the same time, this early warning mechanism enhances patients' sense of security and dependence, and helps to establish a trusting relationship between doctors and patients.

[0060] (3) By assessing the patient's comprehensive medical risk score and psychological state, doctors can develop a personalized treatment plan, which not only includes adjusting the drug treatment plan, but also involves the implementation of psychological intervention measures, such as psychological counseling and relaxation training. This personalized treatment plan takes into account the patient's specific needs and psychological state, making the treatment more targeted and enhancing the patient's treatment compliance. At the same time, the regular feedback mechanism can monitor the patient's response in real time, adjust the treatment plan in a timely manner, ensure the effectiveness and adaptability of the treatment, thereby promoting the patient's comprehensive recovery and improving the quality of life. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 This is a block diagram of a medical risk assessment system based on multi-source data analysis of the present invention.

[0062] Figure 2 This is a schematic diagram of the steps of a medical risk assessment method based on multi-source data analysis of the present invention. DETAILED DESCRIPTION

[0063] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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.

[0064] Example 1

[0065] See also Figure 1 The present invention provides a medical risk assessment system based on multi-source data analysis, including a multi-source data acquisition module, a multi-source data analysis module, a comprehensive assessment module, a medical risk prediction module and an early warning module.

[0066] The multi-source data acquisition module is used to collect physiological data and psychological data of patients, and pre-process the collected physiological data and psychological data of patients to construct a multi-source data set of patients;

[0067] The multi-source data analysis module is used to analyze the patient's physiological data and psychological data in the patient's multi-source data set, obtain the mental health risk score XL and the physiological health risk score SL, and perform correlation analysis on the mental health risk score XL and the physiological health risk score SL to obtain the influence coefficient YX of the patient's psychological state on physiological health;

[0068] The comprehensive assessment module is used to obtain the patient's comprehensive medical risk score CMRI(t) based on the mental health risk score XL, the physical health risk score SL and the influence coefficient of psychology on physical health YX;

[0069] The medical risk prediction module is used to construct a patient medical risk prediction model, predict the patient's future medical risk changes, obtain the patient's comprehensive medical risk score CMRI change curve at a future time point, and obtain the patient's medical risk change rate ΔCMRI(t+j) based on the patient's comprehensive medical risk score CMRI change curve at a future time point;

[0070] The early warning module is used to preset the patient medical risk change threshold JKFY, compare and analyze the patient medical risk change rate ΔCMRI(t+j) and the patient medical risk change threshold JKFY, evaluate the patient's medical risk, and formulate a personalized health management plan.

[0071] During the operation of this system, the multi-source data acquisition module collects the patient's physiological data and psychological data to construct the patient's medical data set. The multi-source data analysis module performs correlation analysis on physiological data and psychology to obtain the influence coefficient YX of the patient's psychological state on physical health. The comprehensive evaluation module obtains the patient's comprehensive medical risk score, and combines it with the medical risk prediction model constructed by the medical risk prediction module to predict the patient's comprehensive medical risk score at future time points. Combined with the set medical risk change threshold JKFY, an early warning mechanism is implemented to prevent the patient's health from further deteriorating, reduce the hospitalization rate and the frequency of emergency events, thereby reducing medical costs and waste of resources.

[0072] Example 2

[0073] Please refer to Figure 1 ,Specifically: the multi-source data acquisition module includes a physiological data acquisition unit, a psychological data acquisition unit and a data preprocessing unit;

[0074] The physiological data acquisition unit is used to collect physiological data of K patients N times a day according to a standardized procedure, including collecting heart rate variability HRV, blood oxygen saturation SpO2 and hemoglobin Hb data of the patients;

[0075] The patient's heart rate variability HRV data is obtained by monitoring the electrical activity of the heart using an electrocardiogram sensor;

[0076] The blood oxygen saturation SpO2 data is obtained by using photoplethysmography;

[0077] The hemoglobin Hb data is obtained by analyzing the patient's blood using a fully automatic blood analyzer;

[0078] The psychological data collection unit is used to collect psychological data from patients using a psychological health assessment tool and a wearable emotion monitoring device, including collecting the patient's psychological stress index PSS, anxiety index HARS and depression index HDRS;

[0079] The patient's psychological stress index PSS is obtained by evaluating the standardized PSS-14 questionnaire;

[0080] The anxiety index HARS is obtained by using the Hamilton Anxiety Rating Scale. Anxiety can lead to increased heart rate and blood pressure. Long-term high anxiety levels are directly related to cardiovascular and gastrointestinal diseases.

[0081] The depression index HDRS is obtained through face-to-face interviews with mental health professionals. Depression is associated with decreased immune system function and endocrine disorders, and can also lead to the occurrence of chronic diseases and aggravation of the course of the disease.

[0082] The data preprocessing unit is used to preprocess the collected patient physiological data and psychological data, detect and remove abnormal values ​​and noise levels, and perform dimensionless processing on the patient's physiological data and psychological data to obtain standardized patient physiological data and psychological data and construct a patient data set.

[0083] In this embodiment, by setting a standardized procedure for collecting physiological data, the consistency and repeatability of data collection are ensured, the reliability of data is improved, and the accuracy of data analysis is enhanced. The unified N-time data collection for multiple patients is helpful to form a large-scale data set, which provides a solid foundation for subsequent analysis and model training. This systematic data collection method is helpful to form a more representative sample in clinical research and patient management, and then support scientific decision-making. The collection of physiological and psychological data is integrated, and the patient's health status can be evaluated in multiple dimensions. This multi-level evaluation can help medical personnel more accurately identify potential health problems, formulate more personalized intervention measures, and improve the overall treatment effect. In addition, the data preprocessing unit significantly improves the quality and reliability of the data by detecting and removing outliers and noise. The dimensionless processing enables data of different dimensions to be compared and analyzed under the same framework, which enhances the availability and consistency of the data. This process not only reduces the error introduced by data noise, but also ensures that the foundation for subsequent analysis and model building is more solid, thereby improving the prediction ability and decision support effect of the overall system.

[0084] Example 3

[0085] Please refer to Figure 1 ,Specifically: the multi-source data analysis module includes a physiological data analysis unit, a psychological data analysis unit and a correlation analysis unit;

[0086] The physiological data analysis unit is used to obtain the patient's physiological health risk score SL according to the patient's physiological data in the patient data set. The patient's physiological health risk score SL is obtained in the following manner:

[0087]

[0088] In the formula, S HRV，i is the heart rate variability index obtained for the patient at the i-th time, S SpO2，i is the blood oxygen saturation index obtained by the patient for the i-th time, S Hb，i is the hemoglobin index of the patient obtained for the i-th time, i = {1, 2, ..., N}, N is the number of data collections in one day, ω1, ω2 and ω3 are all weight coefficients, ω1+ω2+ω3=1;

[0089] The heart rate variability index S obtained for the patient for the i-th time HRV，i , blood oxygen saturation index SSpO2，i and hemoglobin index S Hb，i The method of obtaining is:

[0090]

[0091] In the formula, X HRV，i is the heart rate variability data obtained for the i-th time after data standardization, HRV LL It is the lower limit of the medical standard range of heart rate variability, HRV UL is the upper limit of the medical standard range of heart rate variability, X SpO2，i is the blood oxygen saturation data obtained for the ith time after data standardization, SpO2 LL It is the lower limit of the medical standard range of blood oxygen saturation, SpO2 UL is the upper limit of the medical standard range of blood oxygen saturation, X Hb，i is the hemoglobin data obtained for the ith time after data standardization, Hb LL It is the lower limit of the medical standard range of hemoglobin, Hb UL It is the upper limit of the medical standard range of heart rate variability;

[0092] The psychological data analysis unit is used to obtain the patient's psychological health risk score XL according to the patient's psychological data in the patient data set. The patient's psychological health risk score XL is obtained in the following manner:

[0093]

[0094] In the formula, PSS is the patient's psychological stress index PSS, anxiety index HARS and depression index HDRS, and are weight coefficients,

[0095] In this embodiment, the physiological data analysis unit provides a physiological health risk score SL by comprehensively analyzing the patient's heart rate variability, blood oxygen saturation and hemoglobin level. By standardizing each physiological indicator and combining it with a weight coefficient, the score can not only reflect the health status of a single physiological parameter, but also comprehensively evaluate the patient's physiological health level. This precise evaluation method enables doctors to better identify potential health problems and thus develop more effective treatment plans. The mental health risk score XL obtained by the psychological data analysis unit, combined with the analysis of the physiological health risk score SL, the system can reveal the relationship between the patient's psychological state and physiological health. Such analysis not only helps to identify the impact of psychological factors on physiological health, but also provides data support for the formulation of a comprehensive treatment strategy. This multi-dimensional evaluation method makes medical services more personalized and targeted, and strengthens the overall health management of patients. Based on the comprehensive evaluation of the physiological health score SL and the mental health score XL, the medical team can more effectively develop personalized treatment plans. By dynamically adjusting the weights and analysis of each indicator, doctors can optimize drugs and psychological intervention measures according to the patient's specific conditions, improve the effectiveness of treatment, enhance the patient's treatment compliance, and ultimately promote the patient's physical and mental health and quality of life.

[0096] Example 4

[0097] Please refer to Figure 1 Specifically: the correlation analysis unit is used to perform correlation analysis on the patient's physiological health risk score SL and the patient's mental health risk score XL to obtain the influence coefficient YX of the patient's mental state on the physiological health. The larger the influence coefficient YX is, the greater the influence of the patient's mental state on the physical health is. The influence coefficient YX of the patient's mental state on the physiological health is obtained as follows:

[0098]

[0099] Where COV(XL, SL) is the covariance between the patient's mental health risk score XL and the patient's physical health risk score SL, σ XL，K is the standard deviation of the mental health risk score XL of K patients, σ SL，K is the standard deviation of the physiological health risk score SL of K patients.

[0100] In this embodiment, by obtaining the influence coefficient YX of the patient's mental state on physical health through correlation analysis, the interaction between mental health and physical health can be more deeply understood. This analysis helps medical personnel identify which psychological factors may lead to physical health problems, so that in clinical practice, targeted treatment plans can be formulated more effectively. For example, if a certain anxiety level is highly correlated with a specific physical health problem, the doctor can give priority to psychological intervention during treatment to achieve a more comprehensive treatment effect. Using the influence coefficient YX, the medical team can optimize the patient's personalized treatment strategy. By incorporating mental health factors into physical health assessments, doctors can design more targeted intervention measures for patients, taking into account psychological and physiological factors. When formulating a rehabilitation plan, the medical team can increase psychological counseling and emotional support, thereby improving the patient's overall rehabilitation effect. By regularly monitoring changes in mental health risk scores and physical health risk scores, combined with the influence coefficient YX, potential health risks can be identified in a timely manner. This preventive management can not only intervene at an early stage to prevent the deterioration of health problems, but also encourage patients to actively participate in the maintenance of mental health, thereby improving their quality of life and overall health level. Overall, this analysis method provides a scientific basis for achieving precision medicine and holistic health management, and improves the medical risk management effect of patients.

[0101] Example 5

[0102] Please refer to Figure 1 Specifically: the comprehensive assessment module is used to obtain the patient's comprehensive medical risk score CMRI(t) based on the patient's physical health risk score SL, the patient's mental health risk score XL and the patient's psychological state's influence coefficient on physical health YX. The patient's comprehensive medical risk score CMRI(t) is obtained in the following manner:

[0103] CMRI(t)=α·SL+β·XL·(1+YX);

[0104] In the formula, α and β are the weight coefficients of the patient's physiological health risk score SL and the patient's mental health risk score XL, respectively, and α+β=1.

[0105] In this embodiment, by combining the patient's physiological health risk score SL, mental health risk score XL and the influence coefficient of psychological state on physiological health YX, a multi-dimensional medical risk assessment framework is provided. This integration can ensure that medical personnel have a comprehensive understanding of the patient's health status, not only focusing on physiological indicators, but also paying attention to the influence of psychological factors, thereby achieving more accurate risk assessment, helping to timely discover potential health risks and provide targeted intervention measures. By calculating the comprehensive medical risk score CMRI(t), the physiological and mental health risks are quantified into an easy-to-understand numerical indicator. This quantification method enables medical staff to more intuitively assess the patient's overall health risk, which is convenient for comparison and tracking. At the same time, the dynamic changes of CMRI can help doctors identify the trend of improvement or deterioration of health status, thereby supporting clinical decision-making and optimizing treatment plans. The acquisition of comprehensive medical risk scores provides data support for personalized medical decision-making. Medical personnel can adjust treatment strategies according to the numerical values ​​of CMRI, such as strengthening psychological intervention or adjusting physiological treatment plans to cope with specific health risks. This flexibility not only improves the effectiveness of treatment, but also improves patient satisfaction, enhances the enthusiasm of patients to participate in health management, and promotes the improvement of overall treatment effects.

[0106] Example 6

[0107] Please refer to Figure 1 ,Specifically: the medical risk prediction module includes a model building unit and a risk prediction unit;

[0108] The model building unit is used to build a medical risk prediction model based on a random forest algorithm, collect historical medical data of patients, divide the historical medical data of patients into a test set and a validation set, obtain historical physiological health risk scores and mental health risk scores based on the test set data, input the historical physiological health risk scores and mental health risk scores as feature data into the medical risk prediction model to train the medical risk prediction model, output the patient's comprehensive medical risk score CMRI, use the training set data to evaluate the medical risk prediction model, calculate the predicted mean square error MSE, and optimize the model parameters;

[0109] The above patient historical medical data is obtained from the patient's electronic health record system.

[0110] The risk prediction unit is used to use the patient medical data collected in real time, input the trained medical risk prediction model, and obtain the patient's comprehensive medical risk score CMRI(t+j) at a future time point;

[0111] A patient comprehensive medical risk score curve graph is drawn based on the patient comprehensive medical risk score CMRI(t+j) obtained at the future time point, where the x-axis represents time and the y-axis represents the CMRI value. Based on the patient comprehensive medical risk score curve graph and the current patient comprehensive medical risk score CMRI(t), the patient medical risk change rate ΔCMRI(t+j) is obtained. The patient medical risk change rate ΔCMRI(t+j) is obtained as follows:

[0112] ΔCMRI(t+j)=CMRI(t+j)-CMRI(t);

[0113] Where CMRI(t+j) is the patient's comprehensive medical risk score at a future time point, and t+j represents the predicted future time j.

[0114] In this embodiment, by constructing a medical risk prediction model based on the random forest algorithm, the patient's historical medical data can be effectively used for deep learning and pattern recognition. This machine learning method has the ability to handle complex nonlinear relationships, and can extract key features from a large amount of historical data to improve the accuracy of prediction. An accurate comprehensive medical risk score CMRI helps doctors better understand the patient's health status and provide a scientific basis for clinical decision-making, thereby achieving more accurate health management. By drawing a curve chart of the patient's comprehensive medical risk score, the changing trend of the patient's health risk can be intuitively displayed. By comparing the CMRI at a future time point with the current score, the changes in the patient's health status can be monitored in real time. Doctors and nursing staff can adjust treatment plans in a timely manner and optimize patients' health management strategies based on trend analysis. This dynamic assessment not only helps to identify potential health risks early, but also enhances patients' understanding and trust in the treatment process. Based on the obtained medical risk change rate ΔCMRI(t+j), medical staff can quickly determine whether the patient's health risk is within an acceptable range. Through this indicator, doctors can develop personalized intervention plans, including drug adjustments, increased monitoring frequency, and psychological intervention. This personalized medical intervention not only improves patients' treatment compliance, but also more effectively meets patients' specific health needs, thereby improving patients' overall health outcomes and quality of life.

[0115] Example 7

[0116] Please refer to Figure 1 ,Specifically: the early warning module includes a patient risk early warning unit and a personalized medical plan formulation unit;

[0117] The patient risk warning unit is used to preset the patient medical risk change threshold JKFY. The patient medical risk change threshold JKFY is set by experts based on historical experience and related field indexes. The patient medical risk change rate ΔCMRI(t+j) and the patient medical risk change threshold JKFY are compared and analyzed to evaluate the patient medical risk. The specific evaluation content is as follows:

[0118] If the patient's medical risk change rate ΔCMRI(t+j) is greater than or equal to the patient's medical risk change threshold JKFY, that is, ΔCMRI(t+j)≥JKFY, the early warning mechanism is triggered, and the patient's medical risk report is continuously generated to notify the medical staff until the medical staff responds. The early warning mechanism enables doctors and patients to take action before the health risk increases and avoid the occurrence of critical situations;

[0119] If the patient's medical risk change rate ΔCMRI(t+j) is less than the patient's medical risk change threshold JKFY, that is, ΔCMRI(t+j)<JKFY, the patient's medical data is recorded, and a patient medical risk report is generated and stored in the patient's electronic health record system.

[0120] The personalized medical plan formulation unit is used to notify the patient's attending physician and psychologist to formulate a personalized treatment plan based on the generated patient medical risk report when the patient's medical risk change rate ΔCMRI(t+j) is greater than or equal to the patient's medical risk change threshold JKFY, and trigger the early warning mechanism, including adjusting the patient's drug treatment plan and formulating psychological intervention measures, and increasing the patient's health monitoring frequency;

[0121] The above-mentioned psychological interventions include psychological counseling and relaxation training for patients;

[0122] Record the events and monitoring data that trigger the alarm mechanism in detail, establish a feedback mechanism, regularly review the handling effects of warning events, evaluate the effectiveness of mental health interventions, and adjust relevant treatment strategies.

[0123] In this embodiment, by setting the medical risk change threshold JKFY and implementing an early warning mechanism, the system can identify potential problems at the early stage of patient risk changes. When the medical risk change rate ΔCMRI(t+j) is greater than or equal to the medical risk change threshold JKFY, medical staff will be notified immediately and take intervention measures in time. This early intervention not only helps prevent the patient's health from further deteriorating, but also reduces the frequency of hospitalization rates and emergency events, thereby reducing medical costs and waste of resources. In addition, the formulation of personalized medical plans makes the treatment more targeted at the specific needs of patients. By analyzing the medical risk report, the attending physician and the psychologist can jointly formulate a comprehensive treatment plan. The plan combines drug adjustments and psychological interventions. This comprehensive strategy not only improves the effectiveness of treatment, but also enhances patient compliance, making patients more willing to participate in their own health management, thereby improving the overall treatment effect. At the same time, recording the events that trigger the warning and their related data will help to establish a systematic feedback mechanism. By regularly reviewing the handling effects of warning events, medical staff can evaluate the effectiveness of mental health interventions and adjust treatment strategies based on patient feedback. This continuous monitoring and feedback not only improves the quality of medical services, but also optimizes clinical practice through data-driven decision-making, ensuring that each patient can receive a relatively appropriate treatment plan, thereby enhancing the patient's long-term health and well-being.

[0124] Example 8

[0125] Please refer to Figure 1 and Figure 2 ,Specifically: A medical risk assessment method based on multi-source data analysis, including the following steps,

[0126] S1. Collecting physiological data and psychological data of patients, and preprocessing the collected physiological data and psychological data of patients to construct a multi-source data set of patients;

[0127] S2. Analyze the physiological data and psychological data of the patient in the multi-source data set of the patient to obtain the mental health risk score XL and the physiological health risk score SL, and perform correlation analysis on the mental health risk score XL and the physiological health risk score SL to obtain the influence coefficient YX of the patient's psychological state on physiological health;

[0128] S3. Obtain the patient's comprehensive medical risk score CMRI(t) based on the mental health risk score XL, the physical health risk score SL and the coefficient of psychological influence on physical health YX;

[0129] S4. Construct a patient medical risk prediction model to predict the patient's future medical risk changes, obtain the patient's comprehensive medical risk score CMRI change curve at a future time point, and obtain the patient's medical risk change rate ΔCMRI(t+j) based on the patient's comprehensive medical risk score CMRI change curve at a future time point;

[0130] S5. Preset the patient medical risk change threshold JKFY, compare and analyze the patient medical risk change rate ΔCMRI(t+j) and the patient medical risk change threshold JKFY, evaluate the patient's medical risk, and develop a personalized medical plan.

[0131] In this embodiment, by collecting the patient's physiological data and psychological data, performing correlation analysis, and combining historical medical data, the patient's comprehensive medical risk score at a future time point is predicted, and by setting the medical risk change threshold JKFY, an early warning mechanism is implemented, which can prevent the patient's health from further deteriorating, reduce the hospitalization rate and the frequency of emergency events, thereby reducing medical costs and waste of resources.

[0132] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A medical risk assessment system based on multi-source data analysis, characterized in that: It includes multi-source data collection module, multi-source data analysis module, comprehensive evaluation module, medical risk prediction module and early warning module; The multi-source data acquisition module is used to collect physiological data and psychological data of patients, and pre-process the collected physiological data and psychological data of patients to construct a multi-source data set of patients; The multi-source data analysis module is used to analyze the patient's physiological data and psychological data in the patient's multi-source data set, obtain the mental health risk score XL and the physiological health risk score SL, and perform correlation analysis on the mental health risk score XL and the physiological health risk score SL to obtain the influence coefficient YX of the patient's psychological state on physiological health; The comprehensive assessment module is used to obtain the patient's comprehensive medical risk score CMRI(t) based on the mental health risk score XL, the physical health risk score SL and the influence coefficient of psychology on physical health YX; The medical risk prediction module is used to construct a patient medical risk prediction model, predict the patient's future medical risk changes, obtain the patient's comprehensive medical risk score CMRI change curve at a future time point, and obtain the patient's medical risk change rate ΔCMRI(t+j) based on the patient's comprehensive medical risk score CMRI change curve at a future time point; The early warning module is used to preset the patient medical risk change threshold JKFY, compare and analyze the patient medical risk change rate ΔCMRI(t+j) and the patient medical risk change threshold JKFY, evaluate the patient's medical risk, and formulate a personalized health management plan.

2. A medical risk assessment system based on multi-source data analysis according to claim 1, characterized in that: The multi-source data acquisition module includes a physiological data acquisition unit, a psychological data acquisition unit and a data preprocessing unit; The physiological data acquisition unit is used to collect physiological data of K patients N times a day according to a standardized procedure, including collecting heart rate variability HRV, blood oxygen saturation SpO2 and hemoglobin Hb data of the patients; The patient's heart rate variability HRV data is obtained by monitoring the electrical activity of the heart using an electrocardiogram sensor; The blood oxygen saturation SpO2 data is obtained by using photoplethysmography; The hemoglobin Hb data is obtained by analyzing the patient's blood using a fully automatic blood analyzer; The psychological data collection unit is used to collect psychological data from patients using a psychological health assessment tool and a wearable emotion monitoring device, including collecting the patient's psychological stress index PSS, anxiety index HARS and depression index HDRS; The patient's psychological stress index PSS is obtained by evaluating the standardized PSS-14 questionnaire; The anxiety index HARS is obtained by using the Hamilton Anxiety Rating Scale; The depression index HDRS is obtained through face-to-face interviews with mental health professionals; The data preprocessing unit is used to preprocess the collected patient physiological data and psychological data, detect and remove abnormal values ​​and noise levels, and perform dimensionless processing on the patient's physiological data and psychological data to obtain standardized patient physiological data and psychological data and construct a patient data set.

3. A medical risk assessment system based on multi-source data analysis according to claim 2, characterized in that: The multi-source data analysis module includes a physiological data analysis unit, a psychological data analysis unit and a correlation analysis unit; The physiological data analysis unit is used to obtain the patient's physiological health risk score SL according to the patient's physiological data in the patient data set. The patient's physiological health risk score SL is obtained in the following manner: In the formula, S HRV，i is the heart rate variability index obtained for the patient at the i-th time, S SpO2，i is the blood oxygen saturation index obtained by the patient for the i-th time, S Hb，i is the hemoglobin index of the patient obtained for the i-th time, i = {1, 2, ..., N}, N is the number of data collections in one day, ω1, ω2 and ω3 are weight coefficients; The heart rate variability index S obtained for the patient for the i-th time HRV，i , blood oxygen saturation index S SpO2，i and hemoglobin index S Hb，i The method of obtaining is: Where, X HRV，i is the heart rate variability data obtained for the i-th time after data standardization, HRV LL It is the lower limit of the medical standard range of heart rate variability, HRV UL is the upper limit of the medical standard range of heart rate variability, X SpO2，i is the blood oxygen saturation data obtained for the ith time after data standardization, SpO2 LL It is the lower limit of the medical standard range of blood oxygen saturation, SpO2 UL is the upper limit of the medical standard range of blood oxygen saturation, X Hb，i is the hemoglobin data obtained for the ith time after data standardization, Hb LL It is the lower limit of the medical standard range of hemoglobin, Hb UL It is the upper limit of the medical standard range of heart rate variability; The psychological data analysis unit is used to obtain the patient's psychological health risk score XL according to the patient's psychological data in the patient data set. The patient's psychological health risk score XL is obtained in the following manner: In the formula, PSS is the patient's psychological stress index PSS, anxiety index HARS and depression index HDRS, and are all weight coefficients.

4. A medical risk assessment system based on multi-source data analysis according to claim 3, characterized in that: The correlation analysis unit is used to perform correlation analysis on the patient's physiological health risk score SL and the patient's mental health risk score XL to obtain the influence coefficient YX of the patient's mental state on the physiological health. The influence coefficient YX of the patient's mental state on the physiological health is obtained in the following manner: Where COV(XL, SL) is the covariance between the patient's mental health risk score XL and the patient's physical health risk score SL, σ XL，K is the standard deviation of the mental health risk score XL of K patients, σ SL，K is the standard deviation of the physiological health risk score SL of K patients.

5. The medical risk assessment system based on multi-source data analysis according to claim 1, characterized in that: The comprehensive assessment module is used to obtain the patient's comprehensive medical risk score CMRI(t) based on the patient's physiological health risk score SL, the patient's mental health risk score XL and the patient's psychological state's influence coefficient on physiological health YX. The patient's comprehensive medical risk score CMRI(t) is obtained in the following manner: CMRI(t)=α·SL+β·XL·(1+YX); Where α and β are the weight coefficients of the patient's physiological health risk score SL and the patient's mental health risk score XL, respectively.

6. The medical risk assessment system based on multi-source data analysis according to claim 1, characterized in that: The medical risk prediction module includes a model building unit and a risk prediction unit; The model building unit is used to build a medical risk prediction model based on a random forest algorithm, collect historical medical data of patients, divide the historical medical data of patients into a test set and a validation set, obtain historical physiological health risk scores and mental health risk scores based on the test set data, input the historical physiological health risk scores and mental health risk scores as feature data into the medical risk prediction model to train the medical risk prediction model, output the patient's comprehensive medical risk score CMRI, use the training set data to evaluate the medical risk prediction model, calculate the predicted mean square error MSE, and optimize the model parameters; The above patient historical medical data is obtained from the patient's electronic health record system.

7. A medical risk assessment system based on multi-source data analysis according to claim 6, characterized in that: The risk prediction unit is used to use the patient medical data collected in real time, input the trained medical risk prediction model, and obtain the patient's comprehensive medical risk score CMRI(t+j) at a future time point; A patient comprehensive medical risk score curve graph is drawn based on the patient comprehensive medical risk score CMRI(t+j) obtained at the future time point, where the x-axis represents time and the y-axis represents the CMRI value. Based on the patient comprehensive medical risk score curve graph and the current patient comprehensive medical risk score CMRI(t), the patient medical risk change rate ΔCMRI(t+j) is obtained. The patient medical risk change rate ΔCMRI(t+j) is obtained as follows: ΔCMRI(t+j)=CMRI(t+j)-CMRI(t); Where CMRI(t+j) is the patient's comprehensive medical risk score at a future time point, and t+j represents the predicted future time j.

8. The medical risk assessment system based on multi-source data analysis according to claim 1, characterized in that: The early warning module includes a patient risk early warning unit and a personalized medical plan formulation unit; The patient risk warning unit is used to preset the patient medical risk change threshold JKFY, compare and analyze the patient medical risk change rate ΔCMRI(t+j) and the patient medical risk change threshold JKFY, and evaluate the patient medical risk. The specific evaluation content is as follows: If the patient's medical risk change rate ΔCMRI(t+j) is greater than or equal to the patient's medical risk change threshold JKFY, that is, ΔCMRI(t+j)≥JKFY, the early warning mechanism is triggered, and the patient's medical risk report is continuously generated to notify the medical staff until the medical staff responds; If the patient's medical risk change rate ΔCMRI(t+j) is less than the patient's medical risk change threshold JKFY, that is, ΔCMRI(t+j)<JKFY, the patient's medical data is recorded, and a patient medical risk report is generated and stored in the patient's electronic health record system.

9. A medical risk assessment system based on multi-source data analysis according to claim 8, characterized in that: The personalized medical plan formulation unit is used to notify the patient's attending physician and psychologist to formulate a personalized treatment plan based on the generated patient medical risk report when the patient's medical risk change rate ΔCMRI(t+j) is greater than or equal to the patient's medical risk change threshold JKFY, and trigger the early warning mechanism, including adjusting the patient's drug treatment plan and formulating psychological intervention measures, and increasing the patient's health monitoring frequency; The above-mentioned psychological interventions include psychological counseling and relaxation training for patients; Record the events and monitoring data that trigger the alarm mechanism in detail, establish a feedback mechanism, regularly review the handling effects of warning events, evaluate the effectiveness of mental health interventions, and adjust relevant treatment strategies.

10. A medical risk assessment method based on multi-source data analysis, used to implement a medical risk assessment system based on multi-source data analysis as described in any one of claims 1 to 9, characterized in that: The following steps are included: S1. Collecting physiological data and psychological data of patients, and preprocessing the collected physiological data and psychological data of patients to construct a multi-source data set of patients; S2. Analyze the physiological data and psychological data of the patient in the multi-source data set of the patient to obtain the mental health risk score XL and the physiological health risk score SL, and perform correlation analysis on the mental health risk score XL and the physiological health risk score SL to obtain the influence coefficient YX of the patient's psychological state on physiological health; S3. Obtain the patient's comprehensive medical risk score CMRI(t) based on the mental health risk score XL, the physical health risk score SL and the coefficient of psychological influence on physical health YX; S4. Construct a patient medical risk prediction model to predict the patient's future medical risk changes, obtain the patient's comprehensive medical risk score CMRI change curve at a future time point, and obtain the patient's medical risk change rate ΔCMRI(t+j) based on the patient's comprehensive medical risk score CMRI change curve at a future time point; S5. Preset the patient medical risk change threshold JKFY, compare and analyze the patient medical risk change rate ΔCMRI(t+j) and the patient medical risk change threshold JKFY, evaluate the patient's medical risk, and develop a personalized medical plan.

Images