Multi-mode self-adaptive fusion-based medication guidance system and method for pulmonary hypertension patient

Abstract

The invention relates to the technical field of health management, and discloses a pulmonary hypertension patient medication guidance system and method based on multi-modal adaptive fusion, and the system comprises a construction module which is used for training a support vector machine regression model through the sample data of each medication mode in each body weight interval, and obtaining a plurality of infusion parameter prediction models; the processing module is used for obtaining a target infusion parameter prediction model according to the medication mode and the weight value of the patient, and inputting the current hemodynamic index into the target infusion parameter prediction model to generate initial infusion parameters of treprostinil of the patient in the medication mode; and the guidance module is used for generating a medication guidance scheme of the patient in the medication mode based on the current cardiac function parameters, the environmental parameters and the genotype correction initial infusion parameters. According to the method, personalized medication guidance is provided for the pulmonary hypertension patients, the risks of adverse reaction or poor medication effect and the like caused by individual differences of the patients are reduced, and then doctors are assisted to make more scientific and reasonable medication decisions for the patients.

Description

Technical Field

[0001] The present invention relates to the field of health management technology, and in particular to a medication guidance system and method for pulmonary hypertension patients based on multimodal adaptive fusion. Background Art

[0002] Pulmonary arterial hypertension (PAH) is a disease of the pulmonary arterioles characterized by vascular proliferation, remodeling, in situ thrombosis, progressive occlusion of the pulmonary vascular bed, and progressive increase in pulmonary vascular resistance. Its pathogenesis is complex and its prognosis is poor. Currently, PAH patients are often treated with treprostinil as the main drug. However, the current guidance on treprostinil use in patients with pulmonary hypertension has many drawbacks, mainly reflected in the following aspects: 1) Inadequate consideration of individual differences in medication use. Individual variability among patients with pulmonary hypertension is significant, and different patients have varying responses and tolerance to medications. However, current medication guidance systems often fail to adequately consider individual patient differences, potentially leading to poor drug efficacy and even serious adverse reactions.

[0003] 2) Insufficient application of multimodal data fusion. Traditional medication guidance systems often rely solely on single-modality data (such as clinical symptoms and laboratory test results), while ignoring the potential value of other modal data (such as imaging and physiological sensor data). This may not fully reflect the patient's condition and drug response, thus affecting the accuracy and effectiveness of medication guidance.

[0004] 3) Lack of adaptive adjustment mechanisms. As a patient's condition changes and treatment progresses, their medication needs will also change accordingly. However, current medication guidance systems often lack adaptive adjustment mechanisms, making it impossible to adjust medication regimens in a timely manner based on changes in the patient's condition and treatment effectiveness, which may lead to poor drug effectiveness or unnecessary adverse reactions.

[0005] Therefore, it is urgent to provide a technical solution to solve the above problems. Summary of the Invention

[0006] To solve the above technical problems, the present invention provides a medication guidance system and method for pulmonary hypertension patients based on multimodal adaptive fusion.

[0007] In a first aspect, the present invention provides a medication guidance system for pulmonary hypertension patients based on multimodal adaptive fusion. The technical solution of the system is as follows: A medication guidance system for pulmonary hypertension patients based on multimodal adaptive fusion includes: a construction module, a processing module, and a guidance module; the construction module is used to: construct a support vector machine regression model for predicting the initial infusion parameters of treprostinil required for pulmonary hypertension patients, and iteratively train the support vector machine regression model based on patient sample data corresponding to each preset medication mode in each preset weight range, to obtain an infusion parameter prediction model corresponding to each preset medication mode in each preset weight range; The processing module is configured to: obtain a target infusion parameter prediction model from all infusion parameter prediction models based on the current medication pattern and the weight range to which the current weight value of the target pulmonary hypertension patient belongs, input the current hemodynamic index characteristics of the target pulmonary hypertension patient into the target infusion parameter prediction model, and generate initial infusion parameters of treprostinil for the target pulmonary hypertension patient under the current medication pattern in combination with a first infusion constraint condition corresponding to the current medication pattern; The guidance module is used to: modify the initial infusion parameters based on the current cardiac function parameters, current environmental parameters and CYP2C19 genotype of the target pulmonary hypertension patient, and generate and output the current medication guidance plan for the target pulmonary hypertension patient under the current medication mode in combination with the second infusion constraint condition corresponding to the current medication mode; the current medication guidance plan includes at least: the current infusion rate of treprostinil.

[0008] The beneficial effects of the medication guidance system for pulmonary hypertension patients based on multimodal adaptive fusion of the present invention are as follows: the system of the present invention integrates multimodal data of pulmonary hypertension patients to adaptively adjust the medication plan, thereby providing personalized medication guidance for pulmonary hypertension patients, reducing the risks of adverse reactions or poor medication effects due to individual differences among pulmonary hypertension patients, and then assisting doctors to make more scientific and reasonable treprostinil medication decisions for pulmonary hypertension patients, and better manage the health of pulmonary hypertension patients.

[0009] Integrate multimodal data, implement personalized medication guidance, and adaptively adjust medication regimens to improve the accuracy and effectiveness of medication guidance for patients with pulmonary hypertension.

[0010] On the basis of the above scheme, the medication guidance system for pulmonary hypertension patients based on multimodal adaptive fusion of the present invention can also be improved as follows.

[0011] In an optional manner, the multiple preset medication modes include: subcutaneous infusion mode and intravenous infusion mode; the initial infusion parameter corresponding to the subcutaneous infusion mode is: initial infusion total amount; the initial infusion parameter corresponding to the intravenous infusion mode is: initial infusion rate.

[0012] In an optional manner, all infusion parameter prediction models include: multiple infusion total amount prediction models and multiple infusion rate prediction models; the construction module is specifically used to: Based on the patient sample data corresponding to the subcutaneous infusion pattern in any preset weight interval, the support vector machine regression model is iteratively trained to obtain a total infusion amount prediction model corresponding to the subcutaneous infusion pattern in any preset weight interval, until a total infusion amount prediction model corresponding to the subcutaneous infusion pattern in each preset weight interval is obtained; Based on the patient sample data corresponding to the intravenous infusion pattern in any preset weight range, the support vector machine regression model is iteratively trained to obtain an infusion rate prediction model corresponding to the intravenous infusion pattern in any preset weight range, until an infusion rate prediction model corresponding to the intravenous infusion pattern in each preset weight range is obtained.

[0013] In an optional manner, the current hemodynamic indicator characteristics include: a current systolic blood pressure value, a current heart rate value, and a current pulmonary artery pressure value; the first infusion constraint corresponding to the subcutaneous infusion mode is a first total infusion amount constraint, and the first infusion constraint corresponding to the intravenous infusion mode is a first infusion rate constraint; and the processing module is specifically configured to: constructing an input feature vector according to the current systolic blood pressure value, the current heart rate value, and the current pulmonary artery pressure value of the target pulmonary hypertension patient; When the current medication mode is the subcutaneous infusion mode, the input feature vector is input into the target infusion parameter prediction model, and combined with the first total infusion amount constraint condition, the initial total infusion amount of treprostinil for the target pulmonary hypertension patient in the subcutaneous infusion mode is generated; when the current medication mode is the intravenous infusion mode, the input feature vector is input into the target infusion parameter prediction model, and combined with the first infusion rate constraint condition, the initial infusion rate of treprostinil for the target pulmonary hypertension patient in the intravenous infusion mode is generated.

[0014] In an optional manner, the second infusion constraint corresponding to the subcutaneous infusion mode is a second infusion total amount constraint, and the second infusion constraint corresponding to the intravenous infusion mode is a second infusion rate constraint; the guidance module includes: a first guidance module and a second guidance module; The first guidance module is configured to generate a first correction coefficient based on the current cardiac function parameters of the target pulmonary hypertension patient and in combination with the current medication mode, generate a second correction coefficient based on the current environmental parameters of the target pulmonary hypertension patient, and generate a third correction coefficient based on the CYP2C19 genotype of the target pulmonary hypertension patient; The second guidance module is used for: when the current medication mode is the subcutaneous infusion mode, based on the initial total infusion volume of the target pulmonary hypertension patient in the subcutaneous infusion mode, the first correction coefficient, the second correction coefficient and the third correction coefficient, and combined with the second total infusion volume constraint, obtaining the current total infusion volume, current infusion rate and current infusion interval of the target pulmonary hypertension patient in the subcutaneous infusion mode; when the current medication mode is the intravenous infusion mode, based on the initial infusion rate of the target pulmonary hypertension patient in the intravenous infusion mode, the first correction coefficient, the second correction coefficient and the third correction coefficient, and combined with the second infusion rate constraint, obtaining the current infusion rate of the target pulmonary hypertension patient in the intravenous infusion mode.

[0015] In an optional manner, the current cardiac function parameters include: current ejection fraction, and the current environmental parameters include: current ambient temperature, current ambient humidity, and current circadian rhythm parameters; the first guidance module is specifically used to: when the current medication mode is the subcutaneous infusion mode, input the current ejection fraction of the target pulmonary hypertension patient into the first correction formula corresponding to the subcutaneous infusion mode to obtain the first correction coefficient; when the current medication mode is the intravenous infusion mode, input the current ejection fraction of the target pulmonary hypertension patient into the first correction formula corresponding to the intravenous infusion mode to obtain the first correction coefficient; wherein, the first correction formula corresponding to the subcutaneous infusion mode is: The first correction formula corresponding to the intravenous infusion mode is: is the first correction coefficient corresponding to the subcutaneous infusion mode, is the first correction coefficient corresponding to the intravenous infusion mode, EF is the current ejection fraction, EF baseline is the baseline value of ejection fraction; Construct a compensation function for characterizing environmental influences, and input the current ambient temperature, the current ambient humidity, and the current circadian rhythm parameter into the compensation function to obtain the second correction coefficient; wherein the compensation function is: F env =α×T+β×C; F env is the second correction coefficient, α is the first weight value, β is the second weight value, α+β=1, H is the current ambient humidity, Temp is the current ambient temperature, δ is the temperature coefficient, the unit is ℃ -1 ; C is the current circadian rhythm parameter, T is the current temperature and humidity parameters, t is the current time point corresponding to the 24-hour system, and θ is the phase angle; When the CYP2C19 genotype of the target pulmonary hypertension patient is GG, the first target value is determined as the third correction coefficient; when the CYP2C19 genotype of the target pulmonary hypertension patient is GA, the second target value is determined as the third correction coefficient; when the CYP2C19 genotype of the target pulmonary hypertension patient is AA, the third target value is determined as the third correction coefficient.

[0016] In an optional manner, the second total infusion constraint condition is: V SC ≤2.0ng / kg / min; the second infusion rate constraint is: 0.5ng / kg / min≤VIV≤4.0ng / kg / min; the second guidance module is specifically used to: When the current medication mode is the subcutaneous infusion mode, the initial total infusion volume of the target pulmonary hypertension patient in the subcutaneous infusion mode, the first correction coefficient, the second correction coefficient, and the third correction coefficient are input into the first target formula group, and combined with the second total infusion volume constraint condition, the current total infusion volume, the current infusion rate, and the current infusion interval of the target pulmonary hypertension patient in the subcutaneous infusion mode are obtained; wherein, the first target formula group is: D final is the current total infusion volume of the target pulmonary hypertension patient in the subcutaneous infusion mode, k is the third correction coefficient, is the total initial infusion volume of the target pulmonary hypertension patient in the subcutaneous infusion mode, v SC is the current infusion rate of the target pulmonary hypertension patient in the subcutaneous infusion mode, T infusion is the infusion time of the target pulmonary hypertension patient in the subcutaneous infusion mode, τ interval is the current infusion interval of the target pulmonary hypertension patient in the subcutaneous infusion mode; when the current medication mode is the intravenous infusion mode, the initial infusion rate, the first correction coefficient, the second correction coefficient, and the third correction coefficient of the target pulmonary hypertension patient in the intravenous infusion mode are input into the second target formula, and combined with the second infusion rate constraint condition, the current infusion rate of the target pulmonary hypertension patient in the intravenous infusion mode is obtained; wherein, the second target formula is: v IV is the current infusion rate of the target pulmonary hypertension patient in the intravenous infusion mode, The initial infusion rate for the target pulmonary hypertension patient in the intravenous infusion mode.

[0017] In an optional manner, the method further includes: a determination module; The determination module is used to obtain the subcutaneous tissue thickness of the target pulmonary hypertension patient, and input the subcutaneous tissue thickness, the current ejection fraction, and the current temperature and humidity parameters of the target pulmonary hypertension patient into a dynamic infusion duration formula to obtain the infusion duration of the target pulmonary hypertension patient in the subcutaneous infusion mode; wherein the dynamic infusion duration formula is: T infusion =(10+τ×Thick subQ )×(1+0.02×(EF-EF baseline )+0.01×T);Thick subQ is the thickness of the subcutaneous tissue, τ is the infusion time coefficient, and the unit is min / mm.

[0018] In a second aspect, the present invention provides a method for guiding medication for patients with pulmonary hypertension based on multimodal adaptive fusion. The technical solution of the method is as follows: A support vector machine regression model is constructed for predicting the initial infusion parameters of treprostinil required for patients with pulmonary hypertension, and based on the patient sample data corresponding to each preset medication mode in each preset weight range, the support vector machine regression model is iteratively trained to obtain an infusion parameter prediction model for each preset medication mode in each preset weight range; according to the current medication mode of the target pulmonary hypertension patient and the weight range to which the current weight value belongs, a target infusion parameter prediction model is obtained by matching from all infusion parameter prediction models, the current hemodynamic index characteristics of the target pulmonary hypertension patient are input into the target infusion parameter prediction model, and combined with the first infusion constraint condition corresponding to the current medication mode, the initial infusion parameters of treprostinil for the target pulmonary hypertension patient under the current medication mode are generated; Based on the current cardiac function parameters, current environmental parameters and CYP2C19 genotype of the target pulmonary hypertension patient, the initial infusion parameters are corrected, and combined with the second infusion constraint corresponding to the current medication mode, the current medication guidance plan for the target pulmonary hypertension patient under the current medication mode is generated and output; the current medication guidance plan includes at least: the current infusion rate of treprostinil.

[0019] The beneficial effects of the medication guidance method for pulmonary hypertension patients based on multimodal adaptive fusion of the present invention are as follows: the method of the present invention integrates multimodal data of pulmonary hypertension patients to adaptively adjust the medication plan, thereby providing personalized medication guidance for pulmonary hypertension patients, reducing the risks of adverse reactions or poor medication effects due to individual differences among pulmonary hypertension patients, and then assisting doctors to make more scientific and reasonable treprostinil medication decisions for pulmonary hypertension patients, and better manage the health of pulmonary hypertension patients.

[0020] In a third aspect, the technical solution of a medication guidance device for pulmonary hypertension patients based on multimodal adaptive fusion of the present invention is as follows: The device includes a processor coupled to a memory, wherein at least one computer program is stored in the memory, and the at least one computer program is loaded and executed by the processor so that the device for guiding medication for patients with pulmonary hypertension based on multimodal adaptive fusion implements the method for guiding medication for patients with pulmonary hypertension based on multimodal adaptive fusion as described in the present invention.

[0021] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are specifically listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The accompanying drawings are only used to illustrate the embodiments and are not to be considered as limiting the present invention. In addition, the same reference symbols are used to represent the same components throughout the drawings. In the drawings: Figure 1 This is a schematic structural diagram of an embodiment of a medication guidance system for pulmonary hypertension patients based on multimodal adaptive fusion according to the present invention; Figure 2 Schematic diagram of a flow chart of an embodiment of a method for guiding medication for pulmonary hypertension patients based on multimodal adaptive fusion according to the present invention; Figure 3 This is a structural schematic diagram of an embodiment of a medication guidance device for pulmonary hypertension patients based on multimodal adaptive fusion according to the present invention. DETAILED DESCRIPTION

[0023] The exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be limited to the embodiments set forth herein.

[0024] Figure 1 FIG. 1 shows a schematic structural diagram of an embodiment of a pulmonary hypertension patient medication guidance system 100 based on multimodal adaptive fusion provided by the present invention. Figure 1 As shown, the medication guidance system 100 for pulmonary hypertension patients based on multimodal adaptive fusion includes: a construction module 110, a processing module 120 and a guidance module 130; The construction module 110 is used to: construct a support vector machine regression model for predicting the initial infusion parameters of treprostinil required for patients with pulmonary hypertension, and based on the patient sample data corresponding to each preset medication mode in each preset weight range, iteratively train the support vector machine regression model to obtain an infusion parameter prediction model corresponding to each preset medication mode in each preset weight range.

[0025] The support vector machine regression model includes an objective function and a kernel function. The objective function minimizes the deviation between the predicted value and the true value by finding a hyperplane such that most sample points fall within the epsilon-insensitive tube centered on the hyperplane. The kernel function maps low-dimensional nonlinear problems to a high-dimensional, linearly separable space. Common kernel functions include radial basis function (RBF) and linear kernel functions.

[0026] The multiple preset medication modes in this embodiment include a subcutaneous infusion mode and an intravenous infusion mode. The subcutaneous infusion mode involves injecting the drug directly into the subcutaneous tissue via a needle. Leveraging the subcutaneous tissue's absorption capacity, the drug slowly and continuously enters the bloodstream, making it suitable for patients to self-treat at home. The intravenous infusion mode involves injecting the drug directly into the veins, rapidly distributing it to tissues and organs throughout the body through the bloodstream. It is suitable for treatments requiring rapid onset of action, such as emergency care and intensive care. Treprostinil is suitable for both subcutaneous and intravenous infusion modes. There are multiple preset weight intervals, for example, including: under 40 kg, 40-45 kg, 45-50 kg, 50-55 kg, 55-60 kg, 60-65 kg, 65-70 kg, 70-75 kg, 75-80 kg, 80-85 kg, 85-90 kg, and over 90 kg.

[0027] Each preset medication mode has corresponding patient sample data for multiple preset weight ranges. For example, in the subcutaneous infusion mode, the 45-50kg weight range has corresponding patient sample data. These patient sample data are all data for patients who are taking medication using the subcutaneous infusion mode and weigh 45-50kg. The patient sample data includes hemodynamic index characteristics, medication mode, infusion rate, total infusion volume, medication interval, infusion duration, and other data of patients with pulmonary hypertension who have been treated in the past during the infusion of treprostinil.

[0028] Among them, one preset medication mode corresponds to an infusion parameter prediction model in a preset weight range, and each preset medication mode corresponds to a different infusion parameter prediction model in each preset weight range, that is, the total number of infusion parameter prediction models is the product of the number of preset medication modes and the number of preset weight ranges.

[0029] Specifically, all infusion parameter prediction models include: multiple infusion total amount prediction models and multiple infusion rate prediction models. The construction module 110 is specifically used to: Based on the patient sample data corresponding to the subcutaneous infusion pattern in any preset weight range, the support vector machine regression model is iteratively trained to obtain a total infusion prediction model corresponding to the subcutaneous infusion pattern in any preset weight range, until a total infusion prediction model corresponding to the subcutaneous infusion pattern in each preset weight range is obtained.

[0030] Based on the patient sample data corresponding to the intravenous infusion pattern in any preset weight range, the support vector machine regression model is iteratively trained to obtain an infusion rate prediction model corresponding to the intravenous infusion pattern in any preset weight range, until an infusion rate prediction model corresponding to the intravenous infusion pattern in each preset weight range is obtained.

[0031] Among them, the infusion parameter prediction model corresponding to the subcutaneous infusion mode is the total infusion amount prediction model, and the infusion parameter prediction model corresponding to the intravenous infusion mode is the infusion rate prediction model. The initial infusion parameter corresponding to the subcutaneous infusion mode is: the initial total infusion amount; the initial infusion parameter corresponding to the intravenous infusion mode is: the initial infusion rate. The total infusion amount mentioned in this embodiment is the total infusion amount per unit body weight, and the infusion rate is the infusion rate per unit body weight. The subcutaneous infusion mode mainly adaptively adjusts the infusion time and interval according to the dose to reduce local irritation, so the subcutaneous infusion mode generates the total infusion amount, infusion interval and infusion duration. The infusion duration of the intravenous infusion mode is relatively fixed, which avoids frequent adjustment of the infusion equipment and complies with the clinical continuous infusion specifications, so the intravenous infusion mode only needs to dynamically adjust the infusion rate.

[0032] It should be noted that when the support vector machine regression model is trained using sample data from patients receiving a subcutaneous infusion pattern within any preset weight range, the support vector machine regression model uses a linear kernel function, the model's regularization parameter C∈[0.1, 1.0], and the training error MAE ≤ 6 ng / kg. When the support vector machine regression model is trained using sample data from patients receiving an intravenous infusion pattern within any preset weight range, the support vector machine regression model uses an RBF function, the model's regularization parameter C∈[1, 10], the kernel coefficient γ∈[0.1, 1.0], and the training error MAE ≤ 0.35 ng / kg / min.

[0033] The processing module 120 is used to: obtain a target infusion parameter prediction model from all infusion parameter prediction models based on the current medication pattern and the weight range to which the current weight value of the target pulmonary hypertension patient belongs, input the current hemodynamic index characteristics of the target pulmonary hypertension patient into the target infusion parameter prediction model, and combine the first infusion constraint condition corresponding to the current medication pattern to generate the initial infusion parameters of treprostinil for the target pulmonary hypertension patient under the current medication pattern.

[0034] Among them, the target pulmonary hypertension patient is the pulmonary hypertension patient who needs medication guidance recommendation in this embodiment. The current medication mode is: subcutaneous infusion mode or intravenous infusion mode, which can be determined according to the doctor's advice or the patient's choice. The current weight value is the weight value provided by the patient in real time or measured in real time. Specifically, after clarifying the current medication mode and current weight value of the target pulmonary hypertension patient, the doctor (patient) can click on the operation interface of the visual terminal to select the current medication mode and enter the current weight value to carry out the subsequent medication guidance operation process.

[0035] Assuming the target PH patient uses a subcutaneous infusion mode and weighs 58 kg, the total infusion volume prediction model corresponding to the subcutaneous infusion mode of 55-60 kg is determined as the target infusion parameter prediction model. Current hemodynamic indicator characteristics include: current systolic blood pressure value, current heart rate value, and current pulmonary artery pressure value.

[0036] Specifically, the first total infusion constraint is: the initial total infusion The first infusion rate constraint is: initial infusion rate The processing module 120 is specifically configured to: An input feature vector is constructed according to the current systolic blood pressure value, the current heart rate value, and the current pulmonary artery pressure value of the target pulmonary hypertension patient.

[0037] Among them, the input feature vector X = [BP, HR, PAP]; BP represents the current systolic blood pressure value, BP∈(90, 180) mmHg; HR represents the current heart rate value, HR∈(50, 120) bpm; PAP represents the current pulmonary artery pressure value, PAP∈(15, 180) mmHg.

[0038] When the current medication mode is the subcutaneous infusion mode, the input feature vector is input into the target infusion parameter prediction model, and combined with the first total infusion amount constraint condition, the initial total infusion amount of treprostinil for the target pulmonary hypertension patient in the subcutaneous infusion mode is generated; when the current medication mode is the intravenous infusion mode, the input feature vector is input into the target infusion parameter prediction model, and combined with the first infusion rate constraint condition, the initial infusion rate of treprostinil for the target pulmonary hypertension patient in the intravenous infusion mode is generated.

[0039] The guidance module 130 is used to: modify the initial infusion parameters based on the current cardiac function parameters, current environmental parameters and CYP2C19 genotype of the target pulmonary hypertension patient, and generate and output the current medication guidance plan for the target pulmonary hypertension patient under the current medication mode in combination with the second infusion constraint condition corresponding to the current medication mode; the current medication guidance plan includes at least: the current infusion rate of treprostinil.

[0040] Specifically, the second infusion constraint corresponding to the subcutaneous infusion mode is a second infusion total amount constraint, and the second infusion constraint corresponding to the intravenous infusion mode is a second infusion rate constraint; the guidance module 130 includes: a first guidance module and a second guidance module; The first guidance module is used to: generate a first correction coefficient based on the current cardiac function parameters of the target pulmonary hypertension patient and combined with the current medication pattern, generate a second correction coefficient based on the current environmental parameters of the target pulmonary hypertension patient, and generate a third correction coefficient based on the CYP2C19 genotype of the target pulmonary hypertension patient.

[0041] The current cardiac function parameters include the current ejection fraction EF, and the current environmental parameters include the current ambient temperature Temp, the current ambient humidity H, and the current circadian rhythm parameter C.

[0042] Specifically, the first guidance module is specifically used to: When the current medication mode is the subcutaneous infusion mode, the current ejection fraction of the target pulmonary hypertension patient is input into the first correction formula corresponding to the subcutaneous infusion mode to obtain the first correction coefficient; when the current medication mode is the intravenous infusion mode, the current ejection fraction of the target pulmonary hypertension patient is input into the first correction formula corresponding to the intravenous infusion mode to obtain the first correction coefficient.

[0043] Among them, the first correction formula corresponding to the subcutaneous infusion mode is: The first correction formula corresponding to the intravenous infusion mode is: is the first correction coefficient corresponding to the subcutaneous infusion mode, is the first correction coefficient corresponding to the intravenous infusion mode, EF is the current ejection fraction, EF baseline The default value of the ejection fraction is 50%.

[0044] It should be noted that EDV is the end-diastolic volume (in mL) of the target PH patient, and ESV is the end-systolic volume (in mL) of the target PH patient. Using the U-Net convolutional neural network, the cardiovascular ultrasound images of the target PH patient (the images corresponding to the end of diastole) were automatically segmented, the ventricular contours were extracted, and three-dimensional reconstruction was performed, followed by integration to obtain the end-diastolic volume. Similarly, the cardiovascular ultrasound images of the target PH patient (the images corresponding to the end of systole) were automatically segmented, the ventricular contours were extracted, and three-dimensional reconstruction was performed, followed by integration to obtain the end-systolic volume.

[0045] Construct a compensation function for characterizing environmental influences, and input the current ambient temperature, current ambient humidity, and current circadian rhythm parameters into the compensation function to obtain a second correction coefficient; wherein the compensation function is: F env =α×T+β×C; F env is the second correction coefficient, α is the first weight value, β is the second weight value, α+β=1, H is the current ambient humidity, Temp is the current ambient temperature, δ is the temperature coefficient (the default is 0.05), the unit is ℃ -1 C is the current circadian rhythm parameter, T is the current temperature and humidity parameter, and t is the current time corresponding to the 24-hour system. For example, if it is currently 2:00 PM, then t is 14. θ is the phase angle, and the phase angle θ∈(-π, π) is directly related to the personalized work and rest pattern of the target pulmonary hypertension patient. Through mathematical fitting and clinical data calibration, the circadian rhythm function accurately reflects the patient's biological clock state. It should be noted that when T>1.2, α=0.8 and β=0.2; otherwise, α and β can be adjusted dynamically.

[0046] When the CYP2C19 genotype of the target pulmonary hypertension patient is GG, the first target value is determined as the third correction coefficient; when the CYP2C19 genotype of the target pulmonary hypertension patient is GA, the second target value is determined as the third correction coefficient; when the CYP2C19 genotype of the target pulmonary hypertension patient is AA, the third target value is determined as the third correction coefficient.

[0047] Among them, the value range of the first target value is 1.1-1.3, and the default is 1.2; the value range of the second target value is 0.9-1.1, and the default is 1.0; the value range of the third target value is 0.7-0.9, and the default is 0.8.

[0048] It should be noted that the CYP2C19 genotype specifically refers to: the genotype of the CYP2C19 gene polymorphism site rs4244285. The CYP2C19 genotype can be specifically obtained from the CYP2C19 genotype test report of the target pulmonary hypertension patient. The test report can be detected and generated by gene chip detection technology, fluorescent PCR detection technology, etc., and there is no restriction here. The principle of selecting multimodal data in this embodiment is: ① The systolic blood pressure value can reflect the maximum pressure on the arterial wall during cardiac contraction, directly reflecting the left ventricular pumping function and arterial system resistance; patients with pulmonary hypertension are often accompanied by an increased right ventricular afterload, resulting in limited left ventricular filling, which in turn affects systemic systolic blood pressure. Abnormal systolic blood pressure (too high or too low) may indicate a state of compensatory or decompensated cardiac function. Treprostinil, as a pulmonary artery vasodilator, may cause a decrease in systemic blood pressure. Real-time monitoring of systolic blood pressure can prevent the risk of hypotension and dynamically adjust medication. ② Heart rate directly affects tissue perfusion and oxygen delivery. Treprostinil may indirectly improve cardiac function by reducing pulmonary artery pressure, and heart rate changes can serve as a real-time indicator of treatment responsiveness. ③ Pulmonary hypertension is defined as a resting mean PAP (pulmonary artery pressure) ≥25 mmHg, which is the gold standard for disease diagnosis and staging. Treprostinil reduces PAP by dilating the pulmonary arteries, and real-time PAP monitoring is the most direct basis for medication adjustments. ④ Increased ambient temperature causes vasodilation, accelerating drug absorption and metabolism. Treprostinil clearance increases by approximately 8% for every 5°C increase in temperature. ⑤ High humidity may affect skin permeability (during subcutaneous infusion) or respiratory evaporative heat dissipation (indirectly altering circulatory status). When humidity exceeds 70%, the local absorption rate of the drug decreases by 10%-15%. ⑥ Patients with pulmonary hypertension experience increased hypoxia at night, increasing drug requirements but decreasing metabolic capacity, necessitating a balance between efficacy and safety. ⑦ The FDA recommends genetic testing for drugs metabolized by CYP2C19.

[0049] The second guidance module is used for: when the current medication mode is the subcutaneous infusion mode, based on the initial total infusion volume of the target pulmonary hypertension patient in the subcutaneous infusion mode, the first correction coefficient, the second correction coefficient and the third correction coefficient, and combined with the second total infusion volume constraint, obtaining the current total infusion volume, current infusion rate and current infusion interval of the target pulmonary hypertension patient in the subcutaneous infusion mode; when the current medication mode is the intravenous infusion mode, based on the initial infusion rate of the target pulmonary hypertension patient in the intravenous infusion mode, the first correction coefficient, the second correction coefficient and the third correction coefficient, and combined with the second infusion rate constraint, obtaining the current infusion rate of the target pulmonary hypertension patient in the intravenous infusion mode.

[0050] Among them, the second total infusion constraint is: v SC ≤2.0ng / kg / min; the second infusion rate constraint is: 0.5ng / kg / min≤VIV≤4.0ng / kg / min; the second guidance module is specifically used to: When the current medication mode is the subcutaneous infusion mode, the initial total infusion volume of the target pulmonary hypertension patient in the subcutaneous infusion mode, the first correction coefficient, the second correction coefficient and the third correction coefficient are input into the first target formula group, and combined with the second total infusion volume constraint condition to obtain the current total infusion volume, the current infusion rate and the current infusion interval of the target pulmonary hypertension patient in the subcutaneous infusion mode.

[0051] Among them, the first target formula group is: D final is the current total infusion volume of the target pulmonary hypertension patient in subcutaneous infusion mode, k is the third correction coefficient, is the total initial infusion volume for target pulmonary hypertension patients in subcutaneous infusion mode, v SC is the current infusion rate of the target PH patient in subcutaneous infusion mode, T infusion is the infusion time of the target pulmonary hypertension patient in subcutaneous infusion mode, in minutes (min); τ interval The current infusion interval in subcutaneous infusion mode for the target PH patient, in hours (h).

[0052] When the current medication mode is the intravenous infusion mode, the initial infusion rate, the first correction coefficient, the second correction coefficient and the third correction coefficient of the target pulmonary hypertension patient in the intravenous infusion mode are input into the second target formula, and combined with the second infusion rate constraint condition to obtain the current infusion rate of the target pulmonary hypertension patient in the intravenous infusion mode.

[0053] Among them, the second objective formula is: V IV is the current infusion rate of the target PH patient in intravenous infusion mode, The initial infusion rate for patients with target PH in the intravenous infusion mode.

[0054] It should be noted that the current medication guidance for patients with pulmonary hypertension receiving subcutaneous infusion mode includes the current total infusion volume, current infusion rate, and current infusion interval of treprostinil. The current medication guidance for patients with pulmonary hypertension receiving intravenous infusion mode includes the current infusion rate of treprostinil.

[0055] In an optional manner, the method further includes: a determination module; The determination module is used to obtain the subcutaneous tissue thickness of the target pulmonary hypertension patient, and input the subcutaneous tissue thickness, the current ejection fraction, and the current temperature and humidity parameters of the target pulmonary hypertension patient into the dynamic infusion duration formula to obtain the infusion duration of the target pulmonary hypertension patient in the subcutaneous infusion mode. The dynamic infusion duration formula is: T infusion =(10+τ×Thich subQ )×(1+0.02×(EF-EF baseline )+0.01×T);Thick subQ is the thickness of the subcutaneous tissue, τ is the infusion time coefficient (the default value is 0.5), the unit is min / mm; T infusion ∈[10, 60]min,τ interval ∈[6, 24]h.

[0056] In this embodiment, a high-frequency ultrasonic imaging method is used by default to obtain the subcutaneous tissue thickness of the target pulmonary hypertension patient.

[0057] In an optional manner, the method further includes: a reminder module; The reminder module is used to send the current medication guidance plan of the target pulmonary hypertension patient to the target smart terminal for output reminder.

[0058] The target smart terminal may be, but is not limited to, a mobile phone, a computer, a tablet, or a smart wristband. The output reminder method may include, but is not limited to, graphic reminders and voice reminders.

[0059] To better illustrate the technical solution of this embodiment, the following examples are used for detailed description: 1) The target pulmonary hypertension patient uses the intravenous infusion mode, with a weight of 62 kg. The intravenous infusion mode corresponds to the infusion rate prediction model associated with 60-65 kg. After normalizing the current hemodynamic index characteristics of the target pulmonary hypertension patient (current systolic blood pressure value 130 mmHg, current heart rate value 90 bpm and current pulmonary artery pressure value 28 mmHg), the input feature vector X is constructed and input into the associated infusion rate prediction model to obtain the initial infusion rate The first infusion rate constraint is satisfied. Current ambient temperature Temp = 28°C, current ambient humidity H = 75%, The current time is 14:00, then If the CYP2C19 genotype is AA, then k is 0.8 and EF = 45%. Substituting the above parameters into the corresponding formula, we can get v IV =1.47ng / kg / min, which meets the second infusion rate constraint. The current medication guidance for the target pulmonary hypertension patient is: the current infusion rate v IV =1.47ng / kg / min.

[0060] The technical solution of this embodiment integrates multimodal data of pulmonary hypertension patients to adaptively adjust the medication plan, thereby providing personalized medication guidance for pulmonary hypertension patients, reducing the risks of adverse reactions or poor medication effects due to individual differences among pulmonary hypertension patients, and assisting doctors in making more scientific and reasonable treprostinil medication decisions for pulmonary hypertension patients, thereby better managing the health of pulmonary hypertension patients.

[0061] Figure 2 FIG2 shows a flow chart of an embodiment of a method for guiding medication for pulmonary hypertension patients based on multimodal adaptive fusion provided by the present invention. Figure 2 As shown, the following steps are included: S1. Constructing a support vector machine regression model for predicting the initial infusion parameters of treprostinil required for patients with pulmonary hypertension, and iteratively training the support vector machine regression model based on patient sample data corresponding to each preset medication mode in each preset weight range, to obtain an infusion parameter prediction model corresponding to each preset medication mode in each preset weight range; S2. Based on the current medication pattern and the weight range to which the current weight value of the target pulmonary hypertension patient belongs, a target infusion parameter prediction model is obtained by matching from all infusion parameter prediction models, the current hemodynamic index characteristics of the target pulmonary hypertension patient are input into the target infusion parameter prediction model, and combined with the first infusion constraint condition corresponding to the current medication pattern, an initial infusion parameter of treprostinil for the target pulmonary hypertension patient under the current medication pattern is generated; S3. Based on the current cardiac function parameters, current environmental parameters and CYP2C19 genotype of the target pulmonary hypertension patient, the initial infusion parameters are corrected, and combined with the second infusion constraint corresponding to the current medication mode, the current medication guidance plan for the target pulmonary hypertension patient under the current medication mode is generated and output; the current medication guidance plan includes at least: the current infusion rate of treprostinil.

[0062] In an optional manner, the multiple preset medication modes include: subcutaneous infusion mode and intravenous infusion mode; the initial infusion parameter corresponding to the subcutaneous infusion mode is: initial infusion total amount; the initial infusion parameter corresponding to the intravenous infusion mode is: initial infusion rate.

[0063] In one optional embodiment, all infusion parameter prediction models include: multiple infusion total amount prediction models and multiple infusion rate prediction models; based on the patient sample data corresponding to each preset medication mode in each preset weight range, the support vector machine regression model is iteratively trained to obtain the infusion parameter prediction model corresponding to each preset medication mode in each preset weight range, including: Based on the patient sample data corresponding to the subcutaneous infusion pattern in any preset weight interval, the support vector machine regression model is iteratively trained to obtain a total infusion amount prediction model corresponding to the subcutaneous infusion pattern in any preset weight interval, until a total infusion amount prediction model corresponding to the subcutaneous infusion pattern in each preset weight interval is obtained; Based on the patient sample data corresponding to the intravenous infusion pattern in any preset weight range, the support vector machine regression model is iteratively trained to obtain an infusion rate prediction model corresponding to the intravenous infusion pattern in any preset weight range, until an infusion rate prediction model corresponding to the intravenous infusion pattern in each preset weight range is obtained.

[0064] In an optional manner, the current hemodynamic indicator characteristics include: a current systolic blood pressure value, a current heart rate value, and a current pulmonary artery pressure value; the first infusion constraint corresponding to the subcutaneous infusion mode is a first total infusion constraint, and the first infusion constraint corresponding to the intravenous infusion mode is a first infusion rate constraint; the current hemodynamic indicator characteristics of the target pulmonary hypertension patient are input into the target infusion parameter prediction model, and combined with the first infusion constraint corresponding to the current medication mode, the step of generating the initial infusion parameters of treprostinil for the target pulmonary hypertension patient under the current medication mode includes: constructing an input feature vector according to the current systolic blood pressure value, the current heart rate value, and the current pulmonary artery pressure value of the target pulmonary hypertension patient; When the current medication mode is the subcutaneous infusion mode, the input feature vector is input into the target infusion parameter prediction model, and combined with the first total infusion amount constraint condition, the initial total infusion amount of treprostinil for the target pulmonary hypertension patient in the subcutaneous infusion mode is generated; when the current medication mode is the intravenous infusion mode, the input feature vector is input into the target infusion parameter prediction model, and combined with the first infusion rate constraint condition, the initial infusion rate of treprostinil for the target pulmonary hypertension patient in the intravenous infusion mode is generated.

[0065] In an optional manner, the second infusion constraint corresponding to the subcutaneous infusion mode is a second total infusion amount constraint, and the second infusion constraint corresponding to the intravenous infusion mode is a second infusion rate constraint; based on the current cardiac function parameters, current environmental parameters, and CYP2C19 genotype of the target pulmonary hypertension patient, the initial infusion parameters are modified, and combined with the second infusion constraint corresponding to the current medication mode, the step of generating a current medication guidance plan for the target pulmonary hypertension patient under the current medication mode includes: generating a first correction coefficient based on the current cardiac function parameters of the target pulmonary hypertension patient and in combination with the current medication pattern, generating a second correction coefficient based on the current environmental parameters of the target pulmonary hypertension patient, and generating a third correction coefficient based on the CYP2C19 genotype of the target pulmonary hypertension patient; When the current medication mode is the subcutaneous infusion mode, based on the initial total infusion volume of the target pulmonary hypertension patient in the subcutaneous infusion mode, the first correction coefficient, the second correction coefficient and the third correction coefficient, and combined with the second total infusion volume constraint, the current total infusion volume, current infusion rate and current infusion interval of the target pulmonary hypertension patient in the subcutaneous infusion mode are obtained; when the current medication mode is the intravenous infusion mode, based on the initial infusion rate of the target pulmonary hypertension patient in the intravenous infusion mode, the first correction coefficient, the second correction coefficient and the third correction coefficient, and combined with the second infusion rate constraint, the current infusion rate of the target pulmonary hypertension patient in the intravenous infusion mode is obtained.

[0066] In an optional manner, the current cardiac function parameters include: current ejection fraction, the current environmental parameters include: current ambient temperature, current ambient humidity, and current circadian rhythm parameters; the steps of generating a first correction coefficient based on the current cardiac function parameters of the target pulmonary hypertension patient and in combination with the current medication mode, generating a second correction coefficient based on the current environmental parameters of the target pulmonary hypertension patient, and generating a third correction coefficient based on the CYP2C19 genotype of the target pulmonary hypertension patient include: When the current medication mode is the subcutaneous infusion mode, the current ejection fraction of the target pulmonary hypertension patient is input into the first correction formula corresponding to the subcutaneous infusion mode to obtain the first correction coefficient; when the current medication mode is the intravenous infusion mode, the current ejection fraction of the target pulmonary hypertension patient is input into the first correction formula corresponding to the intravenous infusion mode to obtain the first correction coefficient; wherein, the first correction formula corresponding to the subcutaneous infusion mode is: The first correction formula corresponding to the intravenous infusion mode is: is the first correction coefficient corresponding to the subcutaneous infusion mode, is the first correction coefficient corresponding to the intravenous infusion mode, EF is the current ejection fraction, EF baseline is the baseline value of ejection fraction; Construct a compensation function for characterizing environmental influences, and input the current ambient temperature, the current ambient humidity, and the current circadian rhythm parameter into the compensation function to obtain the second correction coefficient; wherein the compensation function is: F env =α×T+β×C; F env is the second correction coefficient, α is the first weight value, β is the second weight value, α+β=1, H is the current ambient humidity, Temp is the current ambient temperature, δ is the temperature coefficient, the unit is ℃ -1 ; C is the current circadian rhythm parameter, T is the current temperature and humidity parameters, t is the current time point corresponding to the 24-hour system, and θ is the phase angle; When the CYP2C19 genotype of the target pulmonary hypertension patient is GG, the first target value is determined as the third correction coefficient; when the CYP2C19 genotype of the target pulmonary hypertension patient is GA, the second target value is determined as the third correction coefficient; when the CYP2C19 genotype of the target pulmonary hypertension patient is AA, the third target value is determined as the third correction coefficient.

[0067] In an optional manner, the second total infusion amount constraint condition is: SC ≤2.0ng / kg / min; the second infusion rate constraint is: 0.5ng / kg / min≤v IV≤4.0ng / kg / min; when the current medication mode is the subcutaneous infusion mode, based on the initial total infusion volume of the target pulmonary hypertension patient in the subcutaneous infusion mode, the first correction coefficient, the second correction coefficient and the third correction coefficient, and in combination with the second total infusion volume constraint, obtaining the current total infusion volume, current infusion rate and current infusion interval of the target pulmonary hypertension patient in the subcutaneous infusion mode; when the current medication mode is the intravenous infusion mode, based on the initial infusion rate of the target pulmonary hypertension patient in the intravenous infusion mode, the first correction coefficient, the second correction coefficient and the third correction coefficient, and in combination with the second infusion rate constraint, obtaining the current infusion rate of the target pulmonary hypertension patient in the intravenous infusion mode, including: When the current medication mode is the subcutaneous infusion mode, the initial total infusion volume of the target pulmonary hypertension patient in the subcutaneous infusion mode, the first correction coefficient, the second correction coefficient, and the third correction coefficient are input into the first target formula group, and combined with the second total infusion volume constraint condition, the current total infusion volume, the current infusion rate, and the current infusion interval of the target pulmonary hypertension patient in the subcutaneous infusion mode are obtained; wherein, the first target formula group is: D final is the current total infusion volume of the target pulmonary hypertension patient in the subcutaneous infusion mode, k is the third correction coefficient, is the total initial infusion volume of the target pulmonary hypertension patient in the subcutaneous infusion mode, v SC is the current infusion rate of the target pulmonary hypertension patient in the subcutaneous infusion mode, T infusion is the infusion time of the target pulmonary hypertension patient in the subcutaneous infusion mode, τ interval is the current infusion interval of the target pulmonary hypertension patient in the subcutaneous infusion mode; when the current medication mode is the intravenous infusion mode, the initial infusion rate, the first correction coefficient, the second correction coefficient, and the third correction coefficient of the target pulmonary hypertension patient in the intravenous infusion mode are input into the second target formula, and combined with the second infusion rate constraint condition, the current infusion rate of the target pulmonary hypertension patient in the intravenous infusion mode is obtained; wherein, the second target formula is: v IV is the current infusion rate of the target pulmonary hypertension patient in the intravenous infusion mode, The initial infusion rate for the target pulmonary hypertension patient in the intravenous infusion mode.

[0068] In an optional manner, the method further includes: Obtain the subcutaneous tissue thickness of the target pulmonary hypertension patient, and input the subcutaneous tissue thickness, the current ejection fraction, and the current temperature and humidity parameters of the target pulmonary hypertension patient into a dynamic infusion duration formula to obtain the infusion duration of the target pulmonary hypertension patient in the subcutaneous infusion mode; wherein the dynamic infusion duration formula is: T infusion =(10+τ×Thick subQ )×(1+0.02×(EF-EF baseline )+0.01×T);Thick subQ is the thickness of the subcutaneous tissue, τ is the infusion time coefficient, and the unit is min / mm.

[0069] In an optional manner, the method further includes: The current medication guidance plan of the target pulmonary hypertension patient is sent to the target smart terminal for output reminder.

[0070] It should be noted that the beneficial effects of the multimodal adaptive fusion-based medication guidance method for pulmonary hypertension patients provided in the above-mentioned embodiment are the same as the beneficial effects of the multimodal adaptive fusion-based medication guidance system 100 for pulmonary hypertension patients, and will not be repeated here. In addition, the method and system embodiments provided in the above-mentioned embodiment are based on the same concept, and their specific implementation process is detailed in the system embodiment and will not be repeated here.

[0071] Among them, the medication guidance system for pulmonary hypertension patients based on multimodal adaptive fusion of the present invention can be a computer program (including program code) running in a computer device. For example, the medication guidance system for pulmonary hypertension patients based on multimodal adaptive fusion of the present invention is an application software that can be used to execute the corresponding steps in the medication guidance method for pulmonary hypertension patients based on multimodal adaptive fusion of the present invention.

[0072] In some embodiments, the medication guidance system for pulmonary hypertension patients based on multimodal adaptive fusion of the present invention can be implemented by a combination of software and hardware. As an example, the medication guidance system for pulmonary hypertension patients based on multimodal adaptive fusion of the present invention can be a processor in the form of a hardware decoding processor, which is programmed to execute the medication guidance method for pulmonary hypertension patients based on multimodal adaptive fusion of the present invention. For example, the processor in the form of a hardware decoding processor can adopt one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs) or other electronic components.

[0073] The modules described in the embodiments of the present invention may be implemented in software or hardware, and the name of a module does not necessarily limit the module itself.

[0074] A medication guidance device for pulmonary hypertension patients based on multimodal adaptive fusion in an embodiment of the present invention includes a memory, a processor, and a computer program stored on the memory and runnable on the processor. When the processor executes the computer program, it implements any of the above-mentioned medication guidance methods for pulmonary hypertension patients based on multimodal adaptive fusion. That is, a medication guidance device for pulmonary hypertension patients based on multimodal adaptive fusion in an embodiment of the present invention may include but is not limited to: a processor and a memory; a memory for storing a computer program; a processor for executing the medication guidance method for pulmonary hypertension patients based on multimodal adaptive fusion shown in any embodiment of the present invention by calling the computer program.

[0075] In an optional embodiment, a device for guiding medication for pulmonary hypertension patients based on multimodal adaptive fusion is provided. Figure 3 As shown, Figure 3The shown medication guidance device 4000 for pulmonary hypertension patients based on multimodal adaptive fusion includes: a processor 4001 and a memory 4003. The processor 4001 and the memory 4003 are connected, such as through a bus 4002. Optionally, the medication guidance device 4000 for pulmonary hypertension patients based on multimodal adaptive fusion may also include a transceiver 4004, which may be used for data interaction between the medication guidance device for pulmonary hypertension patients based on multimodal adaptive fusion and other devices, such as data transmission and / or data reception. It should be noted that in actual applications, the transceiver 4004 is not limited to one, and the structure of the medication guidance device 4000 for pulmonary hypertension patients based on multimodal adaptive fusion does not constitute a limitation on the embodiments of the present invention.

[0076] Processor 4001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It may implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the present disclosure. Processor 4001 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like.

[0077] Bus 4002 may include a path for transmitting information between the above components. Bus 4002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. Bus 4002 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 In the figure, only one thick line is used to represent the bus 4002, but this does not mean that there is only one bus or one type of bus.

[0078] The memory 4003 may be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited to these.

[0079] The memory 4003 is used to store application code (computer program) for executing the solution of the present invention, and is controlled by the processor 4001. The processor 4001 is used to execute the application code stored in the memory 4003 to implement the content shown in the above method embodiment.

[0080] Among them, the pulmonary hypertension patient medication guidance device 4000 based on multimodal adaptive fusion can also be a terminal device. The terminal device can be any terminal device that can install applications and access web pages through applications, including at least one of smart phones, tablets, laptops, desktop computers, smart speakers, smart watches, smart TVs, and smart car-mounted devices.

[0081] It should be noted that Figure 3 The illustrated medication guidance device 4000 for pulmonary hypertension patients based on multimodal adaptive fusion is merely an example and should not limit the functions and scope of use of the embodiments of the present invention.

[0082] It should be understood that the flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the methods and computer program products according to various embodiments of the present invention. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the module, program segment, or a part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of the boxes in the block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or can be implemented using a combination of dedicated hardware and computer instructions.

[0083] The above description is merely a preferred embodiment of the present invention and an illustration of the technical principles employed. Those skilled in the art should understand that the scope of disclosure involved in the present invention is not limited to the technical solutions formed by the specific combination of the above-mentioned technical features, but also includes other technical solutions formed by any combination of the above-mentioned technical features or their equivalents without departing from the above-mentioned disclosed concepts. For example, a technical solution formed by replacing the above-mentioned features with (but not limited to) technical features with similar functions disclosed in the present invention.

[0084] It should be noted that the terms "first," "second," and the like in the specification and claims of this application are used to distinguish similar objects and to define a specific order or precedence. Where appropriate, the order used for similar objects may be interchanged, such that the embodiments of the present application described herein can be implemented in an order other than the order shown or described.

[0085] Those skilled in the art will appreciate that the present invention may be implemented as a system, method, or computer program product. Therefore, the present invention may be implemented in the following forms: entirely in hardware, entirely in software (including firmware, resident software, microcode, etc.), or in a combination of hardware and software, generally referred to herein as a "circuit," "module," or "system." Furthermore, in some embodiments, the present invention may be implemented in the form of a computer program product embodied in one or more computer-readable media containing computer-readable program code.

[0086] Although the embodiments of the present invention have been shown and described above, it will be understood that the above embodiments are illustrative and are not to be construed as limitations on the present invention. A person skilled in the art may change, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. A medication guidance system for pulmonary hypertension patients based on multimodal adaptive fusion, characterized by: include: Building modules, processing modules, and instructional modules; The construction module is used to: construct a support vector machine regression model for predicting the initial infusion parameters of treprostinil required for patients with pulmonary hypertension, and iteratively train the support vector machine regression model based on patient sample data corresponding to each preset medication mode in each preset weight range, to obtain an infusion parameter prediction model corresponding to each preset medication mode in each preset weight range; The processing module is configured to: obtain a target infusion parameter prediction model from all infusion parameter prediction models based on the current medication pattern and the weight range to which the current weight value of the target pulmonary hypertension patient belongs, input the current hemodynamic index characteristics of the target pulmonary hypertension patient into the target infusion parameter prediction model, and generate initial infusion parameters of treprostinil for the target pulmonary hypertension patient under the current medication pattern in combination with a first infusion constraint condition corresponding to the current medication pattern; The guidance module is configured to: modify the initial infusion parameters based on the current cardiac function parameters, current environmental parameters, and CYP2C19 genotype of the target pulmonary hypertension patient, and generate and output a current medication guidance plan for the target pulmonary hypertension patient under the current medication mode in combination with a second infusion constraint corresponding to the current medication mode; The current medication guidance regimen includes at least: the current infusion rate of treprostinil.

2. The medication guidance system for pulmonary hypertension patients based on multimodal adaptive fusion according to claim 1 is characterized in that: The multiple preset medication modes include: subcutaneous infusion mode and intravenous infusion mode; the initial infusion parameter corresponding to the subcutaneous infusion mode is: initial infusion total amount; the initial infusion parameter corresponding to the intravenous infusion mode is: initial infusion rate.

3. The medication guidance system for pulmonary hypertension patients based on multimodal adaptive fusion according to claim 2 is characterized in that: All infusion parameter prediction models include: multiple infusion total amount prediction models and multiple infusion rate prediction models; the construction module is specifically used to: Based on the patient sample data corresponding to the subcutaneous infusion pattern in any preset weight interval, the support vector machine regression model is iteratively trained to obtain a total infusion amount prediction model corresponding to the subcutaneous infusion pattern in any preset weight interval, until a total infusion amount prediction model corresponding to the subcutaneous infusion pattern in each preset weight interval is obtained; Based on the patient sample data corresponding to the intravenous infusion pattern in any preset weight range, the support vector machine regression model is iteratively trained to obtain an infusion rate prediction model corresponding to the intravenous infusion pattern in any preset weight range, until an infusion rate prediction model corresponding to the intravenous infusion pattern in each preset weight range is obtained.

4. The medication guidance system for pulmonary hypertension patients based on multimodal adaptive fusion according to claim 2 or 3, characterized in that: The current hemodynamic indicator characteristics include: a current systolic blood pressure value, a current heart rate value, and a current pulmonary artery pressure value; the first infusion constraint corresponding to the subcutaneous infusion mode is a first infusion total amount constraint, and the first infusion constraint corresponding to the intravenous infusion mode is a first infusion rate constraint; the processing module is specifically configured to: constructing an input feature vector according to the current systolic blood pressure value, the current heart rate value, and the current pulmonary artery pressure value of the target pulmonary hypertension patient; When the current medication mode is the subcutaneous infusion mode, the input feature vector is input into the target infusion parameter prediction model, and combined with the first total infusion amount constraint condition, the initial total infusion amount of treprostinil for the target pulmonary hypertension patient in the subcutaneous infusion mode is generated; when the current medication mode is the intravenous infusion mode, the input feature vector is input into the target infusion parameter prediction model, and combined with the first infusion rate constraint condition, the initial infusion rate of treprostinil for the target pulmonary hypertension patient in the intravenous infusion mode is generated.

5. The medication guidance system for pulmonary hypertension patients based on multimodal adaptive fusion according to claim 2 is characterized in that: The second infusion constraint corresponding to the subcutaneous infusion mode is a second infusion total amount constraint, and the second infusion constraint corresponding to the intravenous infusion mode is a second infusion rate constraint; the guidance module includes: a first guidance module and a second guidance module; The first guidance module is configured to generate a first correction coefficient based on the current cardiac function parameters of the target pulmonary hypertension patient and in combination with the current medication mode, generate a second correction coefficient based on the current environmental parameters of the target pulmonary hypertension patient, and generate a third correction coefficient based on the CYP2C19 genotype of the target pulmonary hypertension patient; The second guidance module is used for: when the current medication mode is the subcutaneous infusion mode, based on the initial total infusion volume of the target pulmonary hypertension patient in the subcutaneous infusion mode, the first correction coefficient, the second correction coefficient and the third correction coefficient, and combined with the second total infusion volume constraint, obtaining the current total infusion volume, current infusion rate and current infusion interval of the target pulmonary hypertension patient in the subcutaneous infusion mode; when the current medication mode is the intravenous infusion mode, based on the initial infusion rate of the target pulmonary hypertension patient in the intravenous infusion mode, the first correction coefficient, the second correction coefficient and the third correction coefficient, and combined with the second infusion rate constraint, obtaining the current infusion rate of the target pulmonary hypertension patient in the intravenous infusion mode.

6. The medication guidance system for pulmonary hypertension patients based on multimodal adaptive fusion according to claim 5 is characterized in that: The current cardiac function parameters include: current ejection fraction; the current environmental parameters include: current ambient temperature, current ambient humidity, and current circadian rhythm parameters; the first guidance module is specifically used to: When the current medication mode is the subcutaneous infusion mode, the current ejection fraction of the target pulmonary hypertension patient is input into the first correction formula corresponding to the subcutaneous infusion mode to obtain the first correction coefficient; when the current medication mode is the intravenous infusion mode, the current ejection fraction of the target pulmonary hypertension patient is input into the first correction formula corresponding to the intravenous infusion mode to obtain the first correction coefficient; wherein, the first correction formula corresponding to the subcutaneous infusion mode is: The first correction formula corresponding to the intravenous infusion mode is: is the first correction coefficient corresponding to the subcutaneous infusion mode, is the first correction coefficient corresponding to the intravenous infusion mode, EF is the current ejection fraction, EF baseline is the baseline value of ejection fraction; Construct a compensation function for characterizing environmental influences, and input the current ambient temperature, the current ambient humidity, and the current circadian rhythm parameter into the compensation function to obtain the second correction coefficient; wherein the compensation function is: F env =α×T+β×C; F env is the second correction coefficient, α is the first weight value, β is the second weight value, α+β=1, H is the current ambient humidity, Temp is the current ambient temperature, δ is the temperature coefficient, the unit is ℃ -1 ; C is the current circadian rhythm parameter, T is the current temperature and humidity parameters, t is the current time point corresponding to the 24-hour system, and θ is the phase angle; When the CYP2C19 genotype of the target pulmonary hypertension patient is GG, the first target value is determined as the third correction coefficient; when the CYP2C19 genotype of the target pulmonary hypertension patient is GA, the second target value is determined as the third correction coefficient; when the CYP2C19 genotype of the target pulmonary hypertension patient is AA, the third target value is determined as the third correction coefficient.

7. The medication guidance system for pulmonary hypertension patients based on multimodal adaptive fusion according to claim 6 is characterized in that: The second total infusion amount constraint condition is: vSC ≤ 2.0 ng / kg / min; the second infusion rate constraint condition is: 0.5 ng / kg / min ≤ vIV ≤ 4.0 ng / kg / min; the second guidance module is specifically used to: When the current medication mode is the subcutaneous infusion mode, the initial total infusion volume of the target pulmonary hypertension patient in the subcutaneous infusion mode, the first correction coefficient, the second correction coefficient, and the third correction coefficient are input into the first target formula group, and combined with the second total infusion volume constraint condition, the current total infusion volume, the current infusion rate, and the current infusion interval of the target pulmonary hypertension patient in the subcutaneous infusion mode are obtained; wherein, the first target formula group is: D final is the current total infusion volume of the target pulmonary hypertension patient in the subcutaneous infusion mode, k is the third correction coefficient, is the total initial infusion volume of the target pulmonary hypertension patient in the subcutaneous infusion mode, v SC is the current infusion rate of the target pulmonary hypertension patient in the subcutaneous infusion mode, T infusion is the infusion time of the target pulmonary hypertension patient in the subcutaneous infusion mode, τ interval the current infusion interval for the target pulmonary hypertension patient in the subcutaneous infusion mode; When the current medication mode is the intravenous infusion mode, the initial infusion rate of the target pulmonary hypertension patient in the intravenous infusion mode, the first correction coefficient, the second correction coefficient, and the third correction coefficient are input into a second target formula, and combined with the second infusion rate constraint condition to obtain the current infusion rate of the target pulmonary hypertension patient in the intravenous infusion mode; wherein, the second target formula is: v IV is the current infusion rate of the target pulmonary hypertension patient in the intravenous infusion mode, The initial infusion rate for the target pulmonary hypertension patient in the intravenous infusion mode.

8. The medication guidance system for pulmonary hypertension patients based on multimodal adaptive fusion according to claim 7 is characterized in that: Also includes: Identify the module; The determination module is used to obtain the subcutaneous tissue thickness of the target pulmonary hypertension patient, and input the subcutaneous tissue thickness, the current ejection fraction, and the current temperature and humidity parameters of the target pulmonary hypertension patient into a dynamic infusion duration formula to obtain the infusion duration of the target pulmonary hypertension patient in the subcutaneous infusion mode; wherein the dynamic infusion duration formula is: T infusion =(10+τ×Thick subQ )×(1+0.02×(EF-EF baseline )+0.01×T);Thick subQ is the thickness of the subcutaneous tissue, τ is the infusion time coefficient, and the unit is min / mm.

9. A medication guidance method for pulmonary hypertension patients based on multimodal adaptive fusion, characterized in that: include: Constructing a support vector machine regression model for predicting the initial infusion parameters of treprostinil required for patients with pulmonary hypertension, and iteratively training the support vector machine regression model based on patient sample data corresponding to each preset medication mode in each preset weight range to obtain an infusion parameter prediction model corresponding to each preset medication mode in each preset weight range; According to the current medication pattern of the target pulmonary hypertension patient and the weight range to which the current weight value belongs, a target infusion parameter prediction model is obtained by matching from all infusion parameter prediction models, the current hemodynamic index characteristics of the target pulmonary hypertension patient are input into the target infusion parameter prediction model, and combined with the first infusion constraint condition corresponding to the current medication pattern, the initial infusion parameters of treprostinil for the target pulmonary hypertension patient under the current medication pattern are generated; Based on the current cardiac function parameters, current environmental parameters, and CYP2C19 genotype of the target pulmonary hypertension patient, the initial infusion parameters are modified, and combined with the second infusion constraint corresponding to the current medication mode, a current medication guidance plan for the target pulmonary hypertension patient under the current medication mode is generated and output; The current medication guidance regimen includes at least: the current infusion rate of treprostinil.

10. A medication guidance device for pulmonary hypertension patients based on multimodal adaptive fusion, characterized in that: The device comprises a processor coupled to a memory, wherein at least one computer program is stored in the memory, and the at least one computer program is loaded and executed by the processor so that the medication guidance device for pulmonary hypertension patients based on multimodal adaptive fusion implements the medication guidance method for pulmonary hypertension patients based on multimodal adaptive fusion as described in claim 9.

Images