Cardiovascular disease risk assessment system based on big data analysis

Abstract

The invention discloses a cardiovascular disease risk assessment system based on big data analysis, which relates to the technical field of medical big data and quantum and comprises a data acquisition module, a preprocessing module, a feature extraction module, a model training module, a risk assessment module and a result display module. Data acquisition integrates multi-source heterogeneous data, quantum encryption is used to guarantee security, and preprocessing is carried out by deep reinforcement learning cleaning and adaptive normalization; feature extraction is combined with quantum principal component analysis and an auto-encoder, and a quantum attention mechanism is introduced; the model training adopts quantum neural network ensemble learning of quantum annealing optimization; introducing fuzzy logic reasoning to correct probability and layering in risk assessment; results are displayed through virtual reality and augmented reality technologies, and report suggestions are automatically generated. The method has the advantages of prominent advantages, comprehensive multi-modal data acquisition, accurate advanced technology preprocessing and feature extraction, efficient and accurate quantum optimization model training, and combination of real-time monitoring and a knowledge graph, provides a basis for prevention and treatment of cardiovascular diseases, and promotes medical intellectualization and precision.

Description

Technical Field

[0001] The present invention relates to the fields of medical big data and quantum technology, and particularly to a cardiovascular disease risk assessment system based on big data analysis. Background Art

[0002] Cardiovascular diseases, as highly prevalent diseases globally, seriously threaten human health and life safety. According to statistics from the World Health Organization, cardiovascular diseases cause a large number of deaths every year, and the incidence rate is on the rise year by year. Traditional methods for assessing the risk of cardiovascular diseases mainly rely on doctors' clinical experience and limited physiological index detections, and there are many limitations.

[0003] Traditional assessment methods often only focus on a few physiological indicators, such as blood pressure, blood lipids, blood sugar, etc., while ignoring the impacts of various aspects such as lifestyle habits, environmental factors, and genetic information on cardiovascular diseases. Smoking, lack of exercise, unreasonable diet, etc. in lifestyle habits, air pollution and noise interference in environmental factors, and individual genetic susceptibility are all closely related to the occurrence and development of cardiovascular diseases. However, it is difficult for traditional methods to comprehensively integrate these complex factors for accurate assessment.

[0004] In addition, the data sources of traditional assessment methods are limited, mainly relying on routine hospital examinations. This method not only has a small amount of data but also is not updated in a timely manner, and cannot reflect the changes in patients' health conditions in real time. In terms of data processing, traditional methods lack effective means to process large-scale, multi-dimensional, and heterogeneous data, and it is difficult to mine the information and laws hidden behind the data.

[0005] With the development of information technology, big data analysis technology has gradually been applied to the medical field. Big data can integrate data from multiple channels such as hospital information systems, wearable devices, and genetic testing institutions, providing richer and more comprehensive information for the assessment of cardiovascular disease risks. However, there are still some problems in the current cardiovascular disease risk assessment system based on big data.

[0006] In the data collection link, the fusion and secure transmission of multi-source heterogeneous data are a major challenge. The data formats, standards, and qualities of different data sources vary greatly. How to effectively integrate them and ensure the security and integrity of the data during the transmission process is an urgent problem to be solved. In the data preprocessing stage, existing cleaning and normalization methods may not be able to handle complex data noise and outliers, and there is a lack of a comprehensive assessment of data quality.

[0007] In terms of feature extraction, existing technologies have difficulty in mining deep correlations between data and cannot accurately screen out features that are highly correlated with cardiovascular disease risk. During model training, traditional machine learning models are inefficient in processing large-scale data and have limited generalization capabilities, making them difficult to adapt to different patient groups and complex disease conditions. The interpretability and credibility of risk assessment results also need to be improved, and doctors and patients often find it difficult to understand the basis behind the assessment results.

[0008] Therefore, it is of great practical significance to develop a cardiovascular disease risk assessment system and method based on big data analysis. The system needs to be able to integrate multi-source heterogeneous data, use advanced technologies for data processing, feature extraction and model training, provide accurate, reliable and highly interpretable risk assessment results, and provide strong support for the prevention and treatment of cardiovascular diseases. Summary of the invention

[0009] The present invention proposes a cardiovascular disease risk assessment system based on big data analysis to solve the problems mentioned in the above-mentioned prior art.

[0010] In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme: a cardiovascular disease risk assessment system based on big data analysis, comprising: Data collection module: Multimodal data fusion technology integrates hospital information, wearable devices, genetic testing institutions, media health comments and environmental data, and uses quantum encryption technology to ensure security. The improved data integrity assessment formula is: , is the importance weight of the i-th category data, is the amount of complete data of the i-th category, is the total amount of category i; Data preprocessing module: Use deep reinforcement learning algorithm to clean data, adopt adaptive normalization method, select normalization function according to data distribution characteristics, and the data quality evaluation formula based on information entropy is: , is the probability distribution of the i-th type of data, is the weight of the i-th category data; Feature extraction module: Combine quantum principal component analysis with deep autoencoders, use the parallelism of quantum computing to extract features, introduce quantum entanglement technology, explore feature associations, and use formulas based on quantum states. Assess relevance, is the quantum state of the jth characteristic, is the quantum state of disease risk, m is the number of features; Model training module: Adopt a quantum neural network ensemble learning model optimized by the quantum annealing algorithm. Determine the model parameter combination through this algorithm, combine transfer learning technology, initialize the current model with the model parameters of related diseases, and use cross-validation and quantum genetic algorithm to optimize the parameters. The model performance evaluation formula is , is the quantum state density matrix predicted by the model, is the real quantum state density matrix, and F is the quantum entanglement fidelity function; Risk assessment module: Input the features into the model to obtain the risk probability of the patient suffering from cardiovascular disease. The risk stratification formula is , is the membership degree of the i-th fuzzy rule, , are the weight coefficients, is the risk probability predicted by the model, is the subjective risk perception score of the patient, and k is the number of fuzzy rules; Result display module: Adopt virtual reality and augmented reality technologies to display the cardiovascular disease risk assessment results.

[0011] Furthermore, it also includes a real-time monitoring module. This module collects the patient's physiological data through implantable biosensors, adopts an anomaly detection algorithm based on quantum random walk to quickly discover abnormal changes in the physiological data. The anomaly detection accuracy evaluation formula based on the quantum bit error rate is , where is the number of quantum bit errors occurring during the detection process, is the total number of quantum bits; when an anomaly is detected, the system sends warning messages to the patient and the doctor through quantum communication technology.

[0012] Furthermore, it also includes a knowledge graph module. This module constructs a cardiovascular disease knowledge graph based on quantum graph neural networks, processes complex relationships and uncertainties of knowledge, integrates various medical knowledge, encodes and stores the knowledge using the quantum hashing algorithm, and uses the knowledge accuracy evaluation formula of the knowledge graph based on quantum entanglement entropy to evaluate the knowledge accuracy, where is the quantum entanglement entropy of the knowledge graph, is the maximum possible quantum entanglement entropy; Furthermore, the data acquisition module uses blockchain technology to ensure the security and traceability of data. Introduce smart contracts in the blockchain network to automatically verify the legality and integrity of data. Data records are encrypted and stored in the blockchain, and use the blockchain data security evaluation formula based on quantum signature to evaluate the security level, where is the amount of data verified by quantum signature, is the total data volume.

[0013] Furthermore, the data preprocessing module adopts a data augmentation method based on the quantum generative adversarial network QGAN, utilizes the superposition and entanglement characteristics of quantum states to generate synthetic data that is more similar to real data, evaluates the quality of the generated data through quantum measurement technology, and adopts a data augmentation effect evaluation formula based on quantum fidelity to evaluate the augmentation effect, where is the quantum state density matrix of the synthetic data, is the quantum state density matrix of the real data, and F is the quantum fidelity function).

[0014] Furthermore, the feature extraction module introduces a quantum attention mechanism to enable the model to focus on features related to cardiovascular disease risk, and assigns attention weights to different features through an attention weight assignment formula based on the quantum state transfer probability to assign attention weights to different features, where is the quantum state of the i-th feature, is the quantum state guided by attention, and m is the number of features.

[0015] Furthermore, the model training module adopts federated learning technology to share model parameters among medical institutions without revealing the original data of patients, introduces quantum key distribution technology to ensure the security of model parameter transmission, and the federated learning model performance consistency evaluation formula based on quantum entanglement swapping is where is the model quantum state density matrix of the k-th medical institution, is the average model quantum state density matrix of all medical institutions, F is the quantum entanglement swapping fidelity function, and s is the number of medical institutions, ensuring the consistency of model performance among different medical institutions from the quantum level.

[0016] Furthermore, the method of the cardiovascular disease risk assessment system based on big data analysis includes the following steps: Data collection step: Collect multi-source heterogeneous data of patients, including social media and environmental data, by means of multi-modal data fusion technology, transmit data using quantum encryption, adopt an improved data integrity evaluation formula to ensure data integrity, and verify data legality through blockchain smart contracts; Data preprocessing step: Clean data using a deep reinforcement learning algorithm, adopt adaptive normalization and a data augmentation method based on QGAN, and evaluate the processing effect using a data quality evaluation formula based on information entropy and a data augmentation effect evaluation formula based on quantum fidelity; Feature extraction step: Combine QPCA with a deep autoencoder, introduce a quantum attention mechanism and a feature correlation analysis technology based on quantum entanglement, and screen features using a feature correlation evaluation formula based on quantum states; Model training steps: adopt a quantum neural network ensemble learning model optimized based on quantum annealing algorithm, combine transfer learning and federated learning technology, use multi-objective optimization and quantum genetic algorithm to adjust parameters, and use the model performance evaluation formula based on quantum entanglement fidelity to evaluate performance; Risk assessment steps: input the features into the model to obtain the risk probability, introduce fuzzy logic reasoning to modify the probability, and use the risk stratification formula based on fuzzy comprehensive evaluation to perform risk stratification; Result presentation steps: Use VR and AR technologies to immersively display the assessment results, and use natural language generation technology to provide detailed reports and personalized prevention recommendations.

[0017] Furthermore, it also includes a real-time monitoring step, which collects physiological data in real time through implantable nanobiosensors, uses an anomaly detection algorithm based on quantum random walk to detect anomalies, uses an anomaly detection accuracy evaluation formula based on quantum bit error rate to evaluate the effect, and automatically issues warnings through quantum communication technology.

[0018] Furthermore, it also includes the steps of knowledge graph construction and application, constructing a knowledge graph based on quantum graph neural network, using quantum hash algorithm to store knowledge, using knowledge accuracy evaluation formula based on quantum entanglement entropy to ensure knowledge accuracy, and performing quantum reasoning and interpretation during the evaluation process.

[0019] Compared with the prior art, the present invention has the following beneficial effects: In terms of data collection, multimodal data fusion technology is used to integrate social media, environmental data and other multi-source information, comprehensively covering all factors that affect cardiovascular disease. Quantum encryption and blockchain smart contracts ensure data security and traceability, and the improved data integrity assessment formula ensures that the collected data is complete and accurate, laying a solid foundation for subsequent evaluation.

[0020] In the data preprocessing stage, the deep reinforcement learning algorithm dynamically cleans the data, and the adaptive normalization and quantum generative adversarial network enhance the data quality and generalization. The evaluation formula based on information entropy and quantum fidelity comprehensively measures the processing effect and improves data availability.

[0021] Feature extraction uses a combination of quantum principal component analysis and deep autoencoder, introduces quantum attention mechanism and feature correlation analysis, accurately screens highly correlated features based on the quantum state evaluation formula, and mines deep data correlations.

[0022] Model training uses quantum neural network ensemble learning optimized by quantum annealing, combined with transfer learning and federated learning, multi-objective optimization and quantum genetic algorithm parameter adjustment, and evaluates performance based on quantum entanglement fidelity to improve model accuracy, generalization ability and training efficiency.

[0023] The risk assessment introduces fuzzy logic reasoning to correct probabilities, and is stratified based on fuzzy comprehensive evaluation, resulting in a more scientific and reasonable outcome. The result display adopts virtual reality and augmented reality technologies to generate reports and suggestions in natural language, enhancing intuitiveness and readability.

[0024] The real-time monitoring module uses implantable nano-sensors and quantum random walk algorithms to detect anomalies in a timely manner, and quantum communication early warning ensures timeliness and security. The knowledge graph module constructs a quantum graph neural network graph, and quantum reasoning enhances the credibility of the evaluation.

[0025] In summary, this application can comprehensively, accurately, and real-timely evaluate the risk of cardiovascular diseases, provide a scientific decision-making basis for doctors, offer personalized prevention suggestions for patients, and promote the intelligent and precise development of cardiovascular disease prevention and treatment. Description of the Drawings

[0026] Figure 1 It is a schematic block diagram of a cardiovascular disease risk assessment system based on big data analysis proposed by the present invention; Figure 2 It is a schematic block diagram of a cardiovascular disease risk assessment method based on big data analysis proposed by the present invention. Detailed Embodiments

[0027] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0028] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by terms such as "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", etc. is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of the present invention.

[0029] In addition, the terms "first" and "second" are for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of the said features. In the description of the present invention, "a plurality of" means two or more, unless otherwise specifically defined. In addition, the terms "mounted", "connected" and "coupled" should be understood in a broad sense. For example, it may be a fixed connection, a detachable connection, or an integral connection; it may be a mechanical connection or an electrical connection; it may be directly connected or indirectly connected through an intermediate medium, and it may be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances. The present invention will be further described in detail below with reference to the accompanying drawings.

[0030] Referring to Figures 1 to 2 : A cardiovascular disease risk assessment system based on big data analysis, comprising a data acquisition module, a data preprocessing module, a feature extraction module, a model training module, a risk assessment module and a result display module.

[0031] Data acquisition module: By means of multi-modal data fusion technology, in addition to traditional hospital information systems, wearable devices, and data from genetic testing institutions, it also integrates health-related speech data of users on social media and environmental data (such as air quality, noise level) in the geographical information system. Quantum encryption technology is used to ensure the security of data during transmission and prevent data from being stolen or tampered with. An improved data integrity evaluation formula (where is the importance weight of the i-th type of data, is the number of complete data items of the i-th type, is the total number of data items of the i-th type) is used to ensure the integrity of the collected data, fully considering the importance of different types of data.

[0032] Data preprocessing module: The deep reinforcement learning algorithm is used for data cleaning, which can dynamically adjust the cleaning strategy according to the historical cleaning situation of the data and the current data characteristics, improving the cleaning efficiency and accuracy. In the normalization process, an adaptive normalization method is adopted to automatically select a suitable normalization function according to the distribution characteristics of the data. A data quality evaluation formula based on information entropy (where is the probability distribution of the i-th type of data, is the weight of the i-th type of data) is used to more comprehensively evaluate the quality of the preprocessed data, considering the uncertainty of the data.

[0033] Feature extraction module: Combine quantum principal component analysis (QPCA) with deep autoencoders to quickly extract key features using the parallelism of quantum computing. Introduce feature correlation analysis technology based on quantum entanglement to explore deeper correlations between features. Use feature correlation evaluation formula based on quantum states (in is the quantum state of the jth characteristic, is the quantum state of cardiovascular disease risk, and m is the number of extracted features) to screen out high-correlation features and measure the correlation between features and disease risk from a quantum level.

[0034] Model training module: A quantum neural network ensemble learning model optimized based on the quantum annealing algorithm is used to quickly find the optimal model parameter combination through the quantum annealing algorithm. Combined with transfer learning technology, the model parameters of other related diseases are used to initialize the current model to accelerate the convergence of the model. Multi-objective optimization cross-validation and quantum genetic algorithm are used to tune the model parameters, while considering the accuracy, generalization ability and training speed of the model. The model performance evaluation formula based on quantum entanglement fidelity is used (in is the quantum state density matrix predicted by the model, is the true quantum state density matrix, and F is the quantum entanglement fidelity function) to evaluate the model performance and measure the predictive ability of the model more accurately from a quantum perspective.

[0035] Risk assessment module: The extracted features are input into the trained model to output the patient's risk probability of cardiovascular disease. Fuzzy logic reasoning technology is introduced to correct the risk probability by combining the patient's subjective description and uncertainty information. A risk stratification formula based on fuzzy comprehensive evaluation is used (in is the membership degree of the i-th fuzzy rule, , is the weight coefficient, is the risk probability predicted by the model, is the patient's subjective risk perception score, and k is the number of fuzzy rules) to stratify the patient's risk and conduct risk assessment by considering multiple factors more scientifically.

[0036] Result display module: It adopts virtual reality (VR) and augmented reality (AR) technologies. These technologies can create an immersive experience environment for users, presenting the cardiovascular disease risk assessment results of patients in an intuitive and impactful way. For example, by creating a highly restored virtual cardiovascular system model, patients can clearly see the state of their own cardiovascular system and the specific situation of potential risks, making the originally abstract risks become visual. At the same time, this module also uses natural language generation technology. It can generate a detailed risk analysis report based on the assessment results. This report not only covers basic data such as risk probability, but also deeply analyzes various factors leading to the risks. And the system will also give highly targeted personalized prevention suggestions according to the individual conditions of patients, with easy-to-understand language for patients to understand and take corresponding measures.

[0037] In the present invention, there is also a real-time monitoring module. This module uses implantable biosensors to collect the physiological data of patients in real time, such as myocardial electrical activity, intravascular pressure, etc. These sensors are manufactured using nanotechnology and have the characteristics of high sensitivity and low power consumption. An anomaly detection algorithm based on quantum random walk is adopted to quickly and accurately detect abnormal changes in complex physiological data. An anomaly detection accuracy evaluation formula based on the quantum bit error rate is used (where is the number of qubit errors that occur during the detection process, is the total number of qubits) to evaluate the detection effect and improve the accuracy of detection from the quantum level. When an anomaly is detected, the system automatically sends warning messages to the patient and the doctor through quantum communication technology to ensure the security and timeliness of information transmission.

[0038] In the present invention, there is also a knowledge graph module. This module constructs a cardiovascular disease knowledge graph based on quantum graph neural networks, which can more effectively process the complex relationships and uncertainties between knowledge. Integrate knowledge such as medical literature, clinical guidelines, expert experience, and the latest medical research results, and use quantum hashing algorithms to encode and store the knowledge to improve the retrieval efficiency and security of the knowledge. A knowledge accuracy evaluation formula for the knowledge graph based on quantum entanglement entropy is used (where is the quantum entanglement entropy of the knowledge graph, is the maximum possible quantum entanglement entropy) to ensure the accuracy of the knowledge and measure the reliability of the knowledge from the quantum perspective. During the risk assessment process, the system can perform quantum reasoning and interpretation based on the knowledge provided by the knowledge graph to enhance the credibility of the assessment results.

[0039] In the present invention, the data acquisition module adopts blockchain technology, providing a solid guarantee for the security and traceability of data. Blockchain technology has characteristics such as decentralization and immutability, and can effectively resist the risks of data being maliciously tampered with and illegally obtained. By introducing the key element of smart contracts into the blockchain network, the collected data can be automatically verified for legality and integrity according to pre-set rules. Once problems are found in the data, an alarm will be issued in a timely manner. Each data record is stored on the blockchain after strict encryption processing. This encrypted storage method greatly enhances the confidentiality of data. Moreover, with the distributed ledger feature of the blockchain, users can query and verify the data at any time under authorization, ensuring that the source and flow process of the data are clear and transparent. In order to more accurately evaluate the security level of the data, the present invention uses a blockchain data security evaluation formula based on quantum signature (where is the amount of data verified by quantum signature, is the total amount of data). Through this formula, the proportion of data verified by quantum signature in the total amount of data can be intuitively reflected. Utilizing the non-forgeability of quantum signature further enhances the security of the data, laying a solid security defense line for subsequent data processing and analysis.

[0040] In the present invention, the data preprocessing module adopts a data enhancement method based on quantum generative adversarial network (QGAN), which becomes a key technical means to improve the quality and richness of data. In terms of principle, the superposition and entanglement characteristics of quantum states are the core basis of this method. The superposition of quantum states allows a quantum system to be in a combination of multiple states simultaneously, which means that in the data generation process, various possible situations can be simulated; while quantum entanglement establishes a special correlation between particles, and even if they are far apart, a change in the state of one particle will instantaneously affect the other particle. Based on these two amazing quantum characteristics, QGAN can generate synthetic data that highly approximates real data. These synthetic data not only expand the originally limited data set in terms of quantity, but also are extremely similar to real data in terms of features and distribution, providing richer materials for subsequent analysis and model training. To ensure the reliability of the generated data, quantum measurement technology is used to evaluate its quality. Quantum measurement can accurately detect the microscopic characteristics of the generated data, and through careful detection of various key parameters, ensure that the generated data has a high degree of credibility and meets the strict requirements of subsequent analysis and applications. For the evaluation of the data enhancement effect, the present invention uses a formula based on quantum fidelity (where is the quantum state density matrix of the synthetic data, is the quantum state density matrix of real data, and F is the quantum fidelity function), thus providing a scientific and quantitative evaluation basis for the effectiveness of data augmentation and ensuring the quality and reliability of the entire data preprocessing process.

[0041] In the present invention, the feature extraction module introduces a quantum attention mechanism, which provides strong support for the accurate analysis of the model. Traditional feature extraction methods often have difficulty accurately focusing on key features when dealing with complex cardiovascular disease-related data. The introduction of the quantum attention mechanism has changed this situation. This mechanism enables the model to more keenly capture features highly relevant to the risk of cardiovascular diseases. Through the attention weight assignment formula based on the quantum state transfer probability (where is the quantum state of the i-th feature, is the quantum state guided by attention, and m is the number of features), from a quantum perspective, this calculation method is more accurate. It takes into account the interaction and transfer probability between quantum states and can more precisely measure the importance of each feature for the risk assessment of cardiovascular diseases, thereby reasonably assigning attention weights to different features. In this way, the model can more efficiently screen and utilize key features, improve the accuracy and reliability of the risk assessment of cardiovascular diseases, and lay a solid foundation for subsequent risk assessment and analysis work.

[0042] In the present invention, the model training module adopts federated learning technology to share model parameters among multiple medical institutions without disclosing the original data of patients. The quantum key distribution technology is introduced to ensure the security of model parameters during transmission. The federated learning model performance consistency evaluation formula based on quantum entanglement swapping is used (where is the model quantum state density matrix of the k-th medical institution, is the average model quantum state density matrix of all medical institutions, F is the quantum entanglement swapping fidelity function, and s is the number of medical institutions) to evaluate the consistency of model performance and ensure the consistency of model performance among different medical institutions at the quantum level.

[0043] In the present invention, the cardiovascular disease risk assessment method based on big data analysis includes the following steps: Data collection steps: With the help of multimodal data fusion technology, multi-source heterogeneous data of patients are collected. These data come from a wide range of sources, including not only traditional medical data, but also health-related comments shared by patients on social media, and environmental data in geographic location information systems, such as air quality, noise level, etc. In the data transmission link, quantum encryption technology is used, which is based on the characteristics of quantum states and can ensure that data is not stolen or tampered with during transmission. An improved data integrity assessment formula is adopted to fully consider the importance of different types of data, and the weighted ratio of the number of complete items of each type of data to the total number is calculated to ensure that the collected data is complete. In addition, the legitimacy of the data is automatically verified through blockchain smart contracts. Smart contracts verify the data according to preset rules, and only data that meets the rules can be recognized and stored.

[0044] Data preprocessing steps: Use deep reinforcement learning algorithms to clean data. The algorithm can dynamically adjust the cleaning strategy according to the historical cleaning status and current data characteristics of the data, thereby improving the cleaning efficiency and accuracy. Adopting an adaptive normalization method, it will automatically select the most appropriate normalization function based on the distribution characteristics of the data, so that the data can be better suited for subsequent analysis. Based on the data enhancement method of QGAN (quantum generative adversarial network), the superposition and entanglement characteristics of quantum states are used to generate synthetic data that is more similar to real data and expand the data sample. The data quality evaluation formula based on information entropy is used to comprehensively consider the uncertainty of the data and evaluate the quality of the cleaned data; at the same time, the data enhancement effect evaluation formula based on quantum fidelity is used to measure the similarity between synthetic data and real data from the quantum level and evaluate the effect of data enhancement.

[0045] Feature extraction step: Combine QPCA (quantum principal component analysis) with deep autoencoders to take advantage of the parallelism of quantum computing to quickly extract key features, reduce data dimensions while retaining important information. The introduction of the quantum attention mechanism allows the model to pay more attention to features that are highly correlated with cardiovascular disease risk, and assigns attention weights to different features through a formula based on quantum state transfer probability to achieve precise focus. Feature correlation analysis technology based on quantum entanglement can explore deeper correlations between features and discover potential combinations of risk factors. Finally, a feature correlation evaluation formula based on quantum states is used to screen out features that are highly correlated with cardiovascular disease risk, providing high-quality input for subsequent model training.

[0046] Model training steps: Use a quantum neural network ensemble learning model optimized by the quantum annealing algorithm, which can utilize the characteristics of the quantum system to quickly find the optimal combination of model parameters and significantly improve the training efficiency. Combine transfer learning techniques to initialize the current model with the model parameters of other related diseases, reducing the training time and accelerating the convergence speed of the model. At the same time, introduce federated learning techniques to achieve joint training of multi-party data while protecting data privacy, expanding the data scale, and enhancing the generalization ability of the model. Apply multi-objective optimization methods to comprehensively consider multiple objectives such as the accuracy, generalization ability, and training speed of the model. At the same time, use the quantum genetic algorithm to optimize the model parameters, and search for a better parameter space by simulating the biological evolution process. Use a model performance evaluation formula based on quantum entanglement fidelity to measure the similarity between the quantum state predicted by the model and the true quantum state from a quantum perspective, and more accurately evaluate the performance of the model.

[0047] Risk assessment steps: Input the key features obtained from the feature extraction step into the trained model, and the model outputs the risk probability of the patient suffering from cardiovascular disease. To make the risk assessment more in line with the actual situation, introduce fuzzy logic reasoning technology to correct the risk probability by combining the patient's subjective description and uncertain information, making up for the limitations of the model based solely on data calculation. Adopt a risk stratification formula based on fuzzy comprehensive evaluation, comprehensively consider multiple factors such as the risk probability predicted by the model and the patient's subjective risk perception score, and stratify the patient according to the membership degree and weight coefficient of different fuzzy rules, more scientifically and comprehensively evaluating the cardiovascular disease risk level of the patient.

[0048] Result display steps: Use virtual reality (VR) and augmented reality (AR) technologies to display the cardiovascular disease risk assessment results of the patient in an immersive way. For example, create a virtual cardiovascular system model so that the patient can intuitively see the health status and potential risks of their own cardiovascular system. At the same time, use natural language generation technology to generate a detailed risk analysis report based on the risk assessment results. The report content not only includes the specific data and conclusions of the risk assessment, but also provides personalized prevention suggestions, using easy-to-understand language to facilitate the patient to understand their own health status and the measures to be taken.

[0049] In the present invention, a real-time monitoring step is further included. Physiological data is collected in real time through implantable nano-biosensors. These sensors are tiny in volume but powerful in function, capable of accurately obtaining key physiological indicators such as heart rate, blood pressure, and blood glucose concentration, and having extremely low invasiveness to the human body, ensuring continuous data collection with almost no perception by the patient. An anomaly detection algorithm based on quantum random walk is used to analyze the collected data. Quantum random walk utilizes the unique properties of quantum states and can more sensitively capture abnormal patterns in the data. Compared with traditional algorithms, it has higher sensitivity to subtle changes, thus being able to detect potential health risks earlier. An evaluation formula for the accuracy of anomaly detection based on the qubit error rate is used to quantitatively evaluate the effectiveness of the detection algorithm. As the basic unit of quantum information, the analysis of the error rate of qubits can intuitively reflect the reliability of the algorithm's detection results. When an abnormal situation is detected, an automatic warning is sent through quantum communication technology. Quantum communication, with its extremely high security and instantaneity, can ensure that the warning information is accurately and quickly conveyed to relevant personnel, such as medical staff or the patient himself, so as to take timely measures.

[0050] In the present invention, a knowledge graph construction and application step is further included. A knowledge graph based on quantum graph neural network is constructed. The quantum graph neural network combines the advantages of quantum computing and graph neural network and can more efficiently process the complex relationships and uncertainties between knowledge. It can integrate various types of knowledge such as medical literature, clinical guidelines, expert experience, and the latest medical research results, and present them in the form of a graph, making the associations between knowledge clearer and more intuitive. A quantum hashing algorithm is used to store knowledge. This algorithm encodes and stores knowledge using the characteristics of quantum states, not only improving the storage efficiency but also enhancing the security and retrieval efficiency of knowledge. An evaluation formula for knowledge accuracy based on quantum entanglement entropy is used to ensure the accuracy of knowledge. Quantum entanglement entropy can measure the degree of association between elements in the knowledge graph, and by analyzing it, the reliability of knowledge can be judged. During the risk assessment process, quantum reasoning and interpretation are carried out with the help of the knowledge graph. Based on the knowledge associations and logical relationships in the graph, reasonable explanations and bases are provided for the assessment results, enhancing the credibility and interpretability of the assessment results.

[0051] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution of the present invention and its inventive concept, making equivalent substitutions or changes, should be covered by the protection scope of the present invention.

Claims

1. A cardiovascular disease risk assessment system based on big data analysis, characterized in that, include: Data acquisition module: The multi-modal data fusion technology integrates data from hospital information, wearable devices, genetic testing institutions, media health remarks, and the environment, and uses quantum encryption technology to ensure security. The improved data integrity evaluation formula is , is the importance weight of the i-th type of data, is the amount of complete data of the i-th type, is the total amount of the i-th type; Data preprocessing module: Use deep reinforcement learning algorithm for data cleaning, adopt adaptive normalization method, select normalization function according to the distribution characteristics of data, and the data quality evaluation formula based on information entropy is , is the probability distribution of the i-th type of data, is the weight of the i-th type of data; Feature extraction module: Combines quantum principal component analysis with a deep autoencoder to extract features using the parallelism of quantum computing, introduces quantum entanglement technology to mine feature correlations, and adopts a formula based on quantum states Evaluate the correlation, is the quantum state of the j-th feature, is the quantum state of disease risk, and m is the number of features; Model training module: Adopt a quantum neural network ensemble learning model optimized by the quantum annealing algorithm. Determine the model parameter combination through this algorithm, combine transfer learning technology, initialize the current model with the model parameters of related diseases, and use cross-validation and quantum genetic algorithm to optimize the parameters. The model performance evaluation formula is , is the quantum state density matrix predicted by the model, is the true quantum state density matrix, and F is the quantum entanglement fidelity function; Risk assessment module: Input the features into the model to obtain the risk probability of the patient suffering from cardiovascular disease. The risk stratification formula is , is the membership degree of the i-th fuzzy rule, , are the weight coefficients, is the risk probability predicted by the model, is the subjective risk perception score of the patient, and k is the number of fuzzy rules; Result display module: Use virtual reality and augmented reality technology to display the results of cardiovascular disease risk assessment.

2. The cardiovascular disease risk assessment system based on big data analysis according to claim 1, wherein, It also includes a real-time monitoring module. This module collects the physiological data of the patient through an implantable biosensor and uses an anomaly detection algorithm based on quantum random walk to quickly detect abnormal changes in the physiological data. The anomaly detection accuracy evaluation formula based on the quantum bit error rate is , where is the number of qubit errors that occur during the detection process, is the total number of qubits; when an anomaly is detected, the system sends warning messages to the patient and the doctor through quantum communication technology.

3. The cardiovascular disease risk assessment system based on big data analysis according to claim 1, wherein It also includes a knowledge graph module, which constructs a cardiovascular disease knowledge graph based on quantum graph neural networks, processes complex knowledge relationships and uncertainties, integrates various medical knowledge, encodes and stores knowledge using a quantum hashing algorithm, and adopts a knowledge accuracy evaluation formula for the knowledge graph based on quantum entanglement entropy to evaluate the knowledge accuracy, where is the quantum entanglement entropy of the knowledge graph, is the maximum possible quantum entanglement entropy.

4. The cardiovascular disease risk assessment system based on big data analysis according to claim 1, wherein The data acquisition module uses blockchain technology to ensure the security and traceability of data. Smart contracts are introduced into the blockchain network to automatically verify the legality and integrity of data. Data records are encrypted and stored in the blockchain, and a blockchain data security evaluation formula based on quantum signatures is used to evaluate the security level, where is the amount of data verified by quantum signatures, is the total amount of data.

5. The cardiovascular disease risk assessment system based on big data analysis according to claim 1, characterized in that, The data preprocessing module adopts a data augmentation method based on the quantum generative adversarial network QGAN, uses the superposition and entanglement characteristics of quantum states to generate synthetic data that is more similar to real data, evaluates the quality of the generated data through quantum measurement technology, and adopts a data augmentation effect evaluation formula based on quantum fidelity to evaluate the augmentation effect, where is the quantum state density matrix of the synthetic data, is the quantum state density matrix of the real data, and F is the quantum fidelity function).

6. The cardiovascular disease risk assessment system based on big data analysis according to claim 1, wherein The feature extraction module introduces a quantum attention mechanism to enable the model to focus on features related to the risk of cardiovascular diseases. Through the attention weight assignment formula based on the quantum state transition probability attention weights are assigned to different features, where is the quantum state of the i-th feature, is the quantum state guided by attention, and m is the number of features.

7. The cardiovascular disease risk assessment system based on big data analysis according to claim 1, characterized in that The model training module adopts federated learning technology to share model parameters among medical institutions without revealing the original data of patients, and introduces quantum key distribution technology to ensure the security of model parameter transmission. The formula for evaluating the performance consistency of the federated learning model based on quantum entanglement swapping is , where is the model quantum state density matrix of the k-th medical institution, is the average model quantum state density matrix of all medical institutions, F is the quantum entanglement swapping fidelity function, s is the number of medical institutions, ensuring the consistency of model performance among different medical institutions at the quantum level.

8. A method for applying the cardiovascular disease risk assessment system based on big data analysis according to any one of claims 1-7, characterized in that, The following steps are involved: Data collection steps: Use multimodal data fusion technology to collect multi-source heterogeneous data of patients, including social media and environmental data, use quantum encryption to transmit data, use improved data integrity assessment formula to ensure data integrity, and verify the legitimacy of data through blockchain smart contracts; Data preprocessing steps: Use deep reinforcement learning algorithm to clean data, use adaptive normalization and QGAN-based data enhancement methods, and evaluate the processing effect using the data quality evaluation formula based on information entropy and the data enhancement effect evaluation formula based on quantum fidelity; Feature extraction step: Combine QPCA with deep autoencoders, introduce quantum attention mechanism and feature correlation analysis technology based on quantum entanglement, and use feature correlation evaluation formula based on quantum state to screen features; Model training steps: adopt a quantum neural network ensemble learning model optimized based on quantum annealing algorithm, combine transfer learning and federated learning technology, use multi-objective optimization and quantum genetic algorithm to adjust parameters, and use the model performance evaluation formula based on quantum entanglement fidelity to evaluate performance; Risk assessment steps: input the features into the model to obtain the risk probability, introduce fuzzy logic reasoning to modify the probability, and use the risk stratification formula based on fuzzy comprehensive evaluation to perform risk stratification; Result presentation steps: Use VR and AR technologies to immersively display the assessment results, and use natural language generation technology to provide detailed reports and personalized prevention recommendations.

9. The method of the cardiovascular disease risk assessment system based on big data analysis according to claim 8, wherein, It also includes real-time monitoring steps, which collect physiological data in real time through implantable nanobiosensors, detect anomalies using an anomaly detection algorithm based on quantum random walk, evaluate the effect using an anomaly detection accuracy evaluation formula based on quantum bit error rate, and automatically issue warnings through quantum communication technology.

10. The method of the cardiovascular disease risk assessment system based on big data analysis according to claim 8, characterized in that, It also includes the steps of knowledge graph construction and application, constructing a knowledge graph based on quantum graph neural network, using quantum hash algorithm to store knowledge, using knowledge accuracy evaluation formula based on quantum entanglement entropy to ensure knowledge accuracy, and performing quantum reasoning and interpretation during the evaluation process.

Images