Fall risk real-time evaluation system and method based on big data

Abstract

The invention discloses a fall risk real-time assessment system and method based on big data, and relates to the field of risk assessment, and the system comprises a data collection and preprocessing module, a big data processing and integration module, a feature extraction and selection module, a fall risk assessment module and a real-time monitoring and early warning module which are sequentially connected and cooperatively work. Wherein the data acquisition and preprocessing module is responsible for acquiring data from a plurality of sources and converting the data into digital signals; the big data processing and integration module is responsible for secure transmission, convergence and preprocessing of data; the feature extraction and selection module is responsible for extracting features related to the falling risk from the preprocessed data; the tumble risk assessment module is responsible for real-time assessment of tumble risks based on the extracted features; and the real-time monitoring and early warning module is responsible for triggering an early warning signal according to the falling risk assessment result. According to the invention, accurate and real-time evaluation of the falling risk, accurate and timely output of the evaluation result and rapid response of the early warning signal are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of risk assessment, and in particular to a real-time fall risk assessment system and method based on big data. Background Art

[0002] Falls are a serious health problem for people of all ages, especially the elderly and those with physical disabilities. They may lead to serious consequences such as fractures and craniocerebral injuries, which not only affects the quality of life of individuals, but also increases medical costs and social burdens. Therefore, accurate and real-time assessment of fall risk is crucial to prevent falls.

[0003] However, existing fall risk assessment technologies have many limitations. In terms of data collection, traditional methods are not comprehensive enough. Biosensors may not accurately collect skin microbial metabolic signals, and are not sensitive enough to environmental factors, which cannot fully reflect the impact of the environment on fall risk. In addition, with the advent of the big data era, data security and privacy protection have become new challenges. Existing systems may lack effective data encryption technology and cannot guarantee the integrity and security of data. In terms of data processing and feature extraction, existing systems have limited fusion processing capabilities for multi-source data, and it is difficult to integrate data from different data sources, resulting in low data quality. At the same time, feature extraction algorithms may not be able to accurately extract features related to fall risk, or the extracted features may be too high in dimension, affecting the computational efficiency of the assessment model. In terms of assessment models, existing models lack accuracy and real-time performance, fail to fully combine multiple factors such as individual biological characteristics, environmental factors, and historical data, and cannot update assessment results and issue warnings in a timely manner.

[0004] In summary, the existing fall risk assessment technology has shortcomings in data collection, data security, data processing, feature extraction, evaluation model, monitoring and early warning, etc. New technologies are urgently needed to improve the accuracy and real-time performance of fall risk assessment. Summary of the invention

[0005] In view of this, the present invention proposes a real-time fall risk assessment system and method based on big data to solve the problems of poor assessment accuracy and real-time performance in the prior art.

[0006] The specific technical solution of the present invention is as follows: A real-time fall risk assessment system based on big data, comprising the following modules that are connected and work together in sequence: A data acquisition and conversion module is used to collect data from multiple sources and convert it into digital signals suitable for subsequent processing; the module includes a sensor unit, which uses sensors to obtain biological related electronic signals and environmental analog signals respectively, and converts them into digital signals through circuits and converters; Big data processing and integration module, used for secure transmission, aggregation and pre-processing of data; The feature extraction and selection module is used to extract features related to risk assessment from preprocessed data. This module includes the federated learning and generative adversarial network submodules, which build a central node to coordinate the federated learning process of different data sources and use generative adversarial networks to extract risk-related features. It also includes the information bottleneck theory application submodule, which uses the information bottleneck theory as a framework to select key feature sets that are useful and independent of each other for risk assessment, and conducts refined screening in combination with domain knowledge. The fall risk assessment module is used to conduct real-time risk assessment based on the extracted features and obtain the fall risk assessment results. This module contains a hybrid model submodule, which integrates two learning models to build a hybrid model, learns the optimal strategy through the interaction between the agent and the environment, builds the model structure based on prior knowledge and expert experience, uses a large amount of data to train and optimize the model, and introduces transfer learning technology to improve the generalization ability of the model. The real-time monitoring and early warning module is used to trigger early warning signals based on the fall risk assessment results.

[0007] Specifically, for the data acquisition and conversion module, a biosensor array based on the principle of microbial fuel cells is closely attached to the skin to capture the electronic signals generated by microbial metabolism in real time. The output signal of the biosensor array is assumed to be , where t represents time; The environmental sensor made of composite materials is used to collect the analog signal of the environment in real time. The output signal of the environmental sensor is ; The biological signal and the environmental signal are converted into digital signals through circuits and converters. The digital signal of the biological signal is expressed as , the digital signal of the environmental signal is expressed as ; The biological signal conversion process can be expressed as ,in represents the biological signal conversion function; the environmental signal conversion process can be expressed as ,in represents the environmental signal conversion function; In the conversion process, noise suppression technology and bio-electrocatalytic amplification analog-to-digital conversion technology are used. The biological signal after noise suppression is assumed to be ,but ; Assume that the output of the ambient signal after signal conditioning is ,but ; The final digital signal and It is transmitted to the big data processing and integration module for subsequent processing.

[0008] Specifically, suppose the feature vector obtained after feature extraction and selection is , They represent the first, second, and nth eigenvalues ​​in the key feature set that is useful and independent of fall risk assessment after feature extraction and selection modules; the weight vector of deep reinforcement learning is , They represent the first, second, and nth weight values ​​learned in the deep reinforcement learning part respectively; the conditional probability matrix of the Bayesian network is , Represents the conditional probability value in the Bayesian network, and the prior knowledge influence factor is , the expert experience impact factor is , the Bayesian network structure construction is affected by prior knowledge and expert experience. , where K and E are quantities that abstractly represent prior knowledge and expert experience; fall risk assessment results , is the activation function, is the relevant probability quantity, Indicates that under certain conditions, The probability of a certain event occurring related to a feature. These probability values ​​together constitute the conditional probability matrix of the Bayesian network. , used to assist in determining the final fall risk value based on probabilistic reasoning in fall risk assessment, represents the i-th eigenvalue, Represents the i-th weight value.

[0009] Specifically, the federated learning and generative adversarial network sub-modules of the feature extraction and selection module construct a central node to coordinate and manage the federated learning process of different data sources to ensure that the privacy and security of the data are strictly protected. Each data source deploys the TensorFlow federated learning framework to establish a local feature learning model, and communicates with the central node through a secure communication protocol to share the learned features or model parameters. A generative adversarial network is constructed based on federated learning, and the generator and discriminator work together to extract features related to the risk of falling.

[0010] Specifically, suppose there are v data source terminals, which are respectively , the central node is C, the generator of the generative adversarial network is G, the discriminator is D, the features or model parameters learned from the data source are F, and the final feature set related to fall risk is P; In the federated learning phase, each data source Deploy the federated learning framework and build a local feature learning model ,in A function for building a local model; Each data source Communicate with the central node C through a secure communication protocol and transmit the learned features or model parameters Shared to the central node: , is the communication function; The central node C collects the information shared by all data sources to form an information collection , They respectively represent the first, second, and vth key features related to the risk of falling obtained after being processed by the feature extraction and selection module.

[0011] Specifically, based on the information set F obtained by federated learning, a generative adversarial network is constructed, and the generator G generates potential features related to the risk of falling according to the input: ; The discriminator D generates features from the generator G To identify the authenticity and relevance of the fall risk, the two continuously optimize the network parameters through adversarial training: , is the training function, and They are the generator and discriminator after training and optimization respectively; finally, the feature set related to the risk of falling is extracted by the trained and optimized generative adversarial network , is the function for extracting the final features.

[0012] Specifically, the big data processing and integration module includes a secure transmission sub-module, which uses a homomorphic encryption algorithm to encrypt sensitive data to ensure security during transmission, and uses cyclic redundancy check and message authentication code technology at the receiving end to verify the integrity of the data; it also includes a data aggregation and preprocessing sub-module. The data aggregation and preprocessing sub-module is designed with a flexible data interface that can receive data from the data acquisition module and external data sources, and perform format unification, deduplication, and merging processing.

[0013] Specifically, the warning signal generation submodule of the real-time monitoring and warning module adopts bio-electronic-photonic integration technology to design a signal generating circuit inside the wearable device. It is precisely managed by a computer system, and the fall risk level is divided according to the fall risk results. Then, the corresponding warning signal is generated in real time according to the changes in the fall risk level.

[0014] Specifically, is the fall risk level, S is the warning signal, H indicates a high risk level, and L indicates a low risk level. is the function that generates signals when the risk is high, It is a function that generates signals when the risk is low. The execution formula is: .

[0015] This application also proposes a real-time fall risk assessment method based on big data, including the following steps: Step 1, data collection and conversion, collects data from multiple sources, and uses a biosensor array based on the principle of microbial fuel cells to fit closely to the skin to capture electronic signals generated by microbial metabolism in real time. At the same time, environmental sensors made of composite materials are used to collect analog signals from the environment in real time. Subsequently, these signals are converted into digital signals that can be used for subsequent processing through circuits and converters; Step 2, big data processing and integration, securely transmit the collected digital signals, use homomorphic encryption algorithm to encrypt sensitive data, and ensure data security during transmission; at the receiving end, use cyclic redundancy check and message authentication code technology to verify the integrity of the data; then, design a flexible data interface to receive data from the data collection step and external data sources, and unify the format, remove duplicates, and merge them; Step 3: Feature extraction and selection. A central node is built to coordinate and manage the federated learning process of different data sources. Generative adversarial networks are used to extract features related to fall risk. At the same time, the information bottleneck theory is used as the theoretical framework for feature selection and compression to select key feature sets that are useful and independent of each other for fall risk assessment, and the features are refined by combining domain knowledge. Step 4: Fall risk assessment: Based on the extracted features, a hybrid model is constructed by integrating deep reinforcement learning and Bayesian network to conduct real-time assessment of fall risk; the optimal strategy is learned through the interaction between the agent and the environment, and the Bayesian network structure is constructed based on prior knowledge and expert experience; a large amount of data is used for model training and optimization, and transfer learning technology is introduced to improve the generalization ability of the model; Step 5: Real-time monitoring and early warning, continuously obtain the individual's fall risk status. When the fall risk level changes, use integrated technology combined with personalized threshold settings to generate early warning signals to accurately prompt the individual's current fall risk level; and use a multi-channel push mechanism to ensure timely communication of early warning information.

[0016] The beneficial effects of the present invention are: The present invention adopts microbial fuel cell biosensors and environmental sensors, combined with circuits and converters, to achieve accurate real-time data collection and conversion, providing a reliable data basis for fall risk assessment.

[0017] The present invention introduces homomorphic encryption algorithm and cyclic redundancy check technology to comprehensively guarantee the integrity and security of data during transmission, thus enhancing the user's confidence in privacy protection.

[0018] The present invention constructs a federated learning network and utilizes a generative adversarial network and information bottleneck theory to efficiently extract and screen out key features related to the risk of falling, thereby improving computational efficiency.

[0019] The present invention integrates deep reinforcement learning and Bayesian networks to construct a hybrid model, and through transfer learning technology, ensures the accuracy and real-time nature of fall risk assessment, providing support for timely intervention.

[0020] The system of the present invention monitors the fall risk status around the clock, uses integrated technology and personalized threshold settings to quickly generate warning signals, and adopts bio-electronic-photonic integrated technology to achieve rapid transmission of warning signals, thus buying valuable time for preventive measures. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0022] Figure 1 It is a structural schematic diagram of the system of the present invention; Figure 2 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0023] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0024] The present invention proposes a real-time fall risk assessment system and method based on big data.

[0025] See also Figure 1 , a real-time fall risk assessment system based on big data, including the following modules that are connected and work together in sequence: The data acquisition and conversion module is responsible for collecting data from multiple sources and converting them into digital signals that can be used for subsequent processing; it includes: biological and environmental sensing units, which use biosensor arrays based on the principle of microbial fuel cells and environmental sensors made of composite materials, which fit closely to the skin, capture the electronic signals generated by microbial metabolism and collect analog signals from the environment in real time, and convert these signals into digital signals through circuits and converters.

[0026] The big data processing and integration module is responsible for the secure transmission, aggregation and preprocessing of data; it includes: the secure transmission submodule, which uses homomorphic encryption algorithm to encrypt sensitive data to ensure the security of the transmission process, and uses cyclic redundancy check and message authentication code technology to verify the integrity of the data at the receiving end; The data aggregation and preprocessing submodule is designed with a flexible data interface that can receive data from the data acquisition module and external data sources, and perform format unification, deduplication, and merging processing.

[0027] The feature extraction and selection module is responsible for extracting features related to fall risk from the processed data. It includes: the federated learning and generative adversarial network submodule, which builds a central node to coordinate and manage the federated learning process of different data sources and uses the generative adversarial network to extract features related to fall risk; The information bottleneck theory application submodule adopts the information bottleneck theory as the theoretical framework for feature selection and compression, selects the key feature sets that are useful and independent of each other for fall risk assessment, and combines domain knowledge to conduct refined screening of features.

[0028] The fall risk assessment module is responsible for real-time assessment of fall risk based on the extracted features; it includes: a deep reinforcement learning and Bayesian network hybrid model sub-module, which integrates deep reinforcement learning and Bayesian network to build a hybrid model, learns the optimal strategy through the interaction between the intelligent agent and the environment, and builds a Bayesian network structure based on prior knowledge and expert experience. It uses a large amount of data for model training and optimization, and introduces transfer learning technology to improve the generalization ability of the model.

[0029] The real-time monitoring and warning module is responsible for triggering warning signals based on the fall risk assessment results; it includes: The warning signal generation submodule is connected to the fall risk assessment module to continuously obtain the individual's fall risk status. When the fall risk level changes, it uses integrated technology combined with personalized threshold settings to generate a warning signal to accurately prompt the individual's current fall risk level, and ensures the timely communication of warning information through a multi-channel push mechanism.

[0030] For the data acquisition and conversion module, this application strictly designs the biosensor array according to the principle of microbial fuel cells. By deeply studying the effects of different materials and structures on the efficiency of electron transfer, the optimal combination was finally determined. This application uses a metal oxide electrode material with high catalytic activity, which not only has excellent electronic conductivity, but also exhibits excellent chemical stability, thereby significantly improving the efficiency of electron transfer. This application installs the sensor array on the inside of a smart bracelet or belt, closely fitting the skin to ensure that the electronic signals generated by skin microbial metabolism can be efficiently captured.

[0031] When microorganisms metabolize on the surface of the skin, they will produce weak bioelectric signals and a variety of biological metabolites. In order to fully capture these signals, the biosensor array is equipped with a precise current detection circuit with a resolution of up to nanoamperes, so that even weak signal changes can be accurately captured. At the same time, this application adds noise suppression technology to ensure that the collected signals are accurate. In addition, microfluidic chip technology is also used to detect the concentration changes of biological metabolites. This technology is based on the principle of enzyme-linked immunosorbent assay and can simultaneously detect at least 20 key metabolites with a detection accuracy of up to picomolar level. Finally, the collected weak biological signals need to be converted into digital signals for subsequent processing and analysis. This application adopts the analog-to-digital conversion technology of bioelectrocatalytic amplification to perform signal conversion under the control of the computer system. During the conversion process, the computer accurately adjusts the bioelectrocatalytic reaction conditions (such as temperature, pH value, substrate concentration, etc.) according to the preset program, and carefully designs the analog-to-digital conversion circuit and uses a high-precision analog-to-digital converter, which ultimately ensures that the error rate of the digital signal is less than 0.1%, thereby ensuring the accuracy and reliability of the data.

[0032] In the design process of the environmental sensor unit, this application selected graphene and nano-metal oxides with special adsorption properties as composite materials, and determined the best composite method of the two materials through computer simulation technology. This composite method combines the excellent conductivity and mechanical strength of graphene with the sensitive adsorption properties of nano-metal oxides to environmental factors (such as gas concentration, humidity, etc.), thereby significantly improving the sensitivity of the sensor to environmental factors. During the manufacturing process, this application uses micro-nano processing technologies such as photolithography and electron beam evaporation. The manufactured sensor has structures such as microelectrodes and microchannels, which enable the sensor to more sensitively detect environmental factors closely related to the risk of falling, such as small changes in electric fields, small deformations of the ground, and changes in air ion concentrations.

[0033] When the environmental sensor unit starts working, it collects analog signals from the environment in real time. These signals may include changes in physical quantities such as electric field strength, degree of deformation, and ion concentration, which are closely related to factors such as the risk of falling. In order to convert these analog signals into digital signals that can be processed by computers, this application uses high-performance signal conditioning circuits and analog-to-digital converters. The signal conditioning circuit includes components such as low-noise amplifiers and filters, which amplify the signal and remove noise interference. The analog-to-digital converter converts the continuous analog signal into a discrete digital signal for subsequent processing by the computer. In order to improve the quality of the digital signal (i.e., the signal-to-noise ratio), this application optimizes the circuit. By adjusting the parameters of the low-noise amplifier and the filter, the noise suppression ratio is increased to at least 30dB or higher. This means that the sensor can more effectively suppress noise interference, thereby improving the accuracy and reliability of the digital signal.

[0034] Specifically, a biosensor array based on the principle of microbial fuel cells is closely attached to the skin to capture the electronic signals generated by microbial metabolism in real time. The output signal of the biosensor array is , where t represents time; The environmental sensor made of composite materials is used to collect the analog signal of the environment in real time. The output signal of the environmental sensor is ; The biological signal and the environmental signal are converted into digital signals through circuits and converters. The digital signal of the biological signal is expressed as , the digital signal of the environmental signal is expressed as ; The biological signal conversion process can be expressed as ,in represents the biological signal conversion function; the environmental signal conversion process can be expressed as ,in represents the environmental signal conversion function; In the conversion process, noise suppression technology and bio-electrocatalytic amplification analog-to-digital conversion technology are used. The biological signal after noise suppression is assumed to be ,but ; Assume that the output of the ambient signal after signal conditioning is ,but ; The final digital signal and It is transmitted to the big data processing and integration module for subsequent processing.

[0035] For the big data processing and integration module, in the big data processing process, secure data transmission and efficient preprocessing are two crucial links. In order to ensure the security and integrity of the data, this application introduces a secure transmission protocol based on homomorphic encryption. The protocol uses advanced homomorphic encryption algorithms (such as lattice-based homomorphic encryption algorithms) to encrypt sensitive data to ensure the high security of data during transmission. At the same time, after receiving the encrypted data, the data aggregation end will use a combination of cyclic redundancy check (CRC) and message authentication code (MAC) technologies to strictly check whether the data has been tampered with or damaged during transmission, thereby ensuring that the integrity of the data reaches an extremely high level of more than 99%.

[0036] In order to meet the needs of multi-source data fusion in big data processing, the big data processing and integration module also designs multi-source data aggregation interfaces. These interfaces can simultaneously receive data from the data acquisition module and data transmitted from external data sources after homomorphic encryption. After receiving these data, the computer will start a series of complex and sophisticated programs to process them, including key steps such as converting data into a unified format, removing duplicate data items, and merging related data records for subsequent unified processing and analysis.

[0037] In addition, in order to improve the quality and availability of data, this application introduces a data preprocessing submodule based on self-supervised learning. For aggregated big data, this submodule will construct a series of self-supervised tasks in the computer system. For example, for biological signal data (such as electrocardiogram, electroencephalogram, etc.), the submodule will construct a data reconstruction task, requiring the computer system to use part of the data to predict the overall data, so as to learn the overall structure and characteristics of the data. For environmental data (such as temperature, humidity, air quality, etc.), the submodule will construct a comparative learning task, requiring the computer system to compare the data features in different environments and find the differences and similarities between them. The construction of these self-supervised tasks is completely implemented by computer algorithms without human intervention, which greatly improves the efficiency and accuracy of data preprocessing.

[0038] After constructing the self-supervised task, the submodule uses the self-supervised learning algorithm framework of the deep learning-based autoencoder structure to perform data cleaning, denoising, and normalization. In the data cleaning stage, the computer algorithm automatically identifies and removes abnormal data, which may be caused by equipment failure, data transmission errors, etc. By adjusting the algorithm parameters based on experiments and data analysis, it can be ensured that the removal rate of abnormal data reaches more than 95%. In the denoising stage, the computer algorithm learns the inherent structure of the data and removes the noise in it, thereby improving the signal-to-noise ratio and usability of the data. In the normalization stage, the computer algorithm performs linear or nonlinear transformations on the data so that the deviation after normalization is within ±2%.

[0039] For the feature extraction and selection module, this application builds a robust computer device as the central node of federated learning. The device not only has efficient data processing capabilities, but also has a built-in strict security communication mechanism. This device is responsible for coordinating and managing the federated learning process from different data sources to ensure that the privacy and security of the data are strictly protected throughout the process. This application integrates multiple data sources such as medical institutions and health monitoring equipment manufacturers into this federated learning network. TensorFlow federated learning framework or other federated learning frameworks are deployed on each data source to establish a local feature learning model. These models run independently on their respective data sources. There is no need to upload the original data to the central node. Instead, they communicate with the central node through a secure communication protocol to share the learned features or model parameters. In this way, multiple data sources can learn and extract features together without leaking the privacy of their respective data.

[0040] On the basis of federated learning, this application further constructs a generative adversarial network (GAN). GAN consists of two parts: a generator and a discriminator, which work together to extract features related to fall risk. The generator adopts a deep convolutional neural network structure, which can generate potential features related to fall risk based on the input data (preprocessed big data). These features may cover multiple aspects such as the dynamic balance characteristics of biological-environmental synergy and the fall tendency characteristics of individual physiological rhythms. The discriminator uses a multi-layer perceptron structure to discriminate the features generated by the generator, judge their authenticity and relevance to the risk of falling. Through adversarial training, the generator and the discriminator compete and learn from each other, and continuously optimize their respective network parameters. In the end, the computer can extract more accurate features related to fall risk from the preprocessed big data. These features are represented in the form of high-dimensional vectors of 200-1000 dimensions, providing effective input for subsequent risk assessment models.

[0041] For the extracted high-dimensional feature vectors, this application uses the information bottleneck theory as the theoretical framework for feature selection and compression. By using the Shannon entropy calculation method, the mutual information between the features and the risk of falling and the redundant information between the features are accurately calculated. The mutual information reflects the strength of the correlation between the features and the risk of falling, while the redundant information reveals the degree of similarity between the features. This information provides strong support for a deeper understanding of the information relationship between the features and the risk of falling and the degree of redundancy between the features.

[0042] Based on the calculated mutual information and redundant information, this application sets a reasonable information threshold to screen features that are useful for fall risk assessment. At the same time, an efficient screening algorithm is designed, which can significantly reduce the number of features (at least 40%) while retaining more than 90% of the evaluation information. The screening algorithm carefully selects a set of key features that are useful and independent of each other for fall risk assessment based on mutual information and redundant information. Through feature selection, the input dimension of the subsequent evaluation model was successfully reduced, thereby significantly improving the computational efficiency of the model. This not only speeds up the training and prediction of the model, but also reduces the demand for computing resources.

[0043] Specifically, suppose there are v data source terminals, which are respectively , the central node is C, the generator of the generative adversarial network is G, the discriminator is D, the features or model parameters learned from the data source are F, and the final feature set related to fall risk is P; In the federated learning phase, each data source Deploy the federated learning framework and build a local feature learning model ,in A function for building a local model; Each data source Communicate with the central node C through a secure communication protocol and transmit the learned features or model parameters Shared to the central node: , is the communication function; The central node C collects the information shared by all data sources to form an information collection , They respectively represent the first, second, and vth key features related to the risk of falling obtained after being processed by the feature extraction and selection module.

[0044] Based on the information set F obtained by federated learning, a generative adversarial network is constructed, and the generator G generates potential features related to the risk of falling according to the input: ; The discriminator D generates features from the generator G To identify the authenticity and relevance of the fall risk, the two continuously optimize the network parameters through adversarial training: , is the training function, and They are the generator and discriminator after training and optimization respectively; finally, the feature set related to the risk of falling is extracted by the trained and optimized generative adversarial network , is the function for extracting the final features.

[0045] For the fall risk assessment module, a hybrid model integrating deep reinforcement learning and Bayesian network is designed to combine the advantages of both to improve the accuracy and real-time performance of fall risk assessment. In the deep reinforcement learning part, an intelligent agent is defined, which adopts the advanced neural network structure of deep Q network (DQN). The core task of the intelligent agent is to learn how to select the optimal action based on the current environmental state (including individual biological characteristics, environmental factors, etc., which contain the key feature set obtained from the feature extraction and selection module) to maximize the long-term cumulative reward. At the same time, this application designs a reward function, which is the key to evaluating the quality of the intelligent agent's actions, and motivates the intelligent agent to learn a better evaluation strategy based on prediction accuracy and fall event avoidance. In the Bayesian network part, this application constructs a network structure based on prior knowledge and expert experience, and determines the probabilistic relationship between each node (such as biological characteristics, environmental factors, etc.), which reflects how each factor works together on the risk of falling. Through expert knowledge or data learning, this application determines the conditional probability table for the nodes in the Bayesian network, providing a solid foundation for probabilistic reasoning.

[0046] Specifically, suppose the feature vector obtained after feature extraction and selection is , They represent the first, second, and nth eigenvalues ​​in the key feature set that is useful and independent of fall risk assessment after feature extraction and selection modules; the weight vector of deep reinforcement learning is , They represent the first, second, and nth weight values ​​learned in the deep reinforcement learning part respectively; the conditional probability matrix of the Bayesian network is , Represents the conditional probability value in the Bayesian network, and the prior knowledge influence factor is , the expert experience impact factor is , the Bayesian network structure construction is affected by prior knowledge and expert experience. , where K and E are quantities that abstractly represent prior knowledge and expert experience; fall risk assessment results , is the activation function, is the relevant probability quantity, Indicates that under certain conditions, The probability of a certain event occurring related to a feature. These probability values ​​together constitute the conditional probability matrix of the Bayesian network. , used to assist in determining the final fall risk value based on probabilistic reasoning in fall risk assessment, represents the i-th eigenvalue, Represents the i-th weight value.

[0047] Specifically, is the fall risk level, S is the warning signal, H indicates a high risk level, and L indicates a low risk level. is the function that generates signals when the risk is high, It is a function that generates signals when the risk is low. The execution formula is: .

[0048] In order to train this hybrid model, this application collects a large amount of clinical data and practical application feedback data, which covers the biological characteristics of different individuals, environmental factors, and fall event records, etc., providing rich samples for model training. During the deep reinforcement learning training process, the agent interacts with the environment, selects actions according to the current environmental state, and obtains feedback according to the reward function. The computer continuously adjusts the agent's strategy based on these feedbacks, so that it gradually learns the optimal fall risk assessment strategy. The training of the Bayesian network may involve the adjustment of the conditional probability table to better reflect the dependencies in the data. This application adopts a method that combines data learning and expert adjustment to ensure the accuracy and reliability of the network structure. In order to further improve the performance of the model, this application adopts an optimization method based on the simulated annealing algorithm, searches for the optimal solution in the solution space by controlling the temperature parameters, and continuously adjusts the parameters of the hybrid model until the predetermined performance indicators are reached. The accuracy of the optimized model in fall risk assessment reached more than 85%, significantly improving the accuracy of the assessment.

[0049] When the feature extraction and selection module outputs key feature sets, these feature sets are input into the hybrid model in real time. The hybrid model calculates the individual's fall risk probability value (ranging from 0 to 1) in real time in the computer system. This probability value reflects the possibility of an individual falling event in the future, providing strong support for fall prevention. In order to more intuitively understand the individual's fall risk status, this application divides the fall risk into three levels: low risk, medium risk and high risk according to preset thresholds. These thresholds can be adjusted based on clinical experimental data and actual application feedback to ensure the accuracy and practicality of the division. For example, low risk may correspond to a probability value less than 0.3, medium risk corresponds to a probability value between 0.3 and 0.7, and high risk corresponds to a probability value greater than 0.7. The computer can automatically divide the fall risk level according to these thresholds, and output corresponding preventive measures and suggestions to provide targeted support for individuals at different risk levels.

[0050] For the real-time monitoring and warning module, the real-time monitoring and warning module is closely connected with the fall risk assessment module to form a closed-loop, highly integrated monitoring system. The fall risk assessment module will comprehensively consider the individual's various physiological parameters (including but not limited to heart rate, blood pressure, gait stability, etc.), historical fall records, current activities (such as walking speed, whether strenuous exercise is being performed, ambient light conditions, etc.) and possible external factors (such as whether the ground is slippery, whether there are obstacles or steps, etc.) to comprehensively assess the individual's fall risk level. When the individual's fall risk level changes, whether it rises or falls to a certain critical threshold, the system will respond immediately and automatically trigger the warning signal generation mechanism according to the preset rules and algorithms to ensure the timeliness and effectiveness of the warning, and provide valuable reaction time for the individual to take preventive measures.

[0051] In order to achieve rapid and accurate transmission of warning signals, this application adopts bio-electronic-photonic integrated technology. This technology cleverly combines the principles of biology, electronics and photonics, and can realize the generation and efficient transmission of complex signals in wearable devices (such as smart bracelets, smart watches, etc.). Inside the wearable device, a signal generating circuit is designed, which is precisely managed by a computer system and can generate corresponding warning signals in real time according to changes in the risk level of falling. When the individual's fall risk level is high, the circuit will generate high-intensity, high-frequency light signals, electrical stimulation signals, or allow the wearable device to generate vibration signals of a specific pattern under the instruction of the computer system. These signals are designed to quickly attract the attention of individuals and remind them that there is a current risk of falling. On the contrary, when the risk level of falling is low, the intensity and frequency of the generated signal will be reduced accordingly to maintain continuous attention to the risk of falling while avoiding unnecessary interference or panic to the individual. In addition, the present application also realizes the function of converting digital signals into visual signals. By displaying icons or text prompts of the fall risk level in real time on the screens of devices such as smart watches, individuals can intuitively understand their fall risk status, thereby taking corresponding preventive measures more proactively and reducing the probability of fall incidents.

[0052] See also Figure 2 This embodiment also discloses a real-time fall risk assessment method based on big data, comprising the following steps: Step 1, data collection and conversion, collects data from multiple sources, and uses a biosensor array based on the principle of microbial fuel cells to fit closely to the skin to capture electronic signals generated by microbial metabolism in real time. At the same time, environmental sensors made of composite materials are used to collect analog signals from the environment in real time. Subsequently, these signals are converted into digital signals that can be used for subsequent processing through circuits and converters; Step 2, big data processing and integration, securely transmit the collected digital signals, use homomorphic encryption algorithm to encrypt sensitive data, and ensure data security during transmission; at the receiving end, use cyclic redundancy check and message authentication code technology to verify the integrity of the data; then, design a flexible data interface to receive data from the data collection step and external data sources, and unify the format, remove duplicates, and merge them; Step 3: Feature extraction and selection. A central node is built to coordinate and manage the federated learning process of different data sources. Generative adversarial networks are used to extract features related to fall risk. At the same time, the information bottleneck theory is used as the theoretical framework for feature selection and compression to select key feature sets that are useful and independent of each other for fall risk assessment, and the features are refined by combining domain knowledge. Step 4: Fall risk assessment: Based on the extracted features, a hybrid model is constructed by integrating deep reinforcement learning and Bayesian network to conduct real-time assessment of fall risk; the optimal strategy is learned through the interaction between the agent and the environment, and the Bayesian network structure is constructed based on prior knowledge and expert experience; a large amount of data is used for model training and optimization, and transfer learning technology is introduced to improve the generalization ability of the model; Step 5: Real-time monitoring and early warning, continuously obtain the individual's fall risk status. When the fall risk level changes, use integrated technology combined with personalized threshold settings to generate early warning signals to accurately prompt the individual's current fall risk level; and use a multi-channel push mechanism to ensure timely communication of early warning information.

[0053] The beneficial effects of the present invention are: Efficiency and accuracy of data acquisition and conversion: The present invention uses a biosensor array based on the principle of microbial fuel cells and an environmental sensor made of composite materials, which can fit closely to the skin and capture electronic signals generated by microbial metabolism and environmental analog signals in real time. Combined with circuits and converters, the error rate after the signal is converted to a digital signal is accurately controlled to less than 0.1%, which significantly improves the accuracy and reliability of the data and provides a solid foundation for subsequent data analysis and fall risk assessment.

[0054] Security of big data processing and integration: This invention innovatively introduces homomorphic encryption algorithms to encrypt sensitive data, ensuring the security of data during transmission. At the same time, cyclic redundancy check and message authentication code technology are used to effectively verify the integrity of data, prevent data from being tampered with or damaged during transmission, and comprehensively guarantee the integrity and security of data, enhancing users' confidence in privacy protection.

[0055] High accuracy and innovation of feature extraction and selection: This paper successfully integrates feature learning models of multiple data sources by building a federated learning network, which not only protects data privacy but also extracts features related to fall risk. Generative adversarial networks and information bottleneck theory are used to further screen out key feature sets that are useful and independent of each other for fall risk assessment, which reduces the input dimension of subsequent evaluation models and significantly improves computational efficiency.

[0056] Accuracy and real-time performance of fall risk assessment: This paper constructs a hybrid model that combines prior knowledge and expert experience by deeply integrating deep reinforcement learning and Bayesian networks, which greatly improves the accuracy of fall risk assessment. The model is trained and optimized using a large amount of data, and transfer learning technology is introduced to further improve the generalization ability of the model, ensuring the real-time and accuracy of the assessment results, and providing strong support for timely and effective intervention measures.

[0057] Timeliness and effectiveness of real-time monitoring and early warning: The system of the present invention can monitor the individual's fall risk status in real time around the clock. Once the risk level changes, it quickly generates early warning signals using integrated technology combined with personalized threshold settings. The use of advanced bio-electronic-photon integrated technology achieves rapid and accurate transmission of early warning signals, which wins precious reaction time for individuals to take preventive measures in a timely manner, further demonstrating the practical value of the present invention.

[0058] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A real-time fall risk assessment system based on big data, characterized in that: Includes the following modules that are connected in sequence and work together: A data acquisition and conversion module is used to collect data from multiple sources and convert it into digital signals suitable for subsequent processing; the module includes a sensor unit, which uses sensors to obtain biological related electronic signals and environmental analog signals respectively, and converts them into digital signals through circuits and converters; Big data processing and integration module, used for secure transmission, aggregation and pre-processing of data; Feature extraction and selection module, used to extract risk assessment-related features from preprocessed data; This module includes the Federated Learning and Generative Adversarial Network submodules, which build a central node to coordinate the federated learning process of different data sources and use the Generative Adversarial Network to extract risk-related features. It also includes the Information Bottleneck Theory Application submodule, which uses the Information Bottleneck Theory as a framework to select key feature sets that are useful and independent of each other for risk assessment, and conducts refined screening in combination with domain knowledge. The fall risk assessment module is used to conduct real-time risk assessment based on the extracted features and obtain the fall risk assessment results. This module contains a hybrid model submodule, which integrates two learning models to build a hybrid model, learns the optimal strategy through the interaction between the agent and the environment, builds the model structure based on prior knowledge and expert experience, uses a large amount of data to train and optimize the model, and introduces transfer learning technology to improve the generalization ability of the model. The real-time monitoring and early warning module is used to trigger early warning signals based on the fall risk assessment results.

2. The real-time fall risk assessment system based on big data as claimed in claim 1, characterized in that: For the data acquisition and conversion module, a biosensor array based on the principle of microbial fuel cells is used to fit closely to the skin to capture the electronic signals generated by microbial metabolism in real time. The output signal of the biosensor array is set to , where t represents time; The environmental sensor made of composite materials is used to collect the analog signal of the environment in real time. The output signal of the environmental sensor is ; The biological signal and the environmental signal are converted into digital signals through circuits and converters. The digital signal of the biological signal is expressed as , the digital signal of the environmental signal is expressed as ; The biological signal conversion process can be expressed as ,in represents the biological signal conversion function; the environmental signal conversion process can be expressed as ,in represents the environmental signal conversion function; In the conversion process, noise suppression technology and bio-electrocatalytic amplification analog-to-digital conversion technology are used. The biological signal after noise suppression is assumed to be ,but ; Assume that the output of the ambient signal after signal conditioning is ,but ; The final digital signal and It is transmitted to the big data processing and integration module for subsequent processing.

3. The real-time fall risk assessment system based on big data as claimed in claim 1, characterized in that: Suppose the feature vector obtained after feature extraction and selection is , They represent the first, second, and nth eigenvalues ​​in the key feature set that is useful and independent of fall risk assessment after feature extraction and selection modules; the weight vector of deep reinforcement learning is , They represent the first, second, and nth weight values ​​learned in the deep reinforcement learning part respectively; the conditional probability matrix of the Bayesian network is , Represents the conditional probability value in the Bayesian network, and the prior knowledge influence factor is , the expert experience impact factor is , the Bayesian network structure construction is affected by prior knowledge and expert experience. , where K and E are quantities that abstractly represent prior knowledge and expert experience; fall risk assessment results , is the activation function, is the relevant probability quantity, Indicates that under certain conditions, The probability of a certain event occurring related to a feature. These probability values ​​together constitute the conditional probability matrix of the Bayesian network. , used to assist in determining the final fall risk value based on probabilistic reasoning in fall risk assessment, represents the i-th eigenvalue, Represents the i-th weight value.

4. The real-time fall risk assessment system based on big data as claimed in claim 1, characterized in that: The federated learning and generative adversarial network submodules of the feature extraction and selection module construct a central node to coordinate and manage the federated learning process of different data sources to ensure that the privacy and security of the data are strictly protected. Each data source deploys the TensorFlow federated learning framework to establish a local feature learning model, and communicates with the central node through a secure communication protocol to share the learned features or model parameters. A generative adversarial network is constructed based on federated learning, and the generator and discriminator work together to extract features related to the risk of falling.

5. The real-time fall risk assessment system based on big data as claimed in claim 4 is characterized in that: There are v data source terminals, which are respectively , the central node is C, the generator of the generative adversarial network is G, the discriminator is D, the features or model parameters learned from the data source are F, and the final feature set related to fall risk is P; In the federated learning phase, each data source Deploy the federated learning framework and build a local feature learning model ,in A function for building a local model; Each data source Communicate with the central node C through a secure communication protocol and transmit the learned features or model parameters Shared to the central node: , is the communication function; The central node C collects the information shared by all data sources to form an information collection , They respectively represent the first, second, and vth key features related to the risk of falling obtained after being processed by the feature extraction and selection module.

6. The real-time fall risk assessment system based on big data as claimed in claim 5, characterized in that: Based on the information set F obtained by federated learning, a generative adversarial network is constructed, and the generator G generates potential features related to the risk of falling according to the input: ; The discriminator D generates features from the generator G To identify the authenticity and relevance of the fall risk, the two continuously optimize the network parameters through adversarial training: , is the training function, and They are the generator and discriminator after training and optimization respectively; finally, the feature set related to the risk of falling is extracted by the trained and optimized generative adversarial network , is the function for extracting the final features.

7. The real-time fall risk assessment system based on big data as claimed in claim 1, characterized in that: The big data processing and integration module includes a secure transmission submodule, which uses a homomorphic encryption algorithm to encrypt sensitive data to ensure security during transmission, and uses cyclic redundancy check and message authentication code technology at the receiving end to verify the integrity of the data; it also includes a data aggregation and preprocessing submodule, which is designed with a flexible data interface and can receive data from the data acquisition module and external data sources, and perform format unification, deduplication, and merging processing.

8. The real-time fall risk assessment system based on big data as claimed in claim 1, characterized in that: The warning signal generating submodule of the real-time monitoring and warning module adopts bio-electronic-photonic integration technology, designs a signal generating circuit inside the wearable device, is accurately managed by a computer system, divides the fall risk level according to the fall risk results, and then generates a corresponding warning signal in real time according to the change of the fall risk level.

9. The real-time fall risk assessment system based on big data as claimed in claim 1, characterized in that: set up is the fall risk level, S is the warning signal, H indicates a high risk level, and L indicates a low risk level. is the function that generates signals when the risk is high, It is a function that generates signals when the risk is low. The execution formula is: .

10. A real-time fall risk assessment method based on big data, applied to the real-time fall risk assessment system based on big data as claimed in claim 1, characterized in that: The following steps are involved: Step 1, data collection and conversion, collects data from multiple sources, and uses a biosensor array based on the principle of microbial fuel cells to fit closely to the skin to capture electronic signals generated by microbial metabolism in real time. At the same time, environmental sensors made of composite materials are used to collect analog signals from the environment in real time. Subsequently, these signals are converted into digital signals that can be used for subsequent processing through circuits and converters; Step 2, big data processing and integration, securely transmit the collected digital signals, use homomorphic encryption algorithm to encrypt sensitive data, and ensure data security during transmission; at the receiving end, use cyclic redundancy check and message authentication code technology to verify the integrity of the data; then, design a flexible data interface to receive data from the data collection step and external data sources, and unify the format, remove duplicates, and merge them; Step 3: Feature extraction and selection. A central node is built to coordinate and manage the federated learning process of different data sources. Generative adversarial networks are used to extract features related to fall risk. At the same time, the information bottleneck theory is used as the theoretical framework for feature selection and compression to select key feature sets that are useful and independent of each other for fall risk assessment, and the features are refined by combining domain knowledge. Step 4: Fall risk assessment: Based on the extracted features, a hybrid model is constructed by integrating deep reinforcement learning and Bayesian network to conduct real-time assessment of fall risk; the optimal strategy is learned through the interaction between the agent and the environment, and the Bayesian network structure is constructed based on prior knowledge and expert experience; a large amount of data is used for model training and optimization, and transfer learning technology is introduced to improve the generalization ability of the model; Step 5: Real-time monitoring and early warning, continuously obtaining the individual's fall risk status. When the fall risk level changes, the integrated technology combined with personalized threshold settings will generate early warning signals to accurately indicate the individual's current fall risk level; And through a multi-channel push mechanism, ensure the timely communication of early warning information.

Images