Medical image data complete transmission method and system

Abstract

The invention belongs to the technical field of data transmission, and discloses a medical image data complete transmission method and system. The method comprises the following steps: constructing a medical image data compression model, a compressed packet data analysis model and a transmission strategy generation model; using the medical image data compression model to compress the real-time medical image data; analyzing the real-time compressed packet data by using a compressed packet data analysis model, and performing local caching; performing transmission strategy generation on the real-time analysis result by using a transmission strategy generation model; according to a real-time transmission strategy, real-time data transmission is carried out on the real-time compressed packet data, and transmission check is carried out; and if the real-time transmission check result is that missing or failed target real-time compressed packet data exists, returning to the transmission strategy generation step. According to the invention, the problems of low transmission efficiency, low transmission reliability and integrity and low intelligent degree in the prior art are solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of data transmission, and particularly relates to a method and system for complete transmission of medical image data. Background Art

[0002] Medical image data refers to the image materials obtained through imaging techniques during medical diagnosis and treatment. These data are usually generated by various medical imaging devices, such as X-ray films, computed tomography, magnetic resonance imaging, ultrasound examinations, positron emission tomography, etc. With the development of medical imaging technology, the amount of medical image data has increased sharply, posing higher requirements for the transmission of medical image data. The transmission of medical image data is of great significance for remote diagnosis, condition analysis, and treatment.

[0003] The existing medical image data transmission technologies have the following defects:

[0004] 1) Low transmission efficiency: The existing technologies often do not effectively utilize data compression algorithms, resulting in insufficient compression of medical image data before transmission, large data volume, long transmission time, and lack of an intelligent scheduling mechanism during transmission, unable to dynamically adjust the transmission strategy according to the network condition, leading to low transmission efficiency;

[0005] 2) Low transmission reliability and integrity: In the case of unstable network or transmission errors, the existing technologies fail to provide effective data recovery and retransmission mechanisms, resulting in data loss, low reliability, lack of real-time monitoring and verification of the transmission process, unable to detect and handle transmission failures in a timely manner, and low integrity;

[0006] 3) Low intelligence level: The existing technologies often only focus on data transmission and ignore the intelligent analysis of data content, unable to adjust the transmission priority according to the importance or urgency of the data. Summary of the Invention

[0007] In order to solve the problems of low transmission efficiency, low transmission reliability and integrity, and low intelligence level existing in the prior art, the purpose of the present invention is to provide a method and system for complete transmission of medical image data.

[0008] The technical solution adopted by the present invention is as follows:

[0009] A method for complete transmission of medical image data includes the following steps:

[0010] Using artificial intelligence algorithms, construct a medical image data compression model, a compressed package data analysis model, and a transmission strategy generation model;

[0011] Collect real-time medical image data, and use a medical image data compression model to compress the real-time medical image data to obtain real-time compressed packet data;

[0012] Use a compressed packet data analysis model to analyze the real-time compressed packet data to obtain a real-time analysis result, and locally cache the real-time compressed packet data to obtain a real-time local cache record;

[0013] Use a transmission strategy generation model to generate a transmission strategy for the real-time analysis result to obtain a real-time transmission strategy for the corresponding real-time compressed packet data;

[0014] According to the real-time transmission strategy, perform real-time data transmission on the real-time compressed packet data to generate a real-time data transmission record, and perform a transmission check with the real-time local cache record to obtain a real-time transmission check result;

[0015] If the real-time transmission check result indicates that there are missing or failed target real-time compressed packet data, then based on the target real-time compressed packet data, return to the transmission strategy generation step.

[0016] Furthermore, use artificial intelligence algorithms to construct a medical image data compression model, a compressed packet data analysis model, and a transmission strategy generation model, including the following steps:

[0017] Collect a number of historical medical image data, and preprocess the number of historical medical image data to obtain a number of preprocessed historical medical image data;

[0018] According to a number of preprocessed historical medical image data, use a deep learning algorithm to construct a medical image data compression model, and generate a number of historical compressed packet data;

[0019] According to a number of historical compressed packet data, use a deep learning algorithm to construct a compressed packet data analysis model, and generate a number of historical analysis results;

[0020] According to a number of historical analysis results, use a reinforcement learning algorithm to construct a transmission strategy generation model, and generate a number of historical transmission strategy generation experiences.

[0021] Furthermore, the medical image data compression model is constructed based on the CNN-DBN algorithm, and the medical image data compression model includes a graph feature extraction module constructed based on the CNN algorithm and a medical image data compression module constructed based on the DBN algorithm that are connected in sequence;

[0022] The compressed packet data analysis model is constructed based on the RF-MLP algorithm, and the compressed packet data analysis model includes a key feature screening module constructed based on the RF algorithm and a compressed packet data analysis module constructed based on the MLP algorithm that are connected in sequence;

[0023] The transmission strategy generation model is constructed based on the FSSA-DQN algorithm. The transmission strategy generation model includes a parameter optimization module constructed based on the FSSA algorithm and a transmission strategy generation module constructed based on the DQN algorithm, which are connected in sequence. The transmission strategy generation module is provided with an agent, a deep Q-network, and an experience replay pool.

[0024] Furthermore, collecting real-time medical image data and using a medical image data compression model to compress the real-time medical image data to obtain real-time compressed packet data includes the following steps:

[0025] Collect real-time medical image data and preprocess the real-time medical image data to obtain preprocessed real-time medical image data;

[0026] Use the graph feature extraction module of the medical image data compression model to extract the real-time graph features of the preprocessed real-time medical image data;

[0027] Use the medical image data compression module of the medical image data compression model to compress the real-time graph features to obtain real-time compressed packet data.

[0028] Furthermore, the cloud data center uses the medical image data compression module of the medical image data compression model to compress the real-time graph features to obtain real-time compressed packet data, including the following steps:

[0029] Use the medical image data compression module of the medical image data compression model to divide the real-time graph features into sub-blocks to obtain several real-time sub-blocks;

[0030] Encode each real-time sub-block to obtain several real-time compressed codewords, and merge the several real-time compressed codewords to obtain initial real-time compressed packet data;

[0031] Post-process the initial real-time compressed packet data to obtain the final real-time compressed packet data.

[0032] Furthermore, using a compressed packet data analysis model to analyze the real-time compressed packet data to obtain a real-time analysis result and locally cache the real-time compressed packet data to obtain a real-time local cache record includes the following steps:

[0033] Use the key feature screening module of the compressed packet data analysis model to extract several real-time key features of the real-time compressed packet data;

[0034] Use the compressed packet data analysis module of the compressed packet data analysis model to analyze the several real-time key features to obtain a real-time analysis result;

[0035] Locally cache the real-time compressed packet data to obtain a real-time local cache record.

[0036] Further, use the transmission policy generation model to generate a transmission policy for the real-time analysis results, obtaining the real-time transmission policy for the corresponding real-time compressed package data, including the following steps:

[0037] Perform an impact mapping on the real-time analysis results to obtain real-time impact mapping factors, and based on the real-time impact mapping factors, set the optimization objectives of the parameter optimization module of the transmission policy generation model;

[0038] According to the optimization objectives, use the parameter optimization module to perform parameter optimization and generate the optimal network parameters of the deep Q-network of the transmission policy generation module of the transmission policy generation model;

[0039] Based on the optimal network parameters, optimize the deep Q-network to obtain the optimized deep Q-network, and randomly extract several historical transmission policy generation experiences from the experience replay pool of the transmission policy generation module;

[0040] According to the real-time analysis results, update the state space of the transmission policy generation module to obtain the updated state space;

[0041] According to several historical transmission policy generation experiences, update the action space of the transmission policy generation module to obtain the updated action space;

[0042] Based on the updated state space, updated action space, and optimized deep Q-network, use the agent to perform transmission policy generation to obtain the real-time transmission policy.

[0043] Further, according to the real-time transmission policy, perform real-time data transmission on the real-time compressed package data, generate a real-time data transmission record, and perform a transmission check with the real-time local cache record to obtain the real-time transmission check result, including the following steps:

[0044] Configure a data transmission channel between the data provider and the data receiver, and adjust the data transmission channel according to the real-time transmission policy to obtain the adjusted data transmission channel;

[0045] Based on the adjusted data transmission channel, execute the real-time transmission policy to perform real-time data transmission on the real-time compressed package data and generate a real-time data transmission record;

[0046] Traverse all the real-time compressed package data to obtain several real-time data transmission records, and perform a transmission check on the real-time data transmission records and the corresponding real-time local cache records to obtain the real-time transmission check result.

[0047] Further, if the real-time transmission check result indicates that there are missing or failed target real-time compressed package data, then based on the target real-time compressed package data, return to the transmission policy generation step, including the following steps:

[0048] Analyze the real-time transmission inspection results. If there are missing real-time local cache records or failed real-time data transmission records in the real-time transmission inspection results, proceed to the next step;

[0049] Use the missing real-time local cache records or the real-time compressed package data corresponding to the failed real-time local cache records as the target real-time compressed package data;

[0050] Use the transmission strategy generation model to regenerate the transmission strategy for the target real-time analysis results of the target real-time compressed package data to obtain the target real-time transmission strategy;

[0051] According to the target real-time transmission strategy, perform real-time data transmission on the target real-time compressed package data, generate real-time data transmission records, and re-perform transmission inspection with the corresponding real-time local cache records.

[0052] A medical image data complete transmission system for implementing the medical image data complete transmission method. The system includes a model construction unit, a data compression unit, a compressed package analysis unit, a transmission strategy generation unit, a transmission inspection unit, and a retransmission unit connected in sequence.

[0053] The beneficial effects of the present invention are as follows:

[0054] A medical image data complete transmission method and system provided by the present invention can effectively reduce the data size and shorten the transmission time by using the medical image data compression model, thereby significantly improving the data transmission efficiency. The transmission strategy generation model can dynamically adjust the transmission parameters according to the network conditions, further optimize the transmission process, and reduce the delay; through real-time monitoring and transmission inspection, it can timely discover and handle the omissions or failures in data transmission, ensure the integrity of the data, and provide a retransmission mechanism for the failed transmission records, ensuring the final successful transmission of the data and improving the reliability; the compressed package data analysis model can intelligently analyze the medical image data, extract key information, provide support for the generation of the transmission strategy, and be used to adjust the transmission priority to ensure the priority transmission of important and urgent data, improving the degree of intelligence.

[0055] Other beneficial effects of the present invention will be further described in the specific implementation manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 is a flow chart of the medical image data complete transmission method in the present invention.

[0057] Figure 2 is a structural block diagram of the medical image data complete transmission system in the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0058] The present invention will be further explained below in conjunction with the accompanying drawings and specific embodiments.

[0059] Embodiment 1:

[0060] As Figure 1 shown, this embodiment provides a method for complete transmission of medical image data, including the following steps:

[0061] S1: Use artificial intelligence algorithms to construct a medical image data compression model, a compressed package data analysis model, and a transmission strategy generation model, including the following steps:

[0062] S1-1: Collect a number of historical medical image data, and preprocess the number of historical medical image data to obtain a number of preprocessed historical medical image data;

[0063] The preprocessing includes denoising, image enhancement, and size normalization performed in sequence to improve the quality of medical image data and provide data support for subsequent model construction;

[0064] S1-2: According to a number of preprocessed historical medical image data, use deep learning algorithms to construct a medical image data compression model and generate a number of historical compressed package data;

[0065] The medical image data compression model is constructed based on the Convolutional Neural Networks (CNN)-Deep Belief Nets (DBN) algorithm, and the medical image data compression model includes a graph feature extraction module constructed based on the CNN algorithm and a medical image data compression module constructed based on the DBN algorithm connected in sequence;

[0066] The CNN network of the graph feature extraction module extracts hierarchical features of the image through structures such as convolutional layers and pooling layers. The DBN is a deep learning model composed of multiple Restricted Boltzmann Machines (RBM) layers, which can learn the probability distribution of graph features. The encoding process involves converting sub-blocks into a set of compressed codewords, which are compact representations of graph features. The DBN model can learn an effective representation of graph features, thereby removing redundant information during the encoding process to achieve efficient compression. The encoding process of the DBN model can ensure that all information is retained so that the original data can be fully restored during decompression to achieve complete data transmission;

[0067] S1-3: According to a number of historical compressed package data, use deep learning algorithms to construct a compressed package data analysis model and generate a number of historical analysis results;

[0068] The compressed package data analysis model is constructed based on the Random Forest (RF)-Multilayer Perceptron (MLP) algorithm, and the compressed package data analysis model includes a key feature screening module constructed based on the RF algorithm and a compressed package data analysis module constructed based on the MLP algorithm that are connected in sequence;

[0069] The key feature screening module screens the feature components of the input compressed package data through the internal Classification And Regression Tree (CART), can process a large number of feature components, generates the key feature importance scores of each feature component, selects the most stable and most discriminative key feature components according to the key feature importance scores, and the trained key feature screening module can directly perform key screening on the newly input compressed package data according to the selected key features to obtain corresponding several key features; The MLP network, as a fully connected network, can accurately and efficiently predict labels based on the key features;

[0070] S1-4: According to several historical analysis results, use the reinforcement learning algorithm to construct a transmission policy generation model and generate several historical transmission policy generation experiences;

[0071] The transmission policy generation model is constructed based on the Fast Sparrow Search Algorithm (FSSA)-Deep Q Network (DQN) algorithm, and the transmission policy generation model includes a parameter optimization module constructed based on the FSSA algorithm and a transmission policy generation module constructed based on the DQN algorithm that are connected in sequence. The transmission policy generation module is provided with an agent, a deep Q network, and an experience replay pool;

[0072] According to several historical analysis results, use the reinforcement learning algorithm to construct a transmission policy generation model and generate several historical transmission policy generation experiences, including the following steps:

[0073] S1-4-1: Use the FSSA algorithm to construct an initial parameter optimization module;

[0074] S1-4-2: Regard the transmission policy generation problem as the simulation environment of the DQN algorithm, define the state space of the DQN algorithm according to the historical analysis states of several historical analysis results, set actions according to several transmission parameters, and define the action space of the DQN algorithm;

[0075] S1-4-3: Define the reward function of the DQN algorithm according to the influence of the actions set according to the transmission parameters on the historical analysis states, and use the DQN algorithm to construct an initial transmission policy generation module;

[0076] S1-4-4: Based on the state space, action space, and reward function, using several historical analysis results, train the initial parameter optimization module and the initial transmission strategy generation module to obtain the final parameter optimization module and the final transmission strategy generation module, and generate several historical transmission strategy generation experiences;

[0077] Store several historical transmission strategy generation experiences in the experience replay pool, and integrate the final parameter optimization module and the final transmission strategy generation module to obtain the transmission strategy generation model;

[0078] The FSSA algorithm is an optimization algorithm based on swarm intelligence, inspired by the foraging behavior and anti-predation behavior of sparrows. There are two types of sparrows in the algorithm: discoverers and joiners. Discoverers are responsible for finding food, while joiners follow the discoverers to find food. At the same time, the algorithm also simulates vigilant behavior to avoid the threat of predators. In the transmission strategy generation model, the FSSA algorithm is used to optimize the network parameters of the deep Q-network in the DQN network, such as the learning rate, discount factor, network structure parameters, etc. By optimizing these parameters, the performance and generalization ability of the DQN network can be improved; The DQN algorithm is a deep learning algorithm used to solve reinforcement learning problems with decision-making processes. It approximates the Q-function through a deep Q-network. The Q-function is used to evaluate the expected return of taking a specific action in a given state. The agent is used in the transmission strategy generation model and is responsible for selecting the transmission strategy according to the current network state. The deep Q-network is the core of the DQN algorithm and is used to learn the value function between states and actions, that is, the Q-value. The experience replay pool: used to store the experiences (states, actions, rewards, next states) of the agent. These experiences are used in subsequent training processes to break the correlation between data and reduce the instability of training;

[0079] S2: Collect real-time medical image data, and use the medical image data compression model to compress the real-time medical image data to obtain real-time compressed packet data, including the following steps:

[0080] S2-1: Collect real-time medical image data and preprocess the real-time medical image data to obtain preprocessed real-time medical image data;

[0081] S2-2: Use the graph feature extraction module of the medical image data compression model to extract the real-time graph features of the preprocessed real-time medical image data;

[0082] S2-3: Use the medical image data compression module of the medical image data compression model to compress the real-time graph features to obtain real-time compressed packet data, including the following steps:

[0083] S2-3-1: The medical image data compression module using the medical image data compression model divides the real-time graph features into sub-blocks to obtain several real-time sub-blocks;

[0084] S2-3-2: Encode each real-time sub-block to obtain several real-time compressed codewords, and merge the several real-time compressed codewords to obtain the initial real-time compressed packet data;

[0085] S2-3-3: Post-process the initial real-time compressed packet data to obtain the final real-time compressed packet data;

[0086] Through post-processing, the data can be further compressed, reducing the bandwidth required for storage and transmission, ensuring that the compressed data will not be damaged due to errors during transmission or storage, maintaining the reliability of the data. The post-processing step provides the necessary metadata for the decompression process, ensuring that the decompression can accurately restore the original data;

[0087] S3: Use the compressed packet data analysis model to analyze the real-time compressed packet data to obtain real-time analysis results, and locally cache the real-time compressed packet data to obtain real-time local cache records, including the following steps:

[0088] S3-1: Use the key feature screening module of the compressed packet data analysis model to extract several real-time key features of the real-time compressed packet data;

[0089] The key features include the size of the compressed packet data, the transmission timestamp of the compressed packet data, the type of the compressed packet data, the source and destination of the compressed packet data, the priority or label of the compressed packet data, the compression ratio of the compressed packet data, and the integrity check information (such as checksum) of the compressed packet data;

[0090] S3-2: Use the compressed packet data analysis module of the compressed packet data analysis model to analyze the several real-time key features to obtain real-time analysis results;

[0091] The analysis results include the content summary of the compressed packet data, the priority of the compressed packet data, potential security threats, transmission efficiency requirements, the loss or damage situation of the compressed packet data, and the network condition;

[0092] S3-3: Locally cache the real-time compressed packet data to obtain real-time local cache records;

[0093] The local cache records include: the compressed packet file name, the compressed packet size, the compression timestamp, and the compressed packet checksum (used to verify data integrity);

[0094] S4: Use the transmission strategy generation model to generate a transmission strategy for the real-time analysis results to obtain the real-time transmission strategy of the corresponding real-time compressed packet data, including the following steps:

[0095] S4-1: Perform impact mapping on the real-time analysis results to obtain real-time impact mapping factors, and based on the real-time impact mapping factors, set the optimization objective of the parameter optimization module of the transmission policy generation model;

[0096] Impact mapping is a process of converting analysis results into impact factors, which can represent the importance, urgency, or transmission priority of data, including factors such as data sensitivity, urgency, data volume, data integrity, system resources, and cost-effectiveness. Real-time impact mapping factors can help determine which data packets require faster transmission or higher transmission quality;

[0097] The optimization objective is usually set based on the impact mapping factors. For example, minimize transmission delay, maximize transmission success rate, or optimize bandwidth utilization. For example, minimize transmission cost, maximize transmission success rate, or optimize bandwidth utilization; in this embodiment, the optimization objective is to minimize transmission cost;

[0098] S4-2: According to the optimization objective, use the parameter optimization module to perform parameter optimization to generate the optimal network parameters of the deep Q-network of the transmission policy generation module of the transmission policy generation model, improve the adaptability and efficiency of the transmission policy, and better adapt to specific transmission environments and task requirements, including the following steps:

[0099] S4-2-1: According to the optimization objective, set the fitness function of the FSSA algorithm of the parameter optimization module, and set the algorithm parameters and the maximum number of iterations of the FSSA algorithm;

[0100] The formula is:

[0101]

[0102] In the formula, f(X) is the fitness value of the FSSA individual X; A(X) is the transmission resource cost function of the FSSA individual X; T(X) is the transmission time cost function of the FSSA individual X; L(X) is the transmission distance cost function of the FSSA individual X; α, β, are all weight coefficients; X is the FSSA individual reference parameter;

[0103] The algorithm parameters of the FSSA algorithm include that the search space is N×D dimensions, and the search space food is determined as F = [F1, F2,..., F D T and the sparrow position is X = [X h1 , X h2 ,..., X hD T ; where F is the search space food matrix, F1, F2,..., F D ​​All are elements of the food matrix in the search space. X is the sparrow position matrix, X h1 , X h2 ,..., X hD All are elements of the sparrow position matrix. N is the number of individuals in FSSA, D is the dimension of the parameter optimization problem; h is the total number of sparrows;

[0104] S4-2-2: Encode the network parameters of the deep Q network into the individual vectors of the FSSA algorithm of the parameter optimization module. According to the algorithm parameters and individual vectors, use the Circle chaotic mapping sequence for initialization to obtain an initial FSSA population including several initial FSSA individuals;

[0105] The formula is:

[0106]

[0107] In the formula, X' c is the initial FSSA individual of the Circle chaotic mapping; X c * is the randomly generated initial FSSA individual;

[0108] S4-2-3: Use the fitness function to obtain the initial fitness values of all initial FSSA populations, and sort the initial FSSA individuals according to the initial fitness values to obtain the initial discoverers, initial joiners, and initial predators;

[0109] S4-2-4: Update the initial FSSA population to obtain an updated FSSA population; the updated FSSA population includes updated discoverers, updated joiners, and updated predators;

[0110] The update formula for the discoverer is:

[0111]

[0112] In the formula, are the c-th discoverer FSSA individuals in the (t + 1)-th and t-th iterations respectively; iter max is the maximum number of iterations; ξ is a random number between 0 and 1; Q is a normally distributed random number; L is a 1×D matrix with all elements being 1; R2 is the warning value; ST is the safety threshold;

[0113] The update formula for the joiner is:

[0114]

[0115] In the formula, are the c-th joiner FSSA individuals in the (t + 1)-th and t-th iterations respectively; The best position occupied by the exposed identity person; Is the current worst position; iter max Is the maximum iteration number threshold; ξ is a random number between 0 and 1; L is a 1×D matrix with all elements being 1 or -1; A + Is the position update parameter;

[0116] The update formula of the predator is:

[0117]

[0118] In the formula, Are the c-th predator FSSA individuals at the (t + 1)-th and t-th iterations respectively; δ is the step size control parameter, and δ = a"·γ", a" is the convergence factor, γ" is a positive real number for step size control that is not 0; Is the current best position; f c 、f g 、f w Are the current, best, and worst fitnesses of the FSSA individual respectively; γ is the minimum constant to prevent the denominator from being 0;

[0119]

[0120] In the formula, a" is the convergence factor; tanh(.) is the hyperbolic tangent function; t is the iteration indicator; t max Is the maximum number of iterations; a max 、a min Are the maximum and minimum values of the convergence factor respectively; λ is the decreasing rate parameter, k" is the decreasing period parameter, λ = -2π, k" = π;

[0121] S4-2-5: Use the dynamic reverse learning algorithm to perform dynamic reverse learning on the updated FSSA population to generate a dynamically reversed FSSA population;

[0122] The formula is:

[0123]

[0124] In the formula, Is the dynamically reversed FSSA individual; γ* is the decreasing inertia coefficient; ub is the upper limit of the search space; lb is the lower limit of the search space; Is the updated FSSA individual;

[0125] S4-2-6: Use the fitness function to obtain the updated fitness values of all FSSA individuals in the updated FSSA population and the dynamically reversed FSSA population, and obtain the optimal individual according to the updated fitness values;

[0126] S4-2-7: Decode the individual vector of the optimal individual to obtain the optimal model parameters of the deep Q-network;

[0127] S4-3: Optimize the deep Q-network based on the optimal network parameters to obtain the optimized deep Q-network, and randomly extract several historical transmission strategy generation experiences from the experience replay pool of the transmission strategy generation module;

[0128] S4-4: Update the state space of the transmission strategy generation module according to the real-time analysis results to obtain the updated state space S' = [s'1,..., s' i" ,..., s' I' , where s' i" is the updated i-th state value, i" is the state indicator, and I' is the total number of state space dimensions;

[0129] S4-5: Update the action space of the transmission strategy generation module according to several historical transmission strategy generation experiences to obtain the updated action space A' = [a'1,..., a' j" ,..., a' I , where a' j" is the updated j-th action value, j" is the action indicator, and I is the total number of action space dimensions;

[0130] S4-6: Based on the updated state space, updated action space, and optimized deep Q-network, use the agent to generate the transmission strategy to obtain the real-time transmission strategy, including the following steps:

[0131] S4-6-1: Use the agent to control the optimized deep Q-network to generate the Q value of each possible transmission parameter adjustment action in the updated action space for each real-time analysis state in the updated state space;

[0132] S4-6-2: According to the reward function, obtain the reward value of each possible real-time inspection route optimization action for each real-time inspection route optimization analysis state, and update the corresponding Q value according to the reward value to obtain the updated Q value;

[0133] The formula is:

[0134] Q(s' p' , a' p' ) = (1 - α") · Q(s p' , a p' ) + α" · (R(s p' , a p' , s' p' ) + γ* · Q max (s p' , a p' ))

[0135] Wherein, Q(s' p' , a' p' ) is the updated Q value corresponding to the updated state value s' p' and the updated action value a' p' ; Q(s p' , a p' ) is the Q value corresponding to the state value s p' and the action value a p' ; α" is the learning rate; Q max (s p' , a p' ) is the highest Q value corresponding to the state value s p' and the action value a p' ; p' is the comprehensive indicator; γ* is the update parameter; R(s p' , a' p' , s' p' ) is the reward value for converting the state value s p' to the updated state value s' p' p' ;

[0136] S4-6-3: Repeat the above Q value update steps until the number of iterations reaches the iteration number threshold, and obtain the final Q value of each possible transmission parameter adjustment action for each real-time analysis state;

[0137] S4-6-4: Use the greedy strategy to select the possible transmission parameter adjustment action corresponding to the highest final Q value as the execution transmission parameter adjustment action for the corresponding real-time analysis state;

[0138] S4-6-5: Integrate the execution transmission parameter adjustment actions of all real-time analysis states, adjust the initialized transmission parameters, and obtain the real-time transmission strategy;

[0139] The adjusted transmission parameters of the transmission strategy include bandwidth allocation scheme, compressed packet data size, transmission rate, retransmission strategy, error correction coding, encryption level, transmission protocol, network routing, timer setting, flow control, and priority setting;

[0140] S5: According to the real-time transmission strategy, perform real-time data transmission on the real-time compressed packet data, generate a real-time data transmission record, and perform a transmission check with the real-time local cache record to obtain a real-time transmission check result, including the following steps:

[0141] S5-1: Configure a data transmission channel between the data provider and the data receiver, and adjust the data transmission channel according to the real-time transmission strategy to obtain an adjusted data transmission channel;

[0142] S5-2: Based on the adjusted data transmission channel, execute the real-time transmission strategy, perform real-time data transmission on the real-time compressed packet data, and generate real-time data transmission records;

[0143] The data transmission records include: transmission status (success, failure, retransmission, etc.), transmission time, data volume transmitted, transmission rate, and any errors and exceptions that occur;

[0144] S5-3: Traverse all the real-time compressed packet data, obtain a number of real-time data transmission records, and perform transmission checks on the real-time data transmission records and the corresponding real-time local cache records to obtain real-time transmission check results;

[0145] S6: If the real-time transmission check result shows that there are missing or failed target real-time compressed packet data, then based on the target real-time compressed packet data, return to the transmission strategy generation step, including the following steps:

[0146] S6-1: Analyze the real-time transmission check result. If the real-time transmission check result shows that there are missing real-time local cache records or failed real-time data transmission records, then proceed to the next step;

[0147] S6-2: Use the real-time compressed packet data corresponding to the missing real-time local cache records or the failed real-time local cache records as the target real-time compressed packet data;

[0148] S6-3: Use the transmission strategy generation model to regenerate the transmission strategy for the target real-time analysis result of the target real-time compressed packet data to obtain the target real-time transmission strategy;

[0149] S6-4: According to the target real-time transmission strategy, perform real-time data transmission on the target real-time compressed packet data, generate real-time data transmission records, and re-perform transmission checks with the corresponding real-time local cache records.

[0150] Embodiment 2:

[0151] As Figure 2 shown, this embodiment provides a medical image data complete transmission system for implementing the medical image data complete transmission method. The system includes a model construction unit, a data compression unit, a compressed packet analysis unit, a transmission strategy generation unit, a transmission check unit, and a retransmission unit that are connected in sequence;

[0152] The model construction unit is used to construct a medical image data compression model, a compressed packet data analysis model, and a transmission strategy generation model using artificial intelligence algorithms;

[0153] The data compression unit is used to collect real-time medical image data and compress the real-time medical image data using the medical image data compression model to obtain real-time compressed packet data;

[0154] The compressed package analysis unit is used to analyze the real-time compressed package data by using the compressed package data analysis model to obtain the real-time analysis result, and locally cache the real-time compressed package data to obtain the real-time local cache record;

[0155] The transmission policy generation unit is used to generate the transmission policy for the real-time analysis result by using the transmission policy generation model to obtain the real-time transmission policy for the corresponding real-time compressed package data;

[0156] The transmission check unit is used to perform real-time data transmission on the real-time compressed package data according to the real-time transmission policy, generate the real-time data transmission record, and perform transmission check with the real-time local cache record to obtain the real-time transmission check result;

[0157] The retransmission unit is used to perform retransmission based on the target real-time compressed package data when the real-time transmission check result indicates that there are missing or failed target real-time compressed package data.

[0158] A method and system for complete transmission of medical image data provided by the present invention can effectively reduce the data size and shorten the transmission time by using the medical image data compression model, thereby significantly improving the data transmission efficiency. The transmission policy generation model can dynamically adjust the transmission parameters according to the network conditions to further optimize the transmission process and reduce the delay; through real-time monitoring and transmission check, it can timely detect and handle the missing or failure in data transmission to ensure the integrity of the data. For the records with transmission failure, a retransmission mechanism is provided to ensure the final successful transmission of the data and improve the reliability; the compressed package data analysis model can intelligently analyze the medical image data, extract key information, provide support for the generation of transmission policies, be used to adjust the transmission priority, ensure the priority transmission of important and urgent data, and improve the degree of intelligence.

[0159] The present invention is not limited to the above optional implementation manners, and anyone can obtain other various forms of products under the inspiration of the present invention. The above specific implementation manners should not be understood as limiting the protection scope of the present invention, and the protection scope of the present invention should be defined by the claims, and the specification can be used to interpret the claims.

Claims

1. A method for complete transmission of medical image data, characterized in that: The steps include: Use artificial intelligence algorithms to build medical imaging data compression models, compression package data analysis models, and transmission strategy generation models; Collecting real-time medical imaging data, and using a medical imaging data compression model to compress the real-time medical imaging data to obtain real-time compressed package data; Use the compressed package data analysis model to analyze the real-time compressed package data to obtain real-time analysis results, and locally cache the real-time compressed package data to obtain real-time local cache records; Use the transmission strategy generation model to generate a transmission strategy for the real-time analysis results to obtain the corresponding real-time transmission strategy for the real-time compressed package data; According to the real-time transmission strategy, the real-time compressed package data is transmitted in real time, a real-time data transmission record is generated, and a transmission check is performed with the real-time local cache record to obtain a real-time transmission check result; If the real-time transmission check result is that there is missing or failed target real-time compressed package data, then based on the target real-time compressed package data, return to the transmission strategy generation step.

2. A method for complete transmission of medical image data according to claim 1, characterized in that: Using artificial intelligence algorithms, we build a medical imaging data compression model, a compression package data analysis model, and a transmission strategy generation model, including the following steps: Collecting a number of historical medical image data, and preprocessing the number of historical medical image data to obtain a number of preprocessed historical medical image data; Based on some pre-processed historical medical imaging data, a medical imaging data compression model is constructed using a deep learning algorithm, and some historical compression package data is generated; Based on several historical compressed package data, a deep learning algorithm is used to build a compressed package data analysis model and generate several historical analysis results; According to several historical analysis results, a transmission strategy generation model is constructed using a reinforcement learning algorithm, and several historical transmission strategy generation experiences are generated.

3. A method for complete transmission of medical image data according to claim 2, characterized in that: The medical image data compression model is constructed based on the CNN-DBN algorithm, and the medical image data compression model includes a graph feature extraction module constructed based on the CNN algorithm and a medical image data compression module constructed based on the DBN algorithm, which are connected in sequence; The compressed package data analysis model is constructed based on the RF-MLP algorithm, and the compressed package data analysis model includes a key feature screening module constructed based on the RF algorithm and a compressed package data analysis module constructed based on the MLP algorithm, which are connected in sequence; The transmission strategy generation model is constructed based on the FSSA-DQN algorithm, and the transmission strategy generation model includes a parameter optimization module constructed based on the FSSA algorithm and a transmission strategy generation module constructed based on the DQN algorithm, which are connected in sequence. The transmission strategy generation module is provided with an intelligent agent, a deep Q network and an experience replay pool.

4. A method for complete transmission of medical image data according to claim 3, characterized in that: Collecting real-time medical image data and compressing the real-time medical image data using a medical image data compression model to obtain real-time compressed package data includes the following steps: Collecting real-time medical imaging data, and preprocessing the real-time medical imaging data to obtain preprocessed real-time medical imaging data; Using the graph feature extraction module of the medical image data compression model, extracting the real-time graph features of the pre-processed real-time medical image data; A medical image data compression module of a medical image data compression model is used to compress real-time image features to obtain real-time compressed package data.

5. A method for complete transmission of medical image data according to claim 4, characterized in that: The cloud data center uses the medical image data compression module of the medical image data compression model to compress the real-time image features to obtain real-time compressed package data, including the following steps: Using a medical image data compression module of a medical image data compression model, the real-time image features are divided into sub-blocks to obtain a plurality of real-time sub-blocks; Encoding each real-time sub-block to obtain a plurality of real-time compressed codewords, and merging the plurality of real-time compressed codewords to obtain initial real-time compressed packet data; The initial real-time compressed package data is post-processed to obtain the final real-time compressed package data.

6. A method for complete transmission of medical image data according to claim 5, characterized in that: Using the compressed package data analysis model, the real-time compressed package data is analyzed to obtain real-time analysis results, and the real-time compressed package data is locally cached to obtain real-time local cache records, including the following steps: Use the key feature screening module of the compressed package data analysis model to extract several real-time key features of the real-time compressed package data; Using the compressed package data analysis module of the compressed package data analysis model, a number of real-time key features are analyzed to obtain real-time analysis results; The real-time compressed package data is locally cached to obtain a real-time local cache record.

7. A method for complete transmission of medical image data according to claim 6, characterized in that: Using the transmission strategy generation model, a transmission strategy is generated for the real-time analysis result to obtain a corresponding real-time transmission strategy for the real-time compressed package data, including the following steps: Performing impact mapping on the real-time analysis results to obtain real-time impact mapping factors, and setting optimization targets for parameter optimization modules of the transmission strategy generation model based on the real-time impact mapping factors; According to the optimization target, a parameter optimization module is used to perform parameter optimization to generate the optimal network parameters of the deep Q network of the transmission strategy generation module of the transmission strategy generation model; Based on the optimal network parameters, the deep Q network is optimized to obtain the optimized deep Q network, and several historical transmission strategy generation experiences are randomly extracted from the experience playback pool of the transmission strategy generation module; According to the real-time analysis result, the state space of the transmission strategy generation module is updated to obtain an updated state space; According to several historical transmission strategy generation experiences, the action space of the transmission strategy generation module is updated to obtain an updated action space; Based on the updated state space, updated action space and optimized deep Q network, the intelligent agent is used to generate the transmission strategy and obtain the real-time transmission strategy.

8. A method for complete transmission of medical image data according to claim 7, characterized in that: According to the real-time transmission strategy, the real-time compressed packet data is transmitted in real time, a real-time data transmission record is generated, and a transmission check is performed with the real-time local cache record to obtain a real-time transmission check result, including the following steps: A data transmission channel is configured between a data provider and a data receiver, and the data transmission channel is adjusted according to a real-time transmission strategy to obtain an adjusted data transmission channel; Based on the adjusted data transmission channel, the real-time transmission strategy is executed to transmit the real-time compressed package data in real time and generate a real-time data transmission record; All real-time compressed packet data are traversed to obtain a number of real-time data transmission records, and the real-time data transmission records are subjected to transmission inspection with corresponding real-time local cache records to obtain real-time transmission inspection results.

9. A method for complete transmission of medical image data according to claim 8, characterized in that: If the real-time transmission check result is that there is missing or failed target real-time compressed package data, then based on the target real-time compressed package data, returning to the transmission strategy generation step includes the following steps: Analyze the real-time transmission check result. If the real-time transmission check result shows that there is a missing real-time local cache record or a failed real-time data transmission record, proceed to the next step. The real-time compressed package data corresponding to the missed real-time local cache record or the failed real-time local cache record is used as the target real-time compressed package data; Using the transmission strategy generation model, regenerate the transmission strategy for the target real-time analysis result of the target real-time compressed package data to obtain the target real-time transmission strategy; According to the target real-time transmission strategy, the target real-time compressed package data is transmitted in real time, a real-time data transmission record is generated, and the transmission check is re-performed with the corresponding real-time local cache record.

10. A complete medical image data transmission system, used to implement the complete medical image data transmission method according to any one of claims 1 to 9, characterized in that: The system comprises a model building unit, a data compression unit, a compression package analysis unit, a transmission strategy generation unit, a transmission inspection unit and a retransmission unit which are connected in sequence.

Images