A remote diagnosis system

Abstract

The present invention discloses a remote diagnosis system, which belongs to the field of remote diagnosis technology. It uses a pre-deployed intelligent optimization algorithm to adaptively adjust the bit rate of the patient's video to be diagnosed, obtains the transmission bit rate corresponding to the video to be diagnosed, and transmits the video to the physician's remote conversation terminal according to the transmission bit rate. It can dynamically adjust the data transmission rate according to the overall network usage status to ensure the real-time and accuracy of remote diagnosis. It is suitable for multiple fields such as medical care, education, and enterprises.

Description

Technical Field

[0001] The present invention belongs to the technical field of remote diagnosis, and in particular relates to a remote diagnosis system. Background Art

[0002] Remote diagnosis refers to a medical service method that transmits a patient's medical information to a remote expert or medical institution through a communication network, and the expert or medical institution provides diagnostic advice to the patient. This method breaks through geographical limitations and enables a more reasonable allocation of medical resources. Remote diagnosis is particularly important in areas with insufficient medical resources. During the remote diagnosis process, a video of the patient to be diagnosed is often collected and then transmitted to the physician. The physician then analyzes and responds to the video to be diagnosed, so that a better diagnosis can be made of the patient's condition. In the existing technology, the video to be diagnosed is often transmitted at a fixed bit rate, which results in low overall network utilization efficiency and network freezes during peak periods, reducing the quality of remote diagnosis. Summary of the Invention

[0003] The present invention provides a remote diagnosis system for solving the technical problems existing in the prior art.

[0004] A remote diagnosis system includes a patient remote conversation terminal, a physician remote conversation terminal, and a transmission code rate automatic allocation module;

[0005] The patient remote conversation terminal is used to collect the video to be diagnosed input by the patient and request the network to transmit the video to be diagnosed to the physician remote conversation terminal;

[0006] The automatic transmission bit rate allocation module is used to adaptively adjust the bit rate of the patient's video to be diagnosed using a pre-deployed intelligent optimization algorithm to obtain a transmission bit rate corresponding to the video to be diagnosed, and transmit the video to be diagnosed to the physician's remote conversation terminal according to the transmission bit rate;

[0007] The physician remote conversation terminal is used to receive the video to be diagnosed and play the video to be diagnosed so that the physician can perform remote diagnosis on the patient.

[0008] In a possible implementation, the method further includes allowing text information transmission between the patient's remote conversation terminal and the automatic transmission code rate allocation module.

[0009] In one possible implementation, a pre-deployed intelligent optimization algorithm is used to adaptively adjust the bit rate of the patient's video to be diagnosed, to obtain a transmission bit rate corresponding to the video to be diagnosed, including:

[0010] For the video to be diagnosed, the video to be diagnosed is divided into multiple video blocks, and a target bit rate is randomly assigned to each video block. The target bit rates corresponding to the video blocks are formed into a vector to obtain the bit rate vector of a single video to be diagnosed;

[0011] Combining the bit rate vectors corresponding to multiple patients into one vector to obtain a target bit rate vector, and repeatedly obtaining multiple target bit rate vectors;

[0012] For any target bit rate vector, a comprehensive video quality evaluation function is used to obtain the evaluation function value corresponding to each target bit rate vector, and the optimal target bit rate vector is obtained according to the evaluation function value corresponding to each target bit rate vector;

[0013] For any target bitrate vector, according to the optimal target bitrate vector, a multi-vector interaction strategy is used to perform multi-information fusion on the target bitrate vector to obtain the target bitrate vector after multi-information fusion.

[0014] For the target bit rate vector after multi-information fusion, a multi-information learning strategy is used to learn the better regional information of the target bit rate vector to obtain the target bit rate vector after the better regional information learning;

[0015] For the target bitrate vector after learning the better region information, an adaptive unknown region development strategy is used to search the unknown region for the target bitrate vector, and the target bitrate vector after the unknown region search is obtained;

[0016] For the target bit rate vector after the unknown area search, a global mutation development strategy is used to perform a global search on the target bit rate vector to obtain the target bit rate vector after the global search;

[0017] Determine whether the optimization end condition is met. If so, redetermine the optimal target bit rate vector based on the target bit rate vector after the global search, and determine the transmission bit rates corresponding to the videos to be diagnosed of multiple patients based on the redetermined optimal target bit rate vector. Otherwise, return to the step of obtaining the evaluation function value.

[0018] In one possible implementation, for any target bitrate vector, an evaluation function value corresponding to each target bitrate vector is obtained using a comprehensive video quality evaluation function, and an optimal target bitrate vector is obtained based on the evaluation function value corresponding to each target bitrate vector, including:

[0019] For any target bitrate vector, the comprehensive video quality evaluation function is first used to obtain the comprehensive video quality corresponding to the bitrate vector of the video to be diagnosed for each patient:

[0020]

[0021] Among them, QoE represents the overall video quality, b l represents the bit rate corresponding to the lth video block in the video to be diagnosed, L represents the total number of video blocks corresponding to the video to be diagnosed, that is, the video to be diagnosed is divided into video blocks of fixed length during transmission, and L number of video blocks are obtained, T l REBUF represents the freeze time corresponding to the lth video block in the video to be diagnosed, b l-1 represents the bit rate corresponding to the l-1th video block in the video to be diagnosed, α represents the first weighting parameter, and β represents the second weighting parameter;

[0022] Based on the weight parameter corresponding to each patient, the comprehensive video quality corresponding to the bitrate vectors of all patients is weighted and summed to obtain the evaluation function value corresponding to the target bitrate vector. The weight parameter corresponding to each patient is preset to 1 and can be modified by staff and physicians.

[0023] According to the evaluation function value corresponding to each target bit rate vector, the target bit rate vector with the largest evaluation function value is determined as the optimal target bit rate vector.

[0024] In one possible implementation, for a target bitrate vector after multi-information fusion, multi-information fusion is performed on the target bitrate vector based on the optimal target bitrate vector and using a multi-vector interaction strategy to obtain the target bitrate vector after multi-information fusion, including:

[0025]

[0026] in, represents the i-th target bitrate vector in the t-th optimization process, Represents the target bitrate vector after learning the better region information K represents the total number of target bitrate vectors, rand1 represents a random number between (0,1), represents the optimal target bit rate vector, represents the first random target rate vector, represents the second random target rate vector, represents the second random target bit rate vector.

[0027] In one possible implementation, for the target bitrate vector after multi-information fusion, a multi-information learning strategy is used to perform better regional information learning on the target bitrate vector to obtain the target bitrate vector after better regional information learning, including:

[0028]

[0029] in, Represents the jth target bitrate vector in the tth optimization process The corresponding historical optimal value, j = 1, 2, ..., K, K represents the total number of target bitrate vectors, Indicates the first optimal target bit rate vector with the second largest evaluation function value, The second best target bit rate vector with the third largest evaluation function value is represented by represents the optimal target bit rate vector, a1 represents the first random vector, a2 represents the second random vector, and a3 represents the third random vector. The dimensions of the first random vector, the second random vector, and the third random vector are the same as the dimensions of the parameter individual, and each dimension is obtained by (2rand2-1)*2(iter max -t) / iter max Randomly generated, rand2 represents a random number between (0,1), iter max Indicates the preset maximum number of optimizations, b1 represents the first information learning coefficient, b2 represents the second information learning coefficient, and b3 represents the third information learning coefficient. The first information learning coefficient, the second information learning coefficient, and the third information learning coefficient are all generated by 2rand3, and rand3 represents a random number between (0,1). Represents the target bitrate vector after unknown area search

[0030] In one possible implementation, for the target bitrate vector after learning the better region information, an adaptive unknown region development strategy is used to perform an unknown region search on the target bitrate vector to obtain the target bitrate vector after the unknown region search, including:

[0031]

[0032] in, represents the target bitrate vector after learning the mth better region information in the tth optimization process, represents the target rate vector after learning the nth better region information in the tth optimization process, n=1,2,…,K, K represents the total number of target rate vectors, represents the average bit rate vector corresponding to all target bit rate vectors, Represents the target bitrate vector after unknown area search λ represents the information fusion control factor, δ1 represents the first development coefficient and is set to 0.6; δ2 represents the second development coefficient and is set to 1.5; e represents the natural constant, d mn Represents the target bitrate vector and target bitrate vector The Euclidean distance between .

[0033] In one possible implementation, for the target bitrate vector after the unknown region search, a global mutation development strategy is used to perform a global search on the target bitrate vector to obtain the target bitrate vector after the global search, including:

[0034]

[0035] in, represents the kth target bitrate vector in the tth optimization process, Represents the target bitrate vector after global search K represents the total number of target bitrate vectors, rand4 represents a random number between (0,1), and e represents a natural constant. represents the fourth random target bit rate vector, γ represents the adaptive global search control factor, Represents a random number between [-2.5γ, 2.5γ], iter max represents the preset maximum number of optimizations, θ represents the upper limit of the scaling factor, and ξ represents the adjustment parameter of the scaling factor.

[0036] In a possible implementation, determining whether the optimization end condition is met includes: determining whether the current optimization times reaches a preset maximum optimization times; if so, determining that the optimization end condition is met; otherwise, determining that the optimization end condition is not met.

[0037] In a possible implementation, determining transmission bit rates corresponding to videos to be diagnosed of multiple patients according to the re-determined optimal target bit rate vector includes:

[0038] According to the re-determined optimal target bit rate vector, the optimal target bit rate vector is re-split into bit rate vectors corresponding to a plurality of patients;

[0039] According to the bit rate vector corresponding to each patient, the transmission bit rate of each video block in the patient's to-be-diagnosed video is determined.

[0040] The present invention provides a remote diagnosis system that adaptively adjusts the bit rate of a patient's video to be diagnosed by using a pre-deployed intelligent optimization algorithm to obtain a transmission bit rate corresponding to the video to be diagnosed, and transmits the video to be diagnosed to a physician's remote conversation terminal according to the transmission bit rate. The system can dynamically adjust the data transmission rate according to the overall network usage status to ensure the real-time and accuracy of remote diagnosis, and is suitable for multiple fields such as medical care, education, and enterprises. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0042] Figure 1A schematic diagram of the structure of a remote diagnosis system provided by an embodiment of the present invention.

[0043] The above drawings illustrate specific embodiments of the present invention, which will be described in more detail below. These drawings and the accompanying description are not intended to limit the scope of the present invention in any way, but rather to illustrate the concept of the present invention to those skilled in the art by reference to specific embodiments. DETAILED DESCRIPTION

[0044] Exemplary embodiments will be described in detail herein, examples of which are illustrated in the accompanying drawings. In the following description, when referring to the drawings, like numbers in different figures represent like or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all possible embodiments consistent with the present invention. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present invention, as detailed in the appended claims.

[0045] The embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0046] like Figure 1 As shown, an embodiment of the present invention provides a remote diagnosis system, including a patient remote conversation terminal 101, a physician remote conversation terminal 103, and a transmission code rate automatic allocation module 102;

[0047] The patient remote conversation terminal 101 is used to collect the video to be diagnosed input by the patient and request the network to transmit the video to be diagnosed to the physician remote conversation terminal;

[0048] It is worth noting that in addition to inputting the video to be diagnosed, the patient can also input other information (such as text information and voice information). However, other information occupies less network space and can be transmitted well even without optimization. Therefore, the embodiment of the present invention mainly allocates bit rate for the video to be diagnosed.

[0049] The automatic transmission bit rate allocation module 102 is used to adaptively adjust the bit rate of the patient's video to be diagnosed using a pre-deployed intelligent optimization algorithm to obtain a transmission bit rate corresponding to the video to be diagnosed, and transmit the video to the physician's remote conversation terminal according to the transmission bit rate;

[0050] In existing technologies, videos to be diagnosed are often transmitted at a fixed bitrate, resulting in low overall network utilization efficiency and network lag during peak periods, reducing the quality of remote diagnosis. Therefore, embodiments of the present invention employ an intelligent optimization algorithm for adaptive bitrate adjustment, improving overall network utilization efficiency while ensuring a consistent user experience for all.

[0051] The physician remote conversation terminal 103 is used to receive the video to be diagnosed and play the video to be diagnosed so that the physician can perform remote diagnosis on the patient.

[0052] Optionally, the physician remote conversation terminal 103 may also transmit data back to the patient remote conversation terminal 101 to achieve diagnosis.

[0053] The present invention provides a remote diagnosis system that adaptively adjusts the bit rate of a patient's video to be diagnosed by using a pre-deployed intelligent optimization algorithm to obtain a transmission bit rate corresponding to the video to be diagnosed, and transmits the video to be diagnosed to a physician's remote conversation terminal according to the transmission bit rate. The system can dynamically adjust the data transmission rate according to the overall network usage status to ensure the real-time and accuracy of remote diagnosis, and is suitable for multiple fields such as medical care, education, and enterprises.

[0054] In one possible implementation, the system further includes allowing text information to be transmitted between the patient's remote conversation terminal and the automatic transmission bit rate allocation module. It is worth noting that, in addition to the data exemplified in the embodiments of the present invention, other information can also be used as interactive information during the diagnosis process to achieve a more comprehensive diagnosis.

[0055] In one possible implementation, a pre-deployed intelligent optimization algorithm is used to adaptively adjust the bit rate of the patient's video to be diagnosed, to obtain a transmission bit rate corresponding to the video to be diagnosed, including:

[0056] For the video to be diagnosed, the video to be diagnosed is divided into multiple video blocks, and a target bit rate is randomly assigned to each video block. The target bit rates corresponding to the video blocks are formed into a vector to obtain the bit rate vector of a single video to be diagnosed;

[0057] Optionally, the general bit rate is set to multiple gears, not continuous interval data. Therefore, the embodiment of the present invention can map each gear to a sub-interval between (0, 1) (for example, the first gear corresponds to the sub-interval (0, 0.1), then the target bit rate randomly assigned to each video block should be the first gear). After the sub-intervals corresponding to all gears are merged, the interval (0, 1) is obtained, so that continuous optimization can be achieved. In the subsequent use and calculation process, the bit rate corresponding to each dimensional data is determined to achieve optimization. The bit rate vectors corresponding to multiple patients are combined into a vector to obtain the target bit rate vector, and multiple target bit rate vectors are repeatedly obtained;

[0058] For any target bit rate vector, a comprehensive video quality evaluation function is used to obtain the evaluation function value corresponding to each target bit rate vector, and the optimal target bit rate vector is obtained according to the evaluation function value corresponding to each target bit rate vector;

[0059] For any target bitrate vector, according to the optimal target bitrate vector, a multi-vector interaction strategy is used to perform multi-information fusion on the target bitrate vector to obtain the target bitrate vector after multi-information fusion.

[0060] For the target bit rate vector after multi-information fusion, a multi-information learning strategy is used to learn the better regional information of the target bit rate vector to obtain the target bit rate vector after the better regional information learning;

[0061] For the target bitrate vector after learning the better region information, an adaptive unknown region development strategy is used to search the unknown region for the target bitrate vector, and the target bitrate vector after the unknown region search is obtained;

[0062] For the target bit rate vector after the unknown area search, a global mutation development strategy is used to perform a global search on the target bit rate vector to obtain the target bit rate vector after the global search;

[0063] Determine whether the optimization end condition is met. If so, redetermine the optimal target bit rate vector based on the target bit rate vector after the global search, and determine the transmission bit rates corresponding to the videos to be diagnosed of multiple patients based on the redetermined optimal target bit rate vector. Otherwise, return to the step of obtaining the evaluation function value.

[0064] Existing intelligent optimization algorithms often suffer from the problem of falling into local optimality, resulting in suboptimal bitrate allocation and inability to maximize network resource utilization and ensure a better patient experience. Therefore, embodiments of the present invention provide an intelligent optimization algorithm that can optimize overall network utilization based on the value of an evaluation function.

[0065] In one possible implementation, for any target bitrate vector, an evaluation function value corresponding to each target bitrate vector is obtained using a comprehensive video quality evaluation function, and an optimal target bitrate vector is obtained based on the evaluation function value corresponding to each target bitrate vector, including:

[0066] For any target bitrate vector, the comprehensive video quality evaluation function is first used to obtain the comprehensive video quality corresponding to the bitrate vector of the video to be diagnosed for each patient:

[0067]

[0068] Among them, QoE represents the overall video quality, b l represents the bit rate corresponding to the lth video block in the video to be diagnosed, L represents the total number of video blocks corresponding to the video to be diagnosed, that is, the video to be diagnosed is divided into video blocks of fixed length during transmission, and L number of video blocks are obtained, T l REBUF represents the freeze time corresponding to the lth video block in the video to be diagnosed, bl-1 represents the bit rate corresponding to the l-1th video block in the video to be diagnosed, α represents the first weighting parameter, and β represents the second weighting parameter; the freeze time can be obtained through deep learning technology or other existing technologies;

[0069] Based on the weight parameter corresponding to each patient, the comprehensive video quality corresponding to the bitrate vectors corresponding to all patients is weighted and summed to obtain the evaluation function value corresponding to the target bitrate vector; among them, the weight parameter corresponding to each patient is preset to 1, and staff and doctors are allowed to modify it; for example, when patients visit different departments, different weights can be set, so that seriously ill patients can have a better medical experience.

[0070] According to the evaluation function value corresponding to each target bit rate vector, the target bit rate vector with the largest evaluation function value is determined as the optimal target bit rate vector.

[0071] In one possible implementation, for a target bitrate vector after multi-information fusion, multi-information fusion is performed on the target bitrate vector based on the optimal target bitrate vector and using a multi-vector interaction strategy to obtain the target bitrate vector after multi-information fusion, including:

[0072]

[0073] in, represents the i-th target bitrate vector in the t-th optimization process, Represents the target bitrate vector after learning the better region information K represents the total number of target bitrate vectors, rand1 represents a random number between (0,1), represents the optimal target bit rate vector, represents the first random target rate vector, represents the second random target rate vector, represents the second random target bit rate vector.

[0074] The multi-vector interaction strategy provided by the embodiment of the present invention can effectively learn the information of the optimal position and other random position information, thereby ensuring that the target bit rate vector moves towards the optimal position while performing local search, and then simultaneously learns the information of other vectors to ensure the diversification of vector information.

[0075] In one possible implementation, for the target bitrate vector after multi-information fusion, a multi-information learning strategy is used to perform better regional information learning on the target bitrate vector to obtain the target bitrate vector after better regional information learning, including:

[0076]

[0077] in, Represents the jth target bitrate vector in the tth optimization process The corresponding historical optimal value, j = 1, 2, ..., K, K represents the total number of target bitrate vectors, Indicates the first optimal target bit rate vector with the second largest evaluation function value, The second best target bit rate vector with the third largest evaluation function value is represented by represents the optimal target bit rate vector, a1 represents the first random vector, a2 represents the second random vector, and a3 represents the third random vector. The dimensions of the first random vector, the second random vector, and the third random vector are the same as the dimensions of the parameter individual, and each dimension is obtained by (2rand2-1)*2(iter max -t) / iter max Randomly generated, rand2 represents a random number between (0,1), iter max Indicates the preset maximum number of optimizations, b1 represents the first information learning coefficient, b2 represents the second information learning coefficient, and b3 represents the third information learning coefficient. The first information learning coefficient, the second information learning coefficient, and the third information learning coefficient are all generated by 2rand3, and rand3 represents a random number between (0,1). Represents the target bitrate vector after unknown area search

[0078] The multi-information learning strategy provided by the embodiment of the present invention can learn information of multiple better regions, and can effectively improve the search accuracy and search efficiency of the algorithm.

[0079] In one possible implementation, for the target bitrate vector after learning the better region information, an adaptive unknown region development strategy is used to perform an unknown region search on the target bitrate vector to obtain the target bitrate vector after the unknown region search, including:

[0080]

[0081] in, represents the target bitrate vector after learning the mth better region information in the tth optimization process, represents the target rate vector after learning the nth better region information in the tth optimization process, n=1,2,…,K, K represents the total number of target rate vectors, Represents the average bitrate vector corresponding to all target bitrate vectors, that is, each dimension parameter of the average bitrate vector is the average value of the parameters of all target bitrate vectors in that dimension. Represents the target bitrate vector after unknown area search λ represents the information fusion control factor, δ1 represents the first development coefficient and is set to 0.6; δ2 represents the second development coefficient and is set to 1.5; e represents the natural constant, d mn Represents the target bitrate vector and target bitrate vector The Euclidean distance between .

[0082] The adaptive unknown area development strategy provided by the embodiment of the present invention can search based on the center of all vectors, whether in the early stage or the late stage of the algorithm, and can effectively realize the search of unknown areas. In the late stage of the algorithm, the algorithm begins to converge, and the search accuracy of the algorithm can also be guaranteed.

[0083] In one possible implementation, for the target bitrate vector after the unknown region search, a global mutation development strategy is used to perform a global search on the target bitrate vector to obtain the target bitrate vector after the global search, including:

[0084]

[0085] in, represents the kth target bitrate vector in the tth optimization process, Represents the target bitrate vector after global search K represents the total number of target bitrate vectors, rand4 represents a random number between (0,1), and e represents a natural constant. represents the fourth random target bit rate vector, γ represents the adaptive global search control factor, Represents a random number between [-2.5γ, 2.5γ], iter max represents the preset maximum number of optimizations, θ represents the upper limit of the scaling factor, and ξ represents the adjustment parameter of the scaling factor.

[0086] The global variation development strategy provided by the embodiment of the present invention can effectively improve the global search capability of the algorithm, thereby ensuring that the algorithm will not fall into the local optimum, and as the number of optimization times t of the algorithm increases, the search accuracy is gradually increased, which can ensure the smooth convergence of the algorithm, and ultimately improve the rationality and efficiency of bit rate allocation, and improve the efficiency of network resource utilization.

[0087] Optionally, after each search, the target rate vector may be subjected to out-of-bounds processing to ensure that the rate allocation is effective.

[0088] In a possible implementation, determining whether the optimization end condition is met includes: determining whether the current optimization times reaches a preset maximum optimization times; if so, determining that the optimization end condition is met; otherwise, determining that the optimization end condition is not met.

[0089] In a possible implementation, determining transmission bit rates corresponding to videos to be diagnosed of multiple patients according to the re-determined optimal target bit rate vector includes:

[0090] According to the re-determined optimal target bit rate vector, the optimal target bit rate vector is re-split into bit rate vectors corresponding to a plurality of patients;

[0091] According to the bit rate vector corresponding to each patient, the transmission bit rate of each video block in the patient's to-be-diagnosed video is determined.

[0092] Those skilled in the art will readily appreciate other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses, or adaptations of the present invention that follow the general principles of the present invention and include common knowledge or customary techniques in the art not disclosed herein. It should be understood that the present invention is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and variations can be made without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

Claims

1. A remote diagnosis system, characterized in that: It includes a patient remote conversation terminal, a physician remote conversation terminal and a transmission code rate automatic allocation module; The patient remote conversation terminal is used to collect the video to be diagnosed input by the patient and request the network to transmit the video to be diagnosed to the physician remote conversation terminal; The automatic transmission bit rate allocation module is used to adaptively adjust the bit rate of the patient's video to be diagnosed using a pre-deployed intelligent optimization algorithm to obtain a transmission bit rate corresponding to the video to be diagnosed, and transmit the video to be diagnosed to the physician's remote conversation terminal according to the transmission bit rate; The physician remote conversation terminal is used to receive the video to be diagnosed and play the video to be diagnosed so that the physician can perform remote diagnosis on the patient; A pre-deployed intelligent optimization algorithm is used to adaptively adjust the bit rate of the patient's video to be diagnosed, and the corresponding transmission bit rate of the video to be diagnosed is obtained, including: For the video to be diagnosed, the video to be diagnosed is divided into multiple video blocks, and a target bit rate is randomly assigned to each video block. The target bit rates corresponding to the video blocks are formed into a vector to obtain the bit rate vector of a single video to be diagnosed; Combining the bit rate vectors corresponding to multiple patients into one vector to obtain a target bit rate vector, and repeatedly obtaining multiple target bit rate vectors; For any target bit rate vector, a comprehensive video quality evaluation function is used to obtain the evaluation function value corresponding to each target bit rate vector, and the optimal target bit rate vector is obtained according to the evaluation function value corresponding to each target bit rate vector; For any target bitrate vector, according to the optimal target bitrate vector, a multi-vector interaction strategy is used to perform multi-information fusion on the target bitrate vector to obtain the target bitrate vector after multi-information fusion. For the target bit rate vector after multi-information fusion, a multi-information learning strategy is used to learn the better regional information of the target bit rate vector to obtain the target bit rate vector after the better regional information learning; For the target bitrate vector after learning the better region information, an adaptive unknown region development strategy is used to search the unknown region for the target bitrate vector, and the target bitrate vector after the unknown region search is obtained; For the target bit rate vector after the unknown area search, a global mutation development strategy is used to perform a global search on the target bit rate vector to obtain the target bit rate vector after the global search; Determine whether the optimization end condition is met. If so, redetermine the optimal target bit rate vector based on the target bit rate vector after the global search, and determine the transmission bit rate corresponding to the videos to be diagnosed of multiple patients based on the redetermined optimal target bit rate vector. Otherwise, return to obtain the evaluation function value.

2. The remote diagnosis system according to claim 1, wherein: Also includes: Allows text information transmission between the patient's remote conversation terminal and the transmission code rate automatic allocation module.

3. The remote diagnosis system according to claim 1, wherein: For any target bitrate vector, the comprehensive video quality evaluation function is used to obtain the evaluation function value corresponding to each target bitrate vector, and the optimal target bitrate vector is obtained according to the evaluation function value corresponding to each target bitrate vector, including: For any target bitrate vector, the comprehensive video quality evaluation function is first used to obtain the comprehensive video quality corresponding to the bitrate vector of the video to be diagnosed for each patient: in, Indicates the overall video quality, Indicates the video to be diagnosed l The bit rate corresponding to the video blocks, L represents the total number of video blocks corresponding to the video to be diagnosed, that is, the video to be diagnosed is divided into video blocks according to a fixed length during the transmission process, and L number of video blocks are obtained. Indicates the video to be diagnosed l The freeze time corresponding to each video block, Indicates the video to be diagnosed l -The bit rate corresponding to 1 video block, represents the first weighting parameter, represents the second weighting parameter; Based on the weight parameter corresponding to each patient, the comprehensive video quality corresponding to the bitrate vectors of all patients is weighted and summed to obtain the evaluation function value corresponding to the target bitrate vector. The weight parameter corresponding to each patient is preset to 1 and can be modified by staff and physicians. According to the evaluation function value corresponding to each target bit rate vector, the target bit rate vector with the largest evaluation function value is determined as the optimal target bit rate vector.

4. The remote diagnosis system according to claim 3, characterized in that: For the target bitrate vector after multi-information fusion, the target bitrate vector is fused according to the optimal target bitrate vector and a multi-vector interaction strategy is adopted to obtain the target bitrate vector after multi-information fusion, including: in, Indicates the t During the optimization process i target bitrate vector, Represents the target bitrate vector after learning the better region information , i =1,2,…,K, K represents the total number of target rate vectors, Represents a random number between (0,1), represents the optimal target bit rate vector, represents the first random target rate vector, represents the second random target rate vector, represents the second random target bit rate vector.

5. The remote diagnosis system according to claim 4, characterized in that: For the target bitrate vector after multi-information fusion, a multi-information learning strategy is used to learn the optimal regional information of the target bitrate vector, and the target bitrate vector after the optimal regional information learning is obtained, including: in, Indicates the t During the optimization process j Target rate vector The corresponding historical optimal value is j =1,2,…,K, K represents the total number of target rate vectors, Indicates the first optimal target bit rate vector with the second largest evaluation function value, The second best target bit rate vector with the third largest evaluation function value is represented by represents the optimal target bit rate vector, represents the first random vector, represents the second random vector, Represents the third random vector, the dimensions of the first random vector, the second random vector and the third random vector are the same as the dimensions of the parameter individual, and each dimension is represented by Randomly generated, Represents a random number between (0,1), Indicates the preset maximum number of optimizations. represents the first information learning coefficient, represents the second information learning coefficient, Represents the third information learning coefficient, the first information learning coefficient, the second information learning coefficient and the third information learning coefficient are all passed through 2 Produced, and Represents a random number between (0,1), Represents the target bitrate vector after unknown area search .

6. The remote diagnosis system according to claim 5, characterized in that: For the target bitrate vector after learning the better region information, an adaptive unknown region development strategy is used to search the unknown region for the target bitrate vector, and the target bitrate vector after the unknown region search is obtained, including: in, Indicates the t During the optimization process m The target bit rate vector after learning the better region information, Indicates the t During the optimization process n The target bit rate vector after learning the better region information, n =1,2,…,K, K represents the total number of target rate vectors, Represents the average bitrate vector corresponding to all target bitrate vectors, that is, each dimension parameter of the average bitrate vector is the average value of all target bitrate vectors in that dimension parameter. Represents the target bitrate vector after unknown area search , represents the information fusion control factor, represents the first development coefficient and is set to 0.6; represents the second development coefficient and is set to 1.5; e represents the natural constant, Represents the target bitrate vector and target bitrate vector The Euclidean distance between .

7. The remote diagnosis system according to claim 6, characterized in that: For the target bitrate vector after the unknown area search, a global mutation development strategy is used to perform a global search on the target bitrate vector to obtain the target bitrate vector after the global search, including: in, Indicates the t During the optimization process k target bitrate vector, Represents the target bitrate vector after global search , k =1,2,…,K, K represents the total number of target rate vectors, Represents a random number between (0,1), represents a natural constant, represents the fourth random target rate vector, represents the adaptive global search control factor, means [-2.5 , 2.5 ], a random number between Indicates the preset maximum number of optimizations. Indicates the upper limit of the scaling factor, An adjustment parameter representing the scaling factor.

8. The remote diagnosis system according to claim 7, characterized in that: Determining whether the optimization end condition is met includes: determining whether the current optimization times reaches a preset maximum optimization times; if so, determining that the optimization end condition is met; otherwise, determining that the optimization end condition is not met.

9. The remote diagnosis system according to claim 1, wherein: Determining transmission bit rates corresponding to the videos to be diagnosed of the multiple patients according to the re-determined optimal target bit rate vector includes: According to the re-determined optimal target bit rate vector, the optimal target bit rate vector is re-split into bit rate vectors corresponding to a plurality of patients; According to the bit rate vector corresponding to each patient, the transmission bit rate of each video block in the patient's to-be-diagnosed video is determined.

Images