CN108831548B - Remote intelligent medical optimization method, device and system - Google Patents
Remote intelligent medical optimization method, device and system Download PDFInfo
- Publication number
- CN108831548B CN108831548B CN201810645413.3A CN201810645413A CN108831548B CN 108831548 B CN108831548 B CN 108831548B CN 201810645413 A CN201810645413 A CN 201810645413A CN 108831548 B CN108831548 B CN 108831548B
- Authority
- CN
- China
- Prior art keywords
- medical
- information data
- optimization
- remote intelligent
- scheme
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000005457 optimization Methods 0.000 title claims abstract description 138
- 238000000034 method Methods 0.000 title claims abstract description 25
- 238000011156 evaluation Methods 0.000 claims abstract description 43
- 238000013433 optimization analysis Methods 0.000 claims abstract description 27
- 238000013528 artificial neural network Methods 0.000 claims abstract description 13
- 238000012545 processing Methods 0.000 claims description 20
- 238000004458 analytical method Methods 0.000 claims description 8
- 230000005540 biological transmission Effects 0.000 claims description 8
- 239000013598 vector Substances 0.000 claims description 5
- 239000008155 medical solution Substances 0.000 description 21
- 239000000243 solution Substances 0.000 description 9
- 230000008569 process Effects 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000000126 substance Substances 0.000 description 2
- 238000004891 communication Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H40/00—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
- G16H40/60—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices
- G16H40/67—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices for remote operation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- Biomedical Technology (AREA)
- General Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Primary Health Care (AREA)
- Medical Informatics (AREA)
- General Health & Medical Sciences (AREA)
- Epidemiology (AREA)
- Public Health (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides an optimization method, device and system for remote intelligent medical treatment, and belongs to the technical field of intelligent medical treatment. The optimization method comprises the following steps: s1, receiving the remote intelligent medical optimization request, and determining a plurality of comprehensive medical schemes, wherein each comprehensive medical scheme comprises information data corresponding to each remote intelligent medical optimization request; s2, acquiring a historical comprehensive medical scheme; s3, performing optimization analysis according to the current comprehensive medical treatment schemes and historical comprehensive medical treatment schemes and a regular neural network optimization analysis strategy, and selecting an optimal comprehensive medical treatment scheme as a current recommendation scheme; s4, judging whether the first evaluation condition is met, if yes, outputting the current recommendation scheme; otherwise, executing S5; and S5, obtaining each optimized comprehensive medical scheme according to the positive-reward negative-penalty incentive optimization strategy, and then repeating S3 and S4 for iteration until the first evaluation condition is met or the iteration number meets a preset threshold value, and outputting the current recommended scheme.
Description
Technical Field
The invention belongs to the technical field of intelligent medical treatment, and particularly relates to a remote intelligent medical treatment optimization method, device and system.
Background
With the rapid development of the internet of things, the number of remote intelligent medical terminal devices is rapidly increased, and the data volume generated by the remote intelligent medical terminal devices reaches Zezi (ZB)
A rank. The problems of high delay, high flow cost, low recognition accuracy and the like are increasingly prominent in the process of carrying out centralized data processing by utilizing the conventional cloud computing system.
Disclosure of Invention
The invention aims to at least solve one of the technical problems in the prior art and provides an optimization method for remote intelligent medical treatment, which can reduce time delay and flow cost and improve accuracy.
The technical scheme adopted for solving the technical problem of the invention is an optimization method of remote intelligent medical treatment, which comprises the following steps:
s1, receiving remote intelligent medical optimization requests sent by the remote intelligent medical terminals, and determining a plurality of comprehensive medical schemes according to the remote intelligent medical optimization requests, wherein each comprehensive medical scheme comprises information data corresponding to each remote intelligent medical optimization request; wherein the information data comprises: at least one of latency, traffic cost, accuracy; the information data of at least one same remote intelligent medical optimization request in different comprehensive medical schemes are different;
s2, acquiring a historical comprehensive medical scheme, wherein the historical comprehensive medical scheme comprises historical information data;
s3, performing optimization analysis according to the information data and the historical information data in the current comprehensive medical treatment schemes and a preset regular neural network optimization analysis strategy, and selecting the optimal comprehensive medical treatment scheme from the comprehensive medical treatment schemes as a current recommendation scheme;
s4, judging whether a preset first evaluation condition is met or not according to the historical comprehensive medical treatment scheme and the current recommendation scheme, and outputting the current recommendation scheme when the preset first evaluation condition is judged to be met; when it is determined that the preset first evaluation condition is not satisfied, performing step S5;
s5, according to the current comprehensive medical schemes and the historical optimization schemes, optimizing the current comprehensive medical schemes according to a preset positive-reward negative-penalty incentive optimization strategy to obtain the optimized comprehensive medical schemes, then repeating the steps S3 and S4 to iterate until the preset first evaluation condition is judged to be met or the iteration times meet a preset threshold value, and outputting the current recommended scheme.
Preferably, the information data includes: time delay, flow cost, accuracy;
the information data in each comprehensive medical treatment scheme is stored in the form of three-dimensional information vectors as follows:
wherein k is the number of iterations,
information data with coordinates (i, j, t) in the integrated medical treatment plan M for the k-th iteration, and i ═1,2,…m，j＝1,2,…n，t＝1,2,…p，
Are respectively information data
Delay, traffic cost and accuracy of the kth iteration of (1).
Further preferably, the step S3 specifically includes:
determining current minimum information data in all comprehensive medical schemes according to the current information data of all the comprehensive medical schemes;
determining the maximum historical information data according to the historical information data;
and performing optimization analysis according to the current information data, the current minimum information data and the maximum historical information data in each comprehensive medical scheme and a regular neural network optimization analysis strategy, and selecting the optimal comprehensive medical scheme from the current comprehensive medical schemes as a recommendation scheme.
Further preferably, the regular neural network optimization analysis strategy specifically includes:
wherein M isminKFor minimum information data, MmaxGIs the maximum historical information data.
More preferably, the first evaluation condition is specifically:
wherein L isminKIs the minimum time delay, C, in each current information dataminKIs the minimum flow cost, A, in each current information datamaxKIs the maximum accuracy in the current pieces of information data.
Further preferably, the positive-reward negative-penalty incentive optimization strategy specifically includes:
wherein L isminGMinimum delay for history, CminGFor historical minimum traffic cost, AmaxGFor maximum accuracy of history, MminGFor historical minimum information data, MmaxKMaximum information data for the k-th iteration,
Information data with coordinates (i, j, t) in the integrated medical treatment plan M for the (k + 1) th iteration, and i is 1,2, … M, j is 1,2, … n, t is 1,2, … p,
is the average information data of the k iteration,
Is the average time delay of the kth iteration,
Average flow cost for the kth iteration,
Is the average accuracy of the kth iteration.
The technical scheme adopted for solving the technical problem of the invention is a remote intelligent medical optimization device, which comprises:
the system comprises a receiving unit, a processing unit and a processing unit, wherein the receiving unit is used for receiving remote intelligent medical optimization requests sent by remote intelligent medical terminals and determining a plurality of current comprehensive medical schemes according to the remote intelligent medical optimization requests, each comprehensive medical scheme comprises information data corresponding to the remote intelligent medical optimization requests, and at least one same remote intelligent medical optimization request in different comprehensive medical schemes has different information data; the information data includes: at least one of latency, traffic cost, accuracy;
the history acquisition unit is used for acquiring a history comprehensive medical scheme, and the history comprehensive medical scheme comprises history information data;
the analysis unit is used for carrying out optimization analysis according to the information data and the historical information data in the current comprehensive medical schemes and a preset regular neural network optimization analysis strategy and selecting the optimal comprehensive medical scheme from the comprehensive medical schemes as a current recommendation scheme;
the evaluation unit is used for judging whether a preset first evaluation condition is met or not according to the historical comprehensive medical scheme and the current recommendation scheme, and outputting the current recommendation scheme when the preset first evaluation condition is judged to be met;
and the optimization unit is used for optimizing the current comprehensive medical schemes according to the current comprehensive medical schemes and historical optimization schemes and a preset positive-reward negative-penalty incentive optimization strategy according to the preset positive-reward negative-penalty incentive optimization strategy when the preset first evaluation condition is judged to be not met, obtaining the optimized comprehensive medical schemes, sending the optimized comprehensive medical schemes to the analysis unit, and outputting the current recommended scheme until the preset first evaluation condition is judged to be met or the iteration times meet a preset threshold value.
The technical scheme adopted for solving the technical problem of the invention is an optimization system for remote intelligent medical treatment, which comprises the following steps:
the remote intelligent medical optimization device;
and the remote intelligent medical terminals are used for sending remote intelligent medical optimization requests to the receiving unit.
Preferably, the remote intelligent medical optimization system further includes:
the edge processing unit is used for receiving and processing part of remote intelligent medical optimization requests sent by the remote intelligent medical terminal;
the edge processing unit is arranged locally at the remote intelligent medical terminal, and the edge processing unit is different from a remote intelligent medical optimization request received by the remote intelligent medical optimization device.
Preferably, the remote intelligent medical optimization system further includes:
and the network transmission unit is used for transmitting the remote intelligent medical optimization request sent by the remote intelligent medical terminal.
According to the remote intelligent medical optimization method, a plurality of comprehensive medical schemes are determined based on remote intelligent medical optimization requests sent by remote intelligent medical terminals, an optimal comprehensive medical scheme is selected from the plurality of comprehensive medical schemes to be the most recommended scheme to be output according to information data in the comprehensive medical schemes, historical information data in the historical comprehensive medical schemes and the like, and when the output condition is not met, the optimal comprehensive medical scheme (recommended scheme) is determined by performing one-time or multiple-time iterative optimization analysis on the comprehensive medical schemes, so that the finally obtained comprehensive medical scheme has the effects of low time delay, low flow cost, high accuracy and the like.
Drawings
Fig. 1 is a flowchart of an optimization method of remote intelligent medical treatment according to embodiment 1 of the present invention;
fig. 2 is a block diagram of an optimizing apparatus for remote intelligent medical treatment according to embodiment 2 of the present invention;
fig. 3 is a block diagram of an optimization system of remote intelligent medical treatment according to embodiment 3 of the present invention.
Detailed Description
In order to make the technical solutions of the present invention better understood, the present invention will be described in further detail with reference to the accompanying drawings and specific embodiments.
Example 1:
as shown in fig. 1, the present embodiment provides an optimization method for remote intelligent medical treatment.
Specifically, the remote intelligent medical optimization request sent by the remote intelligent medical terminal is taken as an example for explanation.
The optimization method of the remote intelligent medical treatment comprises the following steps:
s1, receiving remote intelligent medical optimization requests sent by the remote intelligent medical terminals, and determining a plurality of current comprehensive medical schemes according to the remote intelligent medical optimization requests, wherein each comprehensive medical scheme comprises information data corresponding to the remote intelligent medical optimization requests, and at least one same remote intelligent medical optimization request in different comprehensive medical schemes has different information data.
The content of the remote intelligent medical optimization request specifically comprises medical communication, physical sign indexes, disease diagnosis and the like. In this step, the remote intelligent medical optimization request may be actively reported through the remote intelligent medical terminal, or may be obtained according to a regularly queried mechanism.
In step S1, a plurality of realizable schemes (integrated medical schemes) are determined by performing integrated processing according to the remote intelligent medical optimization requests sent by the remote intelligent medical terminals, and the integrated medical schemes all have request results for the remote intelligent medical optimization requests of the remote intelligent medical terminals so as to satisfy the remote intelligent medical optimization requests of the remote intelligent medical terminals. However, since the information data (e.g., time delay, flow cost, accuracy, etc.) corresponding to the same remote intelligent medical optimization request are different in different comprehensive medical solutions, that is, the time delay, flow cost, accuracy, etc. of each comprehensive medical solution are different, the whole of each comprehensive medical solution has its own advantages and disadvantages, for example, a certain comprehensive medical solution has a high time delay and a high accuracy, and another comprehensive medical solution has a low time delay, a low flow cost, and a low accuracy. In the embodiment, the overall optimal comprehensive medical treatment scheme is found out by performing one or more times of iterative optimization analysis on each comprehensive medical treatment scheme, so that the finally output comprehensive medical treatment scheme has the effects of low time delay, low flow cost, high accuracy and the like.
In this embodiment, the information data includes at least one of latency, traffic cost, and accuracy. The following specifically describes the optimization method of remote intelligent medical treatment by taking information data including time delay, flow cost and accuracy as an example.
Wherein, each information data in each comprehensive medical scheme is stored in a form of three-dimensional information vector as follows:
k is the number of iterations,
the information data with coordinates (i, j, t) in the integrated medical treatment plan M for the k-th iteration, i is 1,2, … M, j is 1,2, … n, t is 1,2, … p,
are respectively information data
Delay, traffic cost and accuracy of the kth iteration of (1).
And S2, acquiring a historical comprehensive medical scheme, wherein the historical comprehensive medical scheme comprises historical information data.
Correspondingly, historical information data in the historical integrated medical treatment scheme is also stored in the form of a three-dimensional information vector.
And S3, performing optimization analysis according to the information data and the historical information data in the current comprehensive medical treatment schemes and a preset regular neural network optimization analysis strategy, and selecting the optimal comprehensive medical treatment scheme from the comprehensive medical treatment schemes as the current recommended scheme.
Preferably, the method specifically comprises the following steps:
and S31, determining the current minimum information data in all the comprehensive medical schemes according to the current information data of each comprehensive medical scheme.
And S32, determining the maximum historical information data according to the historical information data.
And S33, performing optimization analysis according to the current information data, the current minimum information data and the maximum historical information data in each comprehensive medical scheme and a regular neural network optimization analysis strategy, and selecting the optimal comprehensive medical scheme from the current comprehensive medical schemes as a recommendation scheme.
Further preferably, the regular neural network optimization analysis strategy specifically includes:
wherein M isminKMinimum information data for the kth iteration, MmaxGIs the maximum historical information data. That is, in this step, it is preferable to calculate the Z value corresponding to each comprehensive medical treatment plan in the kth iteration according to all the information data in each comprehensive medical treatment plan, the minimum information data, and the maximum historical information data, where the comprehensive medical treatment plan corresponding to the minimum Z value is the optimal comprehensive medical treatment plan (i.e., recommended plan).
S4, judging whether a preset first evaluation condition is met or not according to the historical comprehensive medical treatment scheme and the current recommendation scheme, and outputting the current recommendation scheme when the preset first evaluation condition is judged to be met; when it is determined that the preset first evaluation condition is not satisfied, step S5 is performed.
In this step, the optimal integrated medical solution (i.e., the current recommended solution) selected in step S4 is evaluated according to the first evaluation condition to determine whether the recommended solution can be output as a final result.
Specifically, in this step, whether the first evaluation condition is satisfied is judged according to the minimum delay, the minimum flow cost and the maximum accuracy in all the information data in the current comprehensive medical treatment scheme and the delay, the flow cost and the accuracy in each information data in the current recommendation scheme, and if so, the current recommendation scheme is output; if not, step S5 is executed to reselect the new recommended solution.
Preferably, the first evaluation condition is specifically:
wherein L isminKIn the k-th iteration (i.e. current)) Minimum delay in each information data, CminKFor the minimum traffic cost, A, in each information data in the kth iteration (i.e. current)maxKIs the maximum accuracy in each information datum in the kth iteration (i.e. current).
S5, according to the current comprehensive medical schemes and the historical optimization schemes, optimizing the current comprehensive medical schemes according to a preset positive-reward negative-penalty incentive optimization strategy to obtain the optimized comprehensive medical schemes, then repeating the steps S3 and S4 to iterate until the preset first evaluation condition is judged to be met or the iteration times meet a preset threshold value, and outputting the current recommended scheme.
The first evaluation condition in step S4 is not satisfied, which indicates that the recommended protocol selected in step S3 does not meet the output requirement, and a new comprehensive medical protocol needs to be reselected as the recommended protocol. Since the recommended plan in step S4 is already the optimal plan selected from all the integrated medical plans in step S3, the other integrated medical plans cannot satisfy the output since the recommended plan in step S4 cannot satisfy the output. Therefore, in this step, current comprehensive medical solutions are optimized according to a preset positive-reward negative-penalty incentive optimization strategy to obtain optimized and updated comprehensive medical solutions, then the step returns to step S3, a recommended solution is selected from the optimized and updated comprehensive medical solutions again, and whether the reselected recommended solution meets the first evaluation condition is judged again through step S4, that is, whether the reselected recommended solution meets the first evaluation condition is judged, that is, whether the first evaluation condition meets the first evaluation condition is judged
Either for output or for optimization again. By iterating over and over in this manner, each integrated medical treatment plan is continuously optimized, and an optimal integrated medical treatment plan (i.e., recommended plan) is finally output. It should be noted that, when the number of iterations reaches a certain threshold d, the selected recommendation is considered to be infinitely close to satisfying the first evaluation condition, so that even if the recommendation still fails to satisfy the first evaluation condition, the recommendation can be output to avoid being absentIterative optimization is performed in a limited number of times, which causes waste of computing resources. The iteration number k needs to satisfy the condition that k is 1,2, …, d, wherein d is preferably 50.
Specifically, in this step, each information data in each comprehensive medical treatment plan can be optimized and updated according to a positive-reward negative-penalty incentive optimization strategy and by combining each information data in the historical comprehensive medical treatment plans, so that each comprehensive medical treatment plan can be optimized and updated. Preferably, the information data in any comprehensive medical scheme can be optimized and updated through a positive reward negative penalty excitation optimization strategy, the regular unsupervised learning factor of the kth iteration is determined through the time delay, the flow cost and the accuracy of the current information data and the average time delay, the average flow cost and the average accuracy of the kth iteration, and the reward concurrent penalty factor of each information data is determined through the average information data of the kth iteration, the maximum information data of the kth iteration, the historical minimum information data and each information data, so that each information data is optimized and updated through the regular unsupervised learning factor and the reward concurrent penalty factor. The positive-reward negative-penalty incentive optimization strategy may specifically be:
wherein the content of the first and second substances,
the information data with the coordinate (i, j, t) in the comprehensive medical treatment plan M for the (k + 1) th iteration, i.e. the information data with the coordinate (i, j, t) in the optimized and updated comprehensive medical treatment plan M,
Information data with coordinates (i, j, t) in the integrated medical treatment protocol M for the kth iteration, i.e. the integrated medical treatment protocol before optimization updatingInformation data having coordinates (i, j, t) in the medical treatment plan M,
Is a regular unsupervised learning factor,
For awarding and punishing factor LminGMinimum delay for history, CminGFor historical minimum traffic cost, AmaxGFor maximum accuracy of history, MminGFor historical minimum information data, MminKMinimum information data for the kth iteration, namely minimum information data in each comprehensive medical scheme before the optimization updating,
Is the average information data of the kth iteration, i.e. the average of all information data in all integrated medical solutions before the optimization update,
Is the average time delay of the kth iteration, i.e. the average time delay of all information data in all integrated medical solutions before the optimization update,
Average flow cost for the kth iteration, i.e. average flow cost for all information data in all integrated medical solutions before optimization update,
The average accuracy of the kth iteration, i.e. the average accuracy of all information data in all integrated medical solutions before the optimization update, is used.
To sum up, in the remote intelligent medical optimization method provided in this embodiment, a plurality of comprehensive medical solutions are determined based on a remote intelligent medical optimization request sent by each remote intelligent medical terminal, an optimal comprehensive medical solution is selected from the plurality of comprehensive medical solutions to be output as a recommended solution according to each information data in each comprehensive medical solution, historical information data in a historical comprehensive medical solution, and the like, and when an output condition is not satisfied, the optimal comprehensive medical solution (recommended solution) is determined by performing one or more iterative optimization analyses on each comprehensive medical solution, so that the finally obtained comprehensive medical solution has the effects of low time delay, low flow cost, high accuracy, and the like.
Example 2:
as shown in fig. 2, the present embodiment provides an optimizing apparatus for remote intelligent medical treatment, which can process a request sent by a remote intelligent medical treatment terminal according to the optimizing method for remote intelligent medical treatment provided in embodiment 1. This remote intelligent medical optimization device includes: the device comprises a receiving unit, a history acquisition unit, an analysis unit, an evaluation unit and an optimization unit.
The receiving unit is used for receiving remote intelligent medical optimization requests sent by the remote intelligent medical terminals and determining a plurality of current comprehensive medical schemes according to the remote intelligent medical optimization requests, each comprehensive medical scheme comprises information data corresponding to the remote intelligent medical optimization requests, and at least one same remote intelligent medical optimization request in different comprehensive medical schemes is different in information data; the information data includes: at least one of latency, traffic cost, accuracy.
Preferably, the information data includes: time delay, flow cost and accuracy. The history acquisition unit is used for acquiring a history comprehensive medical scheme, and the history comprehensive medical scheme comprises history information data.
The analysis unit is used for carrying out optimization analysis according to the information data and the historical information data in the current comprehensive medical schemes and a preset regular neural network optimization analysis strategy, and selecting the optimal comprehensive medical scheme from the comprehensive medical schemes as the current recommendation scheme.
The evaluation unit is used for judging whether a preset first evaluation condition is met or not according to the historical comprehensive medical scheme and the current recommendation scheme, and outputting the current recommendation scheme when the preset first evaluation condition is judged to be met.
When the optimization unit judges that the preset first evaluation condition is not met, the optimization unit optimizes each current comprehensive medical scheme according to each current comprehensive medical scheme and the historical optimization scheme and according to a preset positive-reward negative-penalty incentive optimization strategy to obtain each optimized comprehensive medical scheme, and sends the optimized comprehensive medical schemes to the analysis unit until the preset first evaluation condition is judged to be met or the iteration times meet a preset threshold value, and the current recommended scheme is output.
Preferably, the embodiment further includes an infrastructure unit, configured to provide support of computing resources, IT virtual resources, IT physical resources, and the like to the receiving unit, the history obtaining unit, the analyzing unit, the evaluating unit, and the optimizing unit, so as to ensure operation of the remote intelligent medical optimization apparatus.
The remote intelligent medical optimization device provided by this embodiment determines a plurality of comprehensive medical schemes based on a remote intelligent medical optimization request sent by each remote intelligent medical terminal, selects an optimal comprehensive medical scheme to be the most recommended scheme for output according to each information data in each comprehensive medical scheme, historical information data in the historical comprehensive medical schemes and the like, and determines the optimal comprehensive medical scheme (recommended scheme) by performing one or more times of iterative optimization analysis on each comprehensive medical scheme when the output condition is not satisfied, so that the finally obtained comprehensive medical scheme has the effects of low time delay, low flow cost, high accuracy and the like.
Example 3:
as shown in fig. 3, the present embodiment provides an optimization system for remote intelligent medical treatment, including: the remote intelligent medical optimization device and the remote intelligent medical terminals provided in embodiment 2. The remote intelligent medical terminal can send a remote intelligent medical optimization request to the remote intelligent medical optimization device.
Preferably, the remote intelligent medical optimization system further includes: and the network transmission unit is used for transmitting the remote intelligent medical optimization request sent by the remote intelligent medical terminal to the remote intelligent medical optimization device through the network. The network transmission unit may specifically include: operator base stations, satellites, etc.
Further, the remote intelligent medical optimization system further comprises: and the gateway unit can comprise a plurality of medical gateways and is used for ensuring the safety of network transmission in the remote intelligent medical optimization system.
Preferably, the remote intelligent medical optimization system further comprises an edge processing unit, configured to receive and process a part of the remote intelligent medical optimization request sent by the remote intelligent medical terminal. The edge processing unit can be arranged locally at the remote intelligent medical terminal, has certain remote intelligent medical optimization request processing capacity, and can directly process part of remote intelligent medical optimization requests sent by the remote intelligent medical terminal. The remote intelligent medical optimization requests received by the edge processing unit and the remote intelligent medical optimization device are different, that is, the remote intelligent medical terminal can locally process part of the remote intelligent medical optimization requests, and only the rest remote intelligent medical optimization requests which cannot be processed by the edge processing unit need to be sent to the remote intelligent medical optimization device for processing, so that the problems of high time delay and high flow cost caused in the transmission process of the remote intelligent medical optimization requests are at least partially solved.
In the remote intelligent medical optimization system provided by this embodiment, after the remote intelligent medical terminal sends the remote intelligent medical optimization request, the remote intelligent medical optimization device sends the remote intelligent medical optimization request to the remote intelligent medical optimization device through the network transmission unit, and the remote intelligent medical optimization device finally determines an optimal comprehensive medical scheme based on the remote intelligent medical optimization request sent by each remote intelligent medical terminal, and returns the optimal comprehensive medical scheme to each remote intelligent medical terminal through the network transmission unit. Wherein, the remote intelligent medical optimization device determines a plurality of comprehensive medical schemes based on remote intelligent medical optimization requests sent by various remote intelligent medical terminals in the process of determining the optimal comprehensive medical scheme, according to the information data in the comprehensive medical schemes, the historical information data in the historical comprehensive medical schemes and the like, selecting an optimal comprehensive medical scheme from the comprehensive medical schemes as a recommended scheme to be output, when the output condition is not met, the optimal comprehensive medical treatment scheme (recommended scheme) is determined by performing one or more times of iterative optimization analysis on each comprehensive medical treatment scheme, so that the finally obtained comprehensive medical treatment scheme has the effects of low time delay, low flow cost, high accuracy and the like, and further, the optimization system of the remote intelligent medical treatment can realize low time delay, low flow cost and high accuracy.
It will be understood that the above embodiments are merely exemplary embodiments taken to illustrate the principles of the present invention, which is not limited thereto. It will be apparent to those skilled in the art that various modifications and improvements can be made without departing from the spirit and substance of the invention, and these modifications and improvements are also considered to be within the scope of the invention.
Claims (8)
1. A method for optimizing remote intelligent medical treatment, comprising:
s1, receiving remote intelligent medical optimization requests sent by the remote intelligent medical terminals, and determining a plurality of comprehensive medical schemes according to the remote intelligent medical optimization requests, wherein each comprehensive medical scheme comprises information data corresponding to each remote intelligent medical optimization request; wherein the information data comprises: in time delay, flow cost and accuracy, all information data in each comprehensive medical scheme is stored in a three-dimensional information vector form as follows:
wherein k is the number of iterations,
the information data with coordinates (i, j, t) in the integrated medical treatment plan M for the k-th iteration, i is 1,2, … M, j is 1,2, … n, t is 1,2, … p,
are respectively information data
Delay, flow cost and accuracy of the kth iteration of (1); the information data of at least one same remote intelligent medical optimization request in different comprehensive medical schemes are different;
s2, acquiring a historical comprehensive medical scheme, wherein the historical comprehensive medical scheme comprises historical information data;
s3, performing optimization analysis according to the information data and the historical information data in the current comprehensive medical treatment schemes and a preset regular neural network optimization analysis strategy, and selecting the optimal comprehensive medical treatment scheme from the comprehensive medical treatment schemes as a current recommendation scheme;
s4, judging whether a preset first evaluation condition is met or not according to the historical comprehensive medical treatment scheme and the current recommendation scheme, and outputting the current recommendation scheme when the preset first evaluation condition is judged to be met; when it is determined that the preset first evaluation condition is not satisfied, performing step S5; wherein the first evaluation condition is specifically:
wherein L isminKIs the minimum time delay, C, in each current information dataminKIs the minimum flow cost, A, in each current information datamaxKThe maximum accuracy in each current information data;
s5, according to the current comprehensive medical schemes and the historical optimization schemes, optimizing the current comprehensive medical schemes according to a preset positive-reward negative-penalty incentive optimization strategy to obtain the optimized comprehensive medical schemes, then repeating the steps S3 and S4 to iterate until the preset first evaluation condition is judged to be met or the iteration times meet a preset threshold value, and outputting the current recommended scheme.
2. The remote intelligent medical optimization method according to claim 1, wherein the step S3 specifically includes:
determining current minimum information data in all comprehensive medical schemes according to the current information data of all the comprehensive medical schemes;
determining the maximum historical information data according to the historical information data;
and performing optimization analysis according to the current information data, the current minimum information data and the maximum historical information data in each comprehensive medical scheme and a regular neural network optimization analysis strategy, and selecting the optimal comprehensive medical scheme from the current comprehensive medical schemes as a recommendation scheme.
3. The optimizing method for remote intelligent medical treatment according to claim 2, wherein the regular neural network optimization analysis strategy is specifically:
wherein M isminKFor minimum information data, MmaxGIs the maximum historical information data.
4. The remote intelligent medical optimization method according to claim 1,
the positive-reward negative-penalty incentive optimization strategy specifically comprises the following steps:
wherein L isminGMinimum delay for history, CminGFor historical minimum flowThis, AmaxGFor maximum accuracy of history, MminGFor historical minimum information data, MmaxKMaximum information data for the k-th iteration,
Information data having coordinates (i, j, t) in the integrated medical treatment plan M for the (k + 1) -th iteration, and i is 1,2, L M, j is 1,2, L n, t is 1,2, L p,
is the average information data of the k iteration,
Is the average time delay of the kth iteration,
Average flow cost for the kth iteration,
Is the average accuracy of the kth iteration.
5. An optimizing device for remote intelligent medical treatment, comprising:
the system comprises a receiving unit, a processing unit and a processing unit, wherein the receiving unit is used for receiving remote intelligent medical optimization requests sent by remote intelligent medical terminals and determining a plurality of current comprehensive medical schemes according to the remote intelligent medical optimization requests, each comprehensive medical scheme comprises information data corresponding to the remote intelligent medical optimization requests, and at least one same remote intelligent medical optimization request in different comprehensive medical schemes has different information data; the information data includes: time delay, flow cost and accuracy, and all information data in each comprehensive medical scheme is stored in a three-dimensional information vector form as follows:
wherein k is the number of iterationsThe number of the first and second groups is,
the information data with coordinates (i, j, t) in the integrated medical treatment plan M for the k-th iteration, i is 1,2, … M, j is 1,2, … n, t is 1,2, … p,
are respectively information data
Delay, flow cost and accuracy of the kth iteration of (1);
the history acquisition unit is used for acquiring a history comprehensive medical scheme, and the history comprehensive medical scheme comprises history information data;
the analysis unit is used for carrying out optimization analysis according to the information data and the historical information data in the current comprehensive medical schemes and a preset regular neural network optimization analysis strategy and selecting the optimal comprehensive medical scheme from the comprehensive medical schemes as a current recommendation scheme;
the evaluation unit is used for judging whether a preset first evaluation condition is met or not according to the historical comprehensive medical scheme and the current recommendation scheme, and outputting the current recommendation scheme when the preset first evaluation condition is judged to be met; wherein the first evaluation condition is specifically:
wherein L isminKIs the minimum time delay, C, in each current information dataminKIs the minimum flow cost, A, in each current information datamaxKThe maximum accuracy in each current information data;
and the optimization unit is used for optimizing the current comprehensive medical schemes according to the current comprehensive medical schemes and historical optimization schemes and a preset positive-reward negative-penalty incentive optimization strategy according to the preset positive-reward negative-penalty incentive optimization strategy when the preset first evaluation condition is judged to be not met, obtaining the optimized comprehensive medical schemes, sending the optimized comprehensive medical schemes to the analysis unit, and outputting the current recommended scheme until the preset first evaluation condition is judged to be met or the iteration times meet a preset threshold value.
6. An optimization system for remote intelligent medical treatment, comprising:
the telesmart medical optimization device of claim 5;
and the remote intelligent medical terminals are used for sending remote intelligent medical optimization requests to the receiving unit.
7. The remote intelligent medical optimization system according to claim 6, further comprising:
the edge processing unit is used for receiving and processing part of remote intelligent medical optimization requests sent by the remote intelligent medical terminal;
the edge processing unit is arranged locally at the remote intelligent medical terminal, and the edge processing unit is different from a remote intelligent medical optimization request received by the remote intelligent medical optimization device.
8. The remote intelligent medical optimization system according to claim 6, further comprising:
and the network transmission unit is used for transmitting the remote intelligent medical optimization request sent by the remote intelligent medical terminal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810645413.3A CN108831548B (en) | 2018-06-21 | 2018-06-21 | Remote intelligent medical optimization method, device and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810645413.3A CN108831548B (en) | 2018-06-21 | 2018-06-21 | Remote intelligent medical optimization method, device and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108831548A CN108831548A (en) | 2018-11-16 |
| CN108831548B true CN108831548B (en) | 2021-07-06 |
Family
ID=64141968
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810645413.3A Active CN108831548B (en) | 2018-06-21 | 2018-06-21 | Remote intelligent medical optimization method, device and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108831548B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110176311A (en) * | 2019-05-17 | 2019-08-27 | 北京印刷学院 | A kind of automatic medical proposal recommending method and system based on confrontation neural network |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104796734A (en) * | 2015-03-20 | 2015-07-22 | 四川长虹电器股份有限公司 | Real-time interactive smart television program combined recommendation system and method |
| CN105279288A (en) * | 2015-12-04 | 2016-01-27 | 深圳大学 | Online content recommending method based on deep neural network |
| CN108023840A (en) * | 2017-12-12 | 2018-05-11 | 中国联合网络通信集团有限公司 | OVS network traffics accelerate optimization method and OVS network traffics to accelerate optimization system |
-
2018
- 2018-06-21 CN CN201810645413.3A patent/CN108831548B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104796734A (en) * | 2015-03-20 | 2015-07-22 | 四川长虹电器股份有限公司 | Real-time interactive smart television program combined recommendation system and method |
| CN105279288A (en) * | 2015-12-04 | 2016-01-27 | 深圳大学 | Online content recommending method based on deep neural network |
| CN108023840A (en) * | 2017-12-12 | 2018-05-11 | 中国联合网络通信集团有限公司 | OVS network traffics accelerate optimization method and OVS network traffics to accelerate optimization system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108831548A (en) | 2018-11-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108243216B (en) | Data processing method, end-side device, cloud-side device and end cloud cooperative system | |
| US20190279088A1 (en) | Training method, apparatus, chip, and system for neural network model | |
| CN113038386B (en) | Learning model-based device positioning | |
| CN111079780A (en) | Training method of space map convolution network, electronic device and storage medium | |
| CN114065863B (en) | Federal learning method, apparatus, system, electronic device and storage medium | |
| CN115293358A (en) | Internet of things-oriented clustered federal multi-task learning method and device | |
| CN114595049A (en) | Cloud-edge cooperative task scheduling method and device | |
| CN115471805B (en) | Point cloud processing and deep learning model training method and device and automatic driving vehicle | |
| CN112446462B (en) | Method and device for generating target neural network model | |
| CN108830422B (en) | Intelligent driving optimization method, device and system | |
| CN108831548B (en) | Remote intelligent medical optimization method, device and system | |
| WO2021111456A1 (en) | Moderator for identifying deficient nodes in federated learning | |
| CN108770014B (en) | Calculation evaluation method, system and device of network server and readable storage medium | |
| WO2022252694A1 (en) | Neural network optimization method and apparatus | |
| Kushwaha et al. | Optimal device selection in federated learning for resource-constrained edge networks | |
| CN115983362A (en) | Quantization method, recommendation method and device | |
| CN113516163B (en) | Vehicle classification model compression method, device and storage medium based on network pruning | |
| CN108900594B (en) | Optimization method, device and system of mobile video | |
| CN108848514B (en) | Data communication optimization method and data communication optimizer | |
| CN113506638A (en) | Online consultation theme and machine learning based network training method and system | |
| CN116628181B (en) | User control preference sensing method and system based on Internet of things | |
| CN114531669B (en) | Task unloading method and system based on vehicle edge calculation | |
| CN116936048B (en) | Federal learning hospital selection method, device and storage medium for heterogeneous medical information | |
| CN113839794B (en) | Data processing method, device, equipment and storage medium | |
| CN111050299B (en) | Intelligent vehicle data mutual inspection method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |