CN114530245A - Cloud edge coordination medical system based on edge calculation and federal learning - Google Patents
Cloud edge coordination medical system based on edge calculation and federal learning Download PDFInfo
- Publication number
- CN114530245A CN114530245A CN202210180445.7A CN202210180445A CN114530245A CN 114530245 A CN114530245 A CN 114530245A CN 202210180445 A CN202210180445 A CN 202210180445A CN 114530245 A CN114530245 A CN 114530245A
- Authority
- CN
- China
- Prior art keywords
- edge
- cloud
- model
- data
- medical
- 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.)
- Pending
Links
- 238000004364 calculation method Methods 0.000 title description 15
- 238000003745 diagnosis Methods 0.000 claims abstract description 65
- 238000012549 training Methods 0.000 claims abstract description 49
- 201000010099 disease Diseases 0.000 claims abstract description 33
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims abstract description 33
- 238000004891 communication Methods 0.000 claims abstract description 24
- 230000036541 health Effects 0.000 claims description 17
- 230000005540 biological transmission Effects 0.000 claims description 15
- 238000012544 monitoring process Methods 0.000 claims description 11
- 230000006870 function Effects 0.000 claims description 7
- 238000007781 pre-processing Methods 0.000 claims description 7
- 230000007246 mechanism Effects 0.000 claims description 6
- 238000012856 packing Methods 0.000 claims description 6
- 230000004931 aggregating effect Effects 0.000 claims 1
- 238000013480 data collection Methods 0.000 claims 1
- 238000013473 artificial intelligence Methods 0.000 abstract description 3
- 238000000034 method Methods 0.000 description 6
- 238000012545 processing Methods 0.000 description 4
- 239000003814 drug Substances 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 238000005065 mining Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000036772 blood pressure Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 230000003920 cognitive function Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000002360 explosive Substances 0.000 description 1
- 230000007786 learning performance Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 230000003068 static effect Effects 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
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/60—Protecting data
- G06F21/602—Providing cryptographic facilities or services
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/60—Protecting data
- G06F21/62—Protecting access to data via a platform, e.g. using keys or access control rules
- G06F21/6218—Protecting access to data via a platform, e.g. using keys or access control rules to a system of files or objects, e.g. local or distributed file system or database
- G06F21/6245—Protecting personal data, e.g. for financial or medical purposes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/60—Protecting data
- G06F21/64—Protecting data integrity, e.g. using checksums, certificates or signatures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5061—Partitioning or combining of resources
- G06F9/5072—Grid computing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/20—Ensemble learning
-
- 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
- G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
- G16H10/60—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Abstract
The invention discloses a cloud-side coordination medical system based on edge computing and federal learning, belongs to the technical field of artificial intelligence of the Internet of things, and aims to solve the technical problem of relieving communication pressure and data sharing conflict of cloud computing in the medical system. The cloud end is provided with various diagnosis models for diagnosing diseases; the edge terminal is used for acquiring medical data of a patient as local medical data, providing model training service based on federal learning by being matched with the cloud terminal, and providing disease diagnosis service; the model training service is: the edge terminals perform local model training on the diagnosis model distributed from the cloud terminal based on local medical data, the model obtained by training is subjected to gradient encryption on the last cloud terminal, the cloud terminal aggregates the model gradient and updates the diagnosis model, and the updated diagnosis model is distributed to each edge terminal; the disease diagnosis service is: and predicting diseases based on the diagnosis model issued from the cloud.
Description
Technical Field
The invention relates to the technical field of artificial intelligence of the Internet of things, in particular to a cloud-side coordination medical system based on edge computing and federal learning.
Background
At present, medical and health data platforms in China are different in scale and lack of unified standards and specifications, data present multi-source structures and cross-space-time characteristics, are different in quality and are distributed dispersedly and difficult to release. Medical and wellness data (e.g., electronic health records) contain a large amount of medical information. The analysis and mining of the Chinese medicine can be applied to the fields of disease prediction, medical auxiliary diagnosis, personalized information recommendation, clinical decision support, medicine mode mining and the like. On the one hand, the traditional way of storing and processing health data through cloud computing may result in overhead and load pressure for cloud network communication; on the other hand, because cloud computing uses personal data from multiple medical and healthcare facilities, it may lead to inter-department conflict of interests and personal profiles of patients. In order to mine the data values of different organizations and realize the sharing and integration of fragmented data, the data exchange based on the win-win organization should be implemented, and the integration of fragmented local data is promoted through the innovation of a combined technology research and development and management mechanism. Various mobile devices, intelligent wearable devices and medical health sensors continuously generate massive data, hundreds of millions of users use internet services, which shows the explosive growth trend of edge measurement data and realizes data-driven artificial intelligence. However, problems such as "data islands" still exist.
In order to solve the problem of cloud computing, it is considered to place preliminary data processing on an edge computing server in an edge computing mode and balance communication and computing performance.
Based on the above, how to alleviate the communication pressure of cloud computing and data sharing conflicts in the medical system is a technical problem to be solved.
Disclosure of Invention
Aiming at the defects, the invention provides a cloud-side coordination medical system based on edge computing and federal learning to solve the technical problems of communication pressure and data sharing conflict of cloud computing in the medical system.
The invention discloses a cloud-edge coordination medical system based on edge computing and federal learning, which comprises:
the cloud end is provided with various diagnosis models for diagnosing diseases;
the system comprises a plurality of edge terminals, a plurality of client terminals and a plurality of server terminals, wherein the edge terminals are distributed in each medical department, are used for acquiring medical data of a patient as local medical data, are matched with the cloud terminal, provide model training service based on federal learning and provide disease diagnosis service;
the model training service is as follows: for each type of diagnosis model, the edge terminal carries out local model training on the diagnosis model distributed from the cloud terminal based on local medical data, obtains model gradient encryption and last cloud terminal of the training, gathers the model gradient and updates the diagnosis model, and distributes the updated diagnosis model to each edge terminal;
the disease diagnosis service is: the edge terminal takes local medical data as input and carries out disease prediction based on a diagnosis model issued from the cloud so as to obtain a disease diagnosis result.
Preferably, the cloud stores various diagnosis models for diagnosing diseases through the cloud server, aggregates model gradients through the cloud server to update model parameters, and performs global training on the diagnosis models based on the updated model parameters to obtain updated diagnosis models.
Preferably, the edge end includes:
the data acquisition module is used for acquiring medical data of a patient, and the medical data comprises health monitoring data which changes relatively dynamically and health data which is relatively stable and unchangeable;
the data transmission module supports a wireless communication mode, interacts with the data acquisition module and is used for transmitting medical data;
the medical management module interacts with a user serving as an intelligent medical equipment user and is used for supporting the user to configure medical tasks, and the medical tasks comprise a model training task and a disease diagnosis task;
the edge server is interacted with the data transmission module and the medical management module respectively, is interacted with the cloud end in a wireless communication mode, is used for acquiring and storing local medical data, is used for providing model training service based on the model training task and matched with the cloud end, and is used for providing disease diagnosis service based on the disease diagnosis task and matched with the cloud end.
Preferably, the data acquisition module includes:
the sensor unit is used for acquiring health monitoring data of a patient in real time;
and the data acquisition interface is used for interacting with a third-party medical system and acquiring the monitoring data of the patient from the third-party medical system.
Preferably, the data transmission module has an edge calculation function, and is configured to perform data preprocessing on local medical data and transmit the preprocessed local medical data to an edge server, where the edge server is configured to perform local model training on the diagnosis model based on the preprocessed local medical data;
the data preprocessing is to format the acquired local medical data to obtain the local medical data with uniform format.
Preferably, the data transmission module is a base station with edge storage and edge calculation functions.
Preferably, for each type of diagnosis model, the edge server performs local model training on the diagnosis model distributed from the cloud based on local medical data to obtain a model gradient of the edge end, performs local search on the model gradient of the edge end to perform parameter tuning to obtain an optimal edge end model gradient, and packages, encrypts and signs the optimal edge end model gradient and uploads the optimal edge end model gradient to the cloud.
Preferably, the edge server is connected with the cloud end through a wireless communication module, the edge server and the wireless communication module are matched to serve as edge nodes, the edge nodes determine packing weight through executing a consistency mechanism, and the edge nodes with authority upload the edge nodes to the cloud end after gradient packing, encryption and signature of an edge end model;
and after the cloud terminal authenticates the authority of the edge node, receiving the edge end model gradient uploaded by the edge node after the edge node is packaged, encrypted and signed.
Preferably, the medical management module is configured with a task editor for supporting configuration of medical tasks by a user as a user of the smart medical device.
One of the advantages of the invention is as follows:
1. the data acquisition end provides edge calculation, local medical data acquired through the edge end are stored in the edge end, a diagnosis model is distributed to each edge end, local model training is carried out on the diagnosis model through the local medical data at the edge end, only the model gradient of the edge end needs to be aggregated at the cloud end, the diagnosis model is globally trained according to the updated model gradient to obtain an updated diagnosis model, the updated diagnosis model is distributed to each edge end, and the communication pressure and the calculation pressure of cloud calculation are relieved through the edge calculation;
2. the local medical data acquired by the data acquisition end is only stored in the edge end, and the cloud end does not need to be uploaded, so that the privacy of the medical data of the patient is ensured, and the conflict of data sharing among different departments can be avoided;
3. at the edge end, data acquired by the data acquisition module are transmitted to an edge server through the data transmission module, the edge server provides storage and edge calculation, the data transmission module adopts a base station with edge calculation and storage functions, local medical data are preprocessed through the base station and then transmitted to the edge server, and calculation pressure and communication pressure of the edge server are relieved.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed for the embodiments or the prior art descriptions will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on the drawings without creative efforts.
The invention is further described below with reference to the accompanying drawings.
Fig. 1 is a block diagram of a cloud-edge coordinated medical system based on edge computing and federal learning.
Detailed Description
The present invention is further described in the following with reference to the drawings and the specific embodiments so that those skilled in the art can better understand the present invention and can implement the present invention, but the embodiments are not to be construed as limiting the present invention, and the embodiments and the technical features of the embodiments can be combined with each other without conflict.
The embodiment of the invention provides a cloud-edge coordination medical system based on edge computing and federal learning, which is used for solving the technical problem of relieving the communication pressure of cloud computing and data sharing conflict in the medical system.
Example (b):
in order to solve the problem of cloud computing, the adoption of an edge computing mode to put preliminary data processing on an edge computing server is considered, and the method can be applied to the field of intelligent medical treatment. Under the mobile medical scene based on the mobile edge computing environment, the medical health Internet of things equipment can give complex tasks to edge server nodes, and communication and computing performance are balanced. A large number of medical health intelligence devices and edge nodes can adequately perceive and acquire rich and personalized medical health data for model training. The powerful cognitive function of moving edge computation is represented using federated learning techniques based on raw data rather than using centralized processing. The edge node may also obtain data including computational load, storage space, wireless traffic volume, task queue status, etc.
Meanwhile, the joint learning technology based on the mobile equipment forms a local model by utilizing the mobile equipment, so that inconvenience caused by movement of original data is avoided. Joint learning is a new machine learning technique. It forms a distributed model of large nodes (e.g., mobile devices) sharing a data set using a local model, and updates only the model without downloading the original training data. Based on this technique, privacy protection can be provided while improving learning performance. By designing learning algorithms to achieve better learning effects, some research theories propose an effective reputation incentive mechanism to encourage mobile devices with high reputation and high data quality to participate in model learning. These mechanisms can greatly improve the accuracy of federal learning. Compared with a data parallelization training mode, the federal learning does not need to have independent and identically distributed data samples, and the edge model training can still be carried out under the condition that the data quantity of each node is unbalanced. Non-standard data can also be processed in a very large scale wireless environment. Other calculations may be used in the federal learning module to reduce the number of rounds of communication model.
Based on the positive influence of the edge computing and the federal learning on the medical system, the cloud-edge coordination medical system based on the edge computing and the federal learning comprises a cloud end and edge ends, wherein various diagnosis models for diagnosing diseases are configured on the cloud end, the edge ends are distributed in all medical departments, are used for collecting medical data of patients as local medical data, are used for being matched with the cloud end, provide model training service based on the federal learning and provide disease diagnosis service.
In this embodiment, the model training service is: for each type of diagnosis model, the edge terminal conducts local model training on the diagnosis model distributed from the cloud terminal based on local medical data, obtains model gradients through training, encrypts the last cloud terminal, aggregates the model gradients through the cloud terminal, updates the diagnosis model, and distributes the updated diagnosis model to each edge terminal.
The disease diagnosis service is: the edge terminal takes local medical data as input and carries out disease prediction based on a diagnosis model issued from the cloud so as to obtain a disease diagnosis result.
As a specific implementation of the cloud, the cloud stores various diagnosis models for diagnosing diseases through the cloud server, aggregates model gradients through the cloud server to update model parameters, and performs global training on the diagnosis models based on the updated model parameters to obtain updated diagnosis models.
The edge end is specifically implemented and comprises a data acquisition module, a medical management module and an edge server, wherein the data acquisition module is used for acquiring medical data of a patient, and the medical data comprises health monitoring data with relative dynamic change and health data with relative stability and invariance; the data transmission module supports a wireless communication mode, interacts with the data acquisition module and is used for transmitting medical data; the medical management module interacts with a user serving as an intelligent medical equipment user and is used for supporting the user to configure medical tasks, wherein the medical tasks comprise a model training task and a disease diagnosis task; the edge server is respectively interacted with the data transmission module and the medical management module, is interacted with the cloud in a wireless communication mode, is used for acquiring and storing local medical data, is used for providing model training service based on the model training task and matched with the cloud, and is used for providing disease diagnosis service based on a disease diagnosis task and matched with the cloud.
As one implementation of the medical management module, the medical management module is configured with a task editor for supporting configuration of medical tasks by a user as a user of the intelligent medical device. The medical management module configures a model training task and a disease diagnosis task and sends the medical tasks to the edge server.
As a specific implementation of the data acquisition module, the data acquisition module includes a sensor unit and a data acquisition interface, the sensor unit is configured to acquire health monitoring data of a patient in real time, the health monitoring data related in this embodiment is dynamically changed data, such as heartbeat, pulse, blood pressure, and the like, the data acquisition interface is configured to interact with a third-party medical system and acquire monitoring data of the patient from the third-party medical system, and the monitoring data related in this embodiment is relatively static data, such as height, weight, and the like.
The data transmission module transmits the medical data acquired by the data acquisition module to the edge server as local medical data, and the edge server performs model training on the relevant diagnosis model machine distributed from the cloud by taking the local medical data as input. In view of the fact that medical data acquired by the data acquisition module are not uniform in format and the data may have problems of repeated data, unclear data and the like, when model training is performed on the local medical data, data preprocessing needs to be performed on the local medical data, the format of the local medical data is unified, data cleaning is performed on the data, and repeated and unclear invalid data are removed.
In order to reduce the pressure of data preprocessing of the edge server, the data transmission module has an edge calculation function and is used for performing data preprocessing on local medical data and transmitting the preprocessed local medical data to the edge server, so that the edge server is used for performing local model training on a diagnosis model based on the preprocessed local medical data.
In this embodiment, the data transmission module is a base station having edge storage and edge calculation functions.
The edge server conducts local model training on the diagnosis model distributed from the cloud based on local medical data to obtain a model gradient of an edge end, conducts local search on the model gradient of the edge end to conduct parameter optimization to obtain an optimal edge end model gradient, packages, encrypts and signs the optimal edge end model gradient, and uploads the optimal edge end model gradient to the cloud.
The edge server is connected with the cloud end through the wireless communication module, and the edge server and the wireless communication module are matched to serve as edge nodes. In view of the consideration of data security and data authority, the edge node determines the packing weight by executing a consistency mechanism, and the edge node with the authority carries out gradient packing, encryption and signature on an edge end model and uploads the edge end model to a cloud end; and after the authority of the edge node is authenticated by the cloud, receiving the edge model gradient uploaded by the edge node after the edge node is packaged, encrypted and signed.
In this embodiment, the cloud and the edge perform model training on the diagnostic model through horizontal federal learning. In the horizontal federal learning, the distributed model training based on samples can be regarded as the distributed model training, all data are distributed to different machines, each machine downloads the model from the server, then the model is trained by using local data, and then the parameters which need to be updated are returned to the server; the server aggregates the returned parameters on each machine, updates the model, and feeds back the latest model to each machine. In the process, the same and complete model is arranged under each machine, communication among the machines is not dependent, each machine can be predicted independently during prediction, and the process can be regarded as sample-based distributed model training. While federal learning uses local models to form distributed models of large nodes (e.g., mobile devices) of shared data sets, and only updates the models without downloading the original training data. In this embodiment, the cloud server returns the updated diagnosis models to each edge terminal serving as a participant, and each participating edge terminal downloads the latest model from the server; each edge terminal utilizes a local data training model to encrypt gradient and upload the gradient to a cloud server, and the cloud server aggregates gradient update model parameters uploaded by each edge terminal; the server returns the updated model to each participant; each participant updates its respective model.
Other calculations may be used in federal learning to reduce the number of model exchanges. One effective way to increase the computation is to add the number of local Stochastic Gradient Descent (SGD) trains per round. To further reduce the loss in communication, the ue with higher computational power may decide to perform more training batches.
While the invention has been shown and described in detail in the drawings and in the preferred embodiments, it is not intended to limit the invention to the embodiments disclosed, and it will be apparent to those skilled in the art that various combinations of the code auditing means in the various embodiments described above may be used to obtain further embodiments of the invention, which are also within the scope of the invention.
Claims (9)
1. A cloud-edge coordinated medical system based on edge computing and federal learning, comprising:
the cloud end is provided with various diagnosis models for diagnosing diseases;
the system comprises a plurality of edge terminals, a plurality of client terminals and a plurality of server terminals, wherein the edge terminals are distributed in each medical department, are used for acquiring medical data of a patient as local medical data, are matched with the cloud terminal, provide model training service based on federal learning and provide disease diagnosis service;
the model training service is as follows: for each type of diagnosis model, the edge terminal carries out local model training on the diagnosis model distributed from the cloud terminal based on local medical data, obtains model gradient encryption and last cloud terminal of the training, gathers the model gradient and updates the diagnosis model, and distributes the updated diagnosis model to each edge terminal;
the disease diagnosis service is: the edge terminal takes local medical data as input and carries out disease prediction based on a diagnosis model issued from the cloud so as to obtain a disease diagnosis result.
2. The cloud-edge coordinated medical system based on edge computing and federal learning of claim 1, wherein the cloud stores various diagnostic models for diagnosing diseases through a cloud server, and is used for aggregating model gradients through the cloud server to update model parameters, and performing global training on the diagnostic models based on the updated model parameters to obtain updated diagnostic models.
3. The cloud-edge coordinated medical system based on edge computing and federal learning of claim 1, wherein the edge terminal comprises:
the data acquisition module is used for acquiring medical data of a patient, and the medical data comprises health monitoring data which changes relatively dynamically and health data which is relatively stable and unchangeable;
the data transmission module supports a wireless communication mode, interacts with the data acquisition module and is used for transmitting medical data;
the medical management module interacts with a user serving as an intelligent medical equipment user and is used for supporting the user to configure medical tasks, and the medical tasks comprise a model training task and a disease diagnosis task;
the edge server is interacted with the data transmission module and the medical management module respectively, is interacted with the cloud end in a wireless communication mode, is used for acquiring and storing local medical data, is used for providing model training service based on the model training task and matched with the cloud end, and is used for providing disease diagnosis service based on the disease diagnosis task and matched with the cloud end.
4. The cloud-edge coordinated medical system based on edge computing and federal learning of claim 3, wherein the data collection module comprises:
the sensor unit is used for acquiring health monitoring data of a patient in real time;
and the data acquisition interface is used for interacting with a third-party medical system and acquiring the monitoring data of the patient from the third-party medical system.
5. The cloud-edge coordinated medical system based on edge computing and federal learning of claim 3, wherein the data transmission module has an edge computing function, and is configured to perform data preprocessing on local medical data and transmit the preprocessed local medical data to an edge server, and the edge server is configured to perform local model training on the diagnostic model based on the preprocessed local medical data;
the data preprocessing is to format the acquired local medical data to obtain the local medical data with uniform format.
6. The cloud-edge coordinated medical system based on edge computing and federal learning of claim 5, wherein the data transmission module is a base station with edge storage and edge computing functions.
7. The cloud-edge coordinated medical system based on edge computing and federal learning of claim 3, wherein for each type of diagnosis model, the edge server performs local model training on the diagnosis model distributed from the cloud based on local medical data to obtain a model gradient of the edge, performs local search on the model gradient of the edge to perform parameter tuning to obtain an optimal edge model gradient, and packages, encrypts and signs the optimal edge model gradient and uploads the optimal edge model gradient to the cloud.
8. The cloud-edge coordinated medical system based on edge computing and federal learning of claim 7, wherein the edge server is connected with the cloud end through a wireless communication module, the edge server and the wireless communication module are matched to serve as edge nodes, the edge nodes determine packing weights through executing a consistency mechanism, and the edge nodes with authority limit upload the edge nodes to the cloud end after gradient packing, encryption and signature of an edge end model;
and after the cloud terminal authenticates the authority of the edge node, receiving the edge end model gradient uploaded by the edge node after the edge node is packaged, encrypted and signed.
9. The cloud-edge coordinated medical system based on edge computing and federal learning as claimed in any one of claims 3-8, wherein the medical management module is configured with a task editor for supporting user configuration medical tasks as intelligent medical device users.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210180445.7A CN114530245A (en) | 2022-02-25 | 2022-02-25 | Cloud edge coordination medical system based on edge calculation and federal learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210180445.7A CN114530245A (en) | 2022-02-25 | 2022-02-25 | Cloud edge coordination medical system based on edge calculation and federal learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114530245A true CN114530245A (en) | 2022-05-24 |
Family
ID=81624718
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210180445.7A Pending CN114530245A (en) | 2022-02-25 | 2022-02-25 | Cloud edge coordination medical system based on edge calculation and federal learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114530245A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115148379A (en) * | 2022-06-06 | 2022-10-04 | 电子科技大学 | System and method for realizing intelligent health monitoring of solitary old people by utilizing edge calculation |
| CN115578369A (en) * | 2022-10-28 | 2023-01-06 | 佐健(上海)生物医疗科技有限公司 | Online cervical cell TCT slice detection method and system based on federal learning |
| CN116204793A (en) * | 2023-04-28 | 2023-06-02 | 北京航空航天大学 | Industrial Internet cloud edge model aggregation method based on federal learning |
| CN116596065A (en) * | 2023-07-12 | 2023-08-15 | 支付宝(杭州)信息技术有限公司 | Gradient calculation method and device, storage medium, product and electronic equipment |
| WO2024055979A1 (en) * | 2022-09-14 | 2024-03-21 | 抖音视界有限公司 | Model training method and apparatus, system, and storage medium |
-
2022
- 2022-02-25 CN CN202210180445.7A patent/CN114530245A/en active Pending
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115148379A (en) * | 2022-06-06 | 2022-10-04 | 电子科技大学 | System and method for realizing intelligent health monitoring of solitary old people by utilizing edge calculation |
| WO2024055979A1 (en) * | 2022-09-14 | 2024-03-21 | 抖音视界有限公司 | Model training method and apparatus, system, and storage medium |
| CN115578369A (en) * | 2022-10-28 | 2023-01-06 | 佐健(上海)生物医疗科技有限公司 | Online cervical cell TCT slice detection method and system based on federal learning |
| CN115578369B (en) * | 2022-10-28 | 2023-09-15 | 佐健(上海)生物医疗科技有限公司 | Online cervical cell TCT slice detection method and system based on federal learning |
| CN116204793A (en) * | 2023-04-28 | 2023-06-02 | 北京航空航天大学 | Industrial Internet cloud edge model aggregation method based on federal learning |
| CN116596065A (en) * | 2023-07-12 | 2023-08-15 | 支付宝(杭州)信息技术有限公司 | Gradient calculation method and device, storage medium, product and electronic equipment |
| CN116596065B (en) * | 2023-07-12 | 2023-11-28 | 支付宝(杭州)信息技术有限公司 | Gradient calculation method and device, storage medium, product and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114530245A (en) | Cloud edge coordination medical system based on edge calculation and federal learning | |
| Viswanathan et al. | Research challenges in computation, communication, and context awareness for ubiquitous healthcare | |
| CN106033476B (en) | A kind of increment type figure calculation method under distributed computation mode in cloud computing environment | |
| CN202351878U (en) | Auxiliary modularized cardio-pulmonary function monitoring system applying cloud computing technology | |
| do Nascimento et al. | Covid-19: A digital transformation approach to a public primary healthcare environment | |
| CN111383748A (en) | Medical integrated platform system based on elastic calculation and 5G technology | |
| CN104158914A (en) | Family-oriented remote urine-detecting and health-counseling service system and method | |
| CN109346192A (en) | The system and method that whole process is seen a doctor and serviced on line is provided based on internet | |
| KR101814448B1 (en) | mobile health care system and mobile health dashboard providing system based on components using the same | |
| CN110473623A (en) | Medical service system | |
| CN104753866A (en) | Data transmission method and data transmission device | |
| CN112289437B (en) | Diabetes adjuvant therapy cloud platform system based on edge computing framework | |
| CN107411733A (en) | A kind of heart rate detection and analysis system based on cloud platform | |
| US20220156235A1 (en) | Automatic generation of labeled data in iot systems | |
| CN112820366A (en) | Data processing method, device, system, equipment and storage medium | |
| Neubert et al. | Multi-sensor-fusion approach for a data-science-oriented preventive health management system: concept and development of a decentralized data collection approach for heterogeneous data sources | |
| CN108550399A (en) | A kind of fetal monitoring early warning system and method | |
| CN111429288A (en) | User portrait construction method and device, computer equipment and storage medium | |
| CN105550504B (en) | A kind of portable medical information system and medical information processing method | |
| García-Magariño et al. | Fog computing for assisting and tracking elder patients with neurodegenerative diseases | |
| CN101969471A (en) | Remote transmission self-networking household medical monitoring system | |
| Abdellatif et al. | I-Health: leveraging edge computing and blockchain for epidemic management | |
| Pysmennyi et al. | Edge computing in multi-scope service-oriented mobile healthcare systems | |
| CN106446592A (en) | Medical information sharing system | |
| Falat et al. | Telecommunication infrastructures for telemedicine in smart cities |
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 |