Medical data processing method, controller, computing cluster system and device

Abstract

The invention relates to the technical field of intelligent medical treatment, and discloses a medical data processing method, a controller and a computing cluster system and device. The method comprises the steps that medical data and application program specifying information are received, a specified target application program is determined based on the application program specifying information, and the target application program can process the medical data based on computing resources; determining similar application programs consistent with the type of the target application program; based on the resource usage data of the computing resources, sending the medical data to similar application programs, so that the similar application programs process the medical data based on the computing resources to obtain similar processing results; and obtaining feedback data based on similar processing results, and sending the feedback data. Through the method, feedback data based on similar application programs can be sent, and the accuracy of a diagnosis result is improved.

Description

Technical Field

[0001] The present invention relates to the field of smart medical technology, and in particular to a medical data processing method, controller, computing cluster system and equipment. Background Art

[0002] As artificial intelligence continues to develop in the medical field, there may be multiple manufacturers in the market developing auxiliary diagnosis applications for a certain disease. However, when different application products analyze and process medical images to diagnose a certain disease, there may be different diagnostic results for the same set of medical images. For users, their budget is limited. If users only choose one application product for auxiliary diagnosis, there may be missed diagnoses, resulting in a bad user experience. Summary of the invention

[0003] The embodiments of the present application are intended to solve at least one of the technical problems in the related art to a certain extent. To this end, the embodiments of the present application provide a medical data processing method, a controller, a computing cluster system and a device.

[0004] An embodiment of the present application provides a medical data processing method, the method comprising: receiving medical data and application specified information, and determining a specified target application based on the application specified information, wherein the target application is capable of processing the medical data based on computing resources; determining a similar application of the same type as the target application; sending the medical data to the similar application based on resource usage data of the computing resources, so that the similar application processes the medical data based on the computing resources to obtain similar processing results; obtaining feedback data based on the similar processing results, and sending the feedback data.

[0005] Through the embodiments of the present application, users can receive the diagnosis results of similar applications for auxiliary diagnosis of the medical data uploaded by them. Since the diagnosis results received by users are not obtained based on a single application, missed diagnosis can be effectively avoided and the accuracy of auxiliary diagnosis can be improved. When sending medical data to similar applications for auxiliary diagnosis, the computing resources required by the target application and similar applications when processing medical data are taken into account. Therefore, for users, whether auxiliary diagnosis is based on a single application or based on similar applications, the time spent on the two does not change significantly, and users are almost unaware of the entire auxiliary diagnosis process. The above method can not only provide users with the diagnosis results of similar applications, but also effectively reduce the missed diagnosis rate in auxiliary diagnosis and improve the accuracy of auxiliary diagnosis. It can also improve the utilization efficiency of computing resources in the computing cluster, so that computing resources are optimized in terms of cost-effectiveness.

[0006] In some embodiments, the computing resources include a first computing resource and a second computing resource, and the target application processes the medical data based on the first computing resource; based on the resource usage data of the computing resources, the medical data is sent to similar applications, including: when the resource usage data indicates that the second computing resource can support similar applications to process the medical data, the medical data is sent to similar applications.

[0007] Through the embodiments of the present application, on the one hand, by ensuring that the target application has sufficient computing resources for calculation, it is ensured that the overall time required for processing medical data does not change significantly, avoiding users from experiencing long waits due to insufficient computing resources when selecting the target application for auxiliary diagnosis, thereby improving the user experience. On the other hand, on the basis of ensuring the normal operation of the target application, the remaining computing resources are fully utilized. It avoids the idle waste of computing resources, especially when the total amount of computing cluster resources of the cloud platform is fixed. This reasonable resource allocation method can maximize the efficiency of the entire computing resource system and improve the economic benefits of resources. On the other hand, when the resource usage data of the computing resources indicates that it can support similar applications for calculation, more similar applications with different analysis methods and algorithms from the target application may be introduced into the auxiliary diagnosis process. Different similar applications may have their own advantages, which are helpful to interpret medical data from multiple dimensions and further improve the accuracy of auxiliary diagnosis.

[0008] In some embodiments, when the resource usage data indicates that the usage rate of the second computing resource is lower than a preset threshold, it means that the second computing resource is able to support similar applications to process medical data.

[0009] Through the embodiments of the present application, by setting a preset threshold for the usage rate of computing resources, the computing resources required by similar applications when processing medical data can be effectively taken into account, ensuring that the overall time required to process medical data does not change significantly, avoiding users from spending more time waiting for diagnosis results, and improving user experience.

[0010] In some embodiments, the method further includes: upon receiving newly added medical data and newly added application designation information, determining based on the resource usage data to cancel the processing of the medical data by the same type of application based on the second computing resource.

[0011] Through the embodiments of the present application, whether receiving the medical data and application-specified information submitted by the user first, or receiving the newly added medical data and newly added application-specified information submitted by the user later, the cloud platform gives priority to the computing resources required by the user-specified target application when processing medical data. If the computing resources of the current cloud platform's computing cluster are insufficient to support the user-specified target application to process medical data, the second computing resources are released for the user-specified target application to use by canceling the processing of medical data by similar applications based on the second computing resources. This ensures that the target application specified by the user at different time points has sufficient computing resources for calculation, and that the time spent by the target application in processing medical data does not change significantly, thereby avoiding users from experiencing long waiting times when performing auxiliary diagnosis and improving user experience.

[0012] In some embodiments, the new application designation information is used to specify a new target application, and the computing resources also include a third computing resource; based on the resource usage data, it is determined to cancel the processing of medical data by similar applications based on the second computing resource, including: when the resource usage data indicates that the third computing resource is insufficient to support the new target application to process the new medical data, canceling the processing of medical data by similar applications based on the second computing resource so as to release the second computing resource for the new target application to process the new medical data.

[0013] Through the embodiments of the present application, by canceling the processing of medical data by similar applications based on the second computing resources, the second computing resources are released for use by the target application specified by the user. It is ensured that the newly added target application specified by the user has sufficient computing resources for calculation, and the time spent by the newly added target application in processing medical data does not change significantly, avoiding users from experiencing long waiting times when performing auxiliary diagnosis, and improving user experience.

[0014] In some embodiments, the same type of applications includes a first same type of application and a second same type of application, the second computing resource includes a first sub-computing resource and a second sub-computing resource, the first same type of application processes medical data based on the first sub-computing resource, and the second same type of application processes medical data based on the second sub-computing resource; canceling the same type of application processing medical data based on the second computing resource includes: canceling the first same type of application processing medical data based on the first sub-computing resource; after canceling the first same type of application processing medical data based on the first sub-computing resource, if the resource usage data indicates that the first sub-computing resource and the third computing resource can support the newly added target application to process the newly added medical data, then keeping the second same type of application continuing to process medical data based on the second sub-computing resource.

[0015] Through the embodiments of the present application, the current computing resource usage of the computing cluster can be accurately judged through the resource usage data of the computing resources. This method can not only give priority to the target applications specified by the user at different time nodes for auxiliary diagnosis, but also effectively reduce the waiting time of the user in the auxiliary diagnosis process, thereby significantly improving the user experience. In addition, it is also possible to decide to cancel all or part of the same type of applications based on the resource usage data, and provide users with more auxiliary diagnosis results as much as possible on the basis of ensuring the computing performance of the application, which helps to further improve the accuracy and efficiency of auxiliary diagnosis.

[0016] In some embodiments, obtaining feedback data based on similar processing results includes: when a target processing result is obtained by processing medical data based on computing resources through a target application, obtaining feedback data based on the target processing result and similar processing results.

[0017] Through the embodiments of the present application, the processing results of similar applications and target applications can be obtained, providing diagnostic references in different dimensions. Since similar applications and target applications may have differences in processing methods and algorithms, integrating the diagnostic results of similar applications and target applications can reduce the risk of missed diagnosis or misdiagnosis that may be caused by a single application. In addition, the feedback data can also include an explanation of the diagnostic results, which helps users better understand the basis for diagnosis. Users can refer to the explanations of the diagnostic results of similar applications and target applications to determine which diagnostic result is more reliable or whether there are other influencing factors.

[0018] An embodiment of the present application provides a controller, which implements the steps of the method of any of the above embodiments.

[0019] An embodiment of the present application provides a computing cluster system, which can communicate data with a medical service end, and the computing cluster system includes: an application pool, the application pool includes at least a target application and a similar application of the same type as the target application; a GPU resource pool, which is used to provide computing resources; a resource monitor, which is used to monitor the GPU resource pool to obtain resource usage data of the computing resources; a controller, which is used to execute: receiving medical data and application specified information from the medical service end, and determining a specified target application based on the application specified information, wherein the target application can process the medical data based on the computing resources; determining a similar application of the same type as the target application from the application pool; obtaining the resource usage data of the computing resources from the resource monitor, and sending the medical data to the similar application based on the resource usage data of the computing resources, so that the similar application processes the medical data based on the computing resources to obtain the similar processing results; obtaining feedback data based on the similar processing results, and sending the feedback data to the medical service end.

[0020] Through the embodiments of the present application, the computing cluster system can provide users with diagnostic results of target applications and similar applications based on their computing resource usage data. When there are similar applications and the computing resource usage data is lower than the preset threshold, the user can receive the auxiliary diagnosis results of the medical data uploaded by the similar application, thereby effectively preventing the situation of missed diagnosis due to auxiliary diagnosis based on only a single application, and improving the accuracy of auxiliary diagnosis. At the same time, the computing cluster system has the ability to dynamically allocate computing resources of the GPU resource pool. This makes it possible for the computing tasks submitted by users at different time nodes of the medical service end to not significantly change due to the occupation of computing resources by computing tasks of similar applications, thereby ensuring that the waiting time for users to obtain auxiliary diagnosis results remains stable. Moreover, by specially marking the computing tasks of similar applications, the computing cluster system can flexibly cancel all or part of the computing tasks of similar applications based on the computing resources required for these computing tasks. This operation mode does not add additional burden to the computing cluster system, and enables the system to efficiently apply computing resources when idle. Moreover, by setting a controller at the hardware layer, the controller realizes communication with the medical service end by means of a unified interface standard. The controller can receive the medical data and application-specified information submitted by the user at the medical service end, and has the ability to determine similar applications that are consistent with the target application type. The controller also has the ability to integrate the target processing results and similar processing results, which is conducive to presenting a more comprehensive and accurate diagnosis result to the user, helping the user to evaluate the disease from multiple dimensions, thereby greatly improving the accuracy and completeness of the diagnosis. In addition, the controller also has the ability to obtain information collected by the resource monitor. After the controller obtains this information, it can cancel all or part of the computing tasks of similar applications based on the computing resource usage data of the computing cluster system, when the target application has sufficient computing resources for medical data processing, thereby optimizing the utilization efficiency of computing resources, avoiding conflicts in the use of computing resources of the computing cluster system, and ensuring the stability of the quality of auxiliary diagnosis services.

[0021] An embodiment of the present application provides an electronic device, which includes: a memory, and one or more processors communicatively connected to the memory; the memory stores instructions that can be executed by the one or more processors, and the instructions are executed by the one or more processors to enable the one or more processors to implement the steps of the method of any of the above embodiments.

[0022] An embodiment of the present application provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the method of any of the above embodiments are implemented.

[0023] An embodiment of the present application provides a computer program product, including a computer program. When the computer program is executed by a processor, the steps of the method according to any one of the above embodiments are implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 The diagnosis results of lung nodule images examined by two different applications are provided in one embodiment of the present invention;

[0025] Figure 2 A flowchart of a method for processing medical data provided in an embodiment of the present application;

[0026] Figure 3 A schematic diagram of a process for merging processing results provided in an embodiment of the present application;

[0027] Figure 4 A schematic diagram of a computing cluster system provided in an embodiment of the present application;

[0028] Figure 5 A block diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0029] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and should not be construed as limiting the present invention.

[0030] With the continuous development of artificial intelligence in the medical field, there may be multiple manufacturers in the market developing auxiliary diagnosis applications for a certain disease. However, when different application products analyze and process medical images to diagnose a certain disease, there may be different diagnostic results for the same set of medical images. For users, their budget is limited. If users only choose one application product for auxiliary diagnosis, there may be missed diagnoses, resulting in a poor user experience. Auxiliary diagnosis refers to the use of applications deployed and running on the cloud platform to analyze and process the medical data of medical images to obtain the diagnostic results of medical images. Applications and cloud platforms jointly build an efficient and intelligent diagnostic ecosystem through data sharing, resource integration and functional complementarity.

[0031] Cloud platforms are usually supported by infrastructure provided by basic service operators who have powerful data centers and network resources. Cloud platforms use these resources to build their service architecture and provide functions such as storage, processing and analysis of image data. Cloud platforms can also dynamically allocate computing resources to applications based on the actual needs of applications and the real-time status of the GPU (Graphics Processing Unit) resource pool. Multiple applications can share the computing resources of the cloud platform GPU resource pool to improve the overall utilization of computing resources.

[0032] Applications can use the network services of the cloud platform to establish secure, high-speed, and multi-data format-supported transmission channels. Medical institutions in different regions and at different levels can share medical data and diagnostic results through the cloud platform; applications can provide patients with certain diagnostic result interpretation services through the cloud platform, for example, informing patients of key information in the diagnostic results in an easy-to-understand manner; applications can also mark key locations in the diagnostic results with coordinates and return the coordinates to PACS (Picture Archiving and Communication Systems), meeting the needs of doctors who use applications for auxiliary diagnosis to manually mark and confirm on PACS.

[0033] Applications and cloud platforms not only improve the accuracy and efficiency of diagnosis, but also provide more convenient services for users (including medical institution members and patients). However, for a certain disease, there may be a variety of different applications on the market that can assist in diagnosis, and different applications may produce different auxiliary diagnosis results when analyzing and processing the same set of medical data.

[0034] Figure 1 An embodiment of the present invention provides diagnostic results of lung nodule images examined by two different application programs.

[0035] like Figure 1 As shown in the figure, taking lung nodules as an example, when two different applications analyze and process the same set of lung nodule image data, the number of lesions in the two diagnostic results is different. The number of lesions refers to the number of lesion areas found. Figure 1 The two diagnostic results of pulmonary nodule disease shown in the figure show 7 and 10 nodules respectively. The "7" and "10" here refer to the number of lung lesions. Different numbers of lesions mean different stages or types of disease. The number and distribution of lesions are important factors in assessing tumor staging and prognosis. Figure 1As shown, when different applications are used to process and analyze the same set of lung nodule image data, there are differences in the number of lesions in the identified diagnostic results, which may lead to inaccurate diagnostic results and thus affect subsequent treatment plans.

[0036] Therefore, the present application provides a method for processing medical data, which can effectively improve the accuracy of auxiliary diagnosis results.

[0037] Figure 2 A flowchart of a method for processing medical data provided in an embodiment of the present application.

[0038] like Figure 2 As shown, a medical data processing method 200 provided in an embodiment of the present application includes steps S210-S240.

[0039] Step S210, receiving medical data and application designation information, and determining a designated target application based on the application designation information, wherein the target application can process the medical data based on computing resources.

[0040] For example, medical data refers to medical data of medical images submitted by users from a medical service end, application-specified information refers to instructions for which application the user wants to select to process the submitted medical data, and target application refers to an application determined by the cloud platform based on the instructions. The target application can use computing resources in a GPU resource pool to process the medical data submitted by the user.

[0041] Step S220, determining similar applications that are consistent with the type of the target application.

[0042] Exemplarily, similar applications refer to applications that process the same disease. This step, for example, requires finding similar applications in all applications that have the same disease type as that processed by the target application.

[0043] Step S230, based on the resource usage data of the computing resources, the medical data is sent to the similar application so that the similar application processes the medical data based on the computing resources to obtain the similar processing results.

[0044] Exemplarily, the resource usage data of computing resources represents the usage of the GPU resource pool in the computing cluster of the cloud platform. By analyzing the resource usage data of computing resources and ensuring that the target application and similar applications have sufficient computing resources to process the medical data, the medical data is sent to similar applications, allowing similar applications to process the medical data based on the computing resources and obtain similar processing results.

[0045] Step S240: obtaining feedback data based on the same type of processing results, and sending the feedback data.

[0046] For example, the feedback data may include the diagnosis result obtained by the same type of application processing the medical data, and after the feedback data is obtained, the feedback data is sent to the medical service end. The feedback data may also include an explanation of the diagnosis result.

[0047] In an embodiment of the present application, a user can receive the diagnosis results of the same type of application for auxiliary diagnosis of the medical data uploaded by the user. Since the diagnosis results received by the user are not obtained based on a single application, it can effectively avoid the occurrence of missed diagnosis and improve the accuracy of auxiliary diagnosis. When sending medical data to similar applications for auxiliary diagnosis, since the computing resources required by the target application and similar applications when processing medical data are taken into account, for the user, whether it is based on a single application for auxiliary diagnosis or based on similar applications for auxiliary diagnosis, the time spent by the two does not change significantly, and the user is almost unaware of the entire auxiliary diagnosis process. The above method can not only provide users with the diagnosis results of similar applications, but also effectively reduce the missed diagnosis rate in auxiliary diagnosis and improve the accuracy of auxiliary diagnosis. It can also improve the utilization efficiency of computing resources in the computing cluster, so that the computing resources are optimized in terms of cost-effectiveness.

[0048] In another embodiment of the present application, feedback data is obtained based on similar processing results, including: when a target processing result is obtained by processing medical data based on computing resources through a target application, feedback data is obtained based on the target processing result and similar processing results.

[0049] Exemplarily, the same type of application processes the medical data based on the computing resources and obtains the same type of processing results, and the target application processes the medical data based on the computing resources and obtains the target processing results. Feedback data is obtained based on the same type of processing results and the target processing results, and the feedback data may cover multiple aspects, for example, the feedback data may include the diagnosis results obtained by the same type of application processing the medical data and the diagnosis results obtained by the target application processing the medical data, for example, the feedback data may also include the interpretation of the diagnosis results by the same type of application and the target application.

[0050] It can be seen that through the embodiments of the present application, the processing results of similar applications and target applications can be obtained, providing diagnostic references in different dimensions. Since similar applications and target applications may have differences in processing methods and algorithms, integrating the diagnostic results of similar applications and target applications can reduce the risk of missed diagnosis or misdiagnosis that may be caused by a single application. In addition, the feedback data can also include an explanation of the diagnostic results, which helps users better understand the basis for diagnosis. Users can refer to the explanations of the diagnostic results of similar applications and target applications to determine which diagnostic result is more reliable or whether there are other influencing factors.

[0051] In another embodiment of the present application, the computing resources include a first computing resource and a second computing resource, and the target application processes the medical data based on the first computing resource; based on the resource usage data of the computing resources, the medical data is sent to the same type of application, including: when the resource usage data indicates that the second computing resource can support the same type of application to process the medical data, the medical data is sent to the same type of application.

[0052] Exemplarily, the first computing resource represents the computing resource required by the target application when processing medical data, and the second computing resource may be a resource other than the first computing resource, and similar applications need to perform calculations based on the second computing resource when processing medical data. In the computing cluster of the cloud platform, the total amount of computing resources is fixed. To ensure that users have a high-quality user experience, for example, the time required for the target application to process and analyze medical data can be guaranteed to remain stable. For example, sufficient computing resources are preferentially allocated to the target application to support its auxiliary diagnosis function. When the computing resources are deducted from the first computing resources required by the target application, if the resource usage data of the computing resources indicates that there are still abundant second computing resources that can meet the calculation of similar applications, it indicates that there are still sufficient idle resources in the computing resources for similar applications to use. At this time, the medical data can be sent to similar applications for processing.

[0053] In the embodiments of the present application, on the one hand, by ensuring that the target application has sufficient computing resources for calculation, it is ensured that the overall time required for processing medical data does not change significantly, avoiding the user from having to wait for a long time due to insufficient computing resources when selecting the target application for auxiliary diagnosis, thereby improving the user experience. On the other hand, on the basis of ensuring the normal operation of the target application, the remaining computing resources are fully utilized. It avoids the idle waste of computing resources, especially when the total amount of computing cluster resources of the cloud platform is fixed. This reasonable resource allocation method can maximize the efficiency of the entire computing resource system and improve the economic benefits of resources. On the other hand, when the resource usage data of the computing resources indicates that it can support similar applications for calculation, more similar applications with different analysis methods and algorithms from the target application may be introduced into the auxiliary diagnosis process. Different similar applications may have their own advantages, which are helpful to interpret medical data from multiple dimensions and further improve the accuracy of auxiliary diagnosis.

[0054] In another embodiment of the present application, when the resource usage data indicates that the usage rate of the second computing resource is lower than a preset threshold, it means that the second computing resource can support similar applications to process medical data.

[0055] Exemplarily, a preset threshold can be set for the usage rate of computing resources. When the resource usage data indicates that the usage rate of the second computing resource is lower than the preset threshold, it means that the second computing resource can support similar applications to process medical data and send the medical data to similar applications.

[0056] In an embodiment of the present application, by setting a preset threshold for the utilization rate of computing resources, the computing resources required by similar applications when processing medical data can be effectively taken into account, ensuring that the overall time required to process medical data does not change significantly, avoiding users from spending too much time waiting for diagnosis results, and improving user experience.

[0057] In another embodiment of the present application, the method further includes: upon receiving new medical data and new application designation information, determining based on the resource usage data to cancel the same type of application from processing the medical data based on the second computing resource.

[0058] Exemplarily, newly added medical data means that after the time point of receiving the medical data submitted by the user as mentioned above, medical data submitted by the user is received again. The user may be the same user who submitted the medical data mentioned above, or another different user. When the cloud platform receives newly added medical data and newly added application designation information, for example, priority may be given to the target application specified by the user through the newly added application designation information, and the computing resources required for the target application specified by the newly added application designation information to process the newly added medical data may be determined preferentially. For example, the resource usage data of the computing resources of the current computing cluster indicates that there are sufficient computing resources to support the target application specified by the newly added application designation information in processing the newly added medical data, but there are insufficient computing resources to support similar applications in processing medical data, and the processing of medical data by similar applications based on the second computing resources may be cancelled.

[0059] Therefore, whether receiving the medical data and application-specified information submitted by the user first, or receiving the newly added medical data and newly added application-specified information submitted by the user later, the cloud platform gives priority to the computing resources required by the user-specified target application to process the medical data. If the computing resources of the current cloud platform's computing cluster are insufficient to support the user-specified target application to process the medical data, the second computing resources are released for the user-specified target application to use by canceling the processing of medical data by similar applications based on the second computing resources. Ensure that the target application specified by the user at different time points has sufficient computing resources for calculation, and the time spent by the target application in processing medical data does not change significantly, avoiding users from experiencing long waiting times when performing auxiliary diagnosis, and improving user experience.

[0060] In another embodiment of the present application, the new application designation information is used to specify a new target application, and the computing resources also include a third computing resource; based on the resource usage data, it is determined to cancel the processing of medical data by similar applications based on the second computing resource, including: when the resource usage data indicates that the third computing resource is insufficient to support the new target application to process the new medical data, canceling the processing of medical data by similar applications based on the second computing resource so as to release the second computing resource for the new target application to process the new medical data.

[0061] Exemplarily, the third computing resource represents the computing resources required by the new target application when processing the new medical data. When the resource usage data of the computing resources indicates that the current computing resources are insufficient to support the new target application in processing the new medical data, the processing of medical data by similar applications based on the second computing resources can be cancelled, and computing resources can be provided for the new target application to process the new medical data by releasing the second computing resources.

[0062] In the embodiment of the present application, by canceling the processing of medical data by similar applications based on the second computing resources, the second computing resources are released for use by the target application specified by the user. It is ensured that the newly added target application specified by the user has sufficient computing resources for calculation, and the time spent by the newly added target application in processing medical data does not change significantly, avoiding users from experiencing long waiting time when performing auxiliary diagnosis, and improving user experience.

[0063] In another embodiment of the present application, similar applications include a first similar application and a second similar application, the second computing resource includes a first sub-computing resource and a second sub-computing resource, the first similar application processes medical data based on the first sub-computing resource, and the second similar application processes medical data based on the second sub-computing resource; canceling the processing of medical data by similar applications based on the second computing resource includes: canceling the processing of medical data by the first similar application based on the first sub-computing resource; after canceling the processing of medical data by the first similar application based on the first sub-computing resource, if the resource usage data indicates that the first sub-computing resource and the third computing resource can support the newly added target application to process the newly added medical data, then keeping the second similar application to continue processing the medical data based on the second sub-computing resource.

[0064] Exemplarily, the same type of application may include a first same type of application and a second same type of application, the second computing resource may include a first sub-computing resource and a second sub-computing resource, the first same type of application processes medical data based on the first sub-computing resource, and the second same type of application processes medical data based on the second sub-computing resource. When the computing resources of the computing cluster are insufficient to support the newly added target application to process the newly added medical data, if the resource usage data indicates that the first same type of application is cancelled to process the medical data based on the first sub-computing resource, and the newly added target application can be supported to process the newly added medical data after the first sub-computing resource is released, only the first same type of application can be cancelled to process the medical data based on the first sub-computing resource, and the second same type of application can be kept to continue to process the medical data based on the second sub-computing resource. Then, the target processing results of the target application and the feedback data of the same type of processing results are integrated and sent to the medical service end. Vice versa, if the resource usage data indicates that the newly added target application can be supported to process the newly added medical data after the second sub-computing resource is released, only the second same type of application can be cancelled to process the medical data based on the second sub-computing resource, and the first same type of application can be kept to continue to process the medical data based on the first sub-computing resource.

[0065] In another example, there may be n applications of the same type (n>2), and the n applications of the same type process medical data based on the n sub-computing resources required by each. When the computing resources of the computing cluster are insufficient to support the newly added target application to process the newly added medical data, when the resource usage data indicates that some applications of the same type need to be cancelled from processing medical data, for example, it is necessary to cancel i (1≤i≤n) applications of the same type from the n applications of the same type to process medical data based on the i sub-computing resources required by them, and release the i sub-computing resources to the computing resources of the computing cluster, at which time the computing resources of the computing cluster can support the newly added target application to process the newly added medical data. Keep (ni) applications of the same type processing medical data based on the (ni) sub-computing resources required by each, and send the target processing results of the target application and (ni) applications of the same type and the feedback data of the same processing results to the medical service end.

[0066] In the embodiments of the present application, the current computing resource usage of the computing cluster can be accurately judged by computing resource usage data. This method can not only give priority to the target applications specified by the user at different time nodes for auxiliary diagnosis, but also effectively reduce the waiting time of the user in the auxiliary diagnosis process, thereby significantly improving the user experience. In addition, it is also possible to decide to cancel all or part of the same type of applications based on resource usage data, and provide users with more auxiliary diagnosis results as much as possible on the basis of ensuring the computing performance of the application, which helps to further improve the accuracy and efficiency of auxiliary diagnosis.

[0067] Figure 3 A flowchart of a combined processing result provided in one embodiment of the present application.

[0068] For example, Figure 3 As shown, after the cloud platform receives the medical data and application-specified information submitted by the user, it will determine whether there is a similar application that processes the same disease type as the target application specified in the application-specified information based on the type of medical data. When it is determined that there is no similar application, the target application will process the medical data independently based on computing resources, and then send the independently calculated target processing results as feedback data to the medical service end, and then end the entire computing process.

[0069] When it is determined that there is a similar application that processes the same disease type as the target application specified by the application specified information, it will be determined whether the current resource usage rate is lower than the preset threshold value based on the resource usage data of the computing resources. If the resource usage data indicates that the current resource usage rate is higher than the preset threshold value, it means that the current computing resources are insufficient to support similar applications to process medical data, so that the target application processes the medical data separately based on the computing resources, and then sends the separately calculated target processing result as feedback data to the medical service end, and then ends the entire computing process. If the resource usage data indicates that the current resource usage rate is lower than the preset threshold value, it means that there are sufficient computing resources to support similar applications to process medical data, so that the target application and similar applications process the medical data in parallel based on the computing resources, and merge the target processing result (calculation result) and the similar processing result (calculation result) after parallel calculation, and send the merged calculation result as feedback data to the medical service end, and then end the entire computing process.

[0070] In an embodiment of the present application, a user can receive the diagnosis results of the medical data uploaded by the same type of application for auxiliary diagnosis. Since the diagnosis results received by the user are not obtained based on a single application, it can effectively avoid the occurrence of missed diagnosis and improve the accuracy of auxiliary diagnosis. When sending medical data to similar applications for auxiliary diagnosis, since the computing resources required by the target application and similar applications when processing medical data are taken into account, for the user, whether it is based on a single application for auxiliary diagnosis or based on similar applications for auxiliary diagnosis, the time spent by the two does not change significantly, and the user is almost unaware of the entire auxiliary diagnosis process. This method can not only provide users with the diagnosis results of similar applications, but also effectively reduce the missed diagnosis rate in auxiliary diagnosis and improve the accuracy of auxiliary diagnosis. It can also improve the utilization efficiency of computing resources in the computing cluster, improve the accuracy of the diagnostic results without adding additional burden to the cloud platform, and optimize the computing resources in terms of cost-effectiveness.

[0071] Figure 4 A schematic diagram of a computing cluster system provided in an embodiment of the present application.

[0072] like Figure 4 As shown, an embodiment of the present application provides a computing cluster system 400, which can communicate data with a medical service end, and the computing cluster system 400 includes:

[0073] The application pool 410 includes at least a target application and similar applications of the same type as the target application.

[0074] The GPU resource pool 420 is used to provide computing resources.

[0075] The resource monitor 430 is used to monitor the GPU resource pool 420 to obtain resource usage data of computing resources.

[0076] The controller 440 is used to perform the following steps: receiving medical data and application-specified information from a medical service end, and determining a specified target application based on the application-specified information, wherein the target application can process the medical data based on computing resources. Determine similar applications of the same type as the target application from the application pool 410. Obtain resource usage data of computing resources from the resource monitor 430, and send the medical data to similar applications based on the resource usage data of computing resources, so that similar applications process the medical data based on computing resources to obtain similar processing results. Obtain feedback data based on the similar processing results, and send the feedback data to the medical service end. It can be understood that the controller 440 is used to execute the above method.

[0077] In one example, the computing cluster system 400 is a core component of a cloud platform. By connecting multiple servers together to form a collaborative system, the computing cluster system 400 provides powerful computing capabilities for the cloud platform.

[0078] In one example, when there are similar applications that process the same disease type as the target application, the application pool 410 includes at least the target application and similar applications that are of the same type as the target application. For example, the similar applications may be one or more.

[0079] In one example, the cloud platform can also dynamically allocate computing resources to applications based on the actual needs of the applications and the real-time status of the GPU (Graphics Processing Unit) resource pool 420. Multiple applications can share the computing resources of the cloud platform GPU resource pool 420 to improve the overall utilization of computing resources.

[0080] In one example, the resource monitor 430 in the computing cluster system 400 can monitor and summarize the usage of computing resources of the computing cluster system 400 to obtain the usage data of the current computing resources.

[0081] In one example, the present application adds a controller 440 at the hardware layer. When the calculation request of the medical data and application-specified information submitted by the user at the medical service end reaches the middle of the computing application layer, the controller 440 can determine the same type of application consistent with the type of the target application from the application pool 410. Since the controller 440 adopts a unified interface standard, communication with the medical service end can be achieved. When there is a similar application consistent with the type of the target application, the computing cluster system 400 can return the target processing result and the same type of processing result to the controller 440 for further integration to obtain feedback data, and then the controller 440 sends the feedback data to the medical service end through a unified interface standard. The controller 440 further integrates the target processing result and the same type of processing result. For example, in the process of assisting the diagnosis of pulmonary nodules, for example, the target processing result and the multiple sets of coordinate results of the pulmonary nodules in the same type of processing results can be integrated to form a new result set, and then the feedback data is sent to the medical service end through a unified interface standard.

[0082] The controller 440 can obtain the computing resource usage data collected by the resource monitor 430, and set a preset threshold α for the computing resource usage data. When the computing resource usage data is lower than the preset threshold α, it means that the resource usage rate of the computing cluster system 400 is low, allowing similar applications to process medical data based on the idle computing resources of the computing cluster system 400. At this time, the controller 440 can forward the computing request to similar applications at the same time, and specially mark the computing tasks sent to the similar applications. Since there can be multiple similar applications, there can also be multiple specially marked computing tasks.

[0083] During the calculation period of the above-mentioned target application and similar applications processing medical data based on computing resources, if the computing tasks surge, for example, the user submits new medical data and new application designation information on the medical service end, and the computing resource usage data of the computing cluster system 400 indicates that the remaining computing resources are insufficient to process the new medical data, the controller 440 can cancel all or part of the processing of medical data by similar applications based on computing resources according to the computing resource usage data, and release the computing resources of the canceled computing tasks of the similar applications to the computing tasks of the target application specified by the new application designation information. If all the computing tasks of similar applications are canceled, the target processing results are not integrated, and the target processing results are sent to the medical service end as feedback data. If the computing tasks of similar applications are partially canceled, the target processing results and similar processing results are still integrated, and the target processing results and similar processing results are sent to the medical service end as feedback data.

[0084] In another example, the same type of applications may be application A and application B. When the controller 440 can forward the computing request to application A and application B at the same time, and specially mark the computing tasks sent to application A and application B. Among them, application A processes the medical data based on computing resource A to obtain the same type of processing result A, and application B processes the medical data based on computing resource B to obtain the same type of processing result B. When the computing tasks surge, the controller 440 obtains the computing resource usage data collected by the resource monitor 430. If it is determined that the computing resources A and computing resources B required by application A and application B need to be released to the computing tasks of the target application specified by the newly added application specified information, the target application specified by the newly added application specified information can process the medical data normally, then all the computing tasks in application A and application B are cancelled. If the controller 440 obtains the computing resource usage data collected by the resource monitor 430 and determines that only the computing resources A required by application A need to be released to the computing tasks of the target application specified by the newly added application specified information, then only the computing tasks of application A are cancelled, the computing tasks in application B are kept, and the target processing results and the class processing results B are sent to the medical service end as feedback data. When it is determined that only the computing resources B required by application B need to be released to the computing tasks of the target application specified by the newly added application specified information, only the computing tasks of application B are canceled, the computing tasks in application A are retained, and the target processing results and class processing results A are sent to the medical service end as feedback data.

[0085] In another example, the computing cluster system 400 further includes a container service 450 , which is used to dynamically allocate computing resources of the GPU resource pool 420 .

[0086] Based on the embodiment of the present application, the computing cluster system can provide the user with the diagnosis results of the target application and similar applications based on its computing resource usage data. When there are similar applications and the computing resource usage data is lower than the preset threshold α, the user can receive the auxiliary diagnosis results of the medical data uploaded by the same application, thereby effectively preventing the situation of missed diagnosis due to auxiliary diagnosis based on only a single application, and improving the accuracy of auxiliary diagnosis. At the same time, the computing cluster system has the ability to dynamically allocate computing resources of the GPU resource pool. This makes it possible for the computing tasks submitted by users at different time nodes on the medical service end to not significantly change due to the occupation of computing resources by computing tasks of similar applications, thereby ensuring that the waiting time for users to obtain auxiliary diagnosis results remains stable. Moreover, by specially marking the computing tasks of similar applications, the computing cluster system can flexibly cancel all or part of the computing tasks of similar applications based on the computing resources required for these computing tasks. This mode of operation does not add additional burden to the computing cluster system, and enables the system to efficiently apply computing resources during idle time.

[0087] Moreover, by setting up a controller at the hardware layer, the controller can achieve communication with the medical service end with a unified interface standard. The controller can receive the medical data and application-specified information submitted by the user at the medical service end, and has the ability to determine similar applications that are consistent with the target application type. The controller also has the ability to integrate the target processing results and similar processing results, which is conducive to presenting a more comprehensive and accurate diagnosis result to the user, helping the user to evaluate the disease from multiple dimensions, thereby greatly improving the accuracy and completeness of the diagnosis. In addition, the controller also has the ability to obtain information collected by the resource monitor. After the controller obtains this information, it can cancel all or part of the computing tasks of similar applications based on the computing resource usage data of the computing cluster system, when the target application has sufficient computing resources for medical data processing, thereby optimizing the utilization efficiency of computing resources, avoiding conflicts in the use of computing resources of the computing cluster system, and ensuring the stability of the quality of auxiliary diagnosis services.

[0088] It can be seen that this application is based on the advantage that the image cloud platform can integrate multiple AI applications. Without adding extra burden to the platform, it can perform calculations through similar applications to improve the accuracy of processing results, improve user experience, and improve the utilization rate of platform resources.

[0089] An embodiment of the present application provides a controller, which can implement the steps of the method in any of the above embodiments.

[0090] An embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the method in any one of the above embodiments are implemented.

[0091] An embodiment of the present application provides a computer program product, which includes instructions. When the instructions are executed by a processor of a computer device, the computer device is enabled to perform the steps of the method of any of the above embodiments.

[0092] Figure 5 A block diagram of an electronic device provided in an embodiment of the present application.

[0093] An embodiment of the present application provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements a method in any of the above embodiments when executing the computer program.

[0094] like Figure 5 As shown, for ease of understanding, the embodiment of the present application shows a specific electronic device 500.

[0095] The electronic device 500 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0096] like Figure 5 As shown, the device 500 includes a computing unit 501, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 502 or a computer program loaded from a storage unit 508 into a random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the electronic device 500 can also be stored. The computing unit 501, the ROM 502, and the RAM 503 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0097] Multiple components in the electronic device 500 are connected to the I / O interface 505, including: an input unit 506, such as a keyboard, a mouse, etc.; an output unit 507, such as various types of displays, speakers, etc.; a storage unit 508, such as a disk, an optical disk, etc.; and a communication unit 509, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 509 allows the electronic device 500 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0098] The computing unit 501 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 501 performs the various methods described above. For example, in some embodiments, any one or more of the methods described above may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 508. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 500 via ROM 502 and / or communication unit 509. When the computer program is loaded into RAM 503 and executed by the computing unit 501, any one or more of the steps of the methods described above may be performed. Alternatively, in other embodiments, the computing unit 501 may be configured to perform any one or more of the methods described above in any other appropriate manner (e.g., by means of firmware).

[0099] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purposes of this application, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection portion with one or more wirings (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or otherwise processing in a suitable manner if necessary, and then stored in a computer memory.

[0100] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0101] In the description of the present application, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In the present application, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0102] In the description of the present application, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise", "axial", "radial", "circumferential" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0103] In addition, the terms "first", "second", etc. used in the embodiments of the present application are only used for descriptive purposes and should not be understood as indicating or implying relative importance, or implicitly indicating the number of technical features indicated in the present embodiment. Therefore, the features defined by the terms "first", "second", etc. in the embodiments of the present application can explicitly or implicitly indicate that at least one of the features is included in the embodiment. In the description of the present application, the word "multiple" means at least two or two or more, such as two, three, four, etc., unless otherwise clearly and specifically defined in the embodiments.

[0104] In this application, unless otherwise clearly specified or limited in the embodiments, the terms "installed", "connected", "connected" and "fixed" etc. appearing in the embodiments should be understood in a broad sense. For example, the connection can be a fixed connection, a detachable connection, or an integrated connection. It can be understood that it can also be a mechanical connection, an electrical connection, etc.; of course, it can also be a direct connection, or an indirect connection through an intermediate medium, or it can be the internal connection of two elements, or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to the specific implementation situation.

[0105] In the present application, unless otherwise clearly specified and limited, a first feature being “above” or “below” a second feature may mean that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, a first feature being “above”, “above”, and “above” a second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is higher in level than the second feature. A first feature being “below”, “below”, and “below” a second feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is lower in level than the second feature.

Claims

1. A medical data processing method, characterized in that: The method comprises: receiving medical data and application specifying information, and determining a specified target application based on the application specifying information, wherein the target application is capable of processing the medical data based on computing resources; Identifying similar applications of the same type as the target application; Based on the resource usage data of the computing resources, the medical data is sent to the similar application so that the similar application processes the medical data based on the computing resources to obtain similar processing results; Feedback data is obtained based on the same type of processing results, and the feedback data is sent.

2. The method according to claim 1, characterized in that The computing resources include a first computing resource and a second computing resource, and the target application processes the medical data based on the first computing resource; The sending the medical data to the similar application based on the resource usage data of the computing resource comprises: In a case where the resource usage data indicates that the second computing resource can support the same type of application to process the medical data, the medical data is sent to the same type of application.

3. The method according to claim 2, characterized in that When the resource usage data indicates that the usage rate of the second computing resource is lower than a preset threshold, it means that the second computing resource can support the same type of application to process the medical data.

4. The method according to claim 2, characterized in that: The method further comprises: When newly added medical data and newly added application designation information are received, it is determined based on the resource usage data to cancel the processing of the medical data by the same type of application based on the second computing resource.

5. The method according to claim 4, characterized in that The newly added application designation information is used to designate a newly added target application, and the computing resources further include a third computing resource; and the step of determining based on the resource usage data to cancel the processing of the medical data by the same type of application based on the second computing resource includes: When the resource usage data indicates that the third computing resource is insufficient to support the newly added target application to process the newly added medical data, the processing of the medical data by the same application based on the second computing resource is canceled to release the second computing resource for the newly added target application to process the new medical data.

6. The method according to claim 5, characterized in that The same type of application includes a first same type of application and a second same type of application, the second computing resource includes a first sub-computing resource and a second sub-computing resource, the first same type of application processes the medical data based on the first sub-computing resource, and the second same type of application processes the medical data based on the second sub-computing resource; The cancelling the processing of the medical data by the same type of application based on the second computing resource includes: canceling the processing of the medical data by the first similar application based on the first sub-computing resource; After canceling the first similar application's processing of the medical data based on the first sub-computing resource, if the resource usage data indicates that the first sub-computing resource and the third computing resource can support the newly added target application to process the newly added medical data, then the second similar application is kept to continue processing the medical data based on the second sub-computing resource.

7. The method according to any one of claims 1 to 6, characterized in that: The obtaining of feedback data based on the same type of processing results includes: When the target application processes the medical data based on the computing resources to obtain a target processing result, the feedback data is obtained based on the target processing result and the similar processing results.

8. A controller, characterized in that: The controller is used to implement the steps of the method according to any one of claims 1 to 7.

9. A computing cluster system, characterized in that: The computing cluster system is capable of performing data communication with the medical service end, and the computing cluster system includes: An application pool, the application pool including at least a target application and similar applications of the same type as the target application; GPU resource pool, used to provide computing resources; A resource monitor, configured to monitor the GPU resource pool to obtain resource usage data of the computing resources; Controller, which performs: receiving medical data and application specifying information from the medical service end, and determining a specified target application based on the application specifying information, wherein the target application is capable of processing the medical data based on computing resources; Determine from the application pool an application of the same type as the target application; Acquire resource usage data of the computing resource from the resource monitor, and send the medical data to the similar application based on the resource usage data of the computing resource, so that the similar application processes the medical data based on the computing resource to obtain similar processing results; Feedback data is obtained based on the same type of processing results, and the feedback data is sent to the medical service end.

10. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

12. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images