Multi-campus medical image storage method, system and storage medium

Abstract

The application relates to a multi-hospital area medical image storage method and system and a storage medium. The method realizes the balance between data management efficiency and retrieval performance by detecting inter-hospital network parameters and storage resource states in real time and dynamically determining an archiving mode. Meanwhile, the influence of network fluctuation on operation stability is reduced by using standard communication instruction execution parameter measurement and cooperating with hysteresis switching judgment logic. In addition, by combining intermediate state identifier management and a two-phase commit protocol, the consistency conflict problem under multi-node concurrent modification is solved while ensuring cross-hospital area search accuracy, thereby improving the continuity of image storage business and the reliability of data access in a complex network environment.

Description

Technical Field

[0001] This application relates to the field of medical information technology, and in particular to a method, system and storage medium for storing medical images in multiple hospital areas. Background Technology

[0002] With the deepening implementation of the national hierarchical medical system and the accelerated development of medical groups, multi-campus hospitals, medical alliances, and medical consortia have become the main forms of large public hospitals. As a core business system of medical institutions, Picture Archiving and Communication Systems (PACS) carry TB-level or even PB-level DICOM format image data. The stability and access efficiency of its storage scheme directly affect the timeliness of clinical diagnosis and medical safety.

[0003] Currently, medical image storage in multi-hospital environments mainly involves the following two technical approaches: The first approach is a centralized storage solution, where all medical imaging data generated by each branch hospital is aggregated in real time to the main hospital's central data center via a wide area network for unified management. While this solution facilitates operation, maintenance, and backup, it is highly dependent on inter-hospital network bandwidth and latency. During peak network periods or in complex network topologies such as intercity dedicated lines, the real-time transmission of imaging data often experiences buffering and delays, resulting in excessively long waiting times for clinicians to access images, and even causing a complete halt to examination services at branch hospitals in the event of a network failure.

[0004] The second approach is a distributed storage solution, where image data is stored on local servers in each branch campus, and only the image index is registered with the central server to enable cross-campus retrieval. While this solution reduces real-time dependence on the core network, it presents several challenges. First, storage resources across different campuses cannot be used in a coordinated manner, and management strategies for reusing existing equipment versus purchasing new equipment are difficult to unify. Second, when branch nodes perform image attribute changes, it often leads to state conflicts between the central server's index and local files. Summary of the Invention

[0005] To address the aforementioned issues, this application provides a multi-hospital medical image storage method, system, and storage medium that optimizes the allocation of image storage resources in complex network environments and ensures continuity of clinical access services.

[0006] To achieve the above objectives, in a first aspect, embodiments of this application provide a method for storing medical images across multiple hospital campuses, comprising the following steps: Step S1: Obtain the medical image data to be archived, and probe the network status parameters between the branch node and the aggregation node, as well as the storage status parameters of the aggregation node. Step S2: Process the detected parameters according to the preset switching judgment logic, determine whether the current network bandwidth and storage capacity meet the preset centralized storage conditions, and determine the target storage mode according to the judgment result. The target storage mode includes centralized storage mode or distributed storage mode. Step S3: Perform image archiving according to the determined target storage mode. When the target storage mode is centralized storage mode, the medical image data and metadata are synchronously uploaded to the aggregation node; When the target storage mode is distributed storage mode, the medical image data is stored in the local storage of the branch node, and the metadata containing storage path information is registered to the aggregation node; Step S4: Establish a cross-campus shared access link, obtain the storage location of the target image by querying the aggregation node, and route to the corresponding node to retrieve the image based on the storage location.

[0007] Preferably, the process of detecting network state parameters in step S1 includes: The standard instructions of the medical imaging communication protocol are invoked to send a C-ECHO test request to the aggregation node, and the average network latency is calculated based on the echo response time. The C-STORE service in the standard instruction is invoked to transmit a test file with a preset data volume to the aggregation node, and the real-time network bandwidth is calculated based on the ratio of the preset data volume to the transmission time.

[0008] Preferably, when calculating real-time network bandwidth, the following bandwidth compensation steps are implemented: The time taken for a single transmission is monitored. If the time taken for a single transmission is less than the first preset time, the number of repeated transmissions is increased until the total time of multiple transmissions is not less than the second preset time. The ratio of the total amount of transmitted data to the total transmission time is used as the target network bandwidth value.

[0009] Preferably, the process of determining the target storage mode in step S2 employs dual-threshold hysteresis logic, including: Monitor whether the real-time network bandwidth is continuously higher than the first bandwidth threshold within the first preset period, and whether the remaining storage capacity of the aggregation node is higher than the first capacity threshold. If the determination result is yes, switch the target storage mode to the centralized storage mode. Monitor whether the real-time network bandwidth is continuously lower than the second bandwidth threshold or the remaining storage capacity is lower than the second capacity threshold within the second preset period, and if the determination result is yes, switch the target storage mode to the distributed storage mode. Wherein, the first bandwidth threshold is higher than the second bandwidth threshold.

[0010] Preferably, during the execution of the centralized storage mode in step S3, the following concurrency protection steps are also included: At the aggregation node, a first status identifier is set for the metadata record corresponding to the medical image data. The first status identifier indicates that the image data is in the process of synchronous aggregation. In response to a retrieval request carrying the first status identifier, the retrieval path is redirected to the corresponding branch node; After the image archiving is completed, the first status identifier is updated to the second status identifier, which indicates that the image data has been archived at the aggregation node.

[0011] Preferably, the method also includes a step of synchronizing image attribute changes, ensuring data consistency through a two-phase commit protocol: In the first phase, the branch node sends a change preparation request to the aggregation node, and the aggregation node performs pre-locking and status feedback on the target record. In the second phase, if the feedback status is "allowed", the branch node performs a local update and sends a synchronization commit command to the aggregation node; if the feedback status is "rejected", the branch node performs a local operation rollback.

[0012] Preferably, it also includes an asynchronous compensation step under abnormal conditions: If a branch node detects a network connection loss, it writes the attribute change operation sequence into the local change buffer; if the network connection is restored, it extracts the operation sequence from the change buffer and triggers data consistency synchronization with the aggregation node.

[0013] Preferably, the specific process of establishing a shared access link in step S4 includes: First, the local cache of the branch node is retrieved, and if the target image copy is found, local retrieval is performed; If the local cache is not hit, the storage mode identifier in the metadata is parsed, the image file is retrieved from the aggregation node or the branch node that generated the target image, and the retrieved image file is loaded into the retrieval cache.

[0014] Secondly, embodiments of this application provide a multi-hospital medical image storage system, including a convergence node and multiple branch nodes, wherein the system is configured to perform the steps of the method described in any embodiment of the first aspect.

[0015] Thirdly, embodiments of this application provide a computer-readable storage medium storing computer instructions thereon, characterized in that the computer instructions, when executed by a processor, implement the steps of the method described in any embodiment of the first aspect.

[0016] The multi-hospital medical image storage method, system, and storage medium designed in this application achieve a balance between data management efficiency and retrieval performance by real-time detection of inter-hospital network parameters and storage resource status and dynamic determination of the archiving mode. Simultaneously, by utilizing standard communication command execution parameter measurements and incorporating hysteresis switching judgment logic, the impact of network fluctuations on operational stability is reduced. Furthermore, by combining intermediate state identifier management and a two-phase commit protocol, the accuracy of cross-hospital retrieval is ensured while resolving consistency conflicts under concurrent modifications across multiple nodes, thereby improving the continuity of image storage services and the reliability of data access in complex network environments. Attached Figure Description

[0017] Figure 1 A flowchart illustrating the multi-hospital-area medical image storage method provided in this application embodiment.

[0018] Figure 2 This is a schematic diagram of the architecture of a multi-hospital medical image storage system provided in an embodiment of this application.

[0019] Figure 3 A flowchart for image archiving provided in the embodiments of this application.

[0020] Figure 4 A flowchart illustrating the cross-campus image retrieval process provided in this application embodiment. Detailed Implementation

[0021] The preferred embodiments of this application are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit this application.

[0022] Firstly, embodiments of this application provide a method for storing medical images across multiple hospital campuses. In this embodiment, the method is applied to a distributed archiving system environment including one aggregation node and multiple branch nodes. (Refer to...) Figure 1 The specific process of this method includes the following steps: Step S1: Obtain the medical image data to be archived, and probe the network status parameters between the branch node and the aggregation node, as well as the storage status parameters of the aggregation node.

[0023] Specifically, the branch node receives DICOM format image files generated by imaging equipment from various hospital areas. Before archiving, the system needs to sense the current physical environment. The process of detecting network status parameters includes: calling standard instructions of medical imaging communication protocols such as the DICOM protocol, sending a C-ECHO test request to the aggregation node, and calculating the average network latency between the branch node and the aggregation node based on the echo response time.

[0024] Furthermore, the branch node invokes the C-STORE service in the standard instructions to transmit a test file with a preset data volume, such as a simulated 20MB image file, to the aggregation node, and calculates the real-time network bandwidth based on the ratio of the preset data volume to the transmission time.

[0025] To eliminate the impact of instantaneous network fluctuations, this embodiment also implements a bandwidth compensation step: the time taken for a single transmission is monitored, and if the time taken for a single transmission is less than a first preset duration, such as less than 2 seconds, it is determined that the amount of sample data is insufficient, and the number of repeated transmissions is increased until the total duration of multiple transmissions is not less than a second preset duration, such as not less than 5 seconds. Finally, the ratio of the total amount of transmitted data to the total transmission duration is used as the target network bandwidth value; at the same time, the branch node obtains the remaining disk storage capacity of the aggregation node through a query command as a storage status parameter.

[0026] Step S2: Process the detected parameters according to the preset switching judgment logic, determine whether the current network bandwidth and storage capacity meet the preset centralized storage conditions, and determine the target storage mode based on the judgment result.

[0027] The target storage mode includes a centralized storage mode or a distributed storage mode. In this embodiment, the process of determining the target storage mode adopts dual-threshold hysteresis logic to avoid frequent mode switching at critical points. The specific logic is as follows: Monitor whether the real-time network bandwidth is continuously higher than the first bandwidth threshold (e.g., 800Mbps) within the first preset period, and whether the remaining storage capacity of the aggregation node is higher than the first capacity threshold (e.g., 25%). If both of these conditions are met, the target storage mode is determined to be the centralized storage mode.

[0028] Conversely, if the real-time network bandwidth is continuously lower than the second bandwidth threshold (e.g., 500Mbps) within the second preset period, or the remaining storage capacity is lower than the second capacity threshold (e.g., 20%), then if the determination result is yes, the target storage mode will be switched to distributed storage mode. The first bandwidth threshold is higher than the second bandwidth threshold; this difference forms a buffer to ensure the stability of system operation.

[0029] Step S3: Perform image archiving according to the determined target storage mode.

[0030] See Figure 3When the target storage mode is determined to be centralized storage mode, the branch node starts a synchronization task to synchronously upload the medical image data and associated metadata to the aggregation node for centralized storage; when the target storage mode is determined to be distributed storage mode, the branch node stores the medical image data in the local storage of the branch node in its own hospital area, and registers the metadata containing the physical storage path information of the file to the aggregation node, thereby realizing a layout of centralized index and distributed images.

[0031] Furthermore, to resolve access conflicts during image upload, this embodiment also includes the following concurrency protection steps during the execution of centralized storage mode in step S3: First, at the aggregation node, a first status identifier is set for the metadata record corresponding to the medical image data, for example, marked as "aggregating". This identifier is used to indicate that the image data is in the process of physical synchronization. Then, in response to a retrieval request carrying the first status identifier, the aggregation node determines that the image has not yet been archived locally, and thus redirects the retrieval path to the corresponding branch node that generated the target image, ensuring that doctors can obtain the image in a timely manner.

[0032] After the image is archived, the aggregation node updates the first status identifier to the second status identifier, such as marking it as "aggregated", to indicate that the image data has been archived at the aggregation node and can be directly supplied by the central end.

[0033] In addition, to ensure data consistency, this embodiment also includes a step of synchronizing image attribute changes. That is, when the administrator performs image attribute modification, splitting, or deletion on a branch node, data consistency is ensured through a two-phase commit (2PC) protocol. The first stage is the preparation stage, in which the branch node sends a change preparation request to the aggregation node, and the aggregation node performs a pre-lock on the target record (preventing other users from modifying it at the same time) and reports the preparation status back to the branch node.

[0034] The second stage is the commit stage. If the feedback status is "allowed", the branch node will officially execute the local database update and send a synchronous commit command to the aggregation node, which will then complete the mirror update. If the feedback status is "rejected" (e.g., due to a momentary network interruption or a record being occupied), the branch node will execute a local operation rollback to restore the state before the change.

[0035] In addition, this embodiment also provides an asynchronous compensation step in abnormal conditions: when a branch node detects that the network connection is broken, the generated attribute change operation sequence is temporarily written into a local change buffer, such as the log table of a local relational database; after the network connection is detected to be restored, the operation sequence in the change buffer is automatically extracted and the data consistency synchronization with the aggregation node is retried.

[0036] Step S4: Establish a cross-campus shared access link. Obtain the storage location of the target image by querying the aggregation node, and route the image to the corresponding node based on the storage location. That is, establish a communication connection between the access workstation and the target storage node based on the storage location to obtain the image.

[0037] For specific implementation, please refer to Figure 4 When a clinician initiates a request to access a workstation, the specific process includes: The system first searches the local cache of the branch node. If the target image copy is found, it directly performs local retrieval to save bandwidth. If the local cache is not found, the system parses the storage mode identifier in the metadata of the aggregation node to determine the actual storage location of the image. Then, it retrieves the image file from the aggregation node or the branch node that generated the target image and loads the retrieved image file into the retrieval cache of the branch node for subsequent secondary retrieval.

[0038] Secondly, embodiments of this application provide a multi-hospital-area medical image storage system, referring to... Figure 2 The system includes a convergence node and multiple branch nodes, which are connected via an inter-institutional network. Each branch node is equipped with an image archiving service module to respond to retrieval requests redirected from the convergence node and to send image data streams to the retrieval terminals. The processor in the system is configured to execute the steps of the method described in the first aspect embodiment by invoking a computer program stored in memory.

[0039] Thirdly, embodiments of this application provide a computer-readable storage medium storing computer instructions thereon, characterized in that the computer instructions, when executed by a processor, implement the steps of the method described in any embodiment of the first aspect.

[0040] The multi-hospital medical image storage method, system, and storage medium provided in this application achieve a balance between data management efficiency and retrieval performance by real-time detection of inter-hospital network parameters and storage resource status and dynamic determination of the archiving mode. Simultaneously, by utilizing standard communication command execution parameter measurements and incorporating hysteresis switching judgment logic, the impact of network fluctuations on operational stability is reduced. Furthermore, by combining intermediate state identifier management and a two-phase commit protocol, the accuracy of cross-hospital retrieval is ensured while resolving consistency conflicts under concurrent modifications by multiple nodes, thereby improving the continuity of image storage services and the reliability of data access in complex network environments.

[0041] In the description of this application, it should be noted that the terms "vertical", "up", "down", "horizontal", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0042] In the description of this application, it should also be noted that, unless otherwise expressly specified and limited, the terms "set," "install," "connect," and "link" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0043] Finally, it should be noted that the above descriptions are merely preferred embodiments of this application and are not intended to limit this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for storing medical images across multiple hospital sites, characterized in that, Includes the following steps: Step S1: Obtain the medical image data to be archived, and probe the network status parameters between the branch node and the aggregation node, as well as the storage status parameters of the aggregation node. Step S2: Process the detected parameters according to the preset switching judgment logic, determine whether the current network bandwidth and storage capacity meet the preset centralized storage conditions, and determine the target storage mode according to the judgment result. The target storage mode includes centralized storage mode or distributed storage mode. Step S3: Perform image archiving according to the determined target storage mode. When the target storage mode is centralized storage mode, the medical image data and metadata are synchronously uploaded to the aggregation node; When the target storage mode is distributed storage mode, the medical image data is stored in the local storage of the branch node, and the metadata containing storage path information is registered to the aggregation node; Step S4: Establish a cross-campus shared access link, obtain the storage location of the target image by querying the aggregation node, and route to the corresponding node to retrieve the image based on the storage location.

2. The multi-hospital-area medical image storage method according to claim 1, characterized in that, The process of probing network state parameters in step S1 includes: The standard instructions of the medical imaging communication protocol are invoked to send a C-ECHO test request to the aggregation node, and the average network latency is calculated based on the echo response time. The C-STORE service in the standard instruction is invoked to transmit a test file with a preset data volume to the aggregation node, and the real-time network bandwidth is calculated based on the ratio of the preset data volume to the transmission time.

3. The multi-hospital-area medical image storage method according to claim 2, characterized in that, When calculating real-time network bandwidth, the following bandwidth compensation steps are implemented: The time taken for a single transmission is monitored. If the time taken for a single transmission is less than the first preset time, the number of repeated transmissions is increased until the total time of multiple transmissions is not less than the second preset time. The ratio of the total amount of transmitted data to the total transmission time is used as the target network bandwidth value.

4. The multi-hospital-area medical image storage method according to claim 1, characterized in that, The process of determining the target storage mode in step S2 employs dual-threshold hysteresis logic, including: Monitor whether the real-time network bandwidth is continuously higher than the first bandwidth threshold within the first preset period, and whether the remaining storage capacity of the aggregation node is higher than the first capacity threshold. If the determination result is yes, switch the target storage mode to the centralized storage mode. Monitor whether the real-time network bandwidth is continuously lower than the second bandwidth threshold or the remaining storage capacity is lower than the second capacity threshold within the second preset period, and if the determination result is yes, switch the target storage mode to the distributed storage mode. Wherein, the first bandwidth threshold is higher than the second bandwidth threshold.

5. The multi-hospital-area medical image storage method according to claim 1, characterized in that, During the execution of the centralized storage mode in step S3, the following concurrency protection steps are also included: At the aggregation node, a first status identifier is set for the metadata record corresponding to the medical image data. The first status identifier indicates that the image data is in the process of synchronous aggregation. In response to a retrieval request carrying the first status identifier, the retrieval path is redirected to the corresponding branch node; After the image archiving is completed, the first status identifier is updated to the second status identifier, which indicates that the image data has been archived at the aggregation node.

6. The multi-hospital-area medical image storage method according to claim 1, characterized in that, It also includes the step of synchronizing image attribute changes, ensuring data consistency through a two-phase commit protocol: In the first phase, the branch node sends a change preparation request to the aggregation node, and the aggregation node performs pre-locking and status feedback on the target record. In the second phase, if the feedback status is "allowed", the branch node performs a local update and sends a synchronization commit command to the aggregation node; if the feedback status is "rejected", the branch node performs a local operation rollback.

7. The multi-hospital-area medical image storage method according to claim 6, characterized in that, It also includes asynchronous compensation steps in abnormal states: If a branch node detects a network connection loss, it writes the attribute change operation sequence into the local change buffer; if the network connection is restored, it extracts the operation sequence from the change buffer and triggers data consistency synchronization with the aggregation node.

8. The multi-hospital-area medical image storage method according to claim 1, characterized in that, The specific process of establishing a shared access link in step S4 includes: First, the local cache of the branch node is retrieved, and if the target image copy is found, local retrieval is performed; If the local cache is not hit, the storage mode identifier in the metadata is parsed, the image file is retrieved from the aggregation node or the branch node that generated the target image, and the retrieved image file is loaded into the retrieval cache.

9. A multi-hospital-area medical image storage system, characterized in that, The system includes a convergence node and multiple branch nodes, and is configured to perform the method steps of any one of claims 1 to 8.

10. A computer-readable storage medium storing computer instructions thereon, characterized in that, When the computer instructions are executed by the processor, they implement the method steps of any one of claims 1 to 8.

Images