Medical PACS (Picture Archiving and Communication System) for quickly retrieving image files

Abstract

The invention relates to a medical PACS (Picture Archiving and Communication System) for quickly retrieving image files, belongs to the technical field of image processing, and solves the problem that the conventional medical PACS is very slow in retrieving massive image files. An image storage subsystem in the system comprises a pre-processing module, a data storage module and a metadata storage module, the pre-processing module merges original image files uploaded by image equipment according to various data merging strategies and writes the original image files into a large file in the data storage module; obtaining basic information of the original image file, sending the basic information to the database subsystem, obtaining metadata information, and writing the metadata information into a metadata storage module; the database subsystem stores the basic information of the original image file sent by the pre-processing module by creating a metadata management table; and the image consulting subsystem is used for loading a to-be-retrieved large file and metadata information from the image storage subsystem according to a retrieval request of a client, analyzing an original image file and displaying the original image file. And rapid retrieval of massive image files is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a medical PACS system for quickly accessing image files. Background Art

[0002] With the widespread use of medical imaging equipment, the increase in the number of films taken by filming equipment at a time, and the continuous improvement in the resolution of filming equipment, the data volume of the hospital's medical PACS (picture archiving and communication system) system is expected to increase by 15% each year, double in five years, and show an accelerating growth trend. The annual new film volume of a tertiary general hospital or a tertiary specialized hospital such as chest and pulmonary department is between 50TB and 60TB. According to the relevant regulations on the preservation of electronic medical records, medical institutions are required to keep and preserve medical imaging files for no less than 15 years. Therefore, the online, near-line and offline data in the PCAS system add up to an astonishingly large scale.

[0003] Faced with massive amounts of image files, there are increasingly diverse ways to access images, such as on the cloud and on mobile devices. This has increased the concurrent access volume of the PACS system, prolonged the time for image access, and seriously affected the user experience of doctors and patients.

[0004] In order to improve the speed of image retrieval, a variety of technologies have emerged, such as: adding a front-end server between the client and the back-end storage, reducing the pressure on the back-end storage through the cache capacity of the front-end server, and shortening the access path when reading data to improve certain performance. Use SSD storage media with lower latency to provide higher reading performance; carry out network transformation, replace the original Gigabit network with a 10G network to increase network bandwidth; according to business scenarios, distribute data to multiple clusters, and divert traffic to different clusters when clients access.

[0005] However, the cache capacity of the front-end server is limited. If there is no cache hit, the performance will not be improved, and it can only solve the performance problem of the second and subsequent access. Improving the hardware configuration has improved the end-to-end user experience to a certain extent, but the performance of the back-end storage has grown much faster than the PACS system in recent years, and the performance of the front-end and back-end is seriously unbalanced. Business diversion can only achieve diversion at three large granularities: online, near-line, and offline. What really affects the user experience of doctors and patients is the access to online data.

[0006] In addition, when reviewing images, a large number of concurrent requests from the front end are transmitted through the FTP protocol. Although the performance of the back-end storage is constantly improving, the number of concurrent connections of FTP is limited, resulting in the transmission speed not keeping up with the improvement of the storage system performance. Summary of the invention

[0007] In view of the above analysis, an embodiment of the present invention aims to provide a medical PACS system for quickly accessing image files, so as to solve the problem that the existing medical PACS system is very slow when accessing massive image files.

[0008] The embodiment of the present invention provides a medical PACS system for quickly accessing image files, comprising: an image storage subsystem, a database subsystem and an image access subsystem; wherein the image storage subsystem comprises a pre-processing module, a data storage module and a metadata storage module; the pre-processing module merges the original image files uploaded by the image device into a large file in the data storage module according to a plurality of data merging strategies, obtains basic information of the original image files and sends it to the database subsystem, obtains metadata information and writes it into the metadata storage module; the database subsystem stores the basic information of the original image files sent by the pre-processing module by creating a metadata management table; the image access subsystem is used to load the large file to be accessed and metadata information from the image storage subsystem according to the access request of the client, and parses the original image files and displays them.

[0009] Based on further improvement of the above system, data merging strategies include: merging according to patient ID, merging according to patient ID and examination site, and merging according to examination date and time period.

[0010] Based on the further improvement of the above system, the pre-processing module merges the original image files uploaded by the imaging device into a large file in the data storage module, including:

[0011] Put the received original image files into the first-level merging queue;

[0012] Take out multiple original image files from the first-level merge queue, put them into the corresponding second-level merge queue according to the data merge strategy, and generate a large file prefix name corresponding to each second-level merge queue;

[0013] According to the number of concurrent operations, multiple original image files to be merged are taken out from the corresponding number of secondary merge queues at the same time, and the corresponding large files are found from the data storage module according to the corresponding large file prefix names, and the original image files to be merged are written into the corresponding large files in sequence.

[0014] Based on further improvements of the above system, basic information of the original image file is obtained and sent to the database subsystem, including: after taking out multiple original image files from the first-level merge queue, respectively calculating the first checksum of each original image file, and sending the name and the first checksum of each original image file as the basic data of each original image file to the database subsystem respectively.

[0015] Based on the further improvement of the above system, when writing the original image file to be merged into the corresponding large file, it also includes:

[0016] Calculate the second checksum of the original image file to be merged, compare the second checksum with the first checksum in the basic data, if they are consistent, write the original image file to be merged into the large file corresponding to the data storage module, obtain the metadata information according to the return information of the data storage module and write it into the metadata storage module; otherwise, the pre-processing module sends an error message in an asynchronous manner.

[0017] Based on further improvement of the above system, the metadata information includes: the name of the original image file to be merged, the name of the written large file, the offset position in the large file and the second checksum.

[0018] Based on further improvements of the above system, large files are obtained according to the storage type of the data storage module, including: when the storage type is object storage, the large file is a data object; when the storage type is block storage, the large file is a file system object on a logical volume or partition; when the storage type is file storage, the large file is a file under a directory in the file system.

[0019] Based on further improvement of the above system, the image review subsystem includes: an image retrieval module, a data management module and an image display module; the image retrieval module obtains the original image file from the server of the department to which the client belongs according to the client's retrieval request and returns it to the image display module for display. If the image file cannot be obtained, the retrieval request is sent to the data management module. After receiving the success message returned by the data management module, the original image file is obtained from the server of the department to which the client belongs and returned to the image display module for display.

[0020] Based on further improvement of the above system, the data management module includes: a data loading module and a data parsing module. The data loading module obtains the metadata information to be retrieved from the metadata storage module in the image storage subsystem according to the name of the original image file to be retrieved in the retrieval request, reads the large file to be retrieved from the data storage module according to the metadata information, and returns the large file to be retrieved and the metadata information to the data parsing module; the data parsing module parses the original image file from the large file to be retrieved according to the metadata information, and caches it in the department server to which the client belongs.

[0021] Based on the further improvement of the above system, the data management module also includes a data cleaning module for regularly deleting the original image files whose cache time exceeds a threshold in the servers of each department.

[0022] Compared with the prior art, the present invention can achieve at least one of the following beneficial effects:

[0023] 1. Improve the existing medical PACS system, use the pre-processing module to merge and write the original image files according to a variety of flexible data merging strategies to suit a variety of medical scenarios; convert the large concurrent download requests of massive small files into requests to download a small number of large files, thereby solving the problem of low concurrency of the FTP protocol and the inability to fully utilize the back-end storage performance, greatly shortening the image retrieval time, and using the stored metadata information to quickly parse the original image files.

[0024] 2. Improve processing efficiency by merging writes through multi-level queues; generate large file prefixes based on data merging strategies to improve retrieval efficiency of large files; store metadata information in a variety of ways to ensure the integrity of file writing.

[0025] In the present invention, the above-mentioned technical solutions can also be combined with each other to achieve more preferred combination solutions. Other features and advantages of the present invention will be described in the subsequent description, and some advantages can become obvious from the description, or can be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained through the contents particularly pointed out in the description and the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The drawings are only for the purpose of illustrating particular embodiments and are not to be considered limiting of the present invention. Like reference symbols denote like components throughout the drawings.

[0027] Figure 1 A schematic diagram of the structure of a medical PACS system for quickly accessing image files in an embodiment of the present invention;

[0028] Figure 2 Schematic diagram of the structure of the image review subsystem in the medical PACS system in an embodiment of the present invention. DETAILED DESCRIPTION

[0029] The preferred embodiments of the present invention are described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not used to limit the scope of the present invention.

[0030] A specific embodiment of the present invention discloses a medical PACS system for quickly accessing image files, such as Figure 1As shown, the system includes: an image storage subsystem, a database subsystem and an image review subsystem; wherein the image storage subsystem includes a pre-processing module, a data storage module and a metadata storage module; the pre-processing module merges the original image files uploaded by the imaging device into a large file in the data storage module according to a variety of data merging strategies, obtains the basic information of the original image files and sends it to the database subsystem, obtains the metadata information and writes it into the metadata storage module; the database subsystem stores the basic information of the original image files sent by the pre-processing module by creating a metadata management table; the image review subsystem is used to load the large file and metadata information to be reviewed from the image storage subsystem according to the review request of the client, and parses the original image files and displays them.

[0031] It should be noted that the medical PACS system receives image files (i.e., image data) uploaded by various medical imaging devices (such as CT, MRI equipment, X-ray machines, infrared devices, microscopes, etc.), processes and stores these image files through the image storage subsystem; uniformly stores and manages patient information and their electronic medical record information through various relational databases in the database subsystem, and associates image files in the image storage subsystem; and queries and displays patients and their image files through the image retrieval subsystem.

[0032] In this embodiment, in order to facilitate doctors to quickly review patient information, after receiving the uploaded image file but before storage, the pre-processing module merges the original image file into the corresponding large file in the data storage module according to the data merging strategy, which can not only improve the image review speed, but also reduce the I / O load of the storage system and improve the utilization of storage space.

[0033] Furthermore, the data merging strategies in this embodiment include: merging by patient ID, merging by patient ID and examination site, and merging by examination date and time period. Among them, merging by patient ID is to merge the image files of the same patient into one or more large files; this merging strategy takes into account that in real business scenarios, when doctors in various departments review images, sequentially reviewing the images of a certain patient is the most frequently used scenario.

[0034] Merging by patient ID and examination site means splitting the examination results of the same patient into multiple sites, and then merging them into one or more large files by site. This merging strategy takes into account that department doctors do not need to load the full data of the patient when viewing image files by site, which provides great flexibility.

[0035] Merging by examination date and time period means merging the image files of multiple patients received in the same time period on the same examination date into one or more large files. This merging strategy takes into account that some examination results are available relatively quickly and the number of patients in the hospital is called sequentially. When loading the image file of one patient, the image files of other patients in the same time period are pre-loaded. For example, the examination date is set by day and the time period is set by hour.

[0036] The above three merging strategies have their own adaptive application scenarios, and can be flexibly set according to the type of imaging equipment and the characteristics of department visits during actual application.

[0037] Specifically, the pre-processing module merges the original image files uploaded by the imaging device into a large file in the data storage module, including:

[0038] ① Put the received original image files into the first-level merging queue.

[0039] It should be noted that the header information of the received original image file includes: patient ID, patient name, examination date, examination time and examination part.

[0040] The length of the first-level merge queue is initialized according to a preset configuration item. Exemplarily, the length of the first-level merge queue is initialized to 1024. In a distributed application scenario, there may be multiple first-level merge queues, each queue being used to store different imaging devices or different types of imaging files.

[0041] ② Take out multiple original image files from the first-level merge queue, put them into the corresponding second-level merge queue according to the data merge strategy, and generate a large file prefix name corresponding to each second-level merge queue.

[0042] It should be noted that in this embodiment, the prefix name of the large file to be written to the original image file is generated by the pre-processing module according to the data merging strategy: when the strategy of merging is based on the patient ID, the prefix name is the patient ID; when the strategy of merging is based on the patient ID and the examination site, the prefix name is obtained by concatenating the patient ID and the examination site with a preset connector; when merging is based on the examination date and time period, the prefix name is obtained by concatenating the examination date and time period with a preset connector. Generating the prefix name of the large file in this way facilitates quick retrieval in the data storage module to obtain the required large file.

[0043] ③ According to the number of concurrent operations, multiple original image files to be merged are taken out from the corresponding number of secondary merge queues at the same time, and the corresponding large files are found from the data storage module according to the corresponding large file prefix name, and the original image files to be merged are written into the corresponding large files in sequence.

[0044] It should be noted that since each large file in the data storage module cannot exceed the maximum storage threshold, when it exceeds the threshold, a new large file will be created. The name of the large file is composed of a prefix name and a unique identification ID. For example, the unique identification ID uses a self-incrementing method to generate a globally unique value.

[0045] When writing, quickly search based on the large file prefix name to determine whether there is a large file with the same prefix name but not fully written in the data storage module, and the remaining space is larger than the original image file to be merged. If so, write the original image file to be merged at the end of the large file; otherwise, create a new large file, generate a new unique identification ID and concatenate it with the prefix name as the new large file name, and then write the original image file to be merged at the end of the new large file.

[0046] Preferably, in order to avoid the possibility that the original image file contained in the large file may be missing or the content of the original file may be erroneous due to various anomalies during the merge writing process, this embodiment calculates and compares the checksum of the original image file to promptly discover errors in the merge writing process.

[0047] Specifically, after taking out multiple original image files from the first-level merge queue in step ②, the first checksum of each original image file is calculated respectively, and the name and the first checksum of each original image file are sent to the database subsystem as the basic data of each original image file. The calculation method of the first checksum includes: using MD5 to generate a hash value of the data, or using a cyclic redundancy check method to generate a CRC checksum. The database subsystem records the name of each original image file and its first checksum by creating a metadata management table.

[0048] When writing the original image file to be merged into the corresponding large file in step ③, first calculate the second checksum of the original image file to be merged, and compare the second checksum with the first checksum in the basic data. If they are consistent, write the original image file to be merged into the large file corresponding to the data storage module, and obtain the metadata information according to the return information of the data storage module and write it into the metadata storage module; otherwise, the pre-processing module sends an error message in an asynchronous manner.

[0049] It should be noted that after the original image file to be merged is successfully written to the end of the large file, the data storage module returns the offset position of the original image file in the large file to facilitate subsequent parsing of the original image file from the large file.

[0050] Therefore, the metadata information finally obtained includes: the name of the original image file to be merged, the name of the written large file, the offset position in the large file and the second checksum; the obtained metadata information is written into the metadata storage module.

[0051] It is worth noting that this embodiment does not limit the format of large files. Large files are obtained according to the storage type of the data storage module, including: when the storage type is object storage, the large file is a data object; when the storage type is block storage, the large file is a file system object on a logical volume or partition; when the storage type is file storage, the large file is a file under a directory in the file system.

[0052] Likewise, the metadata storage module flexibly implements the storage of metadata information according to the storage type.

[0053] During implementation, doctors in each department send access requests to the image access subsystem through the client, such as Figure 2 As shown, the image review subsystem includes: an image review module, a data management module and an image display module. Among them, the image review module obtains the original image file from the department server of the client according to the review request of the client and returns it to the image display module for display. If it cannot be obtained, the review request is sent to the data management module. After receiving the success message returned by the data management module, the original image file is obtained from the department server of the client and returned to the image display module for display.

[0054] It should be noted that since the hospital has many departments, multiple department servers are deployed as local servers according to department type or floor to cache retrieved image files to improve loading speed.

[0055] Furthermore, the data management module includes: a data loading module and a data parsing module, wherein the data loading module obtains the metadata information to be reviewed from the metadata storage module in the image storage subsystem according to the name of the original image file to be reviewed in the review request, reads the large file to be reviewed from the data storage module according to the metadata information, and returns the large file to be reviewed and the metadata information to the data parsing module; the data parsing module parses the original image file from the large file to be reviewed according to the metadata information, and caches it in the department server to which the client belongs.

[0056] It should be noted that, when obtaining the metadata information to be reviewed from the metadata storage module in the image storage subsystem according to the name of the image file to be reviewed, not only the metadata information corresponding to the image file to be reviewed is obtained, but also other metadata information with the same "written large file name" is obtained according to the "written large file name" in the metadata information. This facilitates parsing all the original image files contained in the large file according to the offset positions in all the corresponding metadata information when parsing the large file, thereby preloading other original image files, which is convenient for improving the efficiency of doctors comparing other image files of patients, or reviewing image files of other patients in the same time period.

[0057] The large file to be retrieved is read from the data storage module according to the large file name in the metadata information, and all the original image files therein are retrieved by the data parsing module. Preferably, a third checksum is calculated for each of the original image files retrieved, and the third checksum is compared with the second checksum in the obtained metadata information to see if they are consistent, so as to timely discover abnormalities in data transmission. If they are consistent, the resolved original image file is stored in the department server to which the client belongs, and the image retrieval module is notified to retrieve the original image file from the department server to which the client belongs and return it to the image display module for display.

[0058] Furthermore, the data management module also includes a data cleaning module for regularly deleting original image files whose cache time exceeds a threshold in the servers of each department to release temporary storage space.

[0059] It should be noted that the image review subsystem not only reviews and displays image files, but is also used to query and display other information about the patient. Therefore, the image review subsystem also interacts with the database subsystem. For example, when doctors in various departments view the patient's electronic medical records, they send electronic medical record query requests to the image review module in the image review subsystem through the client. The image review module obtains the electronic medical record information from the data management module and returns it to the image display module for display. The data loading module in the data management module queries the electronic medical record information from the database subsystem and returns it to the image review module.

[0060] Compared with the prior art, the medical PACS system for quickly accessing image files provided in this embodiment improves the existing medical PACS system, and uses a pre-processing module to merge and write the original image files according to a variety of flexible data merging strategies to fit a variety of medical scenarios; it converts the request for downloading a large number of small files in large concurrency into a request for downloading a small number of large files, thereby solving the problem that the FTP protocol has low concurrency and cannot give full play to the back-end storage performance, greatly shortens the image access time, and uses the stored metadata information to quickly parse the original image file. Merge and write through multi-level queues to improve processing efficiency; generate large file prefix names according to data merging strategies to improve the retrieval efficiency of large files; and store metadata information in a variety of ways to ensure the integrity of file writing.

[0061] Those skilled in the art will appreciate that all or part of the processes of the above-mentioned embodiments can be implemented by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium, wherein the computer-readable storage medium is a disk, an optical disk, a read-only storage memory, or a random access memory, etc.

[0062] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. A medical PACS system for quickly accessing image files, characterized in that: include: An image storage subsystem, a database subsystem and an image query subsystem; wherein the image storage subsystem includes a pre-processing module, a data storage module and a metadata storage module, wherein the pre-processing module merges the original image files uploaded by the imaging device into a large file in the data storage module according to a variety of data merging strategies, obtains the basic information of the original image files and sends it to the database subsystem, and obtains the metadata information and writes it into the metadata storage module; The database subsystem stores the basic information of the original image file sent by the pre-processing module by creating a metadata management table; the image review subsystem is used to load the large file and metadata information to be reviewed from the image storage subsystem according to the client's review request, and parse the original image file and display it.

2. The medical PACS system for quickly accessing image files according to claim 1 is characterized in that: The data merging strategies include: merging according to patient ID, merging according to patient ID and examination site, and merging according to examination date and time period.

3. The medical PACS system for quickly accessing image files according to claim 2 is characterized in that: The pre-processing module merges the original image files uploaded by the imaging device into a large file in the data storage module, including: Put the received original image files into the first-level merging queue; Taking out a plurality of original image files from the first-level merge queue, putting them into corresponding second-level merge queues according to the data merge strategy, and generating a large file prefix name corresponding to each second-level merge queue; According to the number of concurrent operations, multiple original image files to be merged are taken out from the corresponding number of secondary merge queues at the same time, and the corresponding large files are found from the data storage module according to the corresponding large file prefix names, and the original image files to be merged are written into the corresponding large files in sequence.

4. The medical PACS system for quickly accessing image files according to claim 3 is characterized in that: The basic information of the original image file is obtained and sent to the database subsystem, including: after taking out multiple original image files from the first-level merge queue, respectively calculating the first checksum of each original image file, and sending the name and the first checksum of each original image file as the basic data of each original image file to the database subsystem respectively.

5. The medical PACS system for quickly accessing image files according to claim 4 is characterized in that: When the original image files to be merged are written into the corresponding large files, the process further includes: Calculate the second checksum of the original image file to be merged, compare the second checksum with the first checksum in the basic data, if they are consistent, write the original image file to be merged into the large file corresponding to the data storage module, obtain metadata information according to the return information of the data storage module and write it into the metadata storage module; otherwise, the pre-processing module sends an error message in an asynchronous manner.

6. The medical PACS system for quickly accessing image files according to claim 5 is characterized in that: The metadata information includes: the name of the original image file to be merged, the name of the large file to be written, the offset position in the large file, and the second checksum.

7. The medical PACS system for quickly accessing image files according to claim 5, characterized in that: The large file is obtained according to the storage type of the data storage module, including: when the storage type is object storage, the large file is a data object; when the storage type is block storage, the large file is a file system object on a logical volume or partition; when the storage type is file storage, the large file is a file under a directory in the file system.

8. The medical PACS system for quickly accessing image files according to claim 1 or 6, characterized in that: The image review subsystem includes: an image review module, a data management module and an image display module; the image review module obtains the original image file from the server of the department to which the client belongs according to the review request of the client and returns it to the image display module for display. If the original image file cannot be obtained, the review request is sent to the data management module. After receiving the success message returned by the data management module, the original image file is obtained from the server of the department to which the client belongs and returned to the image display module for display.

9. The medical PACS system for quickly accessing image files according to claim 8, characterized in that: The data management module includes: a data loading module and a data parsing module. The data loading module obtains metadata information to be reviewed from the metadata storage module in the image storage subsystem according to the name of the original image file to be reviewed in the review request, reads the large file to be reviewed from the data storage module according to the metadata information, and returns the large file to be reviewed and the metadata information to the data parsing module; the data parsing module parses the original image file from the large file to be reviewed according to the metadata information, and caches it in the department server to which the client belongs.

10. The medical PACS system for quickly accessing image files according to claim 9, characterized in that: The data management module also includes a data cleaning module for regularly deleting original image files whose cache time exceeds a threshold in the servers of each department.

Images