A cloud film implementation method and system

Abstract

The present invention discloses a cloud film implementation method and system, comprising: obtaining an examination number input by a patient; querying a printing host that prints the patient's film and an archived printed film to be segmented based on the examination number; retrieving a segmentation model library corresponding to the printing host, inputting the archived printed film to be segmented into the segmentation model library, and matching the segmentation specifications of the archived film to be segmented; performing image segmentation on the archived film to be segmented based on the segmentation specifications to obtain a segmented film image, and returning the segmented film image to the patient via a front-end processor. The present invention can segment printed films, effectively reducing the amount of downloaded data and shortening download time.

Description

Technical Field

[0001] The present invention relates to a cloud film implementation method and system, belonging to the technical field of medical devices. Background Art

[0002] With the rapid development of medical technology and hospital informatization, a growing consensus is emerging that the value of medical imaging should extend beyond the radiology department and encompass the entire healthcare system. In addition to providing effective data support for clinical diagnoses within the hospital, medical imaging also plays a crucial role in promoting regional healthcare systems and remote medical consultations. Traditional film, not only is costly and maintenance-intensive, but it's also inconvenient to carry, hindering the flow of regional examinations. The emergence of "cloud film" helps reduce operating costs, break down silos between examination data, and enable interconnected and interoperable examination data.

[0003] The cloud film manufacturers usually implement this method as follows: the patient first scans the QR code on the report. After verifying the patient's basic information, the mobile phone sends a call request to the hospital system front-end processor. After receiving the call request, the front-end processor downloads all the examination images stored in the PACS and transmits them to the mobile phone browsing interface. The advantages of this method are: the film obtained by the patient is a full-sequence film, and the mobile phone interface can display professional scales. The disadvantages are: due to the direct access to the stored data in the PACS system, the patient downloads a large amount of data, consumes more resources, and takes a long time to download. The images have not been confirmed by the technician, and the downloaded images may include a "B or C" image, which interferes with the clinician's diagnosis. Summary of the Invention

[0004] The purpose of the present invention is to overcome the shortcomings of the existing technology and provide a cloud film implementation method and system that can segment printed films, effectively reducing the amount of downloaded data and shortening the download time. To achieve the above purpose, the present invention is implemented by adopting the following technical solutions:

[0005] In a first aspect, the present invention provides a method for implementing a cloud film, comprising:

[0006] Get the examination number entered by the patient;

[0007] According to the examination number, query the printing host that prints the patient film and the archived printed films to be split;

[0008] Retrieve the segmentation model library corresponding to the printing host, input the archived printed film to be segmented into the segmentation model library, and match the segmentation specifications of the archived film to be segmented;

[0009] The archived film to be segmented is segmented based on the segmentation specifications to obtain segmented film images, which are then returned to the patient via the front-end processor.

[0010] In combination with the first aspect, further, the segmentation model library is obtained by the following steps:

[0011] Obtain all printing specification images of the printing host, feature images under different specifications, edge parameters of the printing host, and the IP address of the printing host;

[0012] Calculate the standard correlation matching coefficient R of the feature images under different printing specifications in the corresponding original images (x,y) ;

[0013] The calculated standard correlation matching coefficient is compared with the preset threshold R 0(x,y) Compare and confirm the position of the feature image in the original image;

[0014] Expand the selected feature image by two pixels to the left and right at the feature position coordinates of the film;

[0015] Establish a segmentation model library with different printing hosts, printing specifications, feature images, feature positions and edge parameters.

[0016] In combination with the first aspect, further, selecting a feature image that matches the original image includes:

[0017] When R (x,y) ≥R 0(x,y) , then the feature image matches the original image;

[0018] When R (x,y) <R 0(x,y) , then the feature image does not match the original image, and the feature image and its feature position on the film are deleted.

[0019] In combination with the first aspect, further, the expansion of two pixels is used to expand the feature position of the feature image to increase robustness.

[0020] In combination with the first aspect, further, the matching of the segmentation specifications of the archived film to be segmented includes:

[0021] Obtain a feature image from a segmentation model library corresponding to the printing host;

[0022] By calling the characteristic image positions of different printing specifications under the host, and cutting out images on the image to be segmented according to the characteristic image positions and specifications;

[0023] The captured images of different printing specifications are subjected to OCR text recognition and the recognition results are counted; in response to the OCR similarity between the recognized text result and the feature image, the corresponding number of times of meeting the specifications is increased, and finally the image segmentation specifications are confirmed by calculating the proportion of meeting the specifications.

[0024] In combination with the first aspect, further, the standard correlation matching coefficient R is used1(x,y) Confirm the location of the feature template in the image.

[0025] In a second aspect, the present invention provides a cloud film implementation system, comprising:

[0026] Acquisition module: used to obtain the examination number entered by the patient;

[0027] Query module: used to query the printing host that prints the patient film and the archived printed films to be split according to the examination number;

[0028] Matching module: used to retrieve the segmentation model library corresponding to the printing host, input the archived printed film to be segmented into the segmentation model library, and match the segmentation specifications of the archived film to be segmented;

[0029] Segmentation output module: used to perform image segmentation on the archived film to be segmented based on the segmentation specifications, obtain the segmented film image, and return the segmented film image to the patient through the front-end machine.

[0030] Compared with the prior art, the cloud film implementation method and system provided by the embodiments of the present invention have the following beneficial effects:

[0031] The present invention obtains an examination number input by a patient; based on the examination number, queries the printing host that prints the patient's film and the archived printed film to be segmented; retrieves a segmentation model library corresponding to the printing host, inputs the archived printed film to be segmented into the segmentation model library, and matches the segmentation specifications of the archived film to be segmented; performs image segmentation on the archived film to be segmented based on the segmentation specifications to obtain a segmented film image, which is then returned to the patient via a front-end processor. Images downloaded by patients are processed by the segmentation model library to avoid including "Class B and Class C" images in the downloaded images; the present invention can segment printed films, effectively reducing the amount of downloaded data and shortening download time. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 This is a flow chart of a method for implementing a cloud film provided in Example 1 of the present invention;

[0033] Figure 2 This is the film to be segmented and printed in the cloud film implementation method provided in the first embodiment of the present invention;

[0034] Figure 3 The position distribution of the characteristic image in the printed film to be segmented in the cloud film implementation method provided in the first embodiment of the present invention;

[0035] Figure 4 The printed film to be divided after the edges are trimmed in the cloud film implementation method provided in the first embodiment of the present invention;

[0036] Figure 5 It is a film image after segmentation in a cloud film implementation method provided in the first embodiment of the present invention;

[0037] Figure 6 Schematic diagram of 16 rows of feature images and margin pixels of United Imaging in Example 2 of the present invention;

[0038] Figure 7 This is a characteristic diagram of the threshold distribution of the United Imaging 16-row CT sub-station in Example 2 of the present invention. DETAILED DESCRIPTION

[0039] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention and are not intended to limit the scope of protection of the present invention.

[0040] Example 1:

[0041] An embodiment of the present invention provides a method for implementing a cloud film, including: establishing a segmentation model library and calling the segmentation model library.

[0042] Build a segmentation model library, including:

[0043] Step 1: Obtain all printing specification images of the printing host, feature images under different specifications, edge parameters of the printing host, and the IP address of the printing host.

[0044] Step 2: Calculate the standard correlation matching coefficient R of the feature images under different printing specifications in the corresponding original images (x,y) .

[0045]

[0046]

[0047]

[0048] T'(x',y') represents the template image, I(x,y) represents the original image, w is the pixel width of the template, and h is the pixel height of the template. I'(x+x',y+y') can be iterated to obtain it in Formula 3.

[0049] Step 3: Calculate the standard correlation matching coefficient R (x,y) and the preset threshold R 0(x,y) Compare and confirm the position of the feature image in the original image.

[0050] When R (x,y) ≥R 0(x,y) , then the feature image matches the original image;

[0051] When R (x,y) <R 0(x,y), then the feature image does not match the original image, and the feature image and its feature position on the film are deleted.

[0052] Step 4: Expand the selected feature image by two pixels to the left and right at the feature position coordinates of the film.

[0053] Expanding two pixels is used to expand the feature position of the feature image and increase robustness.

[0054] Step 5: Establish a segmentation model library with different printing hosts, printing specifications, feature images, feature positions and edge parameters.

[0055] like Figure 1 As described above, calling the segmentation model library includes:

[0056] Step 1: Get the examination number entered by the patient.

[0057] The patient checks the examination information on the browsing interface and enters the examination number.

[0058] Step 2: According to the examination number, query the printing host and Figure 2 The archived print films to be split are shown.

[0059] Query the data of the self-service film printer system through the patient's examination number to confirm the host computer that prints the patient's film and the storage path of the printed film.

[0060] Step 3: Retrieve the segmentation model library corresponding to the printing host, input the archived printed film to be segmented into the segmentation model library, and match the segmentation specifications of the archived film to be segmented.

[0061] Call all feature images and corresponding thresholds under the host from the segmentation model library corresponding to the printing host.

[0062] By calling the characteristic image position of different printing specifications under the host, and cutting out the image on the image to be segmented according to the characteristic image position and specifications, such as Figure 3 Specifically, the standard correlation matching coefficient R is used. 1(x,y) Confirm the location of the feature template in the image.

[0063] The captured images of different printing specifications are subjected to OCR text recognition and the recognition results are counted; in response to the OCR similarity between the recognized text result and the feature image, the corresponding number of times of meeting the specifications is increased, and finally the image segmentation specifications are confirmed by calculating the proportion of meeting the specifications.

[0064] After confirming the splitting specifications, first cut the edges of the archived film to be split, such as Figure 4 Then call the segmentation specification parameters to segment the image, such as Figure 5Finally, the segmented film image is returned to the patient through the front-end processor, ultimately realizing the function of cloud film.

[0065] The present invention can segment the printed film, effectively reducing the amount of downloaded data and shortening the downloading time.

[0066] Example 2:

[0067] This embodiment adopts the cloud film implementation method described in the first embodiment, takes the printing host as the United Imaging 16-row sub-station as an example to establish a segmentation model library, and calls the segmentation model library to segment the printed film.

[0068] 1. Establish a segmentation model library:

[0069] Collect all printing specifications of the United Imaging 16-row CT sub-station images and feature images of different specifications, and measure the margin pixels of different specifications of the 16-row CT sub-station. The specific results are as follows Figure 6 shown.

[0070] Calculate the standard correlation matching coefficient between the feature image and the original film; set the threshold of the standard correlation matching coefficient to 0.85, use the threshold to filter out the unmatched data, and record the coordinates of the template on the original film. The specific results are as follows: Figure 7 shown.

[0071] The generated feature image is expanded by two pixels at the feature position coordinates of the film to increase the robustness of the inspection model. For example, the first feature point [336,155] of the 5*5 specification of the United Imaging 16-row CT auxiliary table is expanded to [[334,155], [335,155], [337,155], [338,155], [334,153], [335,153], [337,153], [338,153], [334,154], [335,154], [337,154], [338,154], [334,156], [335,156], [337,156], [338,156], [334,157], [335,157], [337,157], [338,157]].

[0072] 2. Call the segmentation model library to segment the printed film:

[0073] For example, if a patient enters the examination number CT221022017, the examination number independent film printer database confirms that the main unit printing the patient's image is a United Imaging 16-row CT auxiliary station. Based on the characteristic image distribution of all printing specifications of the 16-row auxiliary station, the image to be segmented is intercepted. The intercepted image is subjected to OCR recognition, and the accuracy of the recognition is statistically analyzed.

[0074] Finally, the 5*6 image size that meets the specifications is the largest, so the size of the image to be segmented can be set to 5*6. After confirming the specifications, the image to be segmented is cropped using the margin pixel size, and the segmented image is then transmitted to the front-end processor and finally to the patient's mobile phone interface.

[0075] Example 3:

[0076] An embodiment of the present invention provides a cloud film implementation system, including:

[0077] Acquisition module: used to obtain the examination number entered by the patient;

[0078] Query module: used to query the printing host that prints the patient film and the archived printed films to be split according to the examination number;

[0079] Matching module: used to retrieve the segmentation model library corresponding to the printing host, input the archived printed film to be segmented into the segmentation model library, and match the segmentation specifications of the archived film to be segmented;

[0080] Segmentation output module: used to perform image segmentation on the archived film to be segmented based on the segmentation specifications, obtain the segmented film image, and return the segmented film image to the patient through the front-end machine.

[0081] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0082] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0083] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0084] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0085] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for realizing cloud film, characterized in that: include: Get the examination number entered by the patient; According to the examination number, query the printing host that prints the patient film and the archived printed films to be split; Retrieve the segmentation model library corresponding to the printing host, input the archived printed film to be segmented into the segmentation model library, and match the segmentation specifications of the archived film to be segmented; Perform image segmentation on the archived film to be segmented based on the segmentation specifications to obtain segmented film images, and return the segmented film images to the patient through the front-end processor; The segmentation model library is established by the following steps, including: Step 1: Obtain all printing specification images of the printing host, feature images under different specifications, edge parameters of the printing host, and the IP address of the printing host; Step 2: Calculate the standard correlation matching coefficient of the feature images under different printing specifications in the corresponding original images R (x,y) , （1） （2） （3） represents the template image, represents the original image, w The pixel width of the template, h The height of the template pixel, where It can be iterated to Publication 3 to find out; Step 3: Calculate the standard correlation matching coefficient R (x,y) With the preset threshold R 0(x,y) Compare and confirm the position of the feature image in the original image: when R (x,y) ≥ R 0(x,y) , then the feature image matches the original image; when R (x,y) < R 0(x,y) , then the feature image does not match the original image, and the feature image and its feature position on the film are deleted; Step 4: Expand the selected feature image by two pixels to the left and right at the feature position coordinates of the film; Expanding two pixels is used to expand the feature position of the feature image and increase robustness; Step 5: Establish a segmentation model library with different printing hosts, printing specifications, feature images, feature positions and edge parameters.

2. The method for realizing cloud film according to claim 1, characterized in that: The segmentation specifications of the matched archived film to be segmented include: Obtain a feature image from a segmentation model library corresponding to the printing host; By calling the characteristic image positions of different printing specifications under the host, and cutting out images on the image to be segmented according to the characteristic image positions and specifications; The captured images of different printing specifications are subjected to OCR text recognition and the recognition results are counted; in response to the OCR similarity between the recognized text result and the feature image, the corresponding number of times of meeting the specifications is increased, and finally the image segmentation specifications are confirmed by calculating the proportion of meeting the specifications.

3. The method for realizing cloud film according to claim 2, characterized in that: Standard correlation matching coefficient R 1(x,y) Confirm the location of the feature template in the image.

4. A cloud film implementation system, characterized in that: include: Acquisition module: used to obtain the examination number entered by the patient; Query module: used to query the printing host that prints the patient film and the archived printed films to be split according to the examination number; Matching module: used to retrieve the segmentation model library corresponding to the printing host, input the archived printed film to be segmented into the segmentation model library, and match the segmentation specifications of the archived film to be segmented; Segmentation output module: used to segment the archived film to be segmented based on the segmentation specifications, obtain the segmented film image, and return the segmented film image to the patient through the front-end processor; The segmentation model library is established by the following steps, including: Step 1: Obtain all printing specification images of the printing host, feature images under different specifications, edge parameters of the printing host, and the IP address of the printing host; Step 2: Calculate the standard correlation matching coefficient of the feature images under different printing specifications in the corresponding original images R (x,y) , （1） （2） （3） represents the template image, represents the original image, w The pixel width of the template, h The height of the template pixel, where It can be iterated to Publication 3 to find out; Step 3: Calculate the standard correlation matching coefficient R (x,y) With the preset threshold R 0(x,y) Compare and confirm the position of the feature image in the original image: when R (x,y) ≥ R 0(x,y) , then the feature image matches the original image; when R (x,y) < R 0(x,y) , then the feature image does not match the original image, and the feature image and its feature position on the film are deleted; Step 4: Expand the selected feature image by two pixels to the left and right at the feature position coordinates of the film; Expanding two pixels is used to expand the feature position of the feature image and increase robustness; Step 5: Establish a segmentation model library with different printing hosts, printing specifications, feature images, feature positions and edge parameters.

Images