Medical traditional film printing method based on intelligent film typesetting

Abstract

The invention relates to film printing, in particular to a medical traditional film printing method based on intelligent film typesetting, which comprises the following steps: after a patient completes image shooting, an imaging department encrypts an original image and transmits the original image to a film printing system; the film printing system receives and decrypts the encrypted original image, and matches the original image with the patient information; the film printing system performs intelligent typesetting on the original image, and encrypts and transmits the typeset image to the cloud for storage; a user checks the film which is stored in the cloud and contains the typeset image through code scanning at the client, and applies to print the film; the client generates a film taking two-dimensional code and sends the film taking two-dimensional code to the user terminal; a user sends a film printing request to the film printing system through code scanning on the self-service printer; the film printing system receives the film printing request and sends a printing command to the self-service printer; the defects that intelligent typesetting of the film is difficult and the film printing efficiency is low can be overcome.

Description

Technical Field

[0001] This invention relates to film printing, and more specifically to a method for printing traditional medical films based on intelligent film layout. Background Technology

[0002] Current medical traditional film printing process as follows Figure 4 As shown, where:

[0003] Image acquisition end: After CT / MR / DR and other equipment generate DICOM medical images, they are sent directly to the designated printer via the DICOM Print protocol. However, there are certain differences between the proprietary protocols of equipment manufacturers (such as GE Centricity, Siemens Syngo, etc.), resulting in poor compatibility of printing commands.

[0004] Information management terminal: Patient registration information is stored in the hospital PACS / RIS system (usually using the HL7 standard), which is physically isolated from the image data. The imaging technician needs to manually verify the patient's name and examination number.

[0005] Output execution end: Dry film printers (such as FUJIFILM DryPix, KODAK DryView, etc.) only receive raw images, lack intelligent layout functions, and rely heavily on preset templates.

[0006] Therefore, the traditional medical film printing process has the following main problems:

[0007] 1) Lack of intelligent layout function: The rule engine equipped in the current PACS system lacks intelligent layout function, making it difficult to adaptively optimize according to image parameters. When processing multi-sequence images (such as CT, MR, etc.), it cannot automatically integrate key frames. For example, the delayed phase and venous phase images of abdominal enhanced scans need to be manually screened and merged across sequences. At the same time, it cannot intelligently adapt when images of different resolutions are mixed. For example, the combined printing of high-resolution mammograms (89μm) and low-resolution DR images (200μm) is prone to pixel distortion. During the printing process, the bone window and brain window images of brain CT are not processed in a targeted manner, resulting in the compression and blurring of key lesion information, which affects the accuracy of diagnosis.

[0008] 2) Insufficient system coordination: There is a serious data silo between the PACS system, printing system, printing terminal and patient terminal, and a lack of a unified task scheduling mechanism. This system silo effect will cause resource scheduling chaos. Emergency printing tasks and ordinary printing tasks cannot be automatically allocated according to priority, which affects the efficiency of handling emergency cases. In addition, imaging technicians cannot obtain the information of patients scanning the code to apply for printing in real time, making it difficult to plan printing tasks in advance. Patients can only wait for the film to be printed in the hospital and cannot choose the time of film printing themselves, resulting in a poor medical experience.

[0009] 3) Significant data security risks: The traditional DICOM transmission protocol uses plaintext transmission, and patient privacy data (such as personal identification information, examination results, etc.) lacks encryption protection during transmission, posing an extremely high risk of information leakage and failing to meet the requirements of medical data security and personal privacy protection.

[0010] 4) Low efficiency of manual typesetting: Currently, the typesetting of traditional medical films relies on manual operation by imaging technicians. From setting the size to adjusting the image layout, each case takes an average of more than 15 minutes. The cumbersome manual process not only consumes a lot of human resources, but is also prone to typesetting errors due to human mistakes, increasing rework costs and severely limiting printing efficiency.

[0011] 5) Long patient waiting time: After the examination, patients need to wait for the radiology technician to manually process the film. Not only is the waiting time long, but the manual distribution of film is also prone to information verification errors, causing patients to wait repeatedly or receive the wrong film. The inefficient printing process prolongs the patient's medical treatment time and reduces the hospital's service efficiency and patient satisfaction.

[0012] 6) The closed-loop process of "online pre-layout - offline on-demand printing" has not been realized. Summary of the Invention

[0013] (a) Technical problems to be solved

[0014] In view of the above-mentioned shortcomings of the existing technology, the present invention provides a medical traditional film printing method based on intelligent film layout, which can effectively overcome the defects of the existing technology, such as difficulty in intelligently laying out films and low film printing efficiency.

[0015] (II) Technical Solution

[0016] To achieve the above objectives, the present invention provides the following technical solution:

[0017] The medical traditional film printing method based on intelligent film layout includes the following steps:

[0018] S1. After the patient undergoes examination in the radiology department and the imaging is completed, the radiology department encrypts and transmits the original images to the film printing system.

[0019] S2. The film printing system receives the encrypted original image and decrypts it, then matches the original image with the patient information.

[0020] The S3 film printing system intelligently typesets the original images and then encrypts and transmits the typeset images to the cloud for storage.

[0021] S4. Users can scan a QR code on the client to view the film containing the formatted images stored in the cloud and request to print the film.

[0022] S5. The client generates a QR code for picking up the film and sends the QR code to the user terminal.

[0023] S6. Users send film printing requests to the film printing system by scanning a code at the self-service printer;

[0024] S7. The film printing system receives the film printing request and sends the printing command to the self-service printer to complete the film printing.

[0025] Preferably, encrypted transmission in S1 and S3 includes:

[0026] The input image is encrypted, the encrypted input image is encrypted again using the data encapsulation key DEK, the data encapsulation key DEK is encrypted using the key encryption key KEK, and the key encryption key KEK is distributed.

[0027] Preferably, the step of encrypting the input image, encrypting the encrypted input image again using the data encapsulation key DEK, encrypting the data encapsulation key DEK using the key encryption key KEK, and distributing the key encryption key KEK includes:

[0028] S11. Encrypt the input image based on the AES-256-GCM encryption algorithm;

[0029] S12. Use the data encapsulation key DEK to encrypt the encrypted input image again;

[0030] S13. Based on the SM2-OAEP encryption algorithm, the key encryption key KEK is used to encrypt the data encapsulation key DEK;

[0031] S14. Distribute the key encryption key KEK and store it in hardware security module HSM node A and hardware security module HSM node B respectively to enhance the security of key management and prevent single point of failure.

[0032] Preferably, in S2, the film printing system receives and decrypts the encrypted original image, and matches the original image with patient information, including:

[0033] S21. The film printing system receives the encrypted original image and decrypts it.

[0034] S22. After the radiology department completes the image transmission, a task is automatically generated in the receiving list of the film printing system. This task contains only one original image record. The examination number and receiving time of the original image are identified through ORC.

[0035] S23. The film printing system enters the hospital's PACS system according to the task number, matches the patient information according to the examination number, and generates a list of films to be intelligently typed after the matching is completed.

[0036] Preferably, the film printing system in S3 performs intelligent layout of the original image and encrypts and transmits the layout image to the cloud for storage, including:

[0037] S31. Perform intelligent size layout on the original images in the intelligent layout list to obtain the physical size of the film and the size of a single image.

[0038] S32. Based on the physical dimensions of the film and the dimensions of a single image, construct the objective function and constraints for intelligent position layout;

[0039] S33. Based on the objective function and constraints of intelligent location layout, the Monte Carlo Tree Search (MCTS) algorithm is used to perform intelligent location layout on the images.

[0040] S34. Encrypt the formatted image and transmit it to the cloud for storage.

[0041] Preferably, in step S31, intelligent size layout is performed on the original images in the intelligent layout list to obtain the physical size of the film and the size of a single image, including:

[0042] The film width is calculated using the following formula. physical :

[0043] Width physical =Columns×PixelSpacing x ×0.03937;

[0044] Where Columns represents the number of pixel columns in the horizontal direction of the film, and PixelSpacing represents... x The center-to-center distance between adjacent pixels along the X-axis is 0.03937, which is the conversion factor from millimeters to inches.

[0045] The film height is calculated using the following formula. physical :

[0046] Height physical =Rows×PixelSpacing y ×0.03937;

[0047] Where Rows is the number of pixel rows in the vertical direction of the film, and PixelSpacing is... y The center-to-center distance between adjacent pixels along the Y-axis;

[0048] Based on the physical dimensions of the film and the size of a single image, S32 constructs the objective function and constraints for intelligent positional layout, including:

[0049] Input parameters: image set, film physical size, and single image size;

[0050] Objective function: To maximize film space utilization through algorithm optimization while meeting medical requirements;

[0051] Constraints: Regions of interest (ROIs) must not be cut to ensure that important areas, including lesions, are presented intact. Images on the film must be temporally adjacent, including images from enhanced CT scans, which must be arranged in the order of examination with an image spacing of no less than 2 mm, to prevent accidental cutting of image content during printing.

[0052] Preferably, in step S33, based on the objective function and constraints of intelligent location layout, the Monte Carlo Tree Search (MCTS) algorithm is used to perform intelligent location layout of the image, including:

[0053] Phase 1: Selection

[0054] 1) Tree structure definition operation: When the images are laid out, an initial node is first constructed. This node represents the current blank layout state of the film. Then the images are placed. Each time an image is placed, a new child node is generated. This child node records the placed images and the division of the remaining space after the image is placed.

[0055] 2) Selection strategy operation: At each node, the UCT algorithm is used to calculate the location where the image can be placed;

[0056] Phase 2: Expansion

[0057] 1) Generate candidate action operations: In the current layout state, the remaining blank space is divided by an algorithm;

[0058] 2) Placement rules: Arrange images according to diagnostic priority;

[0059] Phase 3: Simulation

[0060] 1) Rapid valuation strategy operation: Prioritize placing the largest image in the remaining blank space to minimize space waste, and then randomly place important diagnostic images;

[0061] 2) Optimize layout scheme: For each simulated layout scheme, optimize according to the following formula:

[0062] Reward=α·utilization+β·ROI_integrity-γ·gap_violation;

[0063] Where Reward is the reward, α·utilization is the space utilization weighting term, utilization is the space utilization rate of the film, that is, the proportion of the total area of ​​the images placed to the total area of ​​the film, and α is the weighting coefficient of space utilization, α=.0.7;

[0064] β·ROI_integrity is the weighting term for the integrity of the region of interest (ROI), where ROI_integrity is the integrity of the ROI in the image, and β is the weighting coefficient for the integrity of the ROI, β = 0.3;

[0065] γ·gap_violation is the image spacing penalty term. gap_violation is used to measure whether the image spacing meets the requirement of not less than 2mm, to ensure the safe spacing between images and prevent the image content from being accidentally cut during printing. γ is the weighting coefficient of image spacing, γ=0.5;

[0066] Phase 4: Retrospection

[0067] Medical memory bank operation: Each simulated layout scheme is stored in the medical memory bank. When a similar image layout task is encountered again, the stored historical simulated layout scheme will be optimized in stage 1 to improve layout efficiency.

[0068] Preferably, in S4, the user scans a QR code on the client to view a film containing formatted images stored in the cloud and requests to print the film, including:

[0069] S41. The user scans the QR code on the client / paper report to complete the patient identity verification;

[0070] S42. The client requests data from the cloud so that the user can view film containing typed images stored in the cloud.

[0071] S43. The user requests to print film on the client.

[0072] Preferably, in S5, the client generates a QR code for retrieving the image and sends the QR code to the user terminal, including:

[0073] S51. The client generates the core token in the QR code for retrieving the image according to the following formula:

[0074]

[0075] Here, TaskID is the task ID, used to uniquely identify a business task or operation, TS is the timestamp-related value, / / indicates the concatenation operation, and SM4 CTR indicates the SM4 symmetric encryption algorithm working in counter mode. The task ID and the timestamp-related value TS are concatenated into a whole as the input of the SM4 symmetric encryption algorithm, so that each token is associated with the task and time.

[0076] Salt is a random string. The XOR operation is performed, which XORs the SM4 encryption result with the random string Salt to increase the randomness and unpredictability of the token. Base64 indicates that Base64 encoding is performed.

[0077] S52. The client generates a corresponding QR code for retrieving the chip based on the token and sends the QR code to the user terminal.

[0078] The S7 film printing system receives film printing requests and sends print commands to self-service printers to complete film printing, including:

[0079] The film printing system receives film printing requests and sends printing commands and printing data to the self-service printer to complete the film printing and store the printing records at the same time.

[0080] The film printing system automatically marks the anti-counterfeiting watermark "printing time + device ID" when sending printing data. The radiology department can log in to the film printing system to view printing records and printing quantities for daily departmental statistics.

[0081] (III) Beneficial Effects

[0082] Compared with existing technologies, the medical traditional film printing method based on intelligent film layout provided by this invention has the following beneficial effects:

[0083] 1) Intelligent typesetting function: Traditional PACS systems lack intelligent typesetting function, making it difficult to adaptively optimize according to image parameters. When images of different resolutions are mixed, they cannot be intelligently adapted, which easily leads to pixel distortion. Key lesion information may also be compressed and blurred due to the lack of targeted processing of window width and window level. This invention adopts an intelligent typesetting algorithm, combined with the medical optimization logic of Monte Carlo tree search, which can adaptively optimize according to image parameters and automatically adapt to images of different resolutions, effectively avoiding pixel distortion and blurring of key lesion information.

[0084] 2) Intelligent, cloud-based, and self-service film printing process: In the traditional model, there is a serious data silo between the PACS system, printing system, printing terminal, and patient terminal, lacking a unified task scheduling mechanism, and information is not shared between patients and radiology technicians; This invention realizes a closed-loop, cloud-based collaborative process by constructing a "first archive to the cloud, then print on demand" business process. Radiology technicians can obtain patients' printing request information in real time and print films in advance, while patients can choose the printing time themselves, which greatly improves system synergy and patient medical experience;

[0085] 3) Reduce manual intervention and improve work efficiency: Currently, the layout of traditional medical films relies on manual operation by imaging technicians, which is time-consuming, inefficient and prone to errors; This invention automates the setting of image size and layout through intelligent layout algorithms, effectively reducing the layout error rate and greatly improving printing efficiency.

[0086] 4) Enhance anti-counterfeiting technology to ensure the traceability of medical data: Traditional film printing lacks anti-counterfeiting technology, which is not conducive to medical traceability; this invention uses a unique anti-counterfeiting watermark generation method to generate an anti-counterfeiting watermark of "printing time + device ID", providing a reliable basis for the traceability of medical data. Attached Figure Description

[0087] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0088] Figure 1 This is a schematic diagram of the process of the present invention;

[0089] Figure 2 This is a schematic diagram of the encrypted transmission process in this invention;

[0090] Figure 3 For the present invention Figure 2 A detailed flowchart illustrating the encrypted transmission process;

[0091] Figure 4 This is a schematic diagram of the printing process for existing traditional medical films. Detailed Implementation

[0092] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0093] Traditional medical film printing methods based on intelligent film layout, such as Figure 1 As shown, S1, the patient undergoes an examination in the radiology department. After the imaging is completed, the radiology department encrypts and transmits the original images to the film printing system.

[0094] S2. The film printing system receives and decrypts the encrypted original image, and matches the original image with patient information, specifically including:

[0095] S21. The film printing system receives the encrypted original image and decrypts it.

[0096] S22. After the radiology department completes the image transmission, a task is automatically generated in the receiving list of the film printing system. This task contains only one original image record. The examination number and receiving time of the original image are identified through ORC.

[0097] S23. The film printing system enters the hospital's PACS system according to the task number, matches the patient information according to the examination number, and generates a list of films to be intelligently typed after the matching is completed.

[0098] The S3 film printing system intelligently typesets the original images and encrypts and transmits the typeset images to the cloud for storage. Specifically, this includes:

[0099] S31. Perform intelligent size layout on the original images in the intelligent layout list to obtain the physical size of the film and the size of a single image.

[0100] S32. Based on the physical dimensions of the film and the dimensions of a single image, construct the objective function and constraints for intelligent position layout;

[0101] S33. Based on the objective function and constraints of intelligent location layout, the Monte Carlo Tree Search (MCTS) algorithm is used to perform intelligent location layout on the images.

[0102] S34. Encrypt the formatted image and transmit it to the cloud for storage.

[0103] Specifically, in S31, the original images in the intelligent layout list are intelligently size-arranged to obtain the physical dimensions of the film and the dimensions of a single image, including:

[0104] The film width is calculated using the following formula. physical :

[0105] Width physical =Columns×PixelSpacing x ×0.03937;

[0106] Where Columns represents the number of pixel columns in the horizontal direction of the film, and PixelSpacing represents... x The center-to-center distance between adjacent pixels along the X-axis is 0.03937, which is the conversion factor from millimeters to inches.

[0107] The film height is calculated using the following formula. physical :

[0108] Height physical =Rows×PixelSpacing y ×0.03937;

[0109] Where Rows is the number of pixel rows in the vertical direction of the film, and PixelSpacing is... y This represents the center-to-center distance between adjacent pixels along the Y-axis.

[0110] Specifically, S32 constructs the objective function and constraints for intelligent positional layout based on the physical size of the film and the size of a single image, including:

[0111] Input parameters: image set, film physical size, and single image size;

[0112] Objective function: To maximize film space utilization through algorithm optimization while meeting medical requirements;

[0113] Constraints: Regions of interest (ROIs) must not be cut to ensure that important areas, including lesions, are presented intact. Images on the film must be temporally adjacent, including images from enhanced CT scans, which must be arranged in the order of examination with an image spacing of no less than 2 mm, to prevent accidental cutting of image content during printing.

[0114] Specifically, in S33, based on the objective function and constraints of intelligent location layout, the Monte Carlo Tree Search (MCTS) algorithm is used to perform intelligent location layout of the image, including:

[0115] Phase 1: Selection

[0116] 1) Tree structure definition operation: When the images are laid out, an initial node is first constructed. This node represents the current blank layout state of the film. Then the images are placed. Each time an image is placed, a new child node is generated. This child node records the placed image and the division of the remaining space after the image is placed. For example, after the first cranial CT image is placed, the new child node will record that the CT image has been placed, and the remaining space is divided into the blank area around the CT image.

[0117] 2) Selection strategy operation: At each node, the UCT algorithm is used to calculate the location where the image can be placed;

[0118] Phase 2: Expansion

[0119] 1) Generate candidate action operations: In the current layout state, the remaining blank space is divided by an algorithm;

[0120] 2) Placement rules: Arrange the images according to diagnostic priority (e.g., place the anteroposterior images first, then consider placing the lateral images, etc.);

[0121] Phase 3: Simulation

[0122] 1) Rapid valuation strategy operation: Prioritize placing the largest image in the remaining blank space to minimize space waste, and then randomly place important diagnostic images;

[0123] 2) Optimize layout scheme: For each simulated layout scheme, optimize according to the following formula:

[0124] Reward=α·utilization+β·ROI_integrity-γ·gap_violation;

[0125] Where Reward is the reward, α·utilization is the space utilization weighting term, utilization is the space utilization rate of the film, that is, the proportion of the total area of ​​the images placed to the total area of ​​the film, and α is the weighting coefficient of space utilization, α=.0.7;

[0126] β·ROI_integrity is the weighting term for the integrity of the region of interest (ROI), where ROI_integrity is the integrity of the ROI in the image, and β is the weighting coefficient for the integrity of the ROI, β = 0.3;

[0127] γ·gap_violation is the image spacing penalty term. gap_violation is used to measure whether the image spacing meets the requirement of not less than 2mm, to ensure the safe spacing between images and prevent the image content from being accidentally cut during printing. γ is the weighting coefficient of image spacing, γ=0.5;

[0128] Phase 4: Retrospection

[0129] Medical memory bank operation: Each simulated layout scheme is stored in the medical memory bank. When a similar image layout task is encountered again, the stored historical simulated layout scheme will be optimized in stage 1 to improve layout efficiency.

[0130] In the technical solution of this application, encrypted transmission in S1 and S3, such as Figure 2 As shown, it includes:

[0131] The input image is encrypted, the encrypted input image is encrypted again using the data encapsulation key DEK, the data encapsulation key DEK is encrypted using the key encryption key KEK, and the key encryption key KEK is distributed.

[0132] Specifically, the input image is encrypted, the encrypted input image is encrypted again using the data encapsulation key DEK, the data encapsulation key DEK is encrypted again using the key encryption key KEK, and the key encryption key KEK is then distributed, such as... Figure 3 As shown, it includes:

[0133] S11. Encrypt the input image based on the AES-256-GCM encryption algorithm;

[0134] S12. Use the data encapsulation key DEK to encrypt the encrypted input image again;

[0135] S13. Based on the SM2-OAEP encryption algorithm, the key encryption key KEK is used to encrypt the data encapsulation key DEK;

[0136] S14. Distribute the key encryption key KEK and store it in hardware security module HSM node A and hardware security module HSM node B respectively to enhance the security of key management and prevent single point of failure.

[0137] S4. Users can view pre-formatted images stored in the cloud by scanning a QR code on the client side and request to print the film, specifically including:

[0138] S41. The user scans the QR code on the client / paper report to complete the patient identity verification;

[0139] S42. The client requests data from the cloud so that the user can view film containing typed images stored in the cloud.

[0140] S43. The user requests to print film on the client.

[0141] S5. The client generates a QR code for retrieving the image and sends the QR code to the user terminal, specifically including:

[0142] S51. The client generates the core token in the QR code for retrieving the image according to the following formula:

[0143]

[0144] Here, TaskID is the task ID, used to uniquely identify a business task or operation, TS is the timestamp-related value, / / indicates the concatenation operation, and SM4 CTR indicates the SM4 symmetric encryption algorithm working in counter mode. The task ID and the timestamp-related value TS are concatenated into a whole as the input of the SM4 symmetric encryption algorithm, so that each token is associated with the task and time.

[0145] Salt is a random string. The XOR operation is performed, which XORs the SM4 encryption result with the random string Salt to increase the randomness and unpredictability of the token. Base64 indicates that Base64 encoding is performed.

[0146] S52. The client generates a corresponding QR code for retrieving the chip based on the token and sends the QR code to the user terminal.

[0147] S6. Users send film printing requests to the film printing system by scanning a code at the self-service printer.

[0148] S7. The film printing system receives the film printing request and sends the print command to the self-service printer to complete the film printing, specifically including:

[0149] The film printing system receives film printing requests and sends printing commands and printing data to the self-service printer to complete the film printing and store the printing records at the same time.

[0150] The film printing system automatically marks the anti-counterfeiting watermark "printing time + device ID" when sending printing data. The radiology department can log in to the film printing system to view printing records and printing quantities for daily departmental statistics.

[0151] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A medical traditional film printing method based on intelligent film layout, characterized by: Includes the following steps: S1. After the patient undergoes examination in the radiology department and the imaging is completed, the radiology department encrypts and transmits the original images to the film printing system. S2. The film printing system receives the encrypted original image and decrypts it, then matches the original image with the patient information. The S3 film printing system intelligently typesets the original images and then encrypts and transmits the typeset images to the cloud for storage. S4. Users can scan a QR code on the client to view the film containing the formatted images stored in the cloud and request to print the film. S5. The client generates a QR code for picking up the film and sends the QR code to the user terminal. S6. Users send film printing requests to the film printing system by scanning a code at the self-service printer; S7. The film printing system receives the film printing request and sends the printing command to the self-service printer to complete the film printing.

2. The method for transmitting and printing traditional medical films based on intelligent film layout according to claim 1, characterized in that: Encrypted transmissions in S1 and S3 include: The input image is encrypted, the encrypted input image is encrypted again using the data encapsulation key DEK, the data encapsulation key DEK is encrypted using the key encryption key KEK, and the key encryption key KEK is distributed.

3. The method for transmitting and printing traditional medical films based on intelligent film layout according to claim 2, characterized in that: The process of encrypting the input image involves encrypting the encrypted input image again using the data encapsulation key DEK, encrypting the data encapsulation key DEK using the key encryption key KEK, and then distributing the key encryption key KEK. This includes: S11. Encrypt the input image based on the AES-256-GCM encryption algorithm; S12. Use the data encapsulation key DEK to encrypt the encrypted input image again; S13. Based on the SM2-OAEP encryption algorithm, the key encryption key KEK is used to encrypt the data encapsulation key DEK; S14. Distribute the key encryption key KEK and store it in hardware security module HSM node A and hardware security module HSM node B respectively to enhance the security of key management and prevent single point of failure.

4. The method for transmitting and printing traditional medical films based on intelligent film layout according to claim 1, characterized in that: The S2 film printing system receives and decrypts the encrypted original images, then matches the original images with patient information, including: S21. The film printing system receives the encrypted original image and decrypts it. S22. After the radiology department completes the image transmission, a task is automatically generated in the receiving list of the film printing system. This task contains only one original image record. The examination number and receiving time of the original image are identified through ORC. S23. The film printing system enters the hospital's PACS system according to the task number, matches the patient information according to the examination number, and generates a list of films to be intelligently typed after the matching is completed.

5. The method for transmitting and printing traditional medical films based on intelligent film layout according to claim 1, characterized in that: The S3 film printing system intelligently typesets the original images and encrypts and transmits the typeset images to the cloud for storage, including: S31. Perform intelligent size layout on the original images in the intelligent layout list to obtain the physical size of the film and the size of a single image. S32. Based on the physical dimensions of the film and the dimensions of a single image, construct the objective function and constraints for intelligent position layout; S33. Based on the objective function and constraints of intelligent location layout, the Monte Carlo Tree Search (MCTS) algorithm is used to perform intelligent location layout on the images. S34. Encrypt the formatted image and transmit it to the cloud for storage.

6. The method for transmitting and printing traditional medical films based on intelligent film layout according to claim 5, characterized in that: S31 performs intelligent size layout on the original images in the intelligent layout list to obtain the physical dimensions of the film and the dimensions of a single image, including: The film width is calculated using the following formula. physical : Width physical =Columns×PixelSpacing x ×0.03937; Where Columns represents the number of pixel columns in the horizontal direction of the film, and PixelSpacing represents... x The center-to-center distance between adjacent pixels along the X-axis is 0.03937, which is the conversion factor from millimeters to inches. The film height is calculated using the following formula. physical : Height physical =Rows×PixelSpacing y ×0.03937; Where Rows is the number of pixel rows in the vertical direction of the film, and PixelSpacing is... y The center-to-center distance between adjacent pixels along the Y-axis; Based on the physical dimensions of the film and the size of a single image, S32 constructs the objective function and constraints for intelligent positional layout, including: Input parameters: image set, film physical size, and single image size; Objective function: To maximize film space utilization through algorithm optimization while meeting medical requirements; Constraints: Regions of interest (ROIs) must not be cut to ensure that important areas, including lesions, are presented intact. Images on the film must be temporally adjacent, including images from enhanced CT scans, which must be arranged in the order of examination with an image spacing of no less than 2 mm, to prevent accidental cutting of image content during printing.

7. The method for transmitting and printing traditional medical films based on intelligent film layout according to claim 6, characterized in that: In S33, based on the objective function and constraints of intelligent location layout, the Monte Carlo Tree Search (MCTS) algorithm is used to perform intelligent location layout of the image, including: Phase 1: Selection 1) Tree structure definition operation: When the images are laid out, an initial node is first constructed. This node represents the current blank layout state of the film. Then the images are placed. Each time an image is placed, a new child node is generated. This child node records the placed images and the division of the remaining space after the image is placed. 2) Selection strategy operation: At each node, the UCT algorithm is used to calculate the location where the image can be placed; Phase 2: Expansion 1) Generate candidate action operations: In the current layout state, the remaining blank space is divided by an algorithm; 2) Placement rules: Arrange images according to diagnostic priority; Phase 3: Simulation 1) Rapid valuation strategy operation: Prioritize placing the largest image in the remaining blank space to minimize space waste, and then randomly place important diagnostic images; 2) Optimize layout scheme: For each simulated layout scheme, optimize according to the following formula: Reward=α·utilization+β·ROI_integrity-γ·gap_violation; Where Reward is the reward, α·utilization is the space utilization weighting term, utilization is the space utilization rate of the film, that is, the proportion of the total area of ​​the images placed to the total area of ​​the film, and α is the weighting coefficient of space utilization, α=.0.7; β·ROI_integrity is the weighting term for the integrity of the region of interest (ROI), where ROI_integrity is the integrity of the ROI in the image, and β is the weighting coefficient for the integrity of the ROI, β = 0.3; γ·gap_violation is the image spacing penalty term. gap_violation is used to measure whether the image spacing meets the requirement of not less than 2mm, to ensure the safe spacing between images and prevent the image content from being accidentally cut during printing. γ is the weighting coefficient of image spacing, γ=0.5; Phase 4: Retrospection Medical memory bank operation: Each simulated layout scheme is stored in the medical memory bank. When a similar image layout task is encountered again, the stored historical simulated layout scheme will be optimized in stage 1 to improve layout efficiency.

8. The method for transmitting and printing traditional medical films based on intelligent film layout according to claim 1, characterized in that: In S4, users can scan a QR code on the client to view film containing formatted images stored in the cloud and request to print the film, including: S41. The user scans the QR code on the client / paper report to complete the patient identity verification; S42. The client requests data from the cloud so that the user can view film containing typed images stored in the cloud. S43. The user requests to print film on the client.

9. The method for transmitting and printing traditional medical films based on intelligent film layout according to claim 1, characterized in that: In S5, the client generates a QR code for retrieving the image and sends the QR code to the user's terminal, including: S51. The client generates the core token in the QR code for retrieving the image according to the following formula: Here, TaskID is the task ID, used to uniquely identify a business task or operation, TS is the timestamp-related value, / / indicates the concatenation operation, and SM4 CTR indicates the SM4 symmetric encryption algorithm working in counter mode. The task ID TaskID and the timestamp-related value TS are concatenated into a whole as the input of the SM4 symmetric encryption algorithm, so that each token is associated with the task and time. Salt is a random string. The XOR operation is performed, which XORs the SM4 encryption result with the random string Salt to increase the randomness and unpredictability of the token. Base64 indicates that Base64 encoding is performed. S52. The client generates a corresponding QR code for retrieving the chip based on the token and sends the QR code to the user terminal.

10. The method for transmitting and printing traditional medical films based on intelligent film layout according to claim 1, characterized in that: The S7 film printing system receives film printing requests and sends print commands to self-service printers to complete film printing, including: The film printing system receives film printing requests and sends printing commands and printing data to the self-service printer to complete the film printing and store the printing record at the same time. The film printing system automatically marks the anti-counterfeiting watermark "printing time + device ID" when sending printing data. The radiology department can log in to the film printing system to view printing records and printing quantities for daily departmental statistics.

Images