Intelligent management method and system for orthodontic images

Abstract

The invention relates to an intelligent orthodontic image management method and system, and relates to the technical field of data management, and the system comprises an image obtaining module which is used for obtaining orthodontic data of historical patients; the image storage module is used for encrypting and storing the orthodontic data of historical patients; the feature acquisition module is used for acquiring the orthodontic image of the patient to be subjected to orthodontic treatment from the image generation terminal and extracting the oral features of the patient to be subjected to orthodontic treatment from the orthodontic image of the patient to be subjected to orthodontic treatment; the image searching module is used for determining a target historical patient based on the oral features of the to-be-orthodontic patient; the image sharing module is used for calling the orthodontic data of the target historical patient and sending the called orthodontic data of the target historical patient to the doctor terminal; and the scheme generation module is used for generating an auxiliary orthodontic scheme based on the oral features of the patient to be orthodontic and the orthodontic data of the target historical patient, and the method has the advantage of improving the quality of orthodontic image management.

Description

Technical Field

[0001] The present invention relates to the technical field of data management, and in particular to an intelligent management method and system for orthodontic images. Background Art

[0002] Orthodontic imaging is a medical imaging technology based on oral medicine. It aims to provide doctors with detailed information about the patient's oral condition by photographing the teeth, jaws, periodontal tissues and other structures in the mouth. These imaging data help doctors analyze the patient's tooth arrangement, occlusion, jaw morphology, etc., so as to develop a personalized orthodontic treatment plan. Orthodontic imaging mainly includes dental panoramic films: which can show the positional relationship of all teeth in the entire mouth, helping doctors understand the overall arrangement and occlusion of teeth; lateral cephalograms: providing relative position information between the maxilla and mandible, helping doctors analyze the jaw morphology and determine whether there is jaw deformity.

[0003] Orthodontic images involve a large amount of personal information and medical data of patients, and data security and privacy protection have become important issues. Existing orthodontic images are scattered and stored in computers in various examination rooms. The file storage is chaotic, prone to data leakage, and difficult to find and classify, making it difficult to apply valuable medical information to practical applications.

[0004] Therefore, there is a need to provide an intelligent management method and system for orthodontic images to improve the quality of orthodontic image management. Summary of the invention

[0005] The present invention provides an intelligent management system for orthodontic images, comprising: an image acquisition module, used to acquire orthodontic data of historical patients, wherein the orthodontic data at least includes orthodontic images acquired from at least one image generation terminal, oral features and orthodontic plans determined based on the orthodontic images; an image storage module, used to encrypt the orthodontic data of the historical patients and store the encrypted orthodontic data of the historical patients; a feature acquisition module, used to acquire orthodontic images of patients to be orthodontized from at least one image generation terminal, and extract the oral features of the patients to be orthodontized from the orthodontic images of the patients to be orthodontized; an image search module, used to determine a target historical patient based on the oral features of the patients to be orthodontized; an image sharing module, used to retrieve the orthodontic data of the target historical patient, and send the retrieved orthodontic data of the target historical patient to a doctor terminal; a plan generation module, used to generate an auxiliary orthodontic plan based on the oral features of the patients to be orthodontized and the orthodontic data of the target historical patient.

[0006] Furthermore, the image storage module encrypts the orthodontic data of the historical patient, including: encrypting each orthodontic image; segmenting the oral features determined based on the orthodontic image to generate multiple feature slices, splicing the multiple feature slices to generate spliced ​​oral features and a first splicing path corresponding to the spliced ​​oral features, encrypting the first splicing path based on the encryption key corresponding to the historical patient to generate an encrypted first splicing path; segmenting the orthodontic plan to generate multiple plan slices, splicing the multiple plan slices to generate a spliced ​​orthodontic plan and a second splicing path corresponding to the spliced ​​orthodontic plan, encrypting the second splicing path based on the encryption key corresponding to the historical patient to generate an encrypted second splicing path.

[0007] Furthermore, the image storage module encrypts each of the orthodontic images, including: for each of the orthodontic images, segmenting the orthodontic image to generate image slices corresponding to the orthodontic image; splicing the image slices corresponding to multiple orthodontic images to generate multiple spliced ​​orthodontic images and splicing paths of the spliced ​​orthodontic images; calculating the hash value of each of the spliced ​​orthodontic images; encrypting the second splicing path based on the encryption key corresponding to the historical patient to generate an encrypted third splicing path for each of the spliced ​​orthodontic images; and establishing a mapping relationship between the hash value of each spliced ​​orthodontic image and the corresponding encrypted splicing path.

[0008] Furthermore, the image storage module performs image segmentation on the orthodontic image to generate image slices corresponding to the orthodontic image, including: determining the reference retrieval frequency parameters of the historical patient based on the oral features determined by the orthodontic image; determining the data retrieval frequency parameters of the historical patient based on the orthodontic plan; determining the size of the image slices of the orthodontic image of the historical patient based on the reference retrieval frequency parameters and the data retrieval frequency parameters of the historical patient; and performing image segmentation on the orthodontic image based on the size of the image slices of the orthodontic image of the historical patient to generate image slices corresponding to the orthodontic image.

[0009] Furthermore, the image storage module stores the encrypted orthodontic data of historical patients, including: storing the spliced ​​oral features, spliced ​​orthodontic plans and multiple spliced ​​orthodontic images, and generating storage paths of the spliced ​​oral features, spliced ​​orthodontic plans and multiple spliced ​​orthodontic images, encrypting the storage path based on the encryption key corresponding to the historical patient, generating an encrypted storage path, and updating the storage path table, wherein the storage path table is used to record the mapping relationship between the token of the historical patient and the corresponding encrypted storage path; updating the data retrieval auxiliary table, wherein the data retrieval auxiliary table is used to record the mapping relationship between the token of each of the historical patients and the label information of the corresponding orthodontic data, wherein the label information includes the encrypted first splicing path, the encrypted second splicing path and the mapping relationship between the hash value of each spliced ​​orthodontic image and the corresponding encrypted splicing path.

[0010] Furthermore, the feature acquisition module extracts oral features of the patient to be orthodontic from the oral orthodontic images of the patient to be orthodontic, including: establishing and training a feature extraction model; extracting the oral features of the patient to be orthodontic from the oral orthodontic images of the patient to be orthodontic through the trained feature extraction model.

[0011] Furthermore, the image search module determines the target historical patients based on the oral characteristics of the patients to be orthodontic, including: for any two of the historical patients, calculating the orthodontic similarity of the two historical patients based on the oral characteristics and oral orthodontic plans of the two historical patients; clustering multiple historical patients based on the orthodontic similarity of any two historical patients by K-means clustering algorithm to determine multiple patient clusters; for any one of the patient clusters, calculating the matching parameters of the patient cluster based on the oral characteristics of the patients to be orthodontic and the cavity characteristics of the cluster center of the patient cluster; determining the target patient cluster based on the matching parameters corresponding to each of the patient clusters; and determining the target historical patient from the target patient cluster based on the oral characteristics of the patients to be orthodontic.

[0012] Furthermore, the image sharing module retrieves the orthodontic data of the target historical patient, including: obtaining the token of the target historical patient; based on the token of the target historical patient, reading the encrypted storage path corresponding to the target historical patient from the storage path table, decrypting the encrypted storage path based on the encryption key corresponding to the target historical patient, and retrieving the spliced ​​oral features, spliced ​​orthodontic plan and multiple spliced ​​orthodontic images corresponding to the target historical patient based on the decrypted storage path; based on the token of the target historical patient, retrieving the label information corresponding to the target historical patient from the data retrieval auxiliary table; based on the label information corresponding to the target historical patient, performing data restoration on the spliced ​​oral features, spliced ​​orthodontic plan and multiple spliced ​​orthodontic images corresponding to the target historical patient.

[0013] Furthermore, the plan generation module generates an auxiliary orthodontic plan based on the oral characteristics of the patient to be orthodontic and the oral orthodontic data of the target historical patient, including: establishing and training a plan generation model; generating an auxiliary orthodontic plan based on the oral characteristics of the patient to be orthodontic and the oral orthodontic data of the target historical patient through the trained plan generation model.

[0014] The present invention provides an intelligent management method for orthodontic images, comprising: obtaining orthodontic data of historical patients, wherein the orthodontic data at least includes orthodontic images obtained from at least one image generation terminal, oral features and orthodontic plans determined based on the orthodontic images; encrypting the orthodontic data of the historical patients, and storing the encrypted orthodontic data of the historical patients; obtaining orthodontic images of patients to be orthodontized from at least one image generation terminal, and extracting the oral features of the patients to be orthodontized from the orthodontic images of the patients to be orthodontized; determining a target historical patient based on the oral features of the patients to be orthodontized; retrieving the orthodontic data of the target historical patient, and sending the retrieved orthodontic data of the target historical patient to a doctor terminal; generating an auxiliary orthodontic plan based on the oral features of the patients to be orthodontized and the orthodontic data of the target historical patient.

[0015] Compared with the prior art, the method and system for intelligent management of orthodontic images provided in this specification have at least the following beneficial effects:

[0016] 1. Encrypt the oral orthodontic data of historical patients. Encrypted storage ensures the security of patients' oral orthodontic data, reduces the risk of data leakage, and protects patients' privacy. Encryption technology can effectively prevent unauthorized access and tampering, ensure the integrity and authenticity of data, and on this basis, by analyzing the oral orthodontic treatment plans of similar patients, a personalized treatment plan can be quickly generated for the current patient, reducing the time and workload of doctors in diagnosis and treatment plan formulation. Using big data and algorithms to optimize treatment plans can respond to patients' treatment needs more quickly and improve overall treatment efficiency. Based on the successful cases and data of similar patients, the generated personalized treatment plan is more accurate and scientific, which helps to improve the treatment effect and success rate. By drawing on the treatment experience of other patients, some potential treatment risks and complications can be avoided, providing patients with safer and more reliable treatment plans. Patients can obtain personalized and scientific orthodontic treatment plans more quickly, reducing waiting time and unnecessary anxiety. The digital and intelligent treatment plan formulation process is more transparent and understandable, which helps to enhance patients' confidence and satisfaction with treatment. Orthodontic data stored after encryption can be safely shared and transmitted between different medical institutions, promoting the optimal allocation and rational use of medical resources. By sharing treatment data and experience of similar patients, different medical institutions can learn from each other and improve the diagnosis and treatment level and service quality of the entire industry.

[0017] 2. Orthodontic images are segmented and randomly spliced, which breaks the continuity and integrity of the original images. This enhances data security to a certain extent, making it difficult for unauthorized personnel to directly identify or use these image data. Encrypted storage further enhances data security. Even if the data is illegally obtained, it is difficult to decrypt and restore;

[0018] 3. By calculating the orthodontic similarity of two historical patients, two similar historical patients are clustered. Subsequently, the target patient cluster can be determined based on the matching parameters corresponding to each patient cluster. Based on the oral characteristics of the orthodontic patient, the target historical patient can be determined from the target patient cluster, thereby improving the efficiency of finding the target historical patient. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] This specification will be further described in the form of exemplary embodiments, which will be described in detail by the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same number represents the same structure, wherein:

[0020] Figure 1 It is a module diagram of an intelligent management system for orthodontic images shown in one embodiment of the present application;

[0021] Figure 2is a flow chart showing encryption processing of each oral orthodontic image in one embodiment of the present application;

[0022] Figure 3 It is a flow chart of a method for intelligent management of orthodontic images shown in one embodiment of the present application. DETAILED DESCRIPTION

[0023] In order to more clearly illustrate the technical solutions of the embodiments of this specification, the following briefly introduces the drawings required for describing the embodiments.

[0024] Figure 1 is a module diagram of an orthodontic image intelligent management system shown in one embodiment of the present application, such as Figure 1 As shown, an orthodontic image intelligent management system may include an image acquisition module, an image storage module, a feature acquisition module, an image search module, an image sharing module and a solution generation module.

[0025] The image acquisition module can be used to obtain historical orthodontic data of patients.

[0026] The orthodontic data at least includes an orthodontic image acquired from at least one image generation terminal, oral features determined based on the orthodontic image, and an orthodontic plan.

[0027] The image generation terminal may be an X-ray imaging device, a cone beam computed tomography (CBCT) device, a digital camera, etc. The types of orthodontic images may include panoramic dental films, cephalographic films, cone beam computed tomography images, bitewing films, facial photographs, and intraoral photographs.

[0028] Oral characteristics may include tooth arrangement, bite relationship, jaw morphology, facial symmetry, convexity or concavity, etc.

[0029] An orthodontic plan may include correction methods (e.g., preventive correction, blocking correction, appliance correction, or surgical orthodontic correction), treatment steps (e.g., preliminary work such as dental nails, fillings, tooth extractions, and subsequent review plans, etc.).

[0030] The image storage module can be used to encrypt the oral orthodontic data of historical patients and store the encrypted oral orthodontic data of historical patients.

[0031] In some embodiments, the image storage module encrypts the historical patient's orthodontic data, including:

[0032] Encrypt each orthodontic image;

[0033] Segmenting the oral features determined based on the oral orthodontic image to generate a plurality of feature slices, splicing the plurality of feature slices to generate spliced ​​oral features and a first splicing path corresponding to the spliced ​​oral features, encrypting the first splicing path based on an encryption key corresponding to a historical patient to generate an encrypted first splicing path;

[0034] The orthodontic plan is divided to generate multiple plan slices, the multiple plan slices are spliced ​​to generate a spliced ​​orthodontic plan and a second splicing path corresponding to the spliced ​​orthodontic plan, the second splicing path is encrypted based on the encryption key corresponding to the historical patient, and an encrypted second splicing path is generated.

[0035] Specifically, the encryption key corresponding to the historical patient can be generated based on the personalized information of the historical patient, wherein the personalized information can be facial features, unique identification codes, etc. The first splicing path can be encrypted based on the encryption key corresponding to the historical patient through a symmetric encryption algorithm to generate an encrypted first splicing path, and the second splicing path can be encrypted to generate an encrypted second splicing path.

[0036] Figure 2 FIG. 1 is a flowchart of encrypting each orthodontic image in an embodiment of the present application. Figure 2 As shown, in some embodiments, the image storage module encrypts each orthodontic image, including:

[0037] For each orthodontic image, the orthodontic image is segmented to generate image slices corresponding to the orthodontic image;

[0038] Splicing image slices corresponding to a plurality of orthodontic images to generate a plurality of spliced ​​orthodontic images and splicing paths of the spliced ​​orthodontic images;

[0039] Calculate the hash value of each stitched orthodontic image;

[0040] Encrypting the second stitching path based on the encryption key corresponding to the historical patient to generate an encrypted third stitching path for each stitched orthodontic image;

[0041] A mapping relationship between the hash value of each spliced ​​orthodontic image and the corresponding encrypted splicing path is established.

[0042] For example, the oral orthodontic data of historical patient A includes a panoramic dental film 1, a lateral cephalogram 1, and a cone beam computed tomography image 1; the oral orthodontic data of historical patient B includes a panoramic dental film 2, a lateral cephalogram 2, and a cone beam computed tomography image 2. The panoramic dental film 1 of historical patient A can be sliced ​​to generate multiple image slices 11, the lateral cephalogram 1 of historical patient A can be sliced ​​to generate multiple image slices 12, the cone beam computed tomography image 1 of historical patient A can be sliced ​​to generate multiple image slices 13, the panoramic dental film 2 of historical patient B can be sliced ​​to generate multiple image slices 21, the lateral cephalogram 2 of historical patient B can be sliced ​​to generate multiple image slices 22, and the cone beam computed tomography image 2 of historical patient B can be sliced ​​to generate multiple image slices 23.

[0043] Multiple image slices 11, 12 and 13 of historical patient A are spliced ​​with multiple image slices 21, 22 and 23 of historical patient B to generate spliced ​​orthodontic image 1, spliced ​​orthodontic image 2, spliced ​​orthodontic image 3, spliced ​​orthodontic image 4, spliced ​​orthodontic image 5 and spliced ​​orthodontic image 6. For example only, the spliced ​​orthodontic image 1 can be formed by splicing a part of the multiple image slices 11, a part of the multiple image slices 22 and a part of the multiple image slices 13 in a random order.

[0044] In some embodiments, the image storage module segments the orthodontic image to generate image slices corresponding to the orthodontic image, including:

[0045] Determine the reference retrieval frequency parameters of historical patients based on oral features determined by orthodontic imaging;

[0046] Determine the frequency parameters for data retrieval of historical patients based on the orthodontic plan;

[0047] Determine the size of image slices of the historical patient's orthodontic images based on the historical patient's reference retrieval frequency parameter and the data retrieval frequency parameter;

[0048] Based on the size of image slices of the historical patients' orthodontic images, the orthodontic images are segmented to generate image slices corresponding to the orthodontic images.

[0049] Specifically, for any two historical patients, the feature similarity between the two historical patients can be determined based on the oral features of the two historical patients through a feature similarity determination model, wherein the feature similarity determination model can be a convolutional neural network (CNN) model.

[0050] For a certain historical patient, the feature similarity mean and similarity discrete parameter can be calculated based on the feature similarity between the historical patient and each other historical patient, and the data retrieval frequency parameter of the historical patient can be determined according to the feature similarity mean and similarity discrete parameter.

[0051] For example, the feature similarity mean and similarity dispersion parameter can be calculated according to the following formula:

[0052]

[0053] Among them, S (i,mean) is the mean similarity of the characteristics of the i-th historical patient, S (i,j) is the feature similarity between the i-th historical patient and the j-th historical patient, J is the total number of historical patients, W i is the discrete parameter of the similarity of the i-th historical patient.

[0054] The reference frequency parameter of historical patients can be calculated according to the following formula:

[0055]

[0056] Among them, f (i,1) The reference frequency parameter of the i-th historical patient, M 1 is the preset parameter, and M 1 Greater than 0, W preset is the preset similarity discrete parameter, S (preset,mean) is the preset feature similarity mean, f (0,1) is the preset frequency.

[0057] It can be understood that the larger the mean value of the feature similarity of the ith historical patient is, the smaller the similarity dispersion parameter is, the greater the possibility that the orthodontic data of the ith historical patient will be retrieved as an auxiliary reference case, and therefore, the larger the reference retrieval frequency parameter is.

[0058] The number of follow-up examinations for the historical patients can be determined based on the oral orthodontic plans of the historical patients, and the data retrieval frequency parameters of the historical patients can be determined based on the number of follow-up examinations for the historical patients.

[0059] The data retrieval frequency parameter of historical patients can be calculated according to the following formula:

[0060]

[0061] Among them, f (i,2) Retrieve the frequency parameter for the data of the i-th historical patient, N i is the number of reexaminations of the i-th historical patient, N preset is the preset number of rechecks, f (0,2) is the preset frequency.

[0062] The size of the image slices of the historical orthodontic imaging of the patient can be calculated according to the following formula:

[0063]

[0064] Among them, V i is the size of the image slice of the orthodontic image of the i-th historical patient, M 2 and M 3 is the preset parameter, and M 2 and M 3 Both are greater than 0, V 0 is the size of the preset image slice, a 1 and a 2 is the preset weight, and a 1 and a 2 Greater than 0, a 1 +a 2 =1.

[0065] It can be understood that the larger the reference retrieval frequency parameter of the historical patient and the larger the data retrieval frequency parameter, the more frequently the oral orthodontic data representing the historical patient will be retrieved. In this case, the image slices will be larger, the number of slices will be reduced, the time required for data restoration will be shorter, and data retrieval will be more convenient.

[0066] In some embodiments, the image storage module stores encrypted historical patient orthodontic data, including:

[0067] Store the spliced ​​oral features, the spliced ​​orthodontic plan and multiple spliced ​​orthodontic images, and generate storage paths for the spliced ​​oral features, the spliced ​​orthodontic plan and multiple spliced ​​orthodontic images, encrypt the storage path based on the encryption key corresponding to the historical patient, generate the encrypted storage path, and update the storage path table, wherein the storage path table is used to record the mapping relationship between the token of the historical patient and the corresponding encrypted storage path;

[0068] Update the data retrieval auxiliary table, wherein the data retrieval auxiliary table is used to record the mapping relationship between the token of each historical patient and the label information of the corresponding orthodontic data, wherein the label information includes the encrypted first stitching path, the encrypted second stitching path and the mapping relationship between the hash value of each stitched orthodontic image and the corresponding encrypted stitching path.

[0069] Specifically, the token of the historical patient can be generated based on the personalized information of the historical patient.

[0070] The feature acquisition module can be used to acquire an oral orthodontic image of a patient to be orthodontized from at least one image generation terminal, and extract oral features of the patient to be orthodontized from the oral orthodontic image of the patient to be orthodontized.

[0071] In some embodiments, the feature acquisition module extracts oral features of the patient to be orthodontically treated from the oral orthodontic image of the patient to be orthodontically treated, including:

[0072] Establish and train a feature extraction model, wherein the feature extraction model may be a ResNet (Residual Neural Network) model;

[0073] The trained feature extraction model is used to extract the oral features of the patient to be orthodontic from the orthodontic images of the patient to be orthodontic.

[0074] The image search module can be used to determine the target historical patients based on the oral features of the patients to be orthodontically treated.

[0075] Specifically include:

[0076] For any two historical patients, the orthodontic similarity between the two historical patients is calculated based on their oral characteristics and orthodontic plans;

[0077] Clustering multiple historical patients based on the orthodontic similarity of any two historical patients using the K-means clustering algorithm to determine multiple patient clusters;

[0078] For any patient cluster, the matching parameters of the patient cluster are calculated based on the oral features of the patients to be orthodontically treated and the cavity features of the cluster center of the patient cluster;

[0079] Determine a target patient cluster based on the matching parameters corresponding to each patient cluster;

[0080] Based on the oral characteristics of the patients to be orthodontically treated, target historical patients are determined from the target patient cluster.

[0081] Specifically, the orthodontic similarity of two historical patients can be calculated based on the oral characteristics and oral orthodontic plans of the two historical patients through a patient similarity determination model, wherein the patient similarity determination model can be a Convolutional Neural Network (CNN) model.

[0082] The feature similarity determination model can be based on the oral features of the patients to be orthodontically treated and the cavity features of the cluster centers of the patient clusters, and the feature similarity between the cluster centers of the patient clusters and the patients to be orthodontically treated is calculated. The greater the feature similarity between the cluster centers of the patient clusters and the patients to be orthodontically treated, the greater the matching parameter corresponding to the patient cluster.

[0083] The patient cluster whose matching parameter is greater than a preset matching parameter threshold may be used as the target patient cluster.

[0084] The image sharing module can be used to retrieve the orthodontic data of the target historical patient and send the retrieved orthodontic data of the target historical patient to the doctor's terminal.

[0085] In some embodiments, the image sharing module retrieves the orthodontic data of the target historical patient, including:

[0086] Get the target historical patient's token;

[0087] Based on the token of the target historical patient, the encrypted storage path corresponding to the target historical patient is read from the storage path table, the encrypted storage path is decrypted based on the encryption key corresponding to the target historical patient, and the spliced ​​oral features, spliced ​​orthodontic plan and multiple spliced ​​orthodontic images corresponding to the target historical patient are retrieved based on the decrypted storage path;

[0088] Based on the token of the target historical patient, retrieve the label information corresponding to the target historical patient from the data retrieval auxiliary table;

[0089] Based on the label information corresponding to the target historical patient, data restoration is performed on the spliced ​​oral features, spliced ​​orthodontic plans, and multiple spliced ​​orthodontic images corresponding to the target historical patient.

[0090] Specifically, the encrypted first splicing path and the encrypted second splicing path are decrypted, and the spliced ​​oral features and the spliced ​​oral orthodontic plan corresponding to the target historical patient are restored.

[0091] For each spliced ​​orthodontic image, based on the hash value of the spliced ​​orthodontic image, the encrypted third splicing path corresponding to the spliced ​​orthodontic image is searched, and the encrypted third splicing path is decrypted.

[0092] Based on the decrypted third stitching path corresponding to each stitched orthodontic image, data restoration is performed simultaneously on the multiple stitched orthodontic images.

[0093] The plan generation module can be used to generate an auxiliary orthodontic plan based on the oral characteristics of the patient to be orthodontically treated and the oral orthodontic data of the target historical patient.

[0094] Specifically include:

[0095] Establishing and training a solution generation model, wherein the solution generation model may be a Generative Adversarial Networks model;

[0096] The trained plan generation model generates an auxiliary orthodontic plan based on the oral characteristics of the patient to be orthodontic and the oral orthodontic data of the target historical patients, providing auxiliary support for subsequent doctors to formulate oral orthodontic plans for the patient to be orthodontic.

[0097] Figure 3 is a flow chart of a method for intelligent management of orthodontic images shown in one embodiment of the present application. Figure 3 As shown, a method for intelligent management of orthodontic images may include the following process.

[0098] Step 310, obtaining historical orthodontic data of a patient, wherein the orthodontic data at least includes an orthodontic image obtained from at least one image generation terminal, oral features determined based on the orthodontic image, and an orthodontic plan;

[0099] Step 320, encrypting the historical patient's orthodontic data and storing the encrypted historical patient's orthodontic data;

[0100] Step 330, obtaining an oral orthodontic image of the patient to be orthodontically treated from at least one image generating terminal, and extracting oral features of the patient to be orthodontically treated from the oral orthodontic image of the patient to be orthodontically treated;

[0101] Step 340, determining a target historical patient based on the oral characteristics of the patient to be orthodontically treated;

[0102] Step 350, retrieve the orthodontic data of the target historical patient, and send the retrieved orthodontic data of the target historical patient to the doctor terminal;

[0103] Step 360, generating an auxiliary orthodontic plan based on the oral characteristics of the patient to be orthodontically treated and the oral orthodontic data of the target patient with a history of orthodontics.

[0104] In some embodiments, a method for intelligent management of orthodontic images may be executed by an intelligent management system for orthodontic images. For more descriptions of a method for intelligent management of orthodontic images, please refer to the relevant description of an intelligent management system for orthodontic images, which will not be repeated here.

[0105] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, as an example and not a limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims

1. An intelligent management system for orthodontic images, characterized in that: include: An image acquisition module, used to acquire historical orthodontic data of a patient, wherein the orthodontic data at least includes an orthodontic image acquired from at least one image generation terminal, oral features determined based on the orthodontic image, and an orthodontic plan; An image storage module, used for encrypting the oral orthodontic data of the historical patient and storing the encrypted oral orthodontic data of the historical patient; A feature acquisition module, used for acquiring an oral orthodontic image of a patient to be orthodontically treated from at least one image generation terminal, and extracting oral features of the patient to be orthodontically treated from the oral orthodontic image of the patient to be orthodontically treated; An image search module, used for determining a target historical patient based on the oral characteristics of the patient to be orthodontically treated; An image sharing module, used for retrieving the oral orthodontic data of the target historical patient, and sending the retrieved oral orthodontic data of the target historical patient to a doctor terminal; A plan generation module is used to generate an auxiliary orthodontic plan based on the oral characteristics of the patient to be orthodontically treated and the oral orthodontic data of the target historical patient.

2. The intelligent orthodontic image management system according to claim 1, characterized in that: The image storage module encrypts the oral orthodontic data of the historical patient, including: Encrypting each of the orthodontic images; Segmenting the oral features determined based on the oral orthodontic image to generate a plurality of feature slices, splicing the plurality of feature slices to generate spliced ​​oral features and a first splicing path corresponding to the spliced ​​oral features, encrypting the first splicing path based on the encryption key corresponding to the historical patient to generate an encrypted first splicing path; The orthodontic plan is divided into multiple plan slices, the multiple plan slices are spliced ​​to generate a spliced ​​orthodontic plan and a second splicing path corresponding to the spliced ​​orthodontic plan, and the second splicing path is encrypting based on the encryption key corresponding to the historical patient to generate an encrypted second splicing path.

3. The intelligent orthodontic image management system according to claim 2, characterized in that: The image storage module performs encryption processing on each of the orthodontic images, including: For each of the orthodontic images, segmenting the orthodontic image to generate image slices corresponding to the orthodontic image; Splicing the image slices corresponding to the plurality of orthodontic images to generate a plurality of spliced ​​orthodontic images and splicing paths of the spliced ​​orthodontic images; Calculating a hash value of each of the spliced ​​orthodontic images; Encrypting the second stitching path based on the encryption key corresponding to the historical patient to generate an encrypted third stitching path for each stitched orthodontic image; A mapping relationship between the hash value of each spliced ​​orthodontic image and the corresponding encrypted splicing path is established.

4. The orthodontic image intelligent management system according to claim 3, characterized in that: The image storage module performs image segmentation on the orthodontic image to generate image slices corresponding to the orthodontic image, including: Determining reference call frequency parameters of the historical patient based on oral features determined by the orthodontic image; Based on the orthodontic plan, determining a frequency parameter for retrieving data of the historical patient; Determining the size of the image slice of the oral orthodontic image of the historical patient based on the reference retrieval frequency parameter and the data retrieval frequency parameter of the historical patient; Based on the size of the image slices of the historical patient's orthodontic image, the orthodontic image is segmented to generate image slices corresponding to the orthodontic image.

5. The orthodontic image intelligent management system according to claim 3, characterized in that: The image storage module stores encrypted historical patient orthodontic data, including: The spliced ​​oral features, the spliced ​​orthodontic plan and the multiple spliced ​​orthodontic images are stored, and a storage path of the spliced ​​oral features, the spliced ​​orthodontic plan and the multiple spliced ​​orthodontic images is generated, the storage path is encrypted based on the encryption key corresponding to the historical patient, an encrypted storage path is generated, and a storage path table is updated, wherein the storage path table is used to record the mapping relationship between the token of the historical patient and the corresponding encrypted storage path; Update the data retrieval auxiliary table, wherein the data retrieval auxiliary table is used to record the mapping relationship between the token of each of the historical patients and the label information of the corresponding orthodontic data, wherein the label information includes the encrypted first stitching path, the encrypted second stitching path and the mapping relationship between the hash value of each stitched orthodontic image and the corresponding encrypted stitching path.

6. An orthodontic image intelligent management system according to any one of claims 1 to 5, characterized in that: The feature acquisition module extracts the oral features of the patient to be orthodontically treated from the oral orthodontic image of the patient to be orthodontically treated, including: Build and train a feature extraction model; The oral features of the patient to be orthodontic are extracted from the oral orthodontic images of the patient to be orthodontic using the trained feature extraction model.

7. The orthodontic image intelligent management system according to claim 6, characterized in that: The image search module determines the target historical patient based on the oral characteristics of the patient to be orthodontically treated, including: For any two of the historical patients, based on the oral characteristics and orthodontic plans of the two historical patients, calculating the orthodontic similarity of the two historical patients; Clustering multiple historical patients based on the orthodontic similarity of any two of the historical patients using a K-means clustering algorithm to determine multiple patient clusters; For any one of the patient clusters, based on the oral features of the patient to be orthodontically treated and the cavity features of the cluster center of the patient cluster, the matching parameters of the patient cluster are calculated; Determine a target patient cluster based on the matching parameters corresponding to each of the patient clusters; Based on the oral characteristics of the patient to be orthodontically treated, a target historical patient is determined from the target patient cluster.

8. The orthodontic image intelligent management system according to claim 5, characterized in that: The image sharing module retrieves the oral orthodontic data of the target historical patient, including: Obtaining a token of the target historical patient; Based on the token of the target historical patient, the encrypted storage path corresponding to the target historical patient is read from the storage path table, the encrypted storage path is decrypted based on the encryption key corresponding to the target historical patient, and the spliced ​​oral features, spliced ​​orthodontic plan and multiple spliced ​​orthodontic images corresponding to the target historical patient are retrieved based on the decrypted storage path; Based on the token of the target historical patient, retrieve the label information corresponding to the target historical patient from the data retrieval auxiliary table; Based on the label information corresponding to the target historical patient, data restoration is performed on the spliced ​​oral features, the spliced ​​orthodontic plan, and the multiple spliced ​​orthodontic images corresponding to the target historical patient.

9. An orthodontic image intelligent management system according to any one of claims 1 to 5, characterized in that: The plan generation module generates an auxiliary orthodontic plan based on the oral characteristics of the patient to be orthodontically treated and the oral orthodontic data of the target historical patient, including: Build and train a solution generation model; An auxiliary orthodontic plan is generated by the trained plan generation model based on the oral characteristics of the patient to be orthodontically treated and the oral orthodontic data of the target historical patient.

10. A method for intelligent management of orthodontic images, characterized in that: include: Acquiring historical orthodontic data of a patient, wherein the orthodontic data at least includes an orthodontic image acquired from at least one image generating terminal, oral features determined based on the orthodontic image, and an orthodontic plan; Encrypting the oral orthodontic data of the historical patient and storing the encrypted oral orthodontic data of the historical patient; Acquire an oral orthodontic image of a patient to be orthodontically treated from at least one image generating terminal, and extract oral features of the patient to be orthodontically treated from the oral orthodontic image of the patient to be orthodontically treated; Determine a target historical patient based on the oral characteristics of the patient to be orthodontically treated; Retrieving the orthodontic data of the target historical patient, and sending the retrieved orthodontic data of the target historical patient to the doctor terminal; An auxiliary orthodontic plan is generated based on the oral characteristics of the patient to be orthodontically treated and the oral orthodontic data of the target historical patient.

Images