Dental implant prediction method, system, device and storage medium based on deep learning

Abstract

The application provides a tooth implantation prediction method and system based on deep learning, a device and a storage medium. The method comprises the following steps: establishing a set of original image samples based on X-ray panoramic images of teeth of a user; obtaining a set of whole period image training samples and a set of periapical image training samples based on the set of original image samples; inputting the whole period image training samples and the periapical training samples into a trained neural network to obtain a tooth implantation prediction result. The application improves the accuracy of tooth implantation prediction, effectively predicts the occurrence of tooth implantation failure, and helps to perform early clinical intervention on potential tooth implantation failure to prevent tooth implantation failure and related complications.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence, and more specifically, to a method, system, device, and storage medium for predicting dental implants based on deep learning. Background Technology

[0002] With the increasing use of dental implants to replace missing teeth, implant failure has become more frequent; this increased failure rate is frustrating for both patients and dentists. It typically manifests as fibrous tissue rather than peri-implant osseointegration of the alveolar bone or significant peri-implant marginal bone loss. Implant failure often leads to complex retreatment, which can cause additional physical and psychological problems for patients. Therefore, predicting implant failure is essential for long-term clinical success.

[0003] Peri-implant disease can lead to implant failure, with peri-implantitis being the most common; this is described as a localized lesion involving loss of marginal bone around the implant. Clinical symptoms of peri-implantitis include gingival swelling and progressive bone loss; the former is detected by probing (bleeding during probing), while the latter can be identified on X-rays with a characteristic crater-like margin of bone loss. Generally, inflammation is the primary cause of peri-implantitis, which can lead to implant failure. In such cases, patients require multiple surgical and non-surgical treatments to restore and regenerate the lost bone before undergoing reimplantation surgery, a laborious and costly process. Therefore, early prediction of peri-implant disease allows dentists to be aware of the likelihood of peri-implantitis and take preventative measures even when only mild symptoms or subtle signs are present.

[0004] Another typical implant failure is due to the development of fibrous tissue between the implant and the surrounding bone, without significant marginal bone loss. Unlike successful osseointegration with intact bone-implant contact, a characteristic feature is the presence of 0.5 mm of peri-implant radiolucency. The histopathological mechanisms of this type of failure are not fully understood. Multiple factors, including the patient's medical condition, surgical procedure, and host bone characteristics, can influence implant outcomes. For example, diabetic patients are more likely to experience prolonged osseointegration with the surrounding bone. In other cases, implant failure is more likely to occur when the bone is very sparse or very dense on panoramic radiographs. Therefore, radiological examination based on fine X-rays may reduce the risk of this type of implant failure.

[0005] The study uses demographic data, alveolar bone assessment, intraoral conditions, lifestyle, and implant placement (with or without bone graft procedures) as inputs to establish an implant prediction model. Among these variables, radiological alveolar bone assessment before and after implantation is essential. In fact, imaging information is related to multiple risk factors that may affect the implant outcome, including insufficient bone volume, poor bone quality, periodontal bone loss, and systemic diseases such as osteopenia / osteoporosis and diabetes that may have a harmful effect on implants. Bone metabolism. Under these conditions, establishing osseointegration between the implant and the surrounding bone may be challenging. However, so far, there is no effective method to objectively evaluate the alveolar bone status around the implant in radiographs. Subjectively, there are significant individual differences in film evaluation in daily clinical practice. These limitations may endanger the implant outcome. For example, the superimposition of three-dimensional bone structures may cause subjective evaluation to overestimate the actual bone bonding status in periapical radiographs. In this case, loading before sufficient osseointegration may lead to future failures.

[0006] AI algorithms can provide powerful diagnostic tools to capture characteristic patterns using radiographs in dentistry. Currently, only a few methods mainly use radiographic films to predict the occurrence of implant failure. Our study aims to develop a deep learning-based method that uses panoramic and periapical images to predict the occurrence of implant failure. Convolutional neural networks are used to extract features from the preprocessed images. The model output indicates whether the implant is successful, fails due to perimplant lesions, or fails due to other reasons.

[0007] Therefore, the present invention provides a deep learning-based dental implant prediction method, system, device, and storage medium. Summary of the Invention

[0008] Aiming at the problems in the prior art, the purpose of the present invention is to provide a deep learning-based dental implant prediction method, system, device, and storage medium, which overcomes the difficulties of the prior art, improves the accuracy of dental implant prediction, effectively predicts the occurrence of dental implant failure, and helps to perform early clinical intervention on potential dental implant failure to prevent dental implant failure and related complications.

[0009] An embodiment of the present invention provides a deep learning-based dental implant prediction method, including the following steps:

[0010] Based on the X-ray panoramic images of the user's teeth, a set of original image samples is established;

[0011] Based on the set of original image samples, a set of full-week image training samples and a set of periapical image training samples are obtained; and

[0012] The implant prediction results are obtained by inputting full-circumference image training samples and periapical training samples into the trained neural network.

[0013] Preferably, the set of original image samples established based on the panoramic X-ray images of the user's teeth includes:

[0014] Collect panoramic X-ray images and implantation status information of the user's teeth that have completed periapical implantation. The implantation status information includes at least the preset number of the implanted teeth in the panoramic X-ray image and the implantation result, which includes whether the implantation was successful or failed.

[0015] A set of original image samples is established based on the panoramic X-ray images of the user's teeth, and a mapping relationship is established between each original image sample and the corresponding implantation status information.

[0016] Preferably, obtaining the set of full-circumference image training samples and the set of periapical image training samples based on the set of original image samples includes:

[0017] Image recognition is performed on the X-ray panoramic image to obtain the image region of each tooth in the image;

[0018] Based on the positional relationship of the teeth in the X-ray panoramic image, a mapping relationship is established with the preset tooth numbers;

[0019] Based on the implantation status information corresponding to the X-ray panoramic image, the image region of the implanted tooth is obtained from the X-ray panoramic image as a periapical image;

[0020] Sets of training samples for full-circumference images and sets of training samples for periapical images are established respectively, and a mapping relationship is established between the periapical images, the X-ray panoramic images, and the planting status information.

[0021] Preferably, the step of obtaining the image region of the corresponding implanted tooth as a periapical image from the X-ray panoramic image based on the implantation status information corresponding to the X-ray panoramic image includes:

[0022] Based on the implantation status information corresponding to the X-ray panoramic image, the corresponding image region of the implanted tooth is obtained from the X-ray panoramic image;

[0023] The periapical image is obtained by taking a screenshot of the X-ray panoramic image with the center of the image area of ​​the implanted tooth and according to a preset window size.

[0024] Preferably, the implant prediction result obtained by inputting the full-circumference image training samples and the periapical training samples into the trained neural network includes:

[0025] The pre-trained deep learning convolutional neural network model is used to perform transfer learning on the full-circumference image training samples and the periapical training samples, train and optimize the deep learning convolutional neural network structure, and save the trained deep learning convolutional neural network model.

[0026] The training samples of the full circumference image of the user's tooth to be identified and the periapical training samples are input into the trained deep convolutional neural network model for calculation to obtain the prediction results of implant failure and peri-implantitis.

[0027] Preferably, the process of training and optimizing the deep learning convolutional neural network structure includes:

[0028] New full-circumference image training samples and / or periapical image training samples are obtained by rotating or flipping the full-circumference image training samples and / or periapical image training samples.

[0029] Preferably, the pre-trained deep learning convolutional neural network model further includes:

[0030] The deep learning convolutional neural network model consists of 50 deep layers, and features of a corresponding scale can be obtained by applying convolutional filters to the same layer.

[0031] Embodiments of the present invention also provide a deep learning-based dental implant prediction system for implementing the above-described deep learning-based dental implant prediction method, wherein the deep learning-based dental implant prediction system includes:

[0032] The original image module is a collection of original image samples built from panoramic X-ray images of the user's teeth;

[0033] The training sample module obtains a set of full-circumference image training samples and a set of periapical image training samples based on the set of original image samples.

[0034] The implant prediction module inputs full-circumference image training samples and periapical training samples into a trained neural network to obtain implant prediction results.

[0035] Embodiments of the present invention also provide a deep learning-based dental implant prediction device, comprising:

[0036] processor;

[0037] A memory in which executable instructions of the processor are stored;

[0038] The processor is configured to execute the steps of the deep learning-based dental implant prediction method described above by executing the executable instructions.

[0039] Embodiments of the present invention also provide a computer-readable storage medium for storing a program that, when executed, implements the steps of the deep learning-based dental implant prediction method described above.

[0040] The purpose of this invention is to provide a deep learning-based method, system, device, and storage medium for predicting dental implants, which improves the accuracy of implant prediction, effectively predicts the occurrence of implant failure, and helps to conduct early clinical intervention for potential implant failure, thereby preventing implant failure and related complications. Attached Figure Description

[0041] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings.

[0042] Figure 1 This is a flowchart of the deep learning-based dental implant prediction method of the present invention.

[0043] Figure 2 This is a schematic diagram illustrating the implementation process of the training sample module of the deep learning-based dental implant prediction method of the present invention.

[0044] Figure 3 This is a schematic diagram illustrating the implementation process of the implant prediction module in the deep learning-based implant prediction method of the present invention.

[0045] Figure 4 This is a schematic diagram of the module of the deep learning-based dental implant prediction system of the present invention.

[0046] Figure 5 This is a schematic diagram of the structure of the deep learning-based dental implant prediction device of the present invention.

[0047] Figure 6 This is a schematic diagram of the structure of a computer-readable storage medium according to an embodiment of the present invention. Detailed Implementation

[0048] The following specific examples illustrate the implementation methods of this application. Those skilled in the art can easily understand the other advantages and effects of this application from the content disclosed herein. This application can also be implemented or applied through other different specific embodiments, and various details in this application can be modified or changed according to different viewpoints and application systems without departing from the spirit of this application. It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other.

[0049] The embodiments of this application will now be described in detail with reference to the accompanying drawings, so that those skilled in the art can easily implement the application. This application may be embodied in many different forms and is not limited to the embodiments described herein.

[0050] In this application, the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics represented in connection with that embodiment or example, which are included in at least one embodiment or example of this application. Furthermore, the specific features, structures, materials, or characteristics represented may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate different embodiments or examples represented in this application, as well as features of different embodiments or examples.

[0051] Furthermore, the terms "first" and "second" are used for illustrative purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the representation of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0052] For the purpose of clearly describing this application, devices that are not relevant to the description are omitted, and the same or similar components throughout the specification are given the same reference numerals.

[0053] Throughout this specification, when it is said that a device is "connected" to another device, this includes not only "direct connection" but also "indirect connection" by placing other components in between. Furthermore, when it is said that a device "comprises" a certain constituent element, unless otherwise stated otherwise, this does not exclude other constituent elements, but rather implies that other constituent elements may be included.

[0054] When we say that a device is "above" another device, this can mean that it is directly above the other device, or it can mean that other devices are present in between. Conversely, when we say that a device is "directly" "above" another device, there are no other devices present in between.

[0055] Although the terms first, second, etc., are used in some instances herein to refer to various elements, these elements should not be limited by these terms. These terms are used only to distinguish one element from another. For example, first interface and second interface, etc., are used. Furthermore, as used herein, the singular forms “a,” “an,” and “the” are intended to also include the plural forms unless the context indicates otherwise. It should be further understood that the terms “comprising,” “including,” indicate the presence of features, steps, operations, elements, components, items, kinds, and / or groups, but do not exclude the presence, occurrence, or addition of one or more other features, steps, operations, elements, components, items, kinds, and / or groups. The terms “or” and “and / or” as used herein are interpreted as inclusive, or mean any one or any combination thereof. Thus, “A, B, or C” or “A, B, and / or C” means “any one of: A; B; C; A and B; A and C; B and C; A, B, and C.” Exceptions to this definition will only occur if the combination of elements, functions, steps, or operations is inherently mutually exclusive in some way.

[0056] The technical terms used herein are for reference only to specific embodiments and are not intended to limit the scope of this application. The singular form used herein includes the plural form unless the statement explicitly indicates otherwise. The word "comprising" as used in the specification means to specify a particular characteristic, region, integer, step, operation, element, and / or component, and does not exclude the presence or addition of other characteristics, regions, integers, steps, operations, elements, and / or components.

[0057] Although not explicitly defined, all terms, including technical and scientific terms used herein, shall have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. Terms defined in commonly used dictionaries shall be further interpreted as having a meaning consistent with the relevant technical literature and the content of this present application, and shall not be over-interpreted as having an ideal or overly formulaic meaning unless otherwise defined.

[0058] Figure 1 This is a flowchart of the deep learning-based dental implant prediction method of the present invention. Figure 1 As shown, the deep learning-based dental implant prediction method of the present invention includes the following steps:

[0059] S110. Establish a set of original image samples based on the panoramic X-ray images of the user's teeth.

[0060] S120. Obtain a set of training samples for the entire circumference of images and a set of training samples for the periapical images based on the set of original image samples.

[0061] S130. Input the full-circumference image training samples and the periapical training samples into the trained neural network to obtain the implant prediction results.

[0062] In a preferred embodiment, S110 includes:

[0063] S111. Collect panoramic X-ray images and implantation status information of the user's teeth after periapical implantation. The implantation status information includes at least the preset number of the implanted teeth in the panoramic X-ray image and the implantation result, which includes whether the implantation was successful or failed, but is not limited to this.

[0064] S112. Establish a set of original image samples based on the panoramic X-ray images of the user's teeth, and establish a mapping relationship between each original image sample and the corresponding implantation status information.

[0065] In a preferred embodiment, S120 includes:

[0066] S121. Perform image recognition on the above X-ray panoramic image to obtain the image region of each tooth in the image.

[0067] S122. Based on the positional relationship of teeth in the X-ray panoramic image, a mapping relationship is established with the preset tooth numbers.

[0068] S123. Based on the implantation status information corresponding to the X-ray panoramic image, obtain the image area of ​​the corresponding implanted tooth from the X-ray panoramic image as the periapical image.

[0069] S124. Establish sets of training samples for full-circumference images and sets of training samples for periapical images, and establish mapping relationships between periapical images, X-ray panoramic images, and planting status information.

[0070] In a preferred embodiment, S123 includes:

[0071] S1231. Based on the implantation status information corresponding to the X-ray panoramic image, obtain the image area of ​​the corresponding implanted tooth from the X-ray panoramic image.

[0072] S1232. Using the center of the image area for dental implantation, take a screenshot from the X-ray panoramic image according to the preset window size to obtain a periapical image.

[0073] In a preferred embodiment, S130 includes:

[0074] S131. Use a pre-trained deep learning convolutional neural network model to perform transfer learning on full-circumference image training samples and periapical training samples, train and optimize the deep learning convolutional neural network structure, and save the trained deep learning convolutional neural network model.

[0075] S132. Input the full-circumference image training sample and the periapical training sample of the user's tooth to be identified into the above-trained deep convolutional neural network model for calculation to obtain the prediction results of implant failure and peri-implantitis.

[0076] In a preferred embodiment, S131 includes:

[0077] New circumferential and / or periapical training images are obtained by rotating or flipping the circumferential and / or periapical training images. The flipping is within a range of 30°, including horizontal and vertical flipping, but not limited to this.

[0078] In a preferred embodiment, S130 includes:

[0079] The aforementioned deep convolutional neural network model is based on the ResNet-50 architecture, which consists of 50 deep layers. By applying convolutional filters of different sizes to the same layer, features of different scales can be obtained. Each epoch randomly divides the training dataset into 32 batches and runs for 1000 epochs with a learning rate optimized by the Adam optimizer. However, the neural network model architecture is not limited to this.

[0080] In implementing this invention, a deep learning-based implant prediction system uses a deep learning model to learn features and proximity details from a mixture of periapical and panoramic images. Preprocessed periapical and panoramic radiographs demonstrate higher accuracy in predicting implant failure due to peri-implant lesions, effectively predicting implant failure. This facilitates early clinical intervention for potential implant failure, thereby preventing implant failure and related complications. The data provided in this study may assist clinicians in making informed decisions. Periapical images are obtained through panoramic X-rays, and the combined application of panoramic and periapical images, along with image rotation and flipping, enhances the dataset's quality. Mapping information such as panoramic X-rays, periapical images, implant number, and implant status increases the features and proximity details learnable by the deep learning convolutional neural network.

[0081] Figure 2 This is a schematic diagram illustrating the implementation process of the training sample module of the deep learning-based dental implant prediction method of the present invention. Figure 2As shown, the user collects the patient's panoramic X-ray images and implant status information from the hospital's database, including the location and number of the implanted tooth among all teeth, and whether the implantation was successful or unsuccessful (but not limited to these). This information serves as the raw data for training the deep learning convolutional neural network model 3. The user crops the patient's panoramic X-ray images to the optimal position to remove irrelevant pixels and meet the requirements of the deep learning convolutional neural network model 3, establishing a set of original image samples. Each original image sample 1 is then mapped to its corresponding implant status information (e.g., panoramic X-ray image, original image sample 1, implant tooth number 122, successful implantation, etc.).

[0082] The user performs image recognition on each original image sample 1 to obtain the location of each implant in the image. Centered on the image region of the implant, a crop is taken from the original image sample 1 according to a preset window size to obtain the periapical image 2, where the preset window size is determined according to the requirements of the deep learning neural network model 3. Sets of original image training samples and periapical image training samples are established separately, and a mapping relationship is established between the X-ray panoramic image, original image sample 1, periapical image 2, and implantation status information (e.g., X-ray panoramic image, original image sample 1, periapical image 2, implant number 122, implantation successful, etc.). New enhanced full-circumference image training samples and / or periapical image training samples are obtained by rotating or flipping the full-circumference image training samples and / or periapical image training samples. The rotation is within a 30° range, and the flipping includes horizontal and vertical flipping, but the enhancement method is not limited to this. Data augmentation can effectively improve data quality with limited data, allowing for better adjustment of the neural network model parameters, thereby improving the accuracy of the neural network model's predictions.

[0083] A pre-trained ResNet-50 model was chosen as the starting point for transfer learning in the model used for dental implant prediction. The ResNet-50 architecture is widely used in various feature extraction applications and effectively addresses the problem that classification performance does not improve, network convergence slows, and accuracy decreases as convolutional neural networks reach a certain depth. The ResNet-50 architecture consists of 50 deep layers. By applying convolutional filters of different sizes to the same layer, features at different scales can be obtained. Each epoch randomly divides the training dataset into 32 batches and runs for 1000 epochs with a learning rate optimized by the Adam optimizer. Transfer learning utilizes existing knowledge to learn new knowledge; its core is finding the similarity between existing and new knowledge. When facing similar tasks, it can leverage existing resources, thereby reducing computational cost. The top-level output of the ResNet-50 model was replaced with the dental implant prediction classification, resulting in only two nodes as the final output. The input-output pairs of the ResNet-50 model were fine-tuned on full-circumferential image training samples and / or periapical image training samples to adapt it to the dental implant prediction task. The enhanced full-circumference image training samples and / or periapical image training samples are input into a pre-trained ResNet-50 architecture deep learning convolutional neural network model 3 for transfer learning, training and optimizing the deep learning convolutional neural network structure. The pre-trained ResNet-50 architecture is used as the basis for transfer learning here, but the model and training method are not limited to it.

[0084] Figure 3 This is a schematic diagram illustrating the implementation process of the implant prediction module in the deep learning-based implant prediction method of this invention. Figure 3 As shown, the user obtains the original image sample 4 by cropping a panoramic X-ray image of a patient's tooth to be identified, and obtains the corresponding periapical image 5 of the implant 422 to be predicted through image recognition and screenshotting. The original image sample 4 and the periapical image 5 are input into the trained deep convolutional neural network model 6 to perform calculations, obtain the confidence scores for each category of implant placement results, and take the result with the highest confidence score as the prediction result.

[0085] In this process, the deep learning-based implant prediction method of this invention uses a deep learning model to learn features and adjacent details from a mixture of periapical and panoramic images. The preprocessed periapical and panoramic radiographs demonstrate higher accuracy in predicting implant failure due to peri-implant lesions, effectively predicting implant failure. This facilitates early clinical intervention for potential implant failure, thereby preventing implant failure and related complications. The data provided by this research may assist clinicians in making decisions and has broad application scenarios in the fields of artificial intelligence and medicine.

[0086] Figure 4This is a schematic diagram of the modules of the deep learning-based dental implant prediction system of the present invention. Figure 4 As shown, embodiments of the present invention also provide a deep learning-based dental implant prediction system for implementing the above-described deep learning-based dental implant prediction method. The deep learning-based dental implant prediction system includes:

[0087] The original image module 51 establishes a set of original image samples based on the panoramic X-ray image of the user's teeth.

[0088] Training sample module 52 obtains a set of full-circumference image training samples and a set of periapical image training samples based on the set of original image samples.

[0089] The implant prediction module 53 inputs the full-circumference image training samples and the periapical training samples into the trained neural network to obtain the implant prediction results.

[0090] In a preferred embodiment, the original image module 51 is configured to acquire panoramic X-ray images of the user's teeth after periapical implantation and implantation status information. The implantation status information includes at least a preset number of the implanted tooth in the panoramic X-ray image and the implantation result, which includes successful or unsuccessful implantation. A set of original image samples is established based on the panoramic X-ray images of the user's teeth, and a mapping relationship is established between each original image sample and the corresponding implantation status information.

[0091] In a preferred embodiment, the training sample module 52 is configured to perform image recognition on the X-ray panoramic image to obtain the image region of each tooth in the image. Based on the positional relationship of the teeth in the X-ray panoramic image, a mapping relationship is established with preset tooth numbers. According to the implantation status information corresponding to the X-ray panoramic image, the image region of the corresponding implant tooth is obtained from the X-ray panoramic image. A cropping is performed on the X-ray panoramic image with the center of the implant tooth's image region and according to a preset window size to obtain a periapical image. Sets of training samples for the whole-circumference image and periapical image are established separately, and a mapping relationship between the periapical image, the X-ray panoramic image, and the implantation status information is established.

[0092] In a preferred embodiment, the implant prediction module 53 is configured to perform transfer learning on circumferential image training samples and periapical training samples using a pre-trained deep learning convolutional neural network model, train and optimize the deep learning convolutional neural network structure, and save the trained deep learning convolutional neural network model. A circumferential image training sample and a periapical training sample of a user's tooth to be identified are input into the trained deep convolutional neural network model for calculation to obtain prediction results for implant failure and peri-implantitis.

[0093] The deep learning-based implant prediction system of this invention can learn features and adjacent details from a mixture of periapical and panoramic images using a deep learning model. Preprocessed periapical and panoramic radiographs show higher accuracy in predicting implant failure due to peri-implant lesions, effectively predicting implant failure. It facilitates early clinical intervention for potential implant failure, thereby preventing implant failure and related complications. The data provided by this study may help clinicians make informed decisions.

[0094] This invention also provides a deep learning-based dental implant prediction device, including a processor and a memory storing executable instructions for the processor. The processor is configured to execute steps of a deep learning-based dental implant prediction method via the execution of the executable instructions.

[0095] As shown above, the deep learning-based implant prediction device of this invention uses a deep learning model to learn features and adjacent details from a mixture of periapical and panoramic images. Preprocessed periapical and panoramic radiographs demonstrate higher accuracy in predicting implant failure due to peri-implant lesions, effectively predicting implant failure. This facilitates early clinical intervention for potential implant failure, thereby preventing implant failure and related complications. The data provided by this study may assist clinicians in making informed decisions.

[0096] Those skilled in the art will understand that various aspects of the present invention can be implemented as systems, methods, or program products. Therefore, various aspects of the present invention can be specifically implemented in the following forms: a completely hardware implementation, a completely software implementation (including firmware, microcode, etc.), or a combination of hardware and software aspects, collectively referred to herein as a "circuit," "module," or "platform."

[0097] Figure 5 This is a schematic diagram of the deep learning-based dental implant prediction device of the present invention. See below for reference. Figure 5 To describe an electronic device 600 according to this embodiment of the present invention. Figure 5 The electronic device 600 shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.

[0098] like Figure 5 As shown, the electronic device 600 is presented in the form of a general-purpose computing device. The components of the electronic device 600 may include, but are not limited to: at least one processing unit 610, at least one storage unit 620, a bus 630 connecting different platform components (including storage unit 620 and processing unit 610), a display unit 640, etc.

[0099] The storage unit stores program code, which can be executed by the processing unit 610 to perform the steps described in the above-described section on the electronic prescription transfer processing method according to various exemplary embodiments of the present invention. For example, the processing unit 610 can perform actions such as... Figure 1 The steps are shown in the figure.

[0100] Storage unit 620 may include a readable medium in the form of a volatile storage unit, such as random access memory (RAM) 6201 and / or cache memory 6202, and may further include a read-only memory (ROM) 6203.

[0101] Storage unit 620 may also include a program / utility 6204 having a set (at least one) program module 6205, such program module 6205 including but not limited to: operating system, one or more application programs, other program modules and program data, each or some combination of these examples may include an implementation of a network environment.

[0102] Bus 630 can represent one or more of several types of bus structures, including a memory cell bus or memory cell controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus using any of the multiple bus structures.

[0103] Electronic device 600 can also communicate with one or more external devices 700 (e.g., keyboard, pointing device, Bluetooth device, etc.), and with one or more devices that enable a user to interact with electronic device 600, and / or with any device that enables electronic device 600 to communicate with one or more other computing devices (e.g., router, modem, etc.). This communication can be performed via input / output (I / O) interface 650. Furthermore, electronic device 600 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 660. Network adapter 660 can communicate with other modules of electronic device 600 via bus 630. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with electronic device 600, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage platforms.

[0104] This invention also provides a computer-readable storage medium for storing a program that, when executed, implements the steps of a deep learning-based dental implant prediction method. In some possible implementations, various aspects of the invention can also be implemented as a program product comprising program code that, when run on a terminal device, causes the terminal device to perform the steps described in the above-described electronic prescription processing method section of this specification according to various exemplary embodiments of the invention.

[0105] As shown above, the program on the computer-readable storage medium of this embodiment, when executed, uses a deep learning model to learn features and adjacent details from a mixed image of periapical and panoramic images. The preprocessed periapical and panoramic images demonstrate higher accuracy in predicting implant failure, effectively predicting the occurrence of implant failure. This facilitates early clinical intervention for potential implant failure, thereby preventing implant failure and related complications. The data provided by this study may assist clinicians in making informed decisions.

[0106] Figure 6 This is a schematic diagram of the structure of the computer-readable storage medium of the present invention. (Reference) Figure 6 As shown, a program product 800 for implementing the above-described method according to an embodiment of the present invention is described. This product may employ a portable compact disc read-only memory (CD-ROM) and include program code, and may run on a terminal device, such as a personal computer. However, the program product of the present invention is not limited thereto. In this document, the readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, apparatus, or device.

[0107] The program product may employ any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of readable storage media include: electrical connections having one or more wires, portable disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0108] Computer-readable storage media may include data signals propagated in baseband or as part of a carrier wave, carrying readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable storage medium may also be any readable medium other than a readable storage medium that can transmit, propagate, or transfer a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination thereof.

[0109] Program code for performing the operations of this invention can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can execute entirely on the user's computing device, partially on the user's device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0110] In summary, the purpose of this invention is to provide a deep learning-based method, system, device, and storage medium for predicting dental implant failure. A deep learning model learns features and adjacent details from a mixture of periapical and panoramic images. Preprocessed periapical and panoramic radiographs demonstrate higher accuracy in predicting peri-implant lesions and failure, effectively predicting implant failure. This facilitates early clinical intervention for potential implant failure, thereby preventing implant failure and related complications. The data provided by this research may assist clinicians in making informed decisions.

[0111] The above description, in conjunction with specific preferred embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various simple deductions or substitutions can be made without departing from the concept of the present invention, and all such modifications and substitutions should be considered within the scope of protection of the present invention.

Claims

1. A deep learning-based dental implant prediction method, characterized by, The method comprises the following steps: S110, establishing a set of original image samples based on X-ray panoramic images of user teeth after periapical implantation; S120, obtaining a set of full-periosteal image training samples and a set of periapical image training samples based on the set of original image samples; S130, inputting the full-periosteal image training samples and the periapical training samples into a pre-trained deep learning convolutional neural network model, performing transfer learning, training and optimizing the structure of the deep learning convolutional neural network, and saving the trained deep learning convolutional neural network model; inputting a full-periosteal image and a periapical image of a user tooth to be identified into the trained deep convolutional neural network model for calculation to obtain a dental implant prediction result; The S110 comprises: S111, collecting X-ray panoramic images of user teeth after periapical implantation and implantation state information, wherein the implantation state information at least includes a preset number of implanted teeth in the X-ray panoramic image and an implantation result, and the implantation result includes implantation success or implantation failure; S112, establishing a set of original image samples based on the X-ray panoramic images of the user teeth, and establishing a mapping relationship between each original image sample and corresponding implantation state information; The S120 comprises: S121, performing image recognition on the X-ray panoramic images to obtain an image area of each tooth in the image; S122, establishing a mapping relationship between the image area of the tooth and the preset tooth number based on the positional relationship of the tooth in the X-ray panoramic image; S123, according to the preset number of implanted teeth contained in the implantation state information corresponding to the X-ray panoramic image, obtaining the image area of the corresponding implanted tooth in the X-ray panoramic image as a periapical image according to the mapping relationship; S124, respectively establishing a set of full-periosteal image training samples and a set of periapical image training samples, and establishing a mapping relationship among the full-periosteal image, the periapical image, the X-ray panoramic image, and the implantation state information; wherein the full-periosteal image is obtained by cropping the X-ray panoramic image. 2.The deep learning-based dental implant prediction method of claim 1, wherein, The S123 comprises: S1231, according to the preset number of implanted teeth contained in the implantation state information corresponding to the X-ray panoramic image, obtaining the image area of the corresponding implanted tooth in the X-ray panoramic image according to the mapping relationship between the image area of the tooth and the preset tooth number; S1232, centering on the image area of the implanted tooth, taking a screenshot in the X-ray panoramic image according to a preset window size to obtain the periapical image. 3.The deep learning-based dental implant prediction method of claim 1, wherein: The full-periosteal image training samples and / or the periapical training samples are rotated or flipped to obtain new full-periosteal image training samples and / or periapical image training samples. 4.The deep learning-based dental implant prediction method of claim 1, wherein: The deep learning convolutional neural network model adopts a ResNet-50 architecture. 5.A deep learning-based dental implant prediction system for implementing the deep learning-based dental implant prediction method of claim 1, characterized by, The method comprises: An original image module, which establishes a set of original image samples based on X-ray panoramic images of user teeth after periapical implantation; The method comprises the following steps: collecting X-ray panoramic images of teeth of a user after periapical implantation and implantation state information, the implantation state information at least including a preset number of implanted teeth in the X-ray panoramic images and an implantation result, the implantation result including implantation success or implantation failure; and establishing a set of original image samples based on the X-ray panoramic images of the teeth of the user, each original image sample being in a mapping relationship with corresponding implantation state information. The training sample module obtains a set of full-periosteal image training samples and a set of periapical image training samples based on the set of original image samples. The method comprises the following steps: performing image recognition on the X-ray panoramic images to obtain an image area of each tooth in the images; establishing a mapping relationship between the image area of the tooth and a preset tooth number based on the positional relationship of the tooth in the X-ray panoramic images; obtaining an image area of a corresponding implanted tooth in the X-ray panoramic images as a periapical image according to the mapping relationship according to the preset number of implanted teeth contained in the implantation state information corresponding to the X-ray panoramic images; and establishing a set of full-periosteal image training samples and a set of periapical image training samples, and establishing a mapping relationship among the full-periosteal image, the periapical image, the X-ray panoramic image, and the implantation state information. The full-periosteal image is obtained by cropping the X-ray panoramic image. The dental implant prediction module inputs the full-periosteal image training sample and the periapical training sample into a pre-trained deep learning convolutional neural network model, performs transfer learning, trains and optimizes the structure of the deep learning convolutional neural network, and saves the trained deep learning convolutional neural network model. The full-periosteal image and the periapical image of a user's teeth to be identified are input into the trained deep convolutional neural network model for calculation to obtain a dental implant prediction result. 6.A deep learning-based dental implant prediction device, characterized by, The method comprises the following steps: a processor; a memory having executable instructions stored therein; wherein the processor is configured to execute the steps of the deep learning-based dental implant prediction method of any one of claims 1 to 4 by executing the executable instructions.

7. A computer-readable storage medium for storing executable instructions, the computer-readable storage medium comprising: The executable instructions, when executed by the processor, implement the deep learning-based dental implant prediction method of any one of claims 1 to 4.

Images