Computer-implemented method for identifying a dental implant

Abstract

A computer-implemented method for identifying a dental implant, comprising: - an image obtaining step comprising of obtaining one or more digital images comprising at least one dental prosthetic such as a locator abutment depicted in the image; - an image processing step comprising of - detecting one or more prosthetic parameters of the at least one prosthetic depicted in the one or more digital images; - predicting one or more prosthetic candidates on the basis of the detected one or more prosthetic parameters using a trained neural network; - an identification step comprising of - selecting one of the predicted one or more prosthetic candidates; - identifying one or more dental implants on the basis of the selected predicted prosthetic candidate; - performing the identification of the one or more dental implants.

Description

[0001] Computer-implemented method for identifying a dental implant

[0002] The present invention relates to the field of digital dentistry, and more particularly to techniques for identifying a dental implant using computer-implemented methods.

[0003] Dentists have to replace a crown or bridge from time to time. This process is often accompanied by removing or replacing the prosthetic, which can be an invasive procedure. When replacing the prosthetic, dentists must assess the existing structure of the implant and decide how they will place the new prosthetic onto the existing implant. This is often done on the basis of visual inspection and experience, which leaves room for human error.

[0004] A wide range of dental implants is furthermore commercially available, each with their own unique characteristics and fitting requirements. This makes it very difficult for a dentist to determine with certainty which type of implant was originally placed. This may result in complications during the procedure, such as an incorrect fit of the new prosthetic or even damage to the underlying bone.

[0005] The drawback of the prior art is that it is almost impossible for a dentist to verify which type of implant was previously placed in the patient. This may result in uncertainty and errors during the procedure, which may in turn affect patient safety and treatment results.

[0006] The object of the present invention is therefore to provide a method which enables a dental implant to be identified.

[0007] According to a first aspect, a computer-implemented method is provided for identifying a dental implant. A dental implant can be understood as an artificial tooth root which is placed in the jawbone in order to support a missing tooth. The method comprises a number of steps. Firstly, the method comprises an image obtaining step wherein one or more digital images are obtained. These images show at least one dental prosthetic, such as a locator abutment. A locator abutment can be understood as an attachment part which is used to attach for instance dentures or a crown to a dental implant. The method subsequently comprises an image processing step. In this step one or more prosthetic parameters are detected from the images of the dental prosthetic. These prosthetic parameters are used to predict one or more prosthetic candidates using a trained neural network. A prosthetic candidate can be understood as a possible combination of dental prosthetics which are suitable on the basis of the detected prosthetic parameters. The method subsequently comprises an identification step. In this step one of the predicted prosthetic candidates is selected. One or more dental implants are then identified on the basis of the selected prosthetic candidate. The identification of the one or more dental implants is then performed. The technical advantage of this method is based on the insight that, by detecting prosthetic parameters and predicting prosthetic candidates on the basis of these parameters, the computer-implemented method is able to accurately identify dental implants. This provides an efficient and accurate method for selecting and identifying dental implants on the basis of the characteristics of the dental prosthetic. This allows dental professionals to save time and improve the quality of the dental treatment.

[0008] The computer-implemented method preferably comprises of identifying one or more dental implants, wherein this identifying comprises of selecting one or more dental implants which are associated with the selected predicted prosthetic. In other words, provision can be made that this method comprises of selecting one or more dental implants which are associated with the selected predicted prosthetic. This means that not just any implant is selected, but implants matching the previously selected prosthetic are specifically sought. A technical advantage thereof is that this can result in a better fit and functionality of the final prosthetic structure. Selecting specifically on the basis of compatibility with the predicted prosthetic makes it possible to reduce problems such as discomfort, pain or poor aesthetic results. In this way more efficient use can also be made of available means, since there is less risk of wastage due to unsuitable implants being selected.

[0009] The selecting of the one or more dental implants preferably comprises of selecting from an association database. An association database can be understood as a centralized storage location containing association data which associate one or more prosthetics with one or more dental implants. This means that the database contains data which describe the connection between different prosthetic devices and the corresponding dental implants. Use of this database enables an accurate and efficient selection process of suitable dental implants to be performed, taking into consideration the specific requirements and characteristics of each predicted prosthetic. An advantage thereof is that this results in an optimized selection of dental implants for each individual patient, this resulting in better treatment results. Use of such an association database moreover contributes to a more efficient use of storage capacity since only data relevant to each specific situation are stored and used.

[0010] The predicting of one or more prosthetic candidates on the basis of the detected one or more prosthetic parameters preferably makes use of a trained neural network. This trained neural network is used to determine a probability score which indicates the probability of the predicted one or more prosthetic candidates corresponding with a prosthetic stored in a prosthetics database. A trained neural network can be understood as a form of artificial intelligence which learns from examples and recognizes patterns, while the prosthetics database can be seen as a centralized storage location in which different types of prosthetic are stored. Provision can be made that making use of this trained neural network to determine the probability score helps in predicting potential prosthetic candidates more accurately and efficiently. Use of this technology enables complex patterns and connections in the data to be recognized and used to make better predictions. A technical advantage thereof is that it leads to better treatment results in that more accurate selections of prosthetic candidates are possible. The one or more prosthetic parameters preferably comprise at least one or a combination of a prosthetic type, a length of the prosthetic, a diameter of the prosthetic, a material, a connecting means between prosthetic and implant, and a connecting means between prosthetic and crown or bridge. These parameters are deemed elements which can affect the choice for determined prosthetic types, and thereby the treatment results. A technical advantage thereof is that, by taking into consideration these parameters in the selection of potential prosthetic candidates, it becomes possible to make more accurate predictions. This leads to better treatment results and also contributes to more efficient use of computing power and storage capacity, since less time and fewer means are required to identify suitable candidates. Provision can further be made that the image processing step comprises of predicting the prosthetic type as a prosthetic parameter. The prosthesis type can have different characteristics which affect both the functional and the aesthetic results of dental implant procedures. The technical advantage thereof is that, by also taking into consideration the prosthetic type in the prediction, even more accurate predictions can be made in respect of the prosthetic types that are likely to be suitable for each individual patient. This does not just enable better treatment results.

[0011] The predicting of the prosthetic type is preferably performed with a neural network trained on the basis of a classification model. A classification model can be understood as a predictive model which is used to predict the class or category of a determined datum, in this case the prosthetic type. Provision can be made that, by making use of such a trained neural network, it is possible to make more accurate and efficient predictions in respect of the prosthetic type. This is because neural networks can recognize complex patterns and connections in the data which may not be noted by human specialists. A technical advantage thereof is that it leads to better treatment results by enabling more accurate selections of prosthetic candidates.

[0012] The image processing step preferably also comprises of predicting the length of the prosthetic as a prosthetic parameter. The length of the prosthetic is a fundamental factor which can affect the functionality and aesthetic of the final dental restoration. Taking this parameter into consideration enables even more accurate and more personalized predictions to be made in respect of the prosthesis types that are likely to be suitable for each individual patient.

[0013] The computer-implemented method preferably comprises an addition step. An addition step can be understood as a process wherein new data are input and integrated in the existing system. Provision can be made that this step comprises of inputting one or more new images of dental prosthetics. The one or more new images are further labelled with one or more prosthetic parameters. This means that specific characteristics, properties or parameters which are relevant to the dental prosthetic are coupled to the relevant images. An association is then made between the new one or more dental prosthetics shown in the image and one or more implants. This means that a connection is made between these two components, probably on the basis of compatibility, functionality, aesthetics or other relevant factors. Finally, the neural network is trained further using these new, labelled images and their associated implants. A technical advantage thereof is that this process makes it possible to continuously stay up to date on recent developments in dental prostheses and implants by means of constant input and training with new data.

[0014] According to a further aspect, a computer program product is provided, comprising instructions which, when the program is run on a computer, make the computer perform the method as described above.

[0015] The invention will now be further described on the basis of an exemplary embodiment shown in the drawing.

[0016] In the drawing, figure 1 shows schematically a flow of an exemplary embodiment of a method for identifying a dental implant.

[0017] In the drawings, figures 2 and 3 show an exemplary embodiment of an implant structure and various abutments.

[0018] The following detailed description relates to determined specific embodiments. The teaching hereof can however be applied in different ways. The same or similar elements are designated in the drawings with the same reference numerals.

[0019] The present invention will be described with reference to specific embodiments. The invention is however not limited thereto, but solely by the claims.

[0020] As used here, the singular forms “a” and “the” comprise both the singular and plural references, unless clearly indicated otherwise by the context.

[0021] The terms “comprising”, “comprises” and “composed of’ as used here are synonymous with “including”. The terms “comprising”, “comprises” and “composed of’ when referring to stated components, elements or method steps also comprise embodiments which “consist of’ the components, elements or method steps.

[0022] The terms first, second, third and so on are further used in the description and in the claims to distinguish between similar elements and not necessarily to describe a sequential or chronological order, unless this is specified. It will be apparent that the thus used terms are mutually interchangeable under appropriate circumstances and that the embodiments of the invention described here can operate in an order other than described or illustrated here.

[0023] Reference in this specification to “one embodiment”, “an embodiment”, “some aspects”, “an aspect” or “one aspect” means that a determined feature, structure or characteristic described with reference to the embodiment or aspect is included in at least one embodiment of the present invention. The manifestations of the sentences “in one embodiment”, “in an embodiment”, “some aspects”, “an aspect” or “one aspect” in different places in this specification thus do not necessarily all refer to the same embodiment or aspects. As will be apparent to a skilled person in this field, the specific features, structures or characteristics can further be combined in any suitable manner in one or more embodiments or aspects. Although some embodiments or aspects described here comprise some but no other features which are included in other embodiments or aspects, combinations of features of different embodiments or aspects are further intended to fall within the context of the invention and to form different embodiments or aspects, as would be apparent to the skilled person. In the appended claims all features of the claimed embodiments or aspects can for instance be used in any combination.

[0024] Figure 1 shows an exemplary embodiment of a flow of a computer-implemented method 100 for identifying a dental implant 201. The implant 201 is shown in figures 2 and 3. A computer-implemented method can be understood as a method which is performed by a computer program. This method makes use of steps for performing specific tasks, such as identifying a dental implant.

[0025] The computer-implemented method 100 comprises an image obtaining step 100 which comprises of obtaining one or more digital images 101, 102, 103. This can mean that one image is obtained, but it can also mean that a plurality of images is obtained from different angles or with different technologies in order to obtain a complete picture of the prosthetic. There are many ways of obtaining images for identifying a dental implant. One of the most common methods is to make digital images using an intraoral camera. This is a small camera which is placed in the patient's mouth in order to make detailed images of the teeth and the gums. Another method of obtaining images is by means of x-radiation. This can be done using a digital x-ray, a panoramic x-ray or a cone beam computed tomography (CBCT) scan. These methods can produce detailed images of the implant and the surrounding tissue. Finally, images can also be obtained by means of photography. This can be done using a digital camera and can be useful for documenting the position and the appearance of the prosthetic.

[0026] The images 101, 102, 103 comprise at least one dental prosthetic, shown in figures 2 and 3, such as a locator abutment. This means that the locator abutment need not necessarily be the only component in the images, but that other dental prosthetics may also be shown. It is however preferred to show only a locator abutment in the images, since this makes the identification process more efficient and accurate. This is because the computer program can then focus fully on identifying specific characteristics of the prosthetic, without being distracted by other dental prosthetics in the image. This results in a highly accurate identification of the prosthetic and minimizes the chance of errors. A locator abutment is a type of dental prosthetic which is used in combination with an implant to hold dentures in place. The locator abutment typically comprises a metal pin which is screwed onto the implant and a rubber ring which is attached to the dentures. An advantage of using only a locator abutment in the images is that the identification process becomes more efficient and more accurate. Showing only the locator abutment enables the computer program to focus fully on identifying specific properties of the locator abutment, without being distracted by other dental prosthetics in the image. This results in a highly accurate identification of the abutment and minimizes the chance of errors.

[0027] The method 100 further comprises an image processing step 200 comprising of detecting 210 one or more prosthetic parameters of the at least one prosthetic depicted in the one or more digital images 101, 102, 103, and of predicting 220 of one or more prosthetic candidates on the basis of the detected one or more prosthetic parameters using a trained neural network. An image processing step 200 can be understood as a process wherein digital images are analysed and processed to obtain determined characteristics or information. This process comprises of detecting 210 one or more prosthetic parameters of the at least one prosthetic depicted in the one or more digital images 101, 102, 103. This means that the computer program must detect specific properties of the dental prosthetic, such as the size, shape, colour and position of the implant.

[0028] The method 100 then predicts 220 one or more prosthetic candidates 221, 222, 223 on the basis of the detected one or more prosthetic parameters using a trained neural network. An advantage of this image processing step is that the process is quick and efficient. Use of a trained neural network enables the computer program to predict prosthetic candidates very quickly and accurately on the basis of the detected parameters. This results in fast and accurate identification of the implant in a further step. One or more prosthetic candidates 221, 222, 223 can be interpreted as a list of one or more possible implant options predicted by method 100 on the basis of the detected prosthetic parameters of the digital images 101, 102, 103. This means that the program can predict one or possibly a plurality of prosthetic options, depending on the detected parameters and the database of known prosthetic candidates. The predicting 220 of one or more prosthetic candidates on the basis of the detected one or more prosthetic parameters using a trained neural network can comprise of determining a probability score of the predicted one or more prosthetic candidates corresponding with a prosthetic stored in a prosthetics database. The higher the score, the greater the chance that the predicted prosthetic candidate corresponds with the prosthetic in the database. This step has the result that the method 100 can predict the most probable prosthetic candidate and can then identify the correct dental implants. Use of a probability score makes the identification process still more accurate and minimizes the chance of errors.

[0029] Method 100 further comprises an identification step 300. The identification step 300 can be understood as the process wherein the method 100 proposes the most probable implant or the most probable implants on the basis of the predicted prosthetic candidates. This comprises of selecting one of the predicted one or more prosthetic candidates, identifying 310 one or more dental implants on the basis of the selected predicted prosthetic candidate and carrying out 320 the identification of the one or more dental implants. An example of the selecting of one of the predicted prosthetic candidates is when the method 100 predicts a plurality of possible implant options, for instance three prosthetic candidates. Identifying 310 one or more dental implants on the basis of the selected predicted prosthetic candidate comprises of using the detected prosthetic parameters and the selected implant option to identify the implant or the implants.

[0030] Identifying 310 the one or more dental implants can comprise of selecting 310 one or more dental implants which are associated with the selected predicted prosthetic. This means that method 100 searches specifically for the prosthetic implants that correspond with the selected predicted prosthetic.

[0031] An advantage of this step is that it makes the identification of the implant even more accurate by focussing on the specific implants that are relevant to the selected prosthetic option. This minimizes the chance of errors and results in a highly accurate identification of the implant. An alternative to this approach would be to identify a wider range of dental implants and then manually selecting which implants correspond with the selected prosthetics option. This would however require more time and effort, and would be less accurate than using a computer program which searches specifically for the implants corresponding with the selected prosthetics option.

[0032] Selecting the predicted prosthetic candidate more preferably comprises of selecting from an association database. This association database comprises association data which associate one or more prosthetics with one or more dental implants. Making use of this association database enables the computer program to predict the most probable prosthetic candidates with very high accuracy and to then identify the correct dental implants. This results in a highly accurate identification of the implant and minimizes the chance of errors.

[0033] The above described one or more prosthetic parameters comprise at least one or a combination of a prosthetic type, a length of the prosthetic, a diameter of the prosthetic, a material, a connecting means between prosthetic and implant, a connecting means between prosthetic and crown or bridge. The prosthetic parameters which are used to predict the prosthetic candidates can differ depending on the specific application and the prosthetic type for which the method is used. The prosthetic type is for instance a locator abutment, a ball abutment, a bar or a frame. The length of the prosthetic is for instance the length of the abutment or a portion of the abutment. A connecting means between prosthetic and implant can also be a prosthetic parameter, for instance screw retention, cement retention or magnetic retention. Further prosthetic parameters can for instance be the shape of the prosthetic, surface properties or the location of the prosthetic in the mouth. Other prosthetic parameters are inter alia the thread type, the seat of the screw, the diameter of the screw head, the metric of the thread, the length of the head, the length of the thread, the diameter of the thread, the prosthetic bed (diameter, design), the pitch, the angle, the lead, the screw head form / connection (screwdriver) and the screw connection in a separate component. Thread type refers to the form and size of the thread which is used in attaching the prosthesis to the implant. Different thread types are used depending on the implant type, the size of the prosthesis and the desired stability. The seat of the screw relates to the manner in which the screw fits onto the implant and provides for a firm and stable connection. The seat of the screw must fit exactly on the implant in order to prevent the prosthesis from coming loose. The diameter of the screw head is the size of the head of the screw which is used to attach the prosthesis onto the implant. The diameter of the screw head must correspond with the size of the implant in order to create a firm connection. Metric of the thread is the size of the thread which is used in attaching the prosthesis onto the implant. The metric of the thread must correspond with the type of implant being used. Length of the head is the length of the screw head which is used to attach the prosthesis onto the implant. The length of the head must correspond with the size of the implant and the prosthesis in order to create a firm connection. Length of the thread is the length of the thread which is used in attaching the prosthesis onto the implant. The length of the thread must correspond with the depth of the implant in order to create a firm connection. Diameter of the thread relates to the width of the thread which is used in attaching the prosthesis onto the implant. The diameter of the thread must correspond with the type of implant being used. Pitch is the distance between the teeth or molars on the prosthetic bed. The pitch must correspond with the size and shape of the missing teeth or molars in order to create a natural appearance. The angle relates to the angle at which the teeth or molars are placed on the prosthetic bed. The angle must correspond with the angle of the adjacent teeth or molars in order to create a natural appearance. In addition to said prosthetic parameters, there are also processing components that can help in determining the position of the implant and the stability thereof. An example of such a component is the Osstell® instrument which is used to measure the stability of the implant using pins. The Osstell® instrument is a measuring instrument which is used to determine the stability of the implant by means of resonance frequency analysis. The instrument consists of a handpiece with a measuring head and a number of different pins of different lengths and diameters. The pins are placed on the implant and a measurement is performed in order to determine the stability of the implant. The Osstell® instrument can also aid in determining the position of the implant by placing the pins at different locations and measuring the stability at every location. This can help in finding the ideal position for the implant and in determining the suitable prosthetic parameters. Other processing components that can aid in determining the position of the implant are for instance x-ray equipment and computer-aided implantology.

[0034] With particular preference the image processing step comprises of predicting the prosthetic type as a prosthetic parameter of the one or more prosthetic parameters. A prosthetic type refers to the specific form and function of the dental prosthetic which is attached to the implant. Predicting the prosthetic type is performed using a trained neural network which is based on a classification model. A classification model is a type of machine learning model used to classify objects on the basis of a series of characteristics. In this case the classification model is used to classify the type of prosthetic on the basis of the characteristics of the prosthetic parameters detected in the digital images. The trained neural network analyses the characteristics of the prosthetic parameters and determines which prosthetic type is most likely to correspond with the detected characteristics. The neural network is trained with a dataset of digital images depicting different prosthetics and / or depicting different views of the same prosthetic and corresponding prosthetic parameters to teach the model which characteristics correspond with which prosthetic type. Predicting the prosthetic type using a trained neural network is a very efficient and accurate method of identifying the prosthetic type. It minimizes the chance of errors and can accelerate the identification process to considerable extent, allowing the patient to receive the correct dental treatment more quickly.

[0035] With particular preference the image processing step further also comprises of predicting the length of the prosthetic as a prosthetic parameter of the one or more prosthetic parameters. Predicting the length of the prosthetic is performed with a neural network trained on the basis of a regression model, which neural network is trained on the basis of varying independent variable parameters such as a prosthetic type and measuring the corresponding length of the prosthetic, wherein the length is a dependent variable parameter. A regression model is a type of machine learning model which is used to predict a dependent variable on the basis of the value of one or more independent variables. In this case the regression model is used to predict the length of the prosthetic on the basis of the characteristics of the prosthetic parameters detected in the digital images, or to extract the length directly from the image. The trained neural network analyses the characteristics of the prosthetic parameters and determines the length of the prosthetic which is most likely to correspond with the detected characteristics. The neural network is trained with a dataset of digital images and corresponding prosthetic parameters to teach the model which characteristics correspond with which length of the prosthetic. Predicting the length of the prosthetic as a prosthetic parameter enables the computer program to predict the prosthetic candidates even more accurately and efficiently and to then identify the correct dental implants. This minimizes the chance of errors and ensures that the patient receives the correct treatment.

[0036] The computer-implemented method 100 preferably also comprises an addition step (not shown) to further improve the trained neural network. The addition step comprises of inputting one or more new images of dental prosthetics, labelling the one or more new images of the dental prosthetics with one or more prosthetic parameters, associating the new one or more dental prosthetics shown in the image with one or more implants, and training the neural network further using the labelled one or more new images and the associated one or more implants. Labelling the new images with the prosthetic parameters and associating the dental prosthetics with the correct implants is essential for training the neural network further and improving the accuracy of the predictions. The addition of new images and implants to the training dataset enables the neural network to learn to recognize and predict new and different prosthetic parameters and implants. Further training of the neural network using these new images and implants results in an increasingly accurate prediction of the prosthetic candidates and identification of the correct dental implants. This has the result that the method continues to improve the more images and implants are added to the training dataset.

[0037] Figure 2 shows an exemplary embodiment of an implant structure. The implant structure comprises different components which function together to support dental prosthetics. The implant structure shown in figure 2 comprises a dental implant 201, a dental prosthetic 202, more specifically a locator abutment, and a crown 203. A dental implant 201 can be understood as an artificial tooth root which is placed in the jawbone 204 in order to support a missing tooth. An example of a dental implant is a titanium helical implant which is biocompatible and designed to promote bone integration, an example being shown in both figure 2 and figure 3. The implant can vary in respect of form, such as cylindrical or conical, depending on the required clinical application and the anatomy of the patient. The surface of the implant can further be treated, for instance by means of sandblasting or acid etching, in order to improve attachment to the jawbone.

[0038] A locator abutment can be understood as an attaching part which is used to attach for instance dentures or a crown to the dental implant 201. Figure 3 shows different embodiments of an abutment 202a-202f, each designed to fulfil specific functional and aesthetic requirements within dental implant implantology. The locator abutment 202 can be manufactured from different materials, including titanium, stainless steel and ceramic material, wherein the choice is determined by the required strength and aesthetic properties. Titanium is often used for its excellent biocompatibility and mechanical strength, while ceramic material is opted for for its natural tooth colour and aesthetic appeal.

[0039] Depending on the design, the locator abutment can be manufactured integrally, wherein the connecting part is integrated with the implant in the abutment itself, for instance abutment 202b- 202f.

[0040] In an alternative embodiment the locator abutment can comprise two separate parts, wherein the connecting part comprises between the implant and the abutment a separate part, such as a screw. Examples of such a two-part structure are abutment 202a, 202d in figure 3, wherein a stainless steel screw 202g is used to attach the abutment firmly onto the implant. This design can be adapted and replaced in simple manner if necessary. Alternatively, the abutment can be secured using a bayonet connection, for instance abutment 202b. It provides a safe connection and enables the abutment to be removed and reattached in simple manner without wear to the screw thread.

[0041] The dental prosthetic is however not limited to a locator abutment, but can also comprise a healing cap 203, as shown in figure 3. A healing cap can be understood as a temporary part which is placed on a dental implant to allow the gums to heal and to promote formation of healthy tissue around the implant. A healing cap can be manufactured from biocompatible materials such as titanium or plastic, and the height and diameter thereof can vary in order to adapt to the individual anatomical requirements of the patient.

[0042] The abutments 202a-202f can be equipped with different types of connecting part with varying designs. A usual form is the screw thread connection, wherein the abutment is provided with an internal or external screw thread which firmly secures the crown. This design provides a reliable mechanical connection which can be released in simple manner for maintenance or replacement of the crown. In addition, a cement connection can be applied, wherein the crown is attached to the abutment with dental cement. This method provides aesthetic advantages since there are no visible connecting components, which is especially favourable in the visible part of the teeth. The cement connection however requires careful preparation to ensure a firm connection and provides less flexibility in terms of removability. An alternative form of connection is the click-on or snap-on connection, which makes use of a mechanism similar to press-studs. This method enables the user to click or snap the crown onto the abutment in simple manner, this providing convenience in respect of maintenance and cleaning, while guaranteeing a firm connection.

[0043] A magnetic connection can further be applied, wherein magnetic elements are integrated in both the abutment and the crown. This connection type provides the advantage that it enables simple removal and reattachment, while providing sufficient force to hold the crown in place during normal mouth functions.

[0044] Spherical connecting forms, often referred to as ball joint connections or ball and socket connections, make use of a spherical element which fits into a corresponding recess in the crown.

[0045] Finally, a bayonet connection can be provided wherein the crown is attached to the abutment by means of a turning and snapping motion. This design provides a strong mechanical connection and enables the crown to be removed and reattached in simple manner without wear to the screw thread.

[0046] As elucidated above, the prosthetic parameters can comprise specific characteristics of the connecting parts, such as the presence of screw thread, the form of cement connections, click-on or snap-mechanisms, magnetic elements, spherical elements and bayonet connections. For example, in a screw thread connection the network will focus on the presence of helical patterns that are characteristic for screw mechanisms. For cement connections, the network will pay attention to the absence of visible connecting components and the smoothness of the surface. Click-on or snap-on connections are detected by searching for patterns that are similar to press-studs, while magnetic connections are recognized by the presence of specific magnetic components. The prosthetic parameters such as the form and the type of connecting parts (for instance screw thread, cement, snap mechanism, magnet, spherical elements or bayonet connections) function as identifying characteristics intended specifically for determined implants. These characteristics are similar to a fingerprint, by which a specific implant can be recognized. When the neural network detects the prosthetic parameters in the obtained digital images, it uses this information to predict which implants correspond with the detected characteristics. The network is trained with an extensive dataset of known implants and their corresponding prosthetic, enabling it to make accurate predictions. After the prediction, the neural network selects the most probable implant candidates on the basis of the detected prosthetic parameters. This enables the dentist to identify the correct implant quickly without having to rely on manual and often time-consuming processes. For the dentist, this means that they can recognize and apply the correct implant in simple and efficient manner. The identification of the correct implant is essential to planning treatments, preparing surgical procedures and providing personalized care to patients.

[0047] The skilled person will appreciate on the basis of the above description that the invention can be embodied in different ways and on the basis of different principles. The invention is not limited here to the above described embodiments. The above described embodiments and the figures are purely illustrative and serve only to increase understanding of the invention. The invention is not therefore limited to the embodiments described herein, but is defined in the claims.

Claims

Claims1. A computer-implemented method for identifying a dental implant, comprising:- an image obtaining step comprising of obtaining one or more digital images comprising at least one dental prosthetic (202, 202a-202g; 203) such as a locator abutment depicted in the image;- an image processing step comprising of- detecting one or more prosthetic parameters of the at least one prosthetic depicted in the one or more digital images;- predicting one or more prosthetic candidates on the basis of the detected one or more prosthetic parameters using a trained neural network;- an identification step comprising of- selecting one of the predicted one or more prosthetic candidates;- identifying one or more dental implants on the basis of the selected predicted prosthetic candidate;- performing the identification of the one or more dental implants.

2. The computer-implemented method according to claim 1, wherein the identifying of the one or more dental implants comprises of selecting one or more dental implants which are associated with the selected predicted prosthetic.

3. The computer-implemented method according to claim 2, wherein the selecting comprises of selecting from an association database, which association database comprises association data which associate one or more prosthetics with one or more dental implants.

4. The computer-implemented method according to any one of the foregoing claims, wherein the predicting of one or more prosthetic candidates on the basis of the detected one or more prosthetic parameters using a trained neural network comprises of determining a probability score of the predicted one or more prosthetic candidates corresponding with a prosthetic stored in a prosthetics database.

5. The computer-implemented method according to any one of the foregoing claims, wherein the one or more prosthetic parameters comprise at least one or a combination of a prosthetic type, a length of the prosthetic, a diameter of the prosthetic, a material, a connecting means between prosthetic and implant, a connecting means between prosthetic and crown or bridge.

6. The computer-implemented method according to claim 5, wherein the image processing step further comprises of predicting the prosthetic type as a prosthetic parameter of the one or more prosthetic parameters.

7. The computer-implemented method according to claim 6, wherein the predicting of the prosthetic type is performed with a neural network trained on the basis of a classification model.

8. The computer-implemented method according to any one of the foregoing claims 5-7, wherein the image processing step further comprises of predicting the length of the prosthetic as a prosthetic parameter of the one or more prosthetic parameters.

9. The computer-implemented method according to claim 8, wherein the predicting of the length of the prosthetic is performed with a neural network trained on the basis of a regression model, which neural network is trained on the basis of varying independent variable parameters such as a prosthetic type and measuring the corresponding length of the prosthetic, wherein the length is a dependent variable parameter.

10. The computer-implemented method according to any one of the foregoing claims, further comprising an addition step comprising of- inputting one or more new images of dental prosthetics;- labelling the one or more new images of the dental prosthetics with one or more prosthetic parameters;- associating the new one or more dental prosthetics shown in the image with one or more implants;- training the neural network further using the labelled one or more new images and the associated one or more implants.

11. A computer program product comprising instructions which, when the program is run on a computer, make the computer perform the method according to any one of the foregoing claims.

Images