Manufacturing of retainers

Abstract

The invention relates to a method for manufacturing a holder, said method comprising the following steps: a) acquiring at least one two-dimensional image of at least a part of the dental arch of a user, referred to as the "updated image", using an image acquisition device at an update time after the end of an orthodontic treatment; b) generating, based on the updated image, a digital three-dimensional model representing the dental arch of the user, referred to as the "updated model"; c) designing, based on the update model and preferably by a computer, a holder adapted to hold the teeth in their configuration at the update time; and d) manufacturing the retainer.

Description

Technical Field

[0001] The present invention relates to methods and apparatus for designing and manufacturing orthodontic retainers. Prior art

[0002] After orthodontic treatment, the position of the teeth may change. This unfavorable change is often called a "relapse." Traditionally, patients visit their orthodontist for regular checkups. They may also visit their dentist, who can detect any flaws in the tooth positioning process.

[0003] However, many patients do not undergo these examinations, which would detect a possible relapse. As a result, the teeth may return to a malocclusion, and correcting this may require further orthodontic treatment that may be as extensive as the initial treatment.

[0004] Visiting the doctor is a heavy burden for patients and a stress for orthodontists. US patent application 2018 / 0204332 describes a method that partially solves this problem. However, at the end of orthodontic treatment, it is necessary to acquire at least one digital three-dimensional reference model of the patient's dental arch. Therefore, the patient must visit the orthodontist.

[0005] To prevent adverse changes in tooth position and therefore limit the number of appointments, patients may be asked to wear retainers to keep their teeth in the position they were in at the end of orthodontic treatment.

[0006] A retainer is typically an orthodontic retaining bracket or archwire, consisting of a metal wire that forms the dental arch and is secured against the teeth, usually bonded against the lingual surface from canine to canine. The shape of the orthodontic appliance is typically determined by the orthodontist by taking a 3D scan of the patient's dental arch after orthodontic treatment has been completed.

[0007] Acquiring a 3D scan of a patient's dental arch is time-consuming and requires in-person patient presence.

[0008] There is a need to make it easier to identify and manufacture orthodontic retainers after orthodontic treatment.

[0009] It is an object of the present invention to at least partially meet this need. Summary of the invention Summary of the invention

[0011] The present invention aims to meet some or all of these needs and, according to one aspect of the invention, does so using a method for manufacturing a retainer comprising the steps of:

[0012] a) preferably at an update time, at the end of orthodontic treatment or after the end of orthodontic treatment, using an image acquisition device to acquire at least one two-dimensional image of at least a portion of the user's dental arch, referred to as an "update image"; preferably several update images are acquired;

[0013] b) generating, preferably from the updated image, a digital three-dimensional model representing the user's dental arch, referred to as the "updated model";

[0014] c) designing a retainer based on the renewal model, the retainer being designed to, when positioned on the teeth, retain the teeth in their configuration at the time of renewal;

[0015] d) Manufacturing the retainer.

[0016] Preferably, in step b), the updated model is obtained by deforming a digital three-dimensional model of the user's dental arch acquired before step a) (referred to as a "reference model") based on the updated image. The reference model is deformed starting from the updated image until an updated model is obtained, the updated model representing the dental arch in a configuration matching the configuration of the dental arch represented on the updated image.

[0017] As will be seen in more detail later in the specification, the updated image can be obtained outside of the orthodontic clinic, and the reference model can be generated before or during orthodontic treatment. Specifically, the retainer can be manufactured without the need for an intraoral scan after orthodontic treatment. In this way, the determination and manufacture of the retainer does not require the user to see a dental care professional after orthodontic treatment has been completed, or even without seeing a dental care professional at all.

[0018] The method according to the invention thus advantageously makes it possible to limit the number of appointments with dental care professionals.

[0019] The method according to the invention may include one or more of the following optional features:

[0020] - The retainer is an archwire or an orthodontic retainer tray, preferably an archwire;

[0021] - Collect multiple update images, preferably 3, preferably 4, preferably 5;

[0022] - capturing a plurality of updated images at different capture angles, the updated images preferably comprising at least one image on the right relative to the user, at least one image on the left relative to the user, and at least one image facing the user;

[0023] - The updated image is collected by the user or his relatives;

[0024] - the acquisition in step a) takes place outside the orthodontic or dental clinic, preferably in the user's home;

[0025] -The image acquisition device is a mobile phone;

[0026] - using in step a) a dental retractor and / or an image acquisition kit as described, for example, in EP 3 391 810;

[0027] - the image acquisition kit is in the form of a box to which the image acquisition device is attached;

[0028] - the housing is tubular and has a first opening designed to be positioned in the mouth of a user and a second opening designed to be attached to an image acquisition device so that the image acquisition device acquires at least one image of the user's teeth through said openings;

[0029] - in step b), the updated model is obtained by deforming the digital three-dimensional model of the dental arch of the user acquired before step a) (referred to as the "reference model") based on the updated image;

[0030] - the reference model is generated from the three-dimensional digital model,

[0031] Preferably generated from a digital three-dimensional model acquired before the start of orthodontic treatment or during orthodontic treatment, for example using a 3D scanner;

[0032] - the reference model is an updated model resulting from the implementation of the method according to the invention, called

[0033] "historical update model", which preferably has been stored in the database after step b);

[0034] - the reference model is a three-dimensional digital model of the user's dental arch generated at the beginning of orthodontic treatment, specifically used to model the dental arch at or during the orthodontic treatment phase, for example more than 2 weeks or more than one month before step b);

[0035] - preferably, the reference model represents the dental arch of the user in a desired theoretical configuration at an intermediate moment of orthodontic treatment or in a desired theoretical configuration at a final moment of orthodontic treatment;

[0036] - preferably, the reference model represents the dental arch in a final configuration corresponding to the theoretical configuration desired at the end of orthodontic treatment;

[0037] - deforming the reference model using metaheuristics or neural networks;

[0038] - in step b), preferably before step b), preferably in step a), more preferably before step a), cutting the reference model into a tooth model, the deformation comprising a displacement of the tooth model, the displacement being equal to the difference in tooth positioning between the tooth configuration in the reference model and the tooth configuration in the updated image;

[0039] - in step b) or preferably before step b), preferably in step a), even more preferably before step a), training at least one neural network from a learning base consisting of historical images and a first historical model similar to the update image and the reference model, respectively, and a second historical model similar to the update model as expected at the output of the neural network, to determine an update model at the output of the neural network based on a reference model and an update image submitted at the input of the neural network;

[0040] - in step b), the updated model is obtained without using a reference model, in particular by combining a tooth model selected according to the updated image;

[0041] - In step c), the design of the orthodontic appliance involves determining the curvature of the archwire;

[0042] - preferably, the archwire is designed to adapt to the shape of the teeth to which it is to be applied, the archwire being generally designed to attach to the inner surface of the user's teeth;

[0043] - steps b) to c) are performed by a computer, preferably automatically, ie without operator intervention, or under the supervision of an operator, such as a technician or dental care professional;

[0044] - Step d) is performed using a 3D printer;

[0045] The method according to the invention comprises, after step d), the following step e):

[0046] e) Shipping the retainer to the user or dental care professional.

[0047] In one embodiment, the updated model is obtained by implementing a dental arch modeling method, preferably obtained via a user's mobile phone, the modeling method comprising the following steps:

[0048] 1) analyzing the updated image to determine at least one tooth region and at least one tooth attribute value associated with the tooth region, preferably the type or number of the tooth to which the tooth region belongs;

[0049] 2) for each tooth region determined in step 1), searching in a historical library containing more than 1000 tooth models called “historical tooth models” associated with the tooth attribute value to find a historical tooth model called “optimal tooth model” having the greatest proximity to the tooth region;

[0050] 3) Rank the set of best tooth models to create a model that best matches the update image, ie, the updated model.

[0051] The modeling method preferably includes one or more of the following features:

[0052] - Steps 1) and 2) of the modeling method are performed for a plurality of update images and preferably, in steps 2) and 3), it is sought to obtain the best tooth model and the model combined in such a way as to best match the set of update images;

[0053] - an attribute of a tooth selected from the group consisting of tooth type, tooth number, tooth thickness, crown height, mesio-palatal width of the tooth, incisal edge mesio-distal deflection index of the tooth, tooth color, tooth shape, tooth contour, clinical status of the tooth (e.g. "caries", "broken tooth"), position of the tooth relative to another tooth or to a fixed reference frame of the dental arch,

[0054] and combinations of these attributes;

[0055] - a plurality of attributes are associated with the tooth region and the historical tooth model;

[0056] - in step 2), the proximity is a shape proximity, assessed by the difference between the shape of the tooth represented on the tooth region and the shape of the historical tooth model, preferably by the difference in shape between the tooth region and a view of the historical tooth model having a shape closest to the tooth region;

[0057] - Finding the view of the historical tooth model whose shape is closest to the tooth region uses a meta-heuristic method, preferably an evolutionary method, preferably simulated annealing or a neural network.

[0058] In one embodiment, the reference model that is deformed to obtain the updated model is a model generated by the user, preferably using his mobile phone.

[0059] Specifically, at a moment before the update time, such as during an orthodontic treatment stage, the user may use the above-described modeling method to define an "earlier" update model from an "earlier" update image, and use the earlier update model as a reference model.

[0060] In order to generate this updated model or so-called reference model, the user can also scan the dental arch using a portable scanner, for example of the LIDAR type (preferably a mobile phone LIDAR) or by analyzing a series of images representing luminous markings. In particular, he can implement a method for generating a digital three-dimensional model of the dental arch, preferably using his mobile phone, comprising the following steps:

[0061] I) projecting at least one light beam onto a dental arch of a user so as to place at least one light marker on the dental arch;

[0062] II) while carrying out step I), moving the dental arch through the light beam and acquiring during said movement a series of "marker" images of said dental arch, each image showing a representation of a projected marker or "projection";

[0063] III) identifying said projections on each of the tagged images and then generating a digital three-dimensional model that best matches the set of projections.

[0064] In one embodiment, a generation method is implemented before the update time to generate a reference model, and the update image is used to update the reference model.

[0065] In one embodiment, a generation method is implemented at update time to generate a reference model, and the update image is used to correct any errors in the reference model.

[0066] Preferably, steps b) to e) are performed by a computer, preferably automatically, ie without operator intervention, or under the supervision of an operator, such as a technician or dental care professional.

[0067] Preferably, when the retainer is an archwire, the retainer is shipped to a dental care professional.

[0068] When the retainer is an archwire, the method preferably comprises:

[0069] - in the design step c), an orthodontic transfer tray for accommodating a retainer and an adhesive substance is designed based on the updated model, so that when the orthodontic transfer tray is positioned on the user's teeth, the retainer is positioned and fixed at a certain position on the user's teeth so that the user's teeth maintain their configuration at the time of the update; and

[0070] - In step d) of manufacturing the archwire, the orthodontic transfer tray is manufactured.

[0071] The adhesive may be made of a composite material. Preferably, the adhesive comprises a combination of silane-coated inorganic particles and a dimethacrylate resin (e.g., "ENAMEL plus HRI Flow HF" type resin). The adhesive may comprise a compound containing "BisGMA" and / or "Bis EMA", such as "Braceplaste".

[0072] In step e), the archwire may be delivered into an orthodontic transfer tray, preferably to a dental care professional or user, preferably to a dental professional.

[0073] The invention also relates to a computer program product comprising instructions which, when the program is executed by a computer, cause the computer to perform the following steps:

[0074] a′) receiving at least one two-dimensional image representing the dental arch of the user after the orthodontic treatment is completed, referred to as an “updated image”;

[0075] b) generating a digital three-dimensional model representing the dental arch of the user, called the "updated model", from the updated image, preferably by deforming the digital three-dimensional model of the dental arch of the user acquired before step a') (called the "reference model") from the updated image;

[0076] c) designing a retainer based on the updated model;

[0077] The updated image received in step a') is acquired by an image acquisition device.

[0078] Preferably, the computer program product comprises, after step c), a step d) comprising sending the manufacturing instructions, preferably by digital communication, to an orthodontic appliance manufacturer or directly to a manufacturing device, such as a 3D printer.

[0079] The present invention also relates to:

[0080] - a computer program product according to the invention,

[0081] - a data medium, such as a memory or a CD-ROM, on which such product is recorded, and

[0082] - Load such product on your computer.

[0083] The medium may be computer memory, a hard drive or a CD ROM.

[0084] The present invention also relates to a design kit for a retainer, the design kit comprising:

[0085] - an image acquisition device, preferably a mobile phone,

[0086] - a computer program product according to the invention, preferably recorded on a medium and loaded into a computer, optionally comprising a code for the step of sending manufacturing instructions;

[0087] The image acquisition device communicates with the computer so as to send updated images acquired by the device to the computer.

[0088] The computer program product may be a dedicated application installed on the image acquisition device, which is preferably a mobile phone.

[0089] The kit may also include a device for manufacturing an orthodontic appliance, such as a 3D printer, configured to receive retainer manufacturing instructions from a computer running the computer program product.

[0090] The kit may also include a transport vehicle designed to transport the retainer from a location where the retainer is manufactured to a delivery location for the retainer, which is preferably the user's home and / or the user's dental care professional's office.

[0091] definition

[0092] The term "user" refers to any person on whom the method according to the invention is carried out, whether that person is ill or not.

[0093] The term "dental care professional" refers to anyone qualified to provide dental care and includes, among other things, orthodontists and dentists.

[0094] "Orthodontic treatment" is all or part of treatment aimed at modifying the configuration of a patient's teeth. Treatment aimed at maintaining the configuration of the teeth (maintenance treatment) is not considered orthodontic treatment.

[0095] A 3D scanner or "scanner" is a device that produces a 3D model of the dental arch.

[0096] The "update time" is the time at which one or more update images are acquired. The duration of this time is short enough that the tooth configuration does not change significantly during this period.

[0097] The term "model" refers to a three-dimensional digital model. A model consists of a set of voxels. A "dental arch model" is a model representing at least a portion of a dental arch, preferably at least 2, preferably at least 3, most preferably at least 4 teeth. Figure 3 is an example of a dental arch model.

[0098] A "dental model" is a three-dimensional digital model of the teeth in the user's dental arch. The dental arch model can be cut to define a dental model of at least some, preferably all, of the teeth represented in the dental arch model. Thus, the dental model is a model within the dental arch model. Figure 4 An example of a dental arch model segmented into tooth models is shown.

[0099] A "recipe" is a series of dental arch models that represent a continuous dental arch configuration. Specifically, a "treatment regimen" or "treatment plan" includes models that represent the configuration of the dental arch at different moments during its treatment. These moments are usually the initial moments before the start of treatment, the middle moments during treatment, and the final moments at the end of treatment. Each regimen model that represents the desired configuration of the dental arch at its intermediate moments is called an "intermediate model". Figure 2 Examples of treatment regimens are shown.

[0100] Measuring the difference between two objects is called "matching" or "fitting." A "best fit" is the result of an optimization process that minimizes the difference.

[0101] Specifically, an updated image best matches the model when the view of the model provides an image that best matches the updated image.

[0102] A "teeth region" is a region of an image that at least partially represents teeth.

[0103] The term "historical dental model" refers to a dental model enriched with its specific embodiments.

[0104] "Closeness" is a measure of one or more differences between the historical tooth model and the tooth region. These differences may include differences in shape, but may also include other differences, such as translucency or color. The maximum closeness may be sought by minimizing several differences successively, or by minimizing a combination of these differences, such as a weighted sum of these differences.

[0105] Therefore, "closeness" is a broader concept than "matching", since matching only measures closeness in terms of shape.

[0106] The arch configurations at intermediate and final moments are theoretical configurations because they are the result of simulations for future moments. They are therefore anticipated or "predicted" and may therefore differ from the actual situation at intermediate or final moments. The model in the observation scheme simulates the effects of arch treatment in chronological order.

[0107] A dental arch configuration is considered "real" when it is the actual configuration of the user's dental arch. A dental arch configuration is considered "theoretical" when it is a "simulated," "predicted," or "expected" configuration of the user's dental arch at a future time.

[0108] One example of software for manipulating tooth models and creating treatment plans is the program Treat described in https: / / en.wikipedia.Org / wiki / Clear_aligners#cite_note-invisalignsystem-10. US5975893A also describes the creation of treatment plans.

[0109] An “image” is a two-dimensional picture, such as a photograph or a video frame. An image is made up of pixels.

[0110] The “acquisition conditions” specify the spatial position, spatial orientation and calibration of the following equipment, such as the values ​​of aperture and / or exposure time and / or focal length and / or sensitivity:

[0111] - a real image acquisition device relative to the user's dental arch (real acquisition conditions), or

[0112] - A virtual image acquisition device relative to the user's dental arch (virtual acquisition conditions).

[0113] The "calibration" of an acquisition device consists of all the calibration parameter values. Calibration parameters are parameters intrinsic to the acquisition device (other than its position and orientation) whose values ​​affect the acquired image. For example, the aperture value is a calibration parameter that modifies the depth of field. The exposure time is a calibration parameter that modifies the brightness (or "exposure") of the image. The focal length is a calibration parameter that modifies the angle of view (i.e. the degree of "zoom"). "Sensitivity" is a calibration parameter that modifies the reaction of the sensor of a digital acquisition device to incident light.

[0114] Preferably, the calibration parameters are selected from the group formed by aperture, exposure time, focus and sensitivity.

[0115] Observations of the model under specified virtual acquisition conditions, particularly when the virtual acquisition device is calibrated at specified angles and distances, are called "views".

[0116] The terms "image of a dental arch", "view of a dental arch", "representation of a dental arch", "scan of a dental arch" or "model of a dental arch" refer to an image, view, representation, scan or model of all or part of a dental arch.

[0117] The model of the user's dental arch is "matched" to the image when the view of the model corresponds to the image, i.e., such that the representations of the teeth in the view are positioned relative to each other similarly to the representation of the teeth on the image. Thus, the contours of the teeth model represented on the view are substantially superimposable and aligned with the contours of the representation of the teeth on the image.

[0118] This view of the model may also be described as "matching" or "superimposably alignable" with the image.

[0119] The term "optical mark" refers to the interaction between the light beam and the dental arch. An optical mark can be a point, a line, a bar or a group of points and / or a group of lines and / or a group of bars. Thus, a line can be continuous or partially interrupted and consist of multiple line segments. It can have a fixed or variable width. The representation of the optical mark, in particular in the updating image, varies depending on the direction from which the mark is viewed.

[0120] The term "light" includes all electromagnetic waves from infrared to ultraviolet.

[0121] "Meta-heuristic" methods are well-known optimization methods. They are preferably selected from the group formed by the following methods:

[0122] - an evolutionary algorithm, preferably selected from the following:

[0123] Evolutionary strategies, genetic algorithms, differential evolution algorithms, distribution estimation algorithms, artificial immune systems, shuffled complex evolutionary algorithms, path reconnection, simulated annealing, ant colony algorithms, particle swarm optimization algorithms, taboo search, and GRASP methods;

[0124] - Kangaroo algorithm, Fletcher-Powell method, noise method, random tunneling, random restart hill climbing, cross entropy method, and

[0125] - Hybrid approaches between the above mentioned metaheuristics.

[0126] Deep learning algorithms are well known to those skilled in the art. They include "neural networks" or "artificial neural networks".

[0127] Those skilled in the art know how to select a neural network according to the task at hand. In particular, the neural network can be selected from the following:

[0128] - Networks specialized for image classification, called "CNNs" ("Convolutional Neural Networks"), such as AlexNet (2012), ZF Net (2013), VGG Net

[0129] (2014), GoogleNet (2015), Microsoft ResNet (2015), Caffe:BAIRReference Caff eNet, BAIR AlexNet, Torch VGG CNN S, VGG CNN M, VGG_CNN_M_2048,

[0130] VGG_CNN_M_1024, VGG CNN M 128, VGG CNN F, VGG

[0131] ILSVRC-2014 16-layer, VGG ILSVRC-2014 19-layer, Network-in-

[0132] Network (Imagenet and CIFAR-10), Google: Inception (V3, V4);

[0133] - Networks specifically designed to localize and detect objects in images, i.e. object detection networks, such as R-CNN (2013), SSD (Single Shot Multibox Detector: Object Detection Networks), Faster R-CNN (Faster Region-based Convolutional Network Approach: Object Detection Networks), Faster R-CNN (2015), SSD (2015), RCF (Richer Convolutional Features for Edge Detection) (2017), SPP-Net (2014),

[0134] OverFeat (Sermanet et al.) (2013), GoogleNet (Szegedy et al.)

[0135] (2015), VGGNet (Simonyan and Zisserman) (2014), R-

[0136] CNN (Girshick et al.) (2014), Fast R-CNN (Girshick et al.)

[0137] (2015), ResNet (He et al.) (2016), Faster R-CNN (Ren et al.) (2016), FPN (Lin et al.) (2016), YOLO

[0138] (Redmon et al.) (2016), SSD (Liu et al.) (2016),

[0139] ResNet v2 (He et al.) (2016), R-FCN (Dai et al.) (2016), ResNeXt (Lin et al.) (2017), DenseNet (Huang et al.)

[0140] (2017), DPN (Chen et al.) (2017), YOL09000

[0141] (Redmon and Farhadi) (2017), Hourglass (Newell et al.)

[0142] (2016), MobileNet (Howard et al.) (2017), DCN (Dai et al.) (2017), RetinaNet (Lin et al.) (2017), Mask R-

[0143] CNN (He et al.) (2017), RefineDet (Zhang et al.) (2018), Cascade RCNN (Cai et al.) (2018), NASNet (Zoph et al.) (2019), ComerNet (Law and Deng) (2018), FSAF

[0144] (Zhu et al.) (2019), SENet (Hu et al.) (2018),

[0145] ExtremeNet (Zhou et al.) (2019), NAS-FPN (Ghiasi et al.)

[0146] (2019), Detnas (Chen et al.) (2019), FCOS (Tian et al.)

[0147] (2019), CenterNet (Duan et al.) (2019), EfficientNet

[0148] (Tan and Le) (2019), AlexNet (Krizhevsky et al.) (2012);

[0149] -Networks specialized for image generation, such as Cycle-Consistent Adversarial Networks

[0150] Consistent Adversarial Network, 2017), Enhanced CycleGAN

[0151] (Augmented CycleGAN, 2018), Deep Photo Style Transfer (Deep

[0152] Photo Style Transfer (2017), FastPhotoStyle (2018),

[0153] pix2pix (2017), Style-Based GAN Generator Architecture

[0154] Generator Architecture for GANs, 2018), SRGAN (2018).

[0155] The above list is not exhaustive.

[0156] Training a neural network consists in improving it using a training library containing information about the two types of objects that the neural network must learn to "match" (ie connect to each other).

[0157] The training may be based on a learning library consisting of records, each record containing a first object of a first type and a corresponding second object of a second type.

[0158] Alternatively, a learning library consisting of records may be used for training, each record containing a first object of a first type or a second object of a second type, but each record containing information related to the type of object it contains. Such training techniques are described, for example, in the article "Unpaired image-to-image translation using cycle-consistent adversarial networks." by Zhu, Jun-Yan et al., 2017 IEEE International Conference on Computer Vision (ICCV).

[0159] Training a neural network with these records teaches it to provide, from any object of the first type, a corresponding object of the second type.

[0160] The quality of the analysis performed by the neural network depends directly on the number of records in the training database. Preferably, the learning database contains more than 10,000 records.

[0161] "Computer" refers to a computer processing unit, which comprises a group of several machines with computer processing capabilities. In particular, the unit may be integrated into a mobile phone, or may be a PC-type computer or a server, for example a server located remotely from the user, such as a computer in the "cloud" or located at the dental care professional's premises. The image acquisition device and the computer then comprise communication means for exchanging information between them, in particular for transmitting updated images.

[0162] Typically, a computer includes a processor, a memory, a human-computer interface (usually including a screen), and a computer that can communicate with the Internet, WIFI, The computer may be connected to a communication module of a mobile phone or telephone network. The software configured to implement the method of the present invention is loaded into the memory of the computer. The computer may also be connected to a printer, such as a 3D printer.

[0163] Unless otherwise indicated, "comprising," "including," and "having" should be interpreted broadly and not limitingly. BRIEF DESCRIPTION OF THE DRAWINGS

[0164] Other features and advantages of the present invention will become apparent from the following detailed description and from an examination of the accompanying drawings, which are provided for illustrative and non-limiting purposes. In the drawings:

[0165] [ Figure 1 ] schematically illustrates a manufacturing method according to the present invention;

[0166] [ Figure 2 ] shows the treatment regimen;

[0167] [ Figure 3 ] shows an example of a dental arch model;

[0168] [ Figure 4 ] shows an example of a dental arch model segmented into tooth models;

[0169] [ Figure 5 ] shows a kit according to the invention;

[0170] [ Figure 6 ] shows a deformation of the reference model according to the present invention. DETAILED DESCRIPTION

[0171] Figure 1 An example of the manufacturing method according to the present invention is shown. The manufacturing method comprises steps a) to d), preferably steps a) to e).

[0172] In step a), an updated image 22, preferably a photograph, representing the dental arch of the user is acquired using the image acquisition device 20 in order to update the reference model 32 representing said dental arch of the user.

[0173] Alternatively, the update image is acquired by video acquisition, wherein a video of the user's dental arch is captured using an image acquisition device. The update image can be extracted from the video.

[0174] Step a) is performed at an update time, which is after orthodontic treatment. Preferably, the collection is performed within six months, preferably within three months, more preferably within one month, and even more preferably, immediately after completion of orthodontic treatment.

[0175] The collection of step a) can be performed by the user, a relative or a dental care professional. Preferably, the collection is performed by the user himself.

[0176] The image acquisition device 20 may be a mobile phone, a tablet computer, a camera or a computer. The image acquisition device is preferably a mobile phone.

[0177] Preferably, a plurality of update images 22 are acquired. The update images may include a front view image, a left view image, and a right view image. The update images may be taken for the upper dental arch, the lower dental arch, or both. The update images may be taken with the user's mouth open or closed. Preferably, the update images are in color, preferably true color.

[0178] In step b), an updated model 42 is generated based on the updated image acquired in step a), the generation of which does not require the user to visit the dental clinic, preferably without the user visiting the dental clinic.

[0179] There are several methods to choose from.

[0180] In one embodiment, the update model is generated directly from the update image, preferably using at least one neural network.

[0181] The at least one neural network is pre-trained such that the neural network learns to output an updated model from one or more updated input images.

[0182] Training is performed using a training library including historical images and historical models, the historical images being equivalent to the update images, and the historical models being equivalent to the update models. The historical models can be acquired using a 3D scanner substantially when the corresponding historical images are acquired, or can be generated manually. The records in the learning database can include historical images as first objects and historical models as second objects. The neural network can be told that for a first object including a historical image, it must match a second object including a historical model. Specifically, in order to train the neural network, it will be provided with a set of first objects and a set of second objects so that it can learn to match the first objects and the second objects. After training, the neural network can determine an update model as output when an update image is obtained as input.

[0183] Preferably, the first object comprises a plurality of update images, and the at least one neural network is trained to generate an update model as output from the plurality of update images provided thereto as input.

[0184] The at least one neural network may be a reconstruction neural network, ie an occupation neural network.

[0185] The occupancy neural network is trained to output a 3D model of at least a portion of a dental arch, the 3D model being constructed from a plurality of updated images representing at least a portion of the dental arch. Preferably, the occupancy neural network takes as input only updated images representing at least a portion of a user's dental arch.

[0186] In another embodiment, a reference model 32 is determined. The reference model may be selected manually, for example from a set of 3D models representing the user's dental arch stored in the database 30. Preferably, the reference model is determined automatically, i.e. without human intervention, preferably by a computer.

[0187] The selected reference model 32 is preferably an intermediate model or a final model created when determining an orthodontic treatment plan. Figure 2 The orthodontic treatment plan is shown in FIG. The reference model 32 preferably corresponds to a 3D model acquired using a 3D scanner before the start of orthodontic treatment, also called an "initial model", which has been modified to represent the theoretical configuration of the teeth at an intermediate moment of orthodontic treatment (also called an "intermediate model") or at the final moment of orthodontic treatment (also called a "final model"). Preferably, the reference model corresponds to the final model.

[0188] The reference model may be selected from a database. Alternatively, the reference model may be sent, for example by a dental care professional or technician, via digital communication, preferably via a dedicated application.

[0189] The reference model 32 is then deformed to obtain an updated model 42 representing the configuration of the user's teeth at the time of the update.

[0190] Figure 6 The process of deforming the reference model using a transformation algorithm is shown. The transformation algorithm receives the reference model 32 and the update image 22 as input, and then deforms the reference model according to the update image to determine an update model 42 that matches the update image.

[0191] The conversion algorithm may be a meta-heuristic method, preferably comprising the following steps:

[0192] - searching for the position, orientation and calibration of the virtual acquisition device, collectively referred to as "virtual acquisition conditions", which best matches the actual acquisition conditions of the updated image;

[0193] - comparing the view of the reference model under virtual acquisition conditions and the updated image;

[0194] - deforming the reference model so that the positioning of the teeth observed on said view under virtual acquisition conditions corresponds to the positioning of the teeth observed on the acquired image.

[0195] Therefore, deformation is performed at the time when the acquisition image is acquired to update the reference model.

[0196] Preferably, the reference model is pre-cut into tooth models, such as Figure 4As shown, the deformation comprises movement of the tooth model. In one embodiment, the deformation involves moving the tooth model whose shape remains fixed.

[0197] Specifically, the search for the virtual acquisition conditions and the comparison may be performed according to the teaching of PCT / EP2015 / 074896.

[0198] Alternatively, the conversion algorithm comprises at least one neural network.

[0199] The at least one neural network is pre-trained such that the neural network learns to output an update model based on an update image provided as input and a reference model.

[0200] Training is performed using a training library containing historical images, a first historical model, and a second historical model, the historical image being similar to the update image, the first historical model being similar to the reference model, and the second historical model being similar to the update model. Preferably, each second historical model is determined from the historical image and the corresponding first historical model, for example manually, or substantially acquired using a 3D scanner when the corresponding historical image is acquired. The records in the learning database may include the historical image and the first historical model as the first object and the second historical model as the second object. The neural network may be informed that for the first object including the historical image and the first historical model, it must match the second object including the second historical model. Specifically, in order to train the neural network, it will be provided with a set of first objects and a set of second objects so that it can learn to match the first object and the second object. After training, the neural network can determine an update model as output when it is given an update image as input and a reference model.

[0201] Specifically, if several update images are available, a PPO-type augmentation network can be used, as described on the following website: https: / / openai.com / blog / openai-baselines-ppo / .

[0202] The PPO neural network may be configured to iteratively modify the positions of the teeth of the 3D model representing at least a portion of the user's dental arch based on at least two updated images provided as input to the neural network, so that the 3D model best matches the image provided as input after modification, the at least two updated images being images representing at least a portion of the user's dental arch acquired at different acquisition angles. Providing multiple updated images acquired at different acquisition angles enables the three-dimensional movement of the teeth to be calculated using the triangulation principle.

[0203] Advantageously, such a neural network does not require any data other than the updated image as input to modify the 3D model available to the neural network.

[0204] In order to train the neural network, it needs to be told the final positions of the teeth that are expected as output. The neural network can then learn how to correct the result of its output (the "output" 3D model) by comparing the positions of the teeth in this "output" model with the expected final positions of the teeth. The deviation between the expected tooth positions and the positions of the teeth in the output model provided by the neural network can be calculated, for example, in millimeters or degrees, or can be a surface tolerance or Euclidean LL distance.

[0205] The accuracy of the occupancy neural network may be lower than that of the PPO neural network, especially when a specific portion of at least a portion of the dental arch is not represented on the plurality of updated images provided as input to the neural network. Advantageously, a previously acquired 3D model of the dental arch of the user may be used to reconstruct the missing portion. The previously acquired 3D model of the dental arch of the user may also provide dimensional information, e.g. in millimeters, which cannot be derived from the plurality of updated images provided as input alone.

[0206] In another embodiment of step b), the updated model is generated based on the updated image, preferably from a set of several updated images and from a library of dental models under a modeling method comprising for each updated image the following steps:

[0207] 1) analyzing the updated image to determine at least one tooth region and at least one tooth attribute value associated with the tooth region, preferably the type or number of the tooth to which the tooth region belongs;

[0208] 2) for each tooth region determined in step 1), searching in a historical library containing more than 1000 tooth models called “historical tooth models” associated with the tooth attribute value to find a historical tooth model called “optimal tooth model” having the greatest proximity to the tooth region;

[0209] 3) Rank the set of best tooth models to create a model that best matches the updated image.

[0210] The tooth model library is preferably created before step a) of the manufacturing method according to the present invention. Preferably, the tooth model library contains more than 2,000, preferably more than 5,000, most preferably more than 10,000 historical tooth models. The greater the number of historical tooth models, the more accurate the combined model.

[0211] The tooth attributes are used to filter the historical tooth model to limit the search to teeth having the same tooth attribute values ​​as the teeth represented by the tooth region in the update image.

[0212] One or more tooth attributes may be associated with the tooth model, the one or more tooth attributes being specifically selected from the group consisting of tooth number; tooth type; tooth shape parameters, such as tooth width, specifically mesio-palatal width, thickness, crown height, incisal edge mesial and distal deflection index or degree of wear; tooth appearance parameters, specifically translucency index or color parameters; parameters related to tooth status, such as "worn", "fractured", "decayed" or "fitted" (i.e. in contact with orthodontic appliances); user's age; or a combination of these attributes.

[0213] Preferably, the modeling method comprises the following optional steps:

[0214] 4) optionally, replacing at least one optimal tooth model with another tooth model (historical tooth model) and repeating in step 3) in order to maximize the match between the assembled model and the analyzed image;

[0215] 5) Optionally, repeat step 1) with another update image and in step 3) and / or step 4) search for a maximum match with the set of update images used.

[0216] If several update images are processed, the accuracy of the combined model can be improved.

[0217] In step 1), the update image is analyzed, preferably by submitting it to a neural network trained to be able to determine at least one tooth region and at least one tooth property value associated with said tooth region.

[0218] At the end of step 1), an updated image is obtained, which is enriched with a description providing for each tooth region a tooth attribute value (eg tooth number) for at least one tooth attribute.

[0219] In step 2), for each tooth region determined in the previous step, the history library is searched to find the history tooth model with the greatest proximity to the tooth region. This tooth model is called the "optimal tooth model".

[0220] Preferably, historical tooth models are sought whose at least one tooth attribute has the same value as the tooth region. In particular, the tooth attribute may relate to a tooth type or a tooth number. In other words, the historical tooth models are filtered to examine in more detail only those tooth models that relate to the same tooth type as the tooth represented on the tooth region.

[0221] Alternatively or preferably, in addition to this comparison of property values, the tooth shape represented on the tooth region may be compared with the shape of the historical tooth model to be evaluated, preferably using a metaheuristic method, preferably an evolutionary method, preferably a simulated annealing method.

[0222] To this end, the historical tooth model to be evaluated is viewed from different angles. Each view thus obtained is compared with an updated image, preferably with a tooth region, in order to determine the "distance" between the view and the updated image or preferably with the tooth region. The distance thus measures the difference between the view and the tooth region.

[0223] For each historical dental model tested, a view is determined which has a minimum distance to the updated image or the tooth region. Each historical dental model examined is thus associated with a specific minimum distance which measures its formal proximity to the tooth region.

[0224] The best historical tooth model is the model that is considered to be closest to the tooth area based on the comparisons made.

[0225] The minimum distances obtained for the different tooth models tested are then compared and the best tooth model is defined using the tooth model with the shortest minimum distance. This best tooth model is therefore the closest match to the updated image.

[0226] The search for the maximum match is preferably performed using a metaheuristic method, preferably an evolutionary method, preferably a simulated annealing method.

[0227] In a preferred embodiment, a first evaluation of the historical tooth model is performed sequentially by comparing the value of at least one tooth attribute (e.g. tooth number) with the corresponding value of the tooth analysis area, followed by a second evaluation by shape comparison. The first fast evaluation advantageously enables the historical tooth model to be screened out, so that only the historical tooth model retained by the first evaluation can be submitted to the second slower evaluation.

[0228] Preferably, several first evaluations are performed before the second evaluation.

[0229] At the end of step 2), an optimal tooth model is associated with each tooth region.

[0230] In step 3), a combined model is created by arranging the optimal tooth models.

[0231] A first rough arrangement may be determined by considering the tooth attribute values ​​of the best tooth model. For example, if the tooth numbers of the best tooth model are those of canines and incisors, these tooth models may be arranged in a dental arch that generally corresponds to the dental arch area bearing these types of teeth.

[0232] The shape of the dental arch can be refined based on the values ​​of other tooth attributes.

[0233] The order of the optimal tooth models is the order of the corresponding tooth regions.

[0234] Furthermore, the maximum proximity or "minimum distance" associated with the optimal tooth model results from the observation of the tooth model along the observation point. In other words, the teeth modeled by this model may also be observed in the updated image. Therefore, the optimal tooth models are preferably oriented so that it is possible to observe their surfaces, which correspond to their corresponding tooth areas on the updated image, from the same observation point.

[0235] In this way, an initial arrangement of the optimal tooth model can be defined.

[0236] Preferably, the first arrangement of the optimal tooth model is then iteratively modified to provide a maximum match with the updated image.

[0237] In another embodiment of step b), the updated model (or the reference model intended to be deformed to generate the updated model) is generated by implementing a method for generating a digital three-dimensional model of a dental arch, the method comprising the following steps:

[0238] I) projecting at least one light beam onto a dental arch of a user so as to place at least one light marker on the dental arch;

[0239] II) while performing step I), moving the dental arch through the light beam and acquiring during said movement a series of "marker" images of said dental arch, each image showing a representation of a projection mark or "updated projection";

[0240] III) identifying said projections on each of the tagged images and then generating a digital three-dimensional model that best matches the set of said updated projections.

[0241] Preferably, a structured light beam is projected in step I).

[0242] In step III), the updated model may be searched using optimization methods and / or one or more neural networks.

[0243] In step III), in order to generate the updated model, the model defined according to the user characteristics or the model from the database, such as the initial model, the intermediate model or the final model, may be modified.

[0244] In step III), the updated model may be determined as the test model obtained at the end of the following cycle of steps i) to iii):

[0245] i) Create the model to be tested, then

[0246] ii) determine the distance representing the difference between the set of updated projections and the model to be tested, then,

[0247] iii) If the representative distance exceeds a predetermined acceptability threshold, modify the model to be tested and return to step i).

[0248] The representative distance can be evaluated from the basic distances, and for the corresponding updated projection, each basic distance is determined by evaluating the difference between the updated projection and the best reference projection, the reference projection being a representation of a virtual light mark on a reference image representing a view of the model to be tested, the virtual light mark being generated by projecting onto the model to be tested a virtual light beam having the same shape as the light beam projected onto the dental arch in step I), the best reference image being a reference image showing a reference projection having the smallest distance from the updated projection.

[0249] The model to be tested may be segmented to define a tooth model, and the modification of the model to be tested comprises moving the tooth model and / or deforming the tooth model.

[0250] The first test model to be modified may be a dental arch model, preferably selected according to user characteristics, or a model from a database, such as an initial model, an intermediate model or a final model.

[0251] In step III), at least one of the following methods is used, preferably two methods are used: an optimization method, an artificial intelligence method, a dimensional evaluation method involving stereoscopic vision, a dimensional evaluation method involving analysis of the shape of the light marking and a dimensional evaluation method involving distance analysis.

[0252] In step c), a retainer 44 is designed based on the updated model 42. The design may be done manually, for example by a technician or dental care professional.

[0253] Alternatively, the design is performed by software, preferably automatically, ie without human intervention.The retainer is designed to keep the user's teeth in the configuration they were in at renewal time.

[0254] To ensure this, the shape of the orthodontic appliance is typically conformed to the shape of the patient's dental arch at the time of update so that the orthodontic appliance does not exert any action tending to move the teeth at the time of update.

[0255] Preferably, the retainer is an archwire designed to be fixed (eg bonded) to the user's teeth. Alternatively, the retainer is an orthodontic retainer tray.

[0256] The design of the retainer is to determine the shape of the appliance so that the teeth maintain their configuration at the time of renewal.

[0257] In step d), the holder 52 is produced.

[0258] In one embodiment, the fabrication is accomplished by stamping on a component that physically represents the updated model. In this case, steps c) and d) are combined. For example, a metal wire may be stamped on a plaster or polymer cast that physically represents the updated model to form an archwire. A polymer tray may also be molded onto such a component.

[0259] The retainer can also be directly manufactured based on the updated model, for example, a 3D model of the retainer is first designed and then manufactured, for example, by printing it through a 3D printer.

[0260] Preferably, when the retainer is an archwire, a transfer tray is also manufactured, which enables the retainer to be easily positioned on the user's teeth. First, the archwire is positioned in the orthodontic transfer tray using an adhesive substance. In a second step, the orthodontic transfer tray is placed in a use position on the user's teeth. This facilitates the transfer and positioning and fixing of the archwire. Specifically, the dental arch can be positioned and fixed in a single step. And, since the orthodontic transfer tray is based on a three-dimensional model of the user's dental arch, the archwire can be accurately positioned. The article "Méthode simple et rapide de contention indirecte" by Christine Muller et al., Orthod Fr 2009; 80: 233-238, EDPSciences, SFODF, 2009 DOI: 10.1051 / orthodfr / 2009006 describes an example of a method for manufacturing an orthodontic transfer tray.

[0261] In optional step e), the manufactured retainer is preferably shipped to a dental care professional or user. Instructions on how to wear and / or position the retainer may be provided with the retainer. Advantageously, the user does not have to see a dental care professional to obtain a 3D model using a 3D scanner, or to receive the retainer and / or position the retainer in their mouth.

[0262] Preferably, step b) and step c) are implemented by a computer, preferably without human intervention. Step d) and / or step e) may also be performed without human intervention. After step c), the computer implementing step c) may send design instructions for manufacturing the retainer to the manufacturing device 50.

[0263] Figure 5 A design kit 1 according to the invention is schematically shown.

[0264] The design kit 1 comprises an image acquisition device 20 (preferably a mobile phone) and a computing system 40 (e.g. a computer), the computing system comprising a computer program product, the computer program product being configured to implement step b) of the manufacturing method according to the present invention, preferably to implement steps b) to c) of the manufacturing method according to the present invention. The computer program product can also be configured to send the design instruction 64 directly to the manufacturing device 50.

[0265] The image acquisition device 20 can communicate with the computing system to transmit the acquired updated image 60 to the computing system 40 via digital communication. In a specific embodiment, the acquisition device and the computing system are integrated into a mobile phone.

[0266] The reference model may be obtained from a database 30 accessible by digital communication with the computing system. The reference model may be transmitted 62 to the computing system, in particular by being downloaded. The database may be located on a remote server.

[0267] The design kit 1 may include a manufacturing device, which is preferably configured to communicate with the computing system. Once the holder is designed, the computing system 40 may send manufacturing instructions 64 to the manufacturing device 50.

[0268] Example

[0269] The update image 22 is preferably captured by a user using a mobile phone 20, preferably from a location remote from the dental or orthodontic office.

[0270] The update image 22 is sent to a computing system 40. The computing system may or may not be integrated into the mobile phone. Preferably, the computing system is a dedicated application installable on the mobile phone.

[0271] The computing system 40 may preferably communicate with a database 30 containing intermediate models, initial models, and final models representing the configuration of the user's dental arch from one or more orthodontic treatment plans. The intermediate and final models represent the theoretical configuration of the user's dental arch. The initial model is a model typically acquired by a dental care professional, preferably using a 3D scanner. In step b), a reference model is determined by selecting a model from the database 30.

[0272] Alternatively, in step b), the reference model 32 is manually selected and sent to the computing system 40, for example by a dental care professional or technician.

[0273] Based on the update image 22 and the reference model 32 , the computing system 40 may determine an update model 42 , as described in step c) of the method according to the invention.

[0274] The computing system 40 may then design a retainer 52 using the updated model 42, optionally with the assistance of an operator or dental care professional, as described in step d) of the method according to the invention. The locations where the updated model is determined (step c)) and the retainer is designed (step d)) may be remote from each other. The computing system may include multiple processors.

[0275] Preferably, the mobile phone comprises a dedicated application, which is configured to implement steps a) to c) according to the present invention, preferably to implement steps a) to d) according to the present invention.

[0276] Step d) of retainer design enables the determination of manufacturing instructions for said retainer. The retainer 52 may be an archwire or an orthodontic retainer tray. Advantageously, these instructions are sent to the manufacturing device 50.

[0277] The manufacturing device produces the retainer according to these instructions. The manufacturing device can be a 3D printer. The 3D printer can be installed at the user's location.

[0278] Once the retainer is manufactured, it can be shipped directly to the user. Preferably, when the retainer is an archwire, it is shipped in a transfer tray that includes, in addition to the archwire, an adhesive for securing the archwire to the user's teeth when the transfer tray is positioned on the user's teeth.

[0279] As has now become clear, the method according to the invention enables the manufacture of a retainer based on one or more updated images, preferably photographs taken by the user, and a reference model acquired when the teeth have not yet reached their final configuration. Thus, the manufacturing method can be carried out without the intervention of an orthodontist specifically for this purpose, and therefore the user does not need to make an appointment with an orthodontist.

[0280] Thus, by capturing a simple extraoral image of the user's teeth, i.e., an image captured using an image capture device that is located outside the user's mouth at the time of capture, retainers can be designed and manufactured without the user having to consult with their dental care professional face to face. In particular, the method according to the present invention enables the manufacture of an archwire intended to be positioned on the inner surface of the user's teeth by capturing an extraoral image that specifically represents the outer surface of the user's teeth.

[0281] Of course, the present invention is not limited to the embodiments described above and exemplified.

Claims

1. A method for modeling a retainer for a patient, the method The following steps are involved: Prior to step a), during orthodontic treatment, generating a reference model representing the dental arch of the user; a) at an update time, at the end of orthodontic treatment or after the end of orthodontic treatment, using an image acquisition device to acquire at least one two-dimensional image of at least a portion of the user's dental arch, referred to as an "update image"; b) generating a digital three-dimensional model representing the dental arch of the user based on the updated image, referred to as the "updated model", the updated model being obtained by deforming the reference model from the updated image; c) designing, based on said updated model and preferably by computer, a retainer suitable for keeping the teeth in their configuration at said update time; d) Manufacturing the retainer.

2. The method according to claim 1, wherein said acquisition of said update image is performed by said user, preferably outside an orthodontic or dental clinic.

3. A method according to any one of the preceding claims, wherein the retainer is an archwire or an orthodontic retainer tray, preferably an archwire.

4. A method according to any one of the preceding claims, wherein the reference model represents the user's dental arch in an expected theoretical configuration at an intermediate moment of the orthodontic treatment or in an expected theoretical configuration at a final moment of the orthodontic treatment, respectively referred to as an "intermediate model" or a "final model".

5. A method according to any one of the preceding claims, wherein the reference model is a final model.

6. The method according to any one of the preceding claims, in, In step b), the reference model Obtained using LIDAR integrated into the user's mobile phone or by implementing a dental arch modeling method, preferably obtained via the user's mobile phone, the modeling method comprising the following steps: 1) analyzing the updated image representing the dental arch of the user to determine at least one tooth region and at least one tooth attribute value associated with the tooth region, preferably the type or number of the tooth to which the tooth region belongs; 2) for each tooth region determined in step 1), searching in a historical library containing more than 1000 tooth models referred to as "historical tooth models" associated with the tooth attribute values ​​to find the historical tooth model referred to as the "optimal tooth model" having the greatest proximity to the tooth region; 3) arranging the set of optimal tooth models to create a model that best matches the image analyzed in step 1), wherein for obtaining the updated model, the image is the updated image, or for obtaining the reference model, the image is an image acquired before the update time.

7. The method according to any one of the preceding claims, in, In step b), the reference model is obtained by implementing a method for generating a digital three-dimensional model of a dental arch, the method comprising the following steps: I) projecting at least one light beam onto a dental arch of the user so as to place at least one optical marker on the dental arch; II) while carrying out step I), moving the dental arch through the light beam and acquiring during said movement a series of "marker" images of the dental arch, each image showing a representation of the projected marker or "projection"; III) identifying said projections on each tagged image and then generating a digital three-dimensional model that best matches said set of projections.

8. The method according to any one of the preceding claims, comprising, after step d), a step e): e) Shipping the retainer, preferably to the user or dental care professional.

9. The method according to any of the preceding claims, wherein the acquisition device is a mobile phone and / or the holder is manufactured using a 3D printer.

Images