PROCESS FOR CONTROLLING ORTHODONTIC RECIDIVISM

Abstract

The invention relates to a method comprising the following steps: - production, with a 3D scanner, of a digital three-dimensional reference model of at least part of the patient's arches, or "initial reference model"; - acquisition, by the patient or a relative of the patient, with a mobile phone, of at least one two-dimensional image of said at least part of the patient's arches, called "updated image", under real acquisition conditions; - search, by modification of the initial reference model from the updated image, for a final reference model corresponding to the positioning of the teeth during the acquisition of the updated image. figure 1

Description

Title of the invention: METHOD FOR CONTROLLING ORTHODONTIC RECIDIVISM Technical field

[0001] The present invention relates to a method for controlling the positioning of a patient's teeth and a computer program for implementing this method. State of the art

[0002] It is advisable for everyone to have their teeth checked regularly, in particular to check that the position of their teeth is not changing unfavorably. During orthodontic treatment, this unfavorable change may in particular lead to a change in treatment. After orthodontic treatment, this unfavorable change, called "relapse", may lead to a restart of treatment. Finally, more generally and independently of any treatment, everyone may wish to monitor any movements of their teeth.

[0003] Traditionally, check-ups are carried out by an orthodontist or a dentist, who alone have the appropriate equipment. These check-ups are therefore expensive. In addition, the visits are restrictive.

[0004] An objective of the present invention is to respond, at least partially, to the above-mentioned problems. Summary of the invention

[0005] The invention provides a method for monitoring the positioning of a patient's teeth, said method comprising the following steps: (a) creation of a digital three-dimensional reference model of the patient's arches, or "initial reference model" and, for each tooth, definition, from the initial reference model, of a digital three-dimensional reference model of said tooth, or "tooth model"; b) acquisition of at least one two-dimensional image of said arches, called “updated image”, under real acquisition conditions; (c) analysis of each updated image and creation, for each updated image, of an updated map relating to discriminating information; d) optionally, determination, for each updated image, of rough virtual acquisition conditions approximating said real acquisition conditions; e) search, for each updated image, for a final reference model corresponding to the positioning of the teeth during the acquisition of the updated image, the search preferably being carried out by means of a metaheuristic method, preferably by simulated annealing, and f) for each tooth model, comparison of the positions of said tooth model in the initial reference model and in the reference model obtained at the end of the previous steps, called the “final reference model”, in order to determine the movement of the teeth between steps a) and b).

[0006] As will be seen in more detail in the remainder of the description, a method according to the invention makes it possible, from a simple image of the teeth, taken without precise prepositioning of the teeth relative to the image acquisition device, for example a photograph taken by the patient, to accurately evaluate the movement of the teeth since the creation of the initial reference model. This evaluation can also be carried out by a simple computer or a mobile telephone.

[0007] Preferably, step e) comprises - a first optimization operation making it possible to search for virtual acquisition conditions which best correspond to the real acquisition conditions in a reference model to be tested determined from the initial reference model, and - a second optimization operation making it possible to search, by testing a plurality of said reference models to be tested, for the reference model best corresponding to the positioning of the patient's teeth during the acquisition of the image updated in step b).

[0008] Preferably, a first optimization operation is performed for each test of a reference model to be tested during the second optimization operation.

[0009] Preferably, the first optimization operation and / or the second optimization operation, preferably the first optimization operation and the second optimization operation implement a metaheuristic method, preferably simulated annealing.

[0010] Preferably, step e) comprises the following steps: el) definition of the reference model to be tested as being the initial reference model then, e2) following the following steps, testing virtual acquisition conditions with the reference model to be tested in order to finely approximate said real acquisition conditions; e21) determination of virtual acquisition conditions to be tested; e22) production of a two-dimensional reference image of the reference model to be tested under said virtual acquisition conditions to be tested; e23) processing the reference image to produce at least one reference map representing said discriminating information; e24) comparing the updated and reference maps so as to determine a value for a first evaluation function, said value for the first evaluation function depending on the differences between said updated and reference maps and corresponding to a decision to continue or stop the search for virtual acquisition conditions approximating said real acquisition conditions with more accuracy than said virtual acquisition conditions to be tested determined at the last occurrence of step e21); e25) if said value for the first evaluation function corresponds to a decision to continue said search, modification of the virtual acquisition conditions to be tested, then resumed at step e22); e3) determining a value for a second evaluation function, said value for the second evaluation function depending on the differences between the updated and reference maps under virtual acquisition conditions best approximating said real acquisition conditions and resulting from the last occurrence of step e2), said value for the second evaluation function corresponding to a decision to continue or stop the search for a reference model approximating the positioning of the teeth during the acquisition of the updated image with more accuracy than said reference model to be tested used at the last occurrence of step e2), and if said value for the second evaluation function corresponds to a decision to continue said search, modifying the reference model to be tested by moving one or more tooth models and then continuing at step e2).

[0011] A method according to the invention may also include one or more of the following optional characteristics: - in step a), an occlusion plane is determined according to the following operations: I. determination of the points of the initial reference model which belong to one arch and which are at a distance from the other arch which is less than a predetermined distance, preferably at a distance less than 3 mm from the other arch, called “contact points”; II. optionally, filtering a portion of the contact points, preferably so that the number of contact points belonging to the upper arch is identical to the number of contact points belonging to the lower arch, preferably by eliminating the points of one arch furthest from the other arch; III. linear regression, preferably by the least squares method, on all the remaining contact points so as to determine the occlusion plane; - in step a), the following operations are carried out: i. projection, in an occlusal plane, of the contact points between the teeth of the patient's upper and lower arches, the contact points and / or the occlusal plane preferably being determined, following steps I to III; ii. determination of the barycenter of the projections of said contact points and creation of a reference frame, in the occlusion plane, centered on said barycenter; iii. determination, in said reference frame, of the parabolic function, presenting the greatest coefficient of correlation with the set of projections of the contact points; iv. rotation of the set of projections of the contact points around the barycenter, and resumption of the previous operation iii until the set of projections of the contact points has covered a determined sector, preferably greater than 90°, greater than 180°, or even approximately 360°; v. identification of the highest correlation coefficient for the set of angular positions of the set of projections of the contact points around the barycenter, and the axis of the corresponding parabolic function; vi. determination of a median longitudinal plane of the initial reference model, said plane passing through said axis and being perpendicular to the occlusion plane; - in step a), a tooth model is at least partially delimited by following the following operations: i'. determination, at least partial, of the inner and outer gingival edges of the arch of the tooth concerned, preferably by analysis of the variations in the orientation of the surface of the initial reference model; ii'. projection, in the occlusal plane, of the inner and outer gingival edges; iii'. identification of the deformations of the projections of the inner and outer gingival edges corresponding to the interdental regions, the peaks of these deformations being called "approximation point" (in an interdental region, the two projections each have a point, the two points pointing substantially towards each other, the end of a point being a convergence point); iv'. determination of the shortest path, on the surface of the initial reference model, between two points of approximation of the inner and outer gingival edges, respectively, of an interdental region, preferably by a metaheuristic method, preferably by simulated annealing, said shortest path at least partially delimiting a tooth model; - an updated image is acquired less than 7 days after step a), then steps c) to f) are implemented from this updated image; - in step b), a hand-held acquisition device is used (and in particular one which is not immobilized, for example by means of a support resting on the ground) and / or the patient's head is not immobilized; - in step b), an individual device chosen from the group formed by a connected camera, a smart watch, a digital tablet, a portable 3D scanner and a computer coupled with an image acquisition system, such as a webcam or a digital camera, is used to implement step b) and, preferably at least one of the steps, preferably all of steps c) to f); - in step b), an image acquisition device providing an updated infrared image is used; - in step b), a spacer is used comprising one, preferably more than two, preferably non-aligned registration marks, and, preferably, the representation of the registration marks on the updated image is used for - in step c), crop the image and / or, - in step d), roughly assess the actual acquisition conditions; - in step c), the discriminating information is chosen from the group consisting of contour information, color information, density information, distance information, brightness information, saturation information, reflection information and combinations of this information; - in step d), data provided by the acquisition device and, preferably, concerning its orientation are used; - an optimization is implemented by a metaheuristic method, preferably by simulated annealing for: - in step a), at least partially determine a gingival edge delimiting a tooth model and / or, - in step e2), search for the virtual acquisition conditions corresponding to the real acquisition conditions and / or, - in step e), search for a reference model corresponding to the updated image; - said metaheuristic method is chosen from the group formed by - evolutionary algorithms, preferably chosen from: evolutionary strategies, genetic algorithms, differential evolution algorithms, distribution estimation algorithms, artificial immune systems, Shuffled Complex Evolution path recomposition, simulated annealing, ant colony algorithms, particle swarm optimization algorithms, taboo search, and the GRASP method; - the kangaroo algorithm, - the Fletcher and Powell method, - the sound effects method, - stochastic tunneling, - climbing hills with random restarts, - the cross-entropy method, and - hybrid methods between the metaheuristic methods cited above.

[0012] The invention also relates to the use of a method according to the invention for - detecting a recurrence, and / or - determine a speed of evolution of a change in the positioning of the teeth, and / or - optimize the date of making an appointment with an orthodontist or dentist, and / or - evaluate the effectiveness of an orthodontic treatment, and / or - assess the evolution of the positioning of teeth towards a theoretical model corresponding to a determined positioning of the teeth, in particular an improved positioning of the teeth, and / or - in dentistry.

[0013] The method can in particular be implemented during an orthodontic treatment, in particular to monitor its progress, step a) being implemented less than 3 months, less than 2 months, less than 1 month, less than a week, less than 2 days after the start of the treatment, that is to say after the fitting of an appliance intended to correct the positioning of the patient's teeth, called an "active retention appliance".

[0014] The method can also be implemented after orthodontic treatment, to check that the positioning of the teeth does not change in an unfavorable way ("relapse"). Step a) is then preferably implemented less than 3 months, less than 2 months, less than 1 month, less than 1 week, less than 2 days after the end of the treatment, that is to say after the installation of an appliance intended to keep the teeth in position, called a "passive retention appliance".

[0015] The invention also relates to: - a computer program, and in particular a specialized application for a mobile telephone, comprising program code instructions for the execution of one or more, preferably all of steps b) to f), when said program is executed by a computer, - a computer medium on which such a program is recorded, for example a memory or a CD-ROM, and - a personal device, in particular a mobile phone or tablet, in which such a program is loaded.

[0016] The invention finally relates to a system comprising - a three-dimensional scanner capable of implementing step a) of a method according to the invention, - a personal device, preferably a mobile phone, loaded with a program according to the invention. Definitions

[0017] By "patient" is meant any person for whom a process is implemented in order to check their teeth, whether this person is sick or not.

[0018] By "dental care professional" we mean a dentist, an orthodontist or an orthodontic laboratory.

[0019] By "dentist" is meant a dentist or dental assistant working under the responsibility of a dentist.

[0020] By "dentition" we mean the patient's two dental arches.

[0021] A "mobile telephone" is a device weighing less than 500 g, equipped with a sensor enabling it to capture images, capable of exchanging data with another device more than 500 km away from the mobile telephone, and capable of displaying said data, and in particular said images.

[0022] The “acquisition conditions” specify the position and orientation in space of an image acquisition device relative to the patient's teeth or to a model of the patient's teeth.

[0023] The "occlusal plane" is the plane which provides the best linear correlation with the set of contact points between the teeth of the upper arch on the one hand and the teeth of the lower arch on the other hand.

[0024] The "median longitudinal plane" is the substantially vertical plane when the patient holds the head upright, which substantially symmetrically separates the right and left parts of each arch.

[0025] A “tablet” is a portable computer with a touch screen.

[0026] A 3D scanner is a device that allows you to obtain a three-dimensional representation of an object.

[0027] By "image" is meant a two-dimensional image, such as a photograph.

[0028] By "comprising a" or "comprising a" or "having a" is meant "comprising at least one", unless otherwise indicated. Brief description of the figures

[0029] Other characteristics and advantages of the invention will become apparent upon reading the detailed description which follows and upon examining the attached drawing in which: - [Fig.l] [Fig.l] represents a flowchart illustrating the implementation of a method according to the invention, - [Fig.2] [Fig.2] represents an example of an initial reference model, - [Fig.3] [Fig.3] illustrates the processing carried out to determine the occlusion plane, - [Fig.4] [Fig.4] (4a-4d) illustrates the step necessary to determine the tooth models in a reference model, - [Fig.5] [Fig.5] (5a-5d) illustrates the acquisition of updated images, as well as the cutting operation, - [Fig.6] [Fig.6] (6a-6b) illustrates the processing of an updated image to determine the outline of the teeth, and - [Fig.7] [Fig.7] schematically illustrates the relative position of registration marks 12 on updated images 14i and 142 of a spreader 10, according to the direction of observation (broken line). Detailed description

[0030] A method according to the invention comprises the steps mentioned above.

[0031] In step a), an initial reference model of the arches, or of a part of the arches of the patient is created (see [Fig.2]).

[0032] The initial reference model is a three-dimensional digital model of the patient's arches, for example of the .stl or .Obj, .DXF 3D, IGES, STEP, VDA, or Point Cloud type. Advantageously, such a model, called "3D", can be observed from any angle.

[0033] For monitoring orthodontic treatment, the initial reference model is preferably prepared at the start of the treatment. The initial reference model may correspond to a positioning of the patient's teeth before the treatment or to a positioning of the patient's teeth that the treatment is intended to achieve. In this case, the initial reference model is conventionally calculated from a first three-dimensional model corresponding to the positioning of the patient's teeth before the treatment.

[0034] For recurrence control, the initial reference model is preferably prepared less than six months, preferably less than three months, more preferably less than one month after the end of the orthodontic treatment, generally immediately after the end of the treatment. It thus corresponds to a substantially optimal positioning of the teeth.

[0035] The initial reference model can also be prepared independently of any treatment, for example because the patient wishes to monitor the movements of his teeth.

[0036] The initial reference model may be prepared from measurements taken on the patient's teeth or on a physical model of their teeth, for example a plaster model.

[0037] The initial reference model is preferably created by means of a professional device, for example by means of a 3D scanner, preferably implemented by a health professional, for example by an orthodontist or an orthodontic laboratory. In an orthodontic practice, the patient or the physical model of his teeth can be advantageously arranged in a precise position and the professional device can be improved. This results in an initial reference model very accurate. The initial reference model preferably provides information on the positioning of the teeth with an error of less than 5 / 10 mm, preferably less than 3 / 10 mm, preferably less than 1 / 10 mm.

[0038] Orientation of the initial reference model:

[0039] Preferably, the orientation of the initial reference model in space is determined, and in particular, preferably, the occlusion plane and the median longitudinal plane.

[0040] The occlusal plane and the median longitudinal plane can be determined manually, approximately. The inventors have however discovered methods for determining these planes by computer processing.

[0041] Preferably, the reference model is a model of the arches with the mouth closed, that is to say in a position in which teeth of the upper arch are in contact with teeth of the lower arch.

[0042] Conventionally, the initial reference model provided by a three-dimensional scanner makes it possible to distinguish the upper arch from the lower arch. Generally, the model is provided in the form of two files corresponding respectively to these arches, and comprising data making it possible to position the models of these arches relative to each other in the occlusion position.

[0043] Preferably, to estimate the contact points between the teeth of the upper and lower arches, all the points of the model of the upper arch and the lower arch which are at a distance less than a predetermined limit are determined, this limit preferably being less than 3 mm, preferably about 2 mm. All the other points of these models are then ignored, which leads to the representation of figure 3b. A linear regression then makes it possible to determine the occlusion plane (“plane 1” in figure 3c).

[0044] The initial reference model can thus be oriented along the occlusion plane (Fig. 3d).

[0045] If the initial reference model does not include data allowing the upper and lower arches to be positioned relative to each other, an occlusion bite is preferably used which shows the contact points between the upper teeth and the lower teeth, then the models of the upper and lower arches are repositioned relative to this occlusion bite.

[0046] The median longitudinal plane is perpendicular to the occlusal plane, but its orientation is not known.

[0047] Preferably, the following procedure is used to determine the orientation of the median longitudinal plane:

[0048] We consider axes [Ox) and [Oy) in the occlusion plane, point O being the barycenter of the normal projections of the contact points on the occlusion plane. - In this reference frame (xOy), we seek the curve, preferably parabolic, presenting the greatest correlation coefficient with all of said projections. - All the projections of the contact points are then moved into the occlusion plane, by rotation around point O, and the previous step is repeated from this new angular position of the projections of the contact points.

[0049] The cycle of the preceding operations is continued, preferably until all the contact points have been rotated 360° around the barycenter O. The correlation coefficients corresponding to the different orientations of all the contact points are then compared. The axis of the curve which leads to the highest correlation coefficient is then considered to be included in the median longitudinal plane, which makes it possible to define the orientation of the latter exactly.

[0050] The orientation in space of the initial reference model is thus perfectly determined, quickly. Creation of tooth models

[0051] In the initial reference model, a part that corresponds to a tooth, or "tooth model", is delimited by a gingival margin that can be decomposed into an inner gingival margin (on the side of the inside of the mouth with respect to the tooth), an outer gingival margin (oriented towards the outside of the mouth with respect to the tooth) and two lateral gingival margins. The gingival margins correspond to regions in which the orientation of the surface defined by the initial reference model undergoes modifications of large amplitudes. These variations in orientation can be identified according to known techniques, for example by identifying changes in direction of the normal to the modeled surface. Figure 4a represents a view of the initial reference model processed to reveal these changes in direction. Figure 4b shows the inner gingival margin that can be extracted by analyzing the image of Figure 4a.

[0052] Several views of the initial reference model are thus analyzed, which makes it possible to determine the inner and outer gingival edges in three dimensions, as shown in Figure 4c.

[0053] Furthermore, in projection in the occlusal plane, the inner and outer gingival contours of an arch approach each other on either side of a tooth. To determine a lateral gingival edge of a tooth, the shortest path is sought, on the surface of the initial reference model, between the two points of the inner and outer gingival edges thus brought together and which are substantially opposite each other. The search for the shortest path between two points on a three-dimensional model uses techniques well-known optimization methods. Preferably, this search results from a metaheuristic method, preferably simulated annealing.

[0054] Two adjacent lateral gingival edges and the parts of the inner and outer gingival edges that connect these lateral gingival edges thus make it possible to delimit a tooth at the gum level. Taking into account that a tooth extends from the gingival contour towards the occlusal plane, it is thus possible to determine the parts of the initial reference model that correspond to the different teeth ("tooth models"). Figure 4d represents all the tooth models of an arch.

[0055] The initial reference model may be stored in a centralized database, grouping the initial reference models of a plurality of patients. This database may be physically installed in a specialized establishment. It may also be installed in a laboratory or an orthodontic practice, which limits the transfer of confidential information.

[0056] In one embodiment, the initial reference model is given to the patient. Preferably, a computer file corresponding to the initial reference model is stored on a removable medium, for example on a USB stick or on an electronic card, preferably on a mobile phone, a tablet or a laptop of the patient, and in particular on the personal device which will preferably be used in steps b) and following. Preferably, the patient or a dental care professional loads the initial reference model into said individual device or makes it available for loading into said individual device. The patient preferably loads the initial reference model from the internet.

[0057] In a preferred embodiment, the reference model is not given to the patient. Preferably, the reference model is only made available to a specialized establishment for implementing steps c) to f). It may remain stored in the establishment in which it was produced in step a) and where, preferably, steps c) to f) are implemented.

[0058] In step b), an updated image of the arches is taken using an image acquisition device. Step b) is preferably carried out by the patient or a relative of the patient, but can be carried out by a dentist. Time of acquisition

[0059] Preferably, the updated image is taken after a time interval At after step a). The time interval At may be predetermined. It may be constant, regardless of the occurrence of the method, i.e. whether this interval relates to the first execution of the method or a subsequent execution. It may be variable, and depend for example on the results obtained following a previous execution of the method. in particular, for the control of recurrence, the time interval At can be all the shorter as this execution has made it possible to detect a significant drift.

[0060] In a preferred embodiment, the time interval At is determined by the orthodontist, according to a schedule of checks. Depending on the evolution of the position of the teeth, the orthodontist can modify this schedule and consequently modify the time interval At. In one embodiment, the method according to the invention is executed several times, the time intervals between each execution being able to be identical or different. The time intervals between two successive executions can all be determined before the first execution to correspond to a schedule of checks drawn up by the orthodontist.

[0061] The time interval At may also be indeterminate and depend, for example, on decisions of the patient. For example, the creation of an updated image may be carried out during a dental appointment or at any time when the patient wishes, or even independently of any orthodontic treatment.

[0062] The time interval At is preferably determined to correspond to a potentially significant change in the positioning of the teeth.

[0063] For example, for the control of recurrence, the time interval At is preferably less than three months during the first year after treatment. After this first year, the time interval At is preferably greater than one month, or even greater than six months or greater than twelve months. In particular for the detection of tooth drift, a time interval of between six months and eighteen months is suitable.

[0064] Preferably, at least one reminder informing the patient of the need to create an updated image is sent to the patient. This reminder may be in paper form or, preferably, in electronic form, for example in the form of an email, an automatic alert from the specialized mobile application or an SMS. Such a reminder may be sent by the orthodontic practice or laboratory or by the dentist or by the patient's specialized mobile application, for example.

[0065] In a preferred embodiment, an updated image is acquired before the teeth have been able to move significantly, substantially at the same time as the creation of the initial reference model, preferably less than 7 days, less than 3 days, less than 1 day after step a), i.e. before the teeth have been able to move significantly. Implementing the method with this updated image advantageously makes it possible to verify that the method does not lead to the detection of any difference between the initial and final reference models, and therefore functions correctly.

[0066] In one embodiment, the updated image may be acquired before the initial reference model is created. For example, steps a) and b) may be performed at the end and at the beginning of an orthodontic treatment, respectively. It is thus in particular possible to evaluate the effectiveness of the treatment in the absence of a 3D model at the start of treatment. The time interval At' separating steps a) and b) in this embodiment can in particular take the values ​​described above for At. Image acquisition device

[0067] Preferably, the image acquisition device is a personal device commonly available commercially, for example a mobile phone, a so-called "connected" camera, a so-called "smart" watch, or "smartwatch", or a tablet or a personal computer, fixed or portable, comprising an image acquisition system, such as a webcam or a camera, preferably a digital camera. Even if the updated image can in particular be created by a dentist, it is preferably created by the patient himself or by one of his relatives.

[0068] The image acquisition apparatus preferably weighs less than 3 kg, less than 2 kg, less than 1 kg, less than 500 g, preferably less than 300 g.

[0069] Step b) can therefore advantageously be carried out at a distance from step a), i.e. in a location different from that in which step a) is carried out, in particular more than 50 m, more than 100 m, more than 1 km from the location where step a) is carried out, in particular outside an orthodontic practice. In one embodiment, step b) is not carried out in a dental practice, an orthodontic practice or an orthodontic laboratory, except, possibly, during a session intended to train the patient.

[0070] Preferably, the updated image is a photograph, in particular a panoramic photograph. In one embodiment, the updated image is extracted from a film.

[0071] In a preferred embodiment, the method uses several updated images to have at least one representation of each tooth, preferably at least three updated images corresponding to a front view, a right view and a left view of the patient's teeth.

[0072] Preferably, in step b), at least one updated image in the closed mouth position and at least one updated image in the open mouth position are taken. The closed mouth image advantageously makes it possible to identify the relative movements between the two arches. The updated open mouth image advantageously makes it possible to clearly identify the contours of the teeth, without the teeth of the upper arch masking the teeth of the lower arch or vice versa.

[0073] Updated images may be taken for either the upper arch, the lower arch, or preferably both arches.

[0074] Several similar images (representing substantially the same teeth) can also be useful in order to find the best score. Depending on the conditions acquisition, discriminating information may in particular lead to different scores depending on the updated image used.

[0075] Preferably, a retractor is used in step b), as shown in Figures 5a and 5c. The primary function of the retractor is to spread the lips apart to improve the visibility of the teeth. Preferably, a retractor is given to the patient, for example, during an appointment with their orthodontist or dentist.

[0076] The image acquisition apparatus preferably provides color images, and / or infrared images of the patient's mouth, or even the patient's face. The infrared images advantageously make it possible to display the teeth with excellent contrast.

[0077] Preferably, the image acquisition device comprises a specialized application for implementing step b), but also, preferably, the following steps, preferably all of the following steps. More preferably, this application manages reminders and informs the patient of the need to create an updated image.

[0078] Preferably, the specialized application is loaded into the image acquisition device from a physical medium such as a USB key or a CD-ROM, or is downloaded from the Internet or over the air. In one embodiment, the specialized application is provided to the patient by the practice and / or the orthodontic laboratory. It may in particular take the form of an application of the type commonly downloaded on Apple® iPhones or devices of any brand implementing Android® operating systems or any other operating system.

[0079] The image acquisition device preferably comprises a camera or a video or infrared camera, which the user, for example the patient or one of his relatives, positions by means of a viewfinder or a screen, before activating it. Deceptive means

[0080] A method according to the invention does not require precise positioning of the image acquisition device relative to the teeth.

[0081] In one embodiment, no positioning constraint is imposed to ensure that the image acquisition apparatus is positioned within 30 cm, 20 cm, 10 cm, or 5 cm of a given location.

[0082] Preferably, the image acquisition device however comprises error-proofing means facilitating its approximate positioning relative to the patient before the acquisition of the updated image.

[0083] The user may be guided by written and / or voice messages for the acquisition. For example, the personal device may announce “take a photo from the front”, emit a signal to inform the user that the photo is acceptable or that on the contrary, he must take another photo, announce “take a photo from the right”, preferably in displaying an arrow to guide the user, etc. The end of the acquisition process may also be announced by the device. The device may also assist with positioning, for example by visual messages (for example by displaying arrows), and / or audio messages (such as a succession of beeps whose frequency increases as the device's positioning improves), and / or written and / or vocal messages (“higher”, lower”, etc.).

[0084] The foolproofing means may in particular comprise reference marks which appear on the viewfinder or the screen. The reference marks may for example comprise a line intended to be aligned with the general direction of the joint between the upper teeth and the lower teeth when the teeth are clenched by the patient, or a vertical line intended to be aligned with the joint between the two upper incisors. The reference marks may also refer to other parts of the patient. For example, they may consist of marks corresponding to the position of the eyes or take the form of an outline in which the patient's mouth or face must be positioned.

[0085] In a preferred embodiment, the reference marks correspond to a reference frame, for example on the retractor. A reference frame may also be carried by a piece bitten by the patient.

[0086] In a preferred embodiment, the foolproofing means are defined, at least partially, from information provided by the initial reference model. For example, following the principles of “augmented reality”, a view of the initial reference model, for example a front view or a right view or a left view of the initial reference model, can be made visible, transparently, on the screen of the image acquisition device during acquisition. It is thus very easy for the patient to approximately superimpose such a view with the teeth that he is to photograph.

[0087] Steps c) and following are preferably carried out either on a personal device of the patient, preferably with the device used in step b), or with an application at a dental care professional, or with a dedicated third-party server.

[0088] In step c), each updated image is analyzed so as to produce, for each updated image, an updated map relating to at least one discriminating information. Redistricting

[0089] The analysis of the image may include a re-cropping of the updated image in order to isolate the relevant part, in particular to remove, at least partially, from the updated image the elements which were not the subject of the initial reference model, such as the patient's nose or eyes or the retractor. This re-cropping, or "cropping", is facilitated by the representation of registration marks on the updated image.

[0090] In particular, preferably, as shown in Figures 5a and 5c, the spacer 10 carries at least three non-aligned reference marks 12. If the spacer is in several parts, for example conventionally in two parts, each part preferably carries at least three non-aligned reference marks.

[0091] The shape of a registration mark, for example an asymmetrical shape, can also be used to locate the position of the spacer on the updated image.

[0092] Preferably, the registration marks have shapes and / or colors that facilitate their identification on an updated image. For example, they may be black while the rest of the spacer is white.

[0093] In one embodiment, the identification marks have shapes and / or colors allowing them to be individually identified. For example, they may each be a different color.

[0094] Identifying the registration marks on the updated image makes it possible to identify the area of ​​the updated image containing the elements that were the subject of the initial reference model, i.e., the teeth and gums. The updated image can then be cropped accordingly. Comparing Figures 5a and 5b, or 5c and 5d, illustrates the effect of re-cropping on an updated image. Updated map

[0095] An updated map represents discriminative information in the reference frame of the updated image. For example, Figure 6b is an updated map relating to the contour of the teeth obtained from the updated image of Figure 6a.

[0096] The discriminating information is preferably chosen from the group consisting of contour information, color information, density information, distance information, brightness information, saturation information, reflection information and combinations of these information.

[0097] A person skilled in the art knows how to process an updated image to reveal the discriminating information. This processing includes, for example, the application of well-known masks or filters, provided with image processing software.

[0098] In optional step d), the actual acquisition conditions during step b) are roughly determined. In other words, the relative position of the image acquisition device at the time when it took the updated image is determined (position of the acquisition device in space and orientation of this device). Step d) advantageously makes it possible to limit the number of tests on virtual acquisition conditions during step e), and therefore makes it possible to considerably accelerate step e).

[0099] One or more heuristic rules are preferably used. For example, preferably, conditions which correspond to a position of the image acquisition device behind the teeth or at a distance from the teeth greater than 1 m are excluded from the virtual acquisition conditions which may be tested in step e).

[0100] In a preferred embodiment, as illustrated in [Fig.7], registration marks represented on the updated image, and in particular registration marks 12 of the spacer, are used to determine a substantially conical region of space delimiting virtual acquisition conditions capable of being tested in step e), or "test cone".

[0101] Specifically, at least three non-aligned reference marks 12 are preferably provided on the spacer 10, for example, and their relative positions on the spacer are precisely measured.

[0102] The registration marks are then located on the updated image, as described above. Simple trigonometric calculations make it possible to approximately determine the direction in which the updated image was taken. A cone oriented in this direction, whose apex is at the level of the spacer and whose apex half-angle is preferably less than 10°, preferably less than 5°, for example 3° can then be defined as a "test cone". The apex half-angle corresponds to a degree of uncertainty. The smaller the apex half-angle, the greater the probability that the virtual acquisition conditions corresponding to the real acquisition conditions are outside the test cone.

[0103] For example, when the updated image is taken perpendicular to the plane of the three registration marks on the spacer, it can be deduced that the acquisition device was substantially in a test cone whose axis is substantially perpendicular to this plane when the updated image was taken. If the relative positions of the three registration marks on the updated image are different from those that the registration marks occupy on the spacer, the axis of the test cone in which the search for the positioning of the acquisition device is limited when acquiring the updated image is inclined relative to the plane of the registration marks, as shown in [Fig.7].

[0104] In a particular embodiment, as illustrated in Figures 5a and 5c, the spacer comprises independent left and right parts, each of which comprises at least three registration marks, preferably at least four registration marks. A left test cone can thus be determined by means of the registration marks of the left part and a right test cone can be determined by means of the registration marks of the right part of the spacer. The virtual acquisition conditions that can be tested can then be limited to positions of the acquisition device in space belonging to these two test cones. It can also be considered that the best evaluation of the position of the acquisition device corresponds to the average position between the best position in the left test cone and the best position in the right search cone.

[0105] The position of the registration marks on the updated image also makes it possible to evaluate the attitude of the acquisition device during the capture of the updated image. For example, if it is known that two registration marks are substantially aligned in a horizontal direction during the acquisition of the updated image, the direction of the line containing these two points on the updated image provides an indication of the orientation of the acquisition device under the actual acquisition conditions.

[0106] Finally, the size of the registration marks on the updated image or their spacing can make it possible to evaluate the distance between the image acquisition device and the teeth during the acquisition of the updated image, and therefore to reduce the test cone to a truncated cone.

[0107] In optional step d), it is also possible to use data provided by the acquisition device and concerning its orientation, for example gyroscopic data.

[0108] Step d) only allows a rough evaluation of the actual acquisition conditions. Step d), however, makes it possible to determine a restricted set of virtual acquisition conditions likely to correspond to the actual acquisition conditions, and, within this set, virtual acquisition conditions constituting the best starting point for step el) described below.

[0109] Step d) also makes it possible to detect updated images that are unsuitable for continuing the method, for example an updated image that does not show the registration marks. Preferably, the method is then repeated at step c) with a new updated image.

[0110] Of course, the different methods that can be implemented in step d) can be combined. [YES] In step e1), we determine that the reference model to be tested is the initial reference model during the first execution of step e2).

[0112] In step e 2), virtual acquisition conditions to be tested are first determined, i.e. a virtual position and orientation likely to correspond to the real position and orientation of the acquisition device during the capture of the updated image. The first virtual acquisition conditions to be tested may be random. Preferably, they are chosen from the limited set determined in step d), and more preferably, correspond to virtual acquisition conditions corresponding, according to step d), to the most promising virtual acquisition conditions, i.e. constituting the best springboard for approaching, as quickly as possible, the real acquisition conditions (step e21)).

[0113] The image acquisition device is then virtually placed in the virtual acquisition conditions to be tested in order to acquire a reference image of the model of reference in these virtual acquisition conditions to be tested. The reference image therefore corresponds to the image that the image acquisition device would have taken if it had been placed, relative to the reference model to be tested, in the virtual acquisition conditions to be tested (step e22)).

[0114] If the updated image was taken at the same time as the reference model was made, and if the virtual acquisition conditions are exactly the real acquisition conditions, the reference image is therefore exactly superimposable on the updated image. The differences between the updated image and the reference image result from errors in the evaluation of the virtual acquisition conditions (if they do not correspond exactly to the real acquisition conditions) and from movements of the teeth between step b) and the reference model to be tested.

[0115] To compare the updated and reference images, the discriminating information on these two images is compared. More precisely, a reference map representing the discriminating information is produced from the reference image (step e23)).

[0116] The updated and reference maps, both relating to the same discriminating information, are then compared and the difference between these two maps is evaluated by means of a score. For example, if the discriminating information is the outline of the teeth, the average distance between the points of the outline of the teeth which appear on the reference image and the points of the corresponding outline which appear on the updated image can be compared, the score being higher the smaller this distance.

[0117] The score can be for example a correlation coefficient.

[0118] The score is then evaluated using a first evaluation function. The first evaluation function makes it possible to decide whether the cycling on step e2) must be continued or stopped. The first evaluation function may, for example, be equal to 0 if the cycling must be stopped or be equal to 1 if the cycling must continue.

[0119] The value of the first evaluation function may depend on the score achieved. For example, it may be decided to continue cycling on step e2) if the score does not exceed a first threshold. For example, if an exact match between the updated and reference images leads to a score of 100%, the first threshold may be, for example, 95%. Of course, the higher the first threshold, the better the accuracy of the evaluation of the virtual acquisition conditions will be if the score manages to exceed this first threshold.

[0120] The value of the first evaluation function may also depend on scores obtained with previously tested virtual acquisition conditions.

[0121] The value of the first evaluation function may also depend on random parameters and / or the number of cycles of step e2) already carried out.

[0122] In particular, it is possible that despite the repetition of the cycles, it is not possible to find virtual acquisition conditions that are sufficiently close to the real acquisition conditions for the score to reach said first threshold. The first evaluation function may then lead to the decision to leave the cycling even though the best score obtained has not reached said first threshold. This decision may result, for example, from a number of cycles greater than a predetermined maximum number.

[0123] A random parameter in the first evaluation function may also allow for continued testing of new virtual acquisition conditions, even though the score appears satisfactory.

[0124] The evaluation functions conventionally used in metaheuristic optimization methods, in particular in simulated annealing methods, can be used for the second evaluation function.

[0125] If the value of the first evaluation function indicates that it is decided to continue the cycling on step e2), the tested virtual acquisition conditions are modified (step e25)) and a cycle is started again (step e2)) consisting of producing a reference image and a reference map, then comparing this reference map with the updated map to determine a score.

[0126] The modification of the virtual acquisition conditions corresponds to a virtual displacement in space and / or to a modification of the orientation of the acquisition device. This modification may be random, provided however that the new virtual acquisition conditions to be tested always belong to the set determined in step d). The modification is preferably guided by heuristic rules, for example by favoring the modifications which, according to an analysis of the previous scores obtained, appear the most favorable for increasing the score.

[0127] The cycling on e2) is continued until the value of the first evaluation function indicates that it is decided to exit this cycling and continue to step e3), for example if the score reaches or exceeds said first threshold.

[0128] The optimization of the virtual acquisition conditions in step e2) is preferably carried out using a metaheuristic method, preferably a simulated annealing algorithm. Such an algorithm is well known for nonlinear optimization.

[0129] If the cycling has been left at step e2), without a satisfactory score having been obtained, for example without the score having been able to reach said first threshold, the method can be stopped (failure situation) or resumed at step c) with new discriminating information and / or with a new updated image. The method can also be continued with the virtual acquisition conditions corresponding to the best score achieved. A warning may be issued to inform the user of the error in the result.

[0130] If cycling has been left at step e2) when a satisfactory score has been obtained, for example because the score has reached or even exceeded said first threshold, the virtual acquisition conditions correspond substantially to the real acquisition conditions. However, differences may remain, in particular if teeth have moved between steps a) and b). The correlation between the updated and reference images can then be further improved by repeating step e2), the reference model to be tested then being modified by moving one or more tooth models (step e3)).

[0131] The search for the reference model best approximating the positioning of the teeth during the acquisition of the updated image can be carried out as the search for the virtual acquisition conditions best approximating the real acquisition conditions (step e2)).

[0132] In particular, the score is evaluated by means of a second evaluation function. The second evaluation function makes it possible to decide whether the cycling on steps e2) and e3) must be continued or stopped. The second evaluation function may, for example, be equal to 0 if the cycling must be stopped or be equal to 1 if the cycling must continue.

[0133] The value of the second evaluation function preferably depends on the best score obtained with the reference model to be tested, i.e. the differences between the updated and reference maps, in the virtual acquisition conditions which best approximate said real acquisition conditions.

[0134] The value of the second evaluation function may also depend on the best score obtained with one or more reference models tested previously.

[0135] For example, it may be decided to continue cycling if the score does not exceed a second minimum threshold. The value of the second evaluation function may also depend on random parameters and / or the number of cycles of steps e2) and e3) already performed.

[0136] The evaluation functions conventionally used in metaheuristic optimization methods, in particular in simulated annealing methods, can be used for the second evaluation function.

[0137] If the value of the second evaluation function indicates that it is decided to continue the cycling on steps e2) and e3), the reference model to be tested is modified and a cycle is started again (steps e2) and e3)) with the new reference model to be tested.

[0138] The modification of the reference model to be tested corresponds to a displacement of one or more tooth models. This modification can be random. The modification is preferably guided by heuristic rules, for example by favoring modifications which, based on an analysis of previous scores obtained, appear most favorable for increasing the score.

[0139] Preferably, the displacement of a tooth model that has the greatest impact on the score is sought, the reference model to be tested is modified by moving this tooth model, then the cycling is continued on steps e2) and e3) so as to optimize the score. It is then possible to search, among the other tooth models, for the one that has the greatest impact on improving the score, and again search for the optimal displacement of this other tooth model on the score. This can be continued with each tooth model.

[0140] Then, it is possible to repeat a cycle on all the tooth models and continue in this way until a score greater than the second threshold is obtained. Of course, other strategies can be used to move one or more tooth models into the reference model to be tested and search for the maximum score.

[0141] The cycling on steps e2) and e3) is continued until the value of the second evaluation function indicates that it is decided to exit this cycling and continue to step f), for example if the score reaches or exceeds said second threshold.

[0142] The search for a reference model with cycling on steps e2) and e3) to search for the positions of the tooth models that optimize the score is preferably performed using a metaheuristic method, preferably a simulated annealing algorithm. Such an algorithm is well known for nonlinear optimization.

[0143] If the cycling has been left on steps e2) and e3) without a satisfactory score having been obtained, for example without the score having been able to reach said second threshold, the method can be stopped (failure situation) or resumed at step c) with new discriminating information and / or with a new updated image.

[0144] If it is decided to restart the method in step c) from another discriminating information and / or another updated image because the first threshold or the second threshold has not been reached, the choice of the new discriminating information and / or the new updated image may depend on the scores obtained previously, in order to favor the discriminating information and / or the updated image which, in light of these scores, appear the most promising.

[0145] New discriminating information, obtained for example by combining other discriminating information already tested, can be used. If necessary, it may also be requested to acquire one or more new updated images. Preferably, indications are provided to guide the positioning of the acquisition device for the capture of this new updated image. For example, the patient can be told that he should take a photo of the right part of his lower arch.

[0146] If cycling has been left on steps e2) and e3) without a satisfactory score having been obtained, the method can also be continued with the reference model and the virtual acquisition conditions corresponding to the best score achieved. A warning can be issued to inform the user of the error in the result.

[0147] If cycling has been left on steps e2) and e3) when a satisfactory score could be obtained, for example because the score has reached or even exceeded said second threshold, the virtual acquisition conditions correspond substantially to the real acquisition conditions and the tooth models in the reference model obtained (called the “final reference model”) are substantially in the position of the patient's teeth at the time of step b).

[0148] In step f), the final reference model, resulting from the optimization by displacement of the tooth models, is compared with the initial reference model. The final reference model corresponds substantially to the updated image. The comparison in step f) therefore makes it possible to observe the differences between the positioning of the teeth in step a) (initial reference model) and during the acquisition of the updated image (step b)). The method thus makes it possible to precisely determine, for each of the teeth, the movements between these two steps.

[0149] By repeating steps b) and following, it is also possible to evaluate the speed of evolution of the position of the teeth, and thus to measure, for example, the effectiveness of an orthodontic treatment. A method according to the invention can for example be used to remotely monitor the evolution of an orthodontic treatment, and thus optimize the appointments of patients with their orthodontists.

[0150] In a preferred embodiment, the control method according to the invention is implemented several times for the same patient, preferably successively with several discriminating information items, preferably more than 2, more than 3, more than 5 discriminating information items for each updated image and / or with several updated images, preferably more than 2, more than 3, more than 5 updated images. The evaluation of the movement of a tooth can thus be refined by taking into account the different scores obtained. The comparison of these scores also makes it possible, where appropriate, to discard the unsatisfactory discriminating information items and / or updated images.

[0151] Depending on the measured displacement, practical information may be generated. If the displacement is small, this practical information may be that no action is to be taken. On the contrary, if one or more teeth have moved significantly, the information may be to schedule a visit to the dentist or orthodontist. Preferably, the practical information depends on the degree of movement of the teeth. In one embodiment, an appointment may be automatically made with the dentist. or the orthodontist, depending on the amplitude and / or nature of the movements detected.

[0152] In one embodiment, the practical information is used to modify the time interval after which the patient should be notified that a new updated image should be created.

[0153] In one embodiment, the individual apparatus can display images, or even a sequence of images showing the positioning of the teeth at different dates. These images can be presented in the form of an animation.

[0154] Preferably, the image acquisition device is a telephone which makes it possible to transmit the results obtained by implementing the method, preferably in a secure manner.

[0155] The communication may for example be carried out, at least in part, by radio waves, preferably according to at least one protocol chosen from the edge, 3G, 4G, udmsa, hpdmsa, bluetooth, and wifi protocols, or by any other protocol, adapted to mobile or nomadic equipment, by wired synchronization with the personal computer, or by optical transmission.

[0156] As is now clearly apparent, a method according to the invention allows precise and efficient control of the positioning of the patient's teeth, substantially without constraint for the patient. In particular, simple photographs taken without any particular precaution, for example with a mobile phone, are sufficient. The patient can therefore easily implement this method.

[0157] Of course, the invention is not limited to the embodiments described above and shown.

[0158] In particular, the method is implemented successively for each of the two arches or simultaneously for the two arches.

[0159] Furthermore, several different devices can also be implemented. For example, the acquisition can be carried out with a mobile phone and the following steps by a desktop computer.

[0160] Finally, the patient is not limited to a human being. In particular, a method according to the invention can be used for another animal.

Claims

Claims

1. Method for generating a digital three-dimensional model of at least part of a patient's arches, said method comprising the following steps: a) producing, with a 3D scanner, a digital three-dimensional reference model of the patient's arches, or "initial reference model" and, for each tooth, defining, from the initial reference model, a digital three-dimensional reference model of said tooth, or "tooth model"; b) acquiring, by the patient, with a mobile telephone, at least one two-dimensional image of said arches, called "updated image", under real acquisition conditions; and e) searching, for each updated image and by moving one or more tooth models, for a final reference model corresponding to the positioning of the teeth during the acquisition of the updated image.

2. Method according to the preceding claim, in which the updated image is analyzed and, for each updated image, an updated map is produced relating to discriminant information chosen from the group consisting of contour information, color information, density information, distance information, brightness information, saturation information, information on reflections and combinations of these information, then the initial reference model is modified according to the following steps: el) definition of the reference model to be tested as being the initial reference model then, e2) according to the following steps, testing virtual acquisition conditions with the reference model to be tested in order to finely approximate said real acquisition conditions; e21) determination of virtual acquisition conditions to be tested;e22) producing a two-dimensional reference image of the reference model to be tested under said virtual acquisition conditions to be tested; e23) processing the reference image to produce at least one reference map representing said discriminating information; e24) comparison of the updated and reference maps so as to determine a value for a first evaluation function, said value for the first evaluation function depending on the differences between said updated and reference maps and corresponding to a decision to continue or stop the search for virtual acquisition conditions approximating said real acquisition conditions with more accuracy than said virtual acquisition conditions to be tested determined at the last occurrence of step e21); e25) if said value for the first evaluation function corresponds to a decision to continue said search, modification of the virtual acquisition conditions to be tested, then resumed at step e22);e3) determining a value for a second evaluation function, said value for the second evaluation function depending on the differences between the updated and reference maps under virtual acquisition conditions best approximating said real acquisition conditions and resulting from the last occurrence of step e2), said value for the second evaluation function corresponding to a decision to continue or stop the search for a reference model approximating the positioning of the teeth during the acquisition of the updated image with more accuracy than said reference model to be tested used at the last occurrence of step e2), and if said value for the second evaluation function corresponds to a decision to continue said search, modifying the reference model to be tested by moving one or more tooth models and then continuing at step e2).;

3. Method according to any one of the preceding claims, in which, for each tooth model, positions of said tooth model in the initial reference model and in the final reference model are compared, in order to determine the movement of the teeth between the time of creation of the initial reference model and the time of creation of the final reference model.

4. Method according to any one of the preceding claims, in which in step b), at least one updated image in the closed mouth position and at least one updated image in the open mouth position are taken.

5. A method according to any preceding claim, wherein said at least one updated image comprises at least three updated images corresponding to a front view, a right view and a left view of the patient's teeth.

6. A method according to any preceding claim, wherein the updated image is extracted from a film.

7. A method according to any preceding claim, wherein, for acquiring the updated image, a retractor is used to separate the patient's lips in order to improve the visibility of the teeth.

Images