Method for Tracking Tooth Movement

Abstract

A method for training a neural network intended to analyze a current patient's dental situation is disclosed, comprising the steps of: A) creating a historical learning database of dental bodies and spatial attributes associated with the dental bodies, the historical learning database including 1,000 or more historical records, each historical record relating to a particular historical patient, the historical record including a set of historical images depicting all of the dental bodies in the historical patient (referred to as "historical dental bodies") and one item of spatial information for the historical patient (referred to as "historical spatial information") including a set of values ​​for the spatial attributes; and B) training the neural network by providing the set of historical images as input and the historical spatial information as output to the neural network.

Description

Technical Field

[0001] The present invention relates to a method for training a neural network intended to analyze a patient's dental situation, a method for analyzing a patient's dental situation implementing such a trained neural network, and a method for determining the amount of movement of a dental body, in particular to a method for monitoring the activity of an active orthodontic appliance or the loss of effectiveness of a passive orthodontic appliance.

Background Art

[0002] The applicant has developed methods for remotely monitoring a patient's dental situation before, during or after orthodontic treatment. These methods rely on comparing a photograph taken by the patient's mobile phone with a view of a three-dimensional digital model of at least one dental arch of the patient's dental arcade. by patient by Specifically, an initial model of the dental arcade is customarily created with a 3D scanner and then segmented into tooth models. After the patient has taken a photograph, the initial model is deformed by moving the tooth models to best match the photograph. The comparison between the initial model and the deformed model thus obtained provides information about the movement of the teeth from. Since the initial model is very precise, the same advantageously also applies to the deformed model, and thus also to the information resulting from the comparison.

[0003] More particularly, the initial model of the dental arcade is customarily created with a 3D scanner and then segmented into tooth models. After the patient has taken a photograph, the initial model is deformed by moving the tooth models to best match the photograph. The comparison between the initial model and the deformed model thus obtained provides information about the movement of the teeth from. Since the initial model is very precise, the same advantageously also applies to the deformed model, and thus also to the information resulting from the comparison. at an initial point in time Specifically, for moving the tooth models, initial point in time provides information about the movement of the teeth from. Since the initial model is very precise, the same advantageously also applies to the deformed model, and thus also to the information resulting from the comparison.

[0004] These methods are described in particular in International Patent Application No. PCT / EP2015 / 074868 or International Patent Application No. PCT / EP2015 / 074859.

[0005] Specifically, for moving the tooth models,they are It requires a significant amount of computing resources. Typically, several hours of computer processing are required to evaluate the dental situation.

[0006] Moreover, these methods require the creation of the initial model, and thus the patient has to visit a dentist and use a 3D scanner. This procedure is expensive and unpleasant for the patient.

[0007] Therefore, there is a need for a method that enables remote monitoring of patients and does not have the aforementioned drawbacks. SUMMARY OF THE INVENTION PROBLEM TO BE SOLVED BY THE INVENTION

[0008] An object of the present invention is to at least partially meet these requirements. MEANS FOR SOLVING THE PROBLEM

[0009] The present invention proposes a method for training a neural network intended to analyze the latest dental situation of a patient, the method comprising: A) creating a historical learning database for a tooth body, for example tooth number 13, and spatial attributes associated with the tooth body, for example the center of gravity of tooth number 13, where; the historical learning database contains 1,000 hyper historical records, each historical record being related to an individual historical patient, and each such historical record comprising a set of historical images, where all of the historical images depict the dental body (referred to as the "historical dental body") in the historical patient , and one item of spatial information (referred to as "historical spatial information") for the historical patient, comprising a set of values for the spatial attribute (i.e., in the example considered, the values determining the position of the center of gravity of tooth number 13 for the historical patient), comprising B) training the neural network by providing the neural network with the set of historical images as input and the historical spatial information as output includes the steps of.

[0010] Therefore, after training, the neural network is compatible with the set of historical images in the historical learning database and can determine the latest item of spatial information for a set of the latest images input into the neural network as input.

[0011] Preferably, the historical learning database is specialized for accurate tooth bodies and for spatial attributes with limited variables, thereby improving its performance capabilities. Preferably, nevertheless, of the historical record the number of historical images is limited, thereby enabling the specialized neural network is to be used later to without supplying many images. require

[0012] The training method according to the present invention preferably has one or more of any of the following features: - The neural network is a convolutional type (CNN, "Convolutional Neural Network"); - The spatial attribute is - The position of one or more points to be noted of the tooth body in a three-dimensional reference frame, for example, the position of one or more points to be noted of the tooth body in a three-dimensional reference frame fixed relative to the dental arch of the patient considered or relative to a point of the tooth of the patient considered; and / or - One or more vectors between a plurality of points to be noted of the tooth body and / or between one point to be noted of the tooth body and another point, in particular one point to be noted of the patient considered, preferably one point to be noted of the oral cavity of the patient considered; defines; - The tooth body is a set of one tooth or less than 5 teeth, preferably less than 4 teeth; - The historical tooth body is one tooth having a predetermined number or , a set of a plurality of teeth including one tooth having a predetermined number and one or two teeth adjacent to the one tooth ; ​- the spatial attribute includes a variable that is less than 30, preferably less than 20, preferably less than 10; - the set of historical images of any historical record includes 2 hyper , preferably 3 hyper , preferably 5 hyper , preferably 6 hyper , and / or includes historical images that are less than 30, preferably less than 20, preferably less than 15, preferably less than 10; - each historical image of the set of historical images of any historical record includes 2 hyper , preferably 4 hyper , preferably 4 hyper , and / or is obtained from a group of potential angles including potential angles that are less than 30, preferably less than 20, preferably less than 10, or an angle selected from standard angles; - at least three historical images of any set of historical images, preferably all historical images of any set of historical images, are obtained at different angles; - the historical image is a true - color photograph depicting a dental scene that can be perceived by the human eye, and / or does not depict a dental retractor, and / or is extra - oral; - the historical image is obtained at various distances from the historical patient and is then cropped around the tooth body depicted in the historical image.

[0013] Accordingly, the present invention relates to a method for analyzing the dental situation of a patient at the latest moment (referred to as the "latest patient"), the method comprising a) training a neural network according to a training method according to the present invention before the latest moment; b) at the latest moment, obtaining, by an image acquisition device, a set of latest images that match the neural network and depict the tooth body of the latest patient (referred to as the "latest tooth body"); c) analyzing the set of latest images by the neural network to obtain an item of spatial information (referred to as the "latest spatial information") for the latest patient, including a set of values for the spatial attribute including the steps of.

[0014] An image of a dental scene, such as a photograph, is the result of projecting this dental scene onto a plane. Analysis of the depiction of the dental body of the dental scene in the image can provide spatial information when the shape of this dental body is known, or when the shape of another object of the dental scene that is depicted and linked to this dental body is known. However, generally, the shape of a patient's dental body, especially teeth, is not known. Therefore, a simple analysis of an image of a dental scene generally does not enable reliable spatial information to be determined for the dental body.

[0015] According to the principle of a 3D scanner, by analyzing several images of the same dental scene, it is possible to provide accurate spatial information regarding the dental body even when the shape of the dental body of the dental scene is not known. However, recognizing the dental body on the image and comparing a plurality of the images requires long and costly operations.

[0016] As will be seen in more detail throughout the remainder of this specification, and quite unexpectedly, the inventors have discovered that by using a neural network, it is possible to obtain spatial information from simple images, especially photographs, with very good accuracy. In particular, it was surprising to discover that without relying on a three-dimensional model of the patient's dental arch, the error in the spatial information can be less than 1 mm, less than 0.5 mm, and even less than 0.3 mm. Such accuracy was actually considered impossible to achieve from images alone before the present invention without performing complex processing, such as processing using a 3D scanner. In particular, it was considered impossible to achieve using a photograph, such as a simple photograph taken by the patient themselves.

[0017] Even when the latest patient is not wearing a dental retractor and there is no need to fix a mobile phone on a support, such as a tripod, and the image is taken with a simple mobile phone without special attention , mostIt has been shown that the new spatial information is reliable. , the test is shown.

[0018] Moreover, whereas hitherto several hours of computer processing were required to evaluate each dental situation, the method according to the invention makes it possible to obtain the most recent items of spatial information in a few seconds.

[0019] Finally, the most recent patient no longer has to go to the dentist. Therefore, the analysis method can be carried out by anyone with a mobile phone, independently of contact with dental healthcare professionals.

[0020] The analysis method according to the invention preferably has one or more of the following optional features: - the image acquisition device is a mobile phone; - the most recent image is a true-color photograph depicting a dental scene that can be recognized by the human eye; - the most recent patient is not wearing a dental retractor in step b); - the most recent image is outside the oral cavity; - the most recent image is acquired at various distances from the most recent patient and is then cropped around the tooth body depicted in the most recent image.

[0021] Preferably, in step a), a plurality of neural networks are trained in individual historical learning databases that differ in terms of different tooth bodies and / or different spatial attributes, to obtain a plurality of specialized neural networks; In step b), a most recent image is acquired and, for each of the plurality of specialized neural networks of the plurality of specialized neural networks, a set of the most recent images is created from the most recent image; In step c), each of the plurality of sets of the most recent images is analyzed by the corresponding specialized neural network to obtain a plurality of most recent items of spatial information, where it can generally be said as static information.

[0022] Preferably, each history learning database contains a tooth having a unique number in the history learning database, or , a group of a plurality of teeth, where the numbers of the plurality of teeth in the group are unique to the historical learning database Regarding.

[0023] Preferably, the spatial attributes are the same for all of the historical learning databases.

[0024] The analysis method allows multiple dental bodies to be analyzed with a neural network specific to each dental body, so that the up-to-date spatial information is accurate and substantial.

[0025] Therefore, the present invention opens up a very wide range of applications.

[0026] In a particularly advantageous manner, the analysis method according to the invention comprises: at different latest moments , can be performed multiple times.

[0027] The present invention is particularly before The latest moment of before After the latest moment of after 1. A method for determining the amount of movement between a current instant of time and a current instant of time, the method comprising the steps of: 1) The analysis method according to the present invention before The latest information on spatial information before The most recent items of information, or more generally, a static before 's' item; 2) The analysis method according to the present invention after The latest information on spatial information after The most recent items of information, or more generally, a static after 's' item; 3) Applicable before The latest spatial information and after More generally, the static before The latest spatial information and the static after The latest spatial information is compared with the before The latest moment and afterobtaining the amount of movement between the latest instant and, where the comparison is in particular between the latest spatial information of the before and the latest spatial information of the after (or more generally between the static information of the before and the static information of the after ) and optionally dividing the difference by the time interval between the latest instant of the before and the latest instant of the after ; 4) Preferably, presenting the amount of movement on a screen of, for example, a personal computer or a mobile phone to, preferably to the latest patient and / or dental healthcare provider.

[0028] The method of determining according to the present invention preferably has one or more of the following optional features: - The amount of movement defines the amplitude and / or speed of translational and / or rotational movement between the latest instant of the before and the latest instant of the after for one or more points of the latest dental body and / or for one or more vectors connecting a plurality of points of the latest dental body or for one or more vectors connecting one or more points of the latest dental body and one or more other points of the oral cavity of the latest patient; - In step 3), the amount of movement is compared with a threshold value and depending on the difference between the amount of movement and the threshold value, the following are determined: - An activity index of the dental orthosis worn by the latest patient; and / or - A fitness index of the dental condition of the latest patient having a predetermined situation defined by the dental orthodontic treatment performed by the latest patient or a situation resulting from the dental orthodontic treatment performed by the latest patient or a situation defined by a dental healthcare provider independently of the dental orthodontic treatment, for example a defined normal situation of the latest patient; - The activity index and / or the fitness index are presented in the form of a graph; - The activity index and / or the fitness index are presented on a screen, for example on a screen of a computer or a mobile phone.​

[0029] Preferably, as described above, a plurality of neural networks are specialized for respective teeth or groups of teeth, where each neural network is specialized for a tooth having, for example, a unique number for it.

[0030] Thus, the method of determining according to the present invention enables the analysis of the development over time of several dental bodies using a neural network that is specific to each dental body. Thus, the amount of movement is accurate and substantial. All these amounts of movement can generally be recognized as dynamic information.

[0031] In a preferred embodiment, the specialization is for each type of tooth. For example, one neural network can be specialized for canines, another network can be specialized for incisors, a third neural network can be specialized for molars, etc.

[0032] The specialization can be done for each tooth number. For example, one neural network can be specialized for tooth number 13, another network can be specialized for tooth number 14, a third neural network can be specialized for tooth number 15, etc.

[0033] The analysis method and the determination method according to the present invention are particularly - detecting or evaluating the position or shape of a tooth, and / or the development of the position or shape of a tooth, and / or the speed of development of the position or shape of a tooth; in particular, - during the period before treatment, i.e., the period prior to orthodontic treatment; - when no orthodontic treatment has been carried out, in particular, to monitor tooth eruption, or to detect the recurrence position or abnormal position of a tooth, or to detect tooth wear, such as tooth wear due to bruxism, or to monitor the opening and closing of the space between two teeth or between three or more teeth, in particular between two adjacent teeth, or to monitor the stability or correction of occlusion; - In the context of orthodontic treatment, in particular, for monitoring the movement of teeth, in particular to an improved position of the teeth, to a predetermined position, or for monitoring tooth eruption, or for creating a space suitable for attaching a dental implant, for example, between two teeth or between three or more teeth, in particular between two adjacent teeth, for monitoring the opening and closing of the space; and / or, - Detecting or evaluating the position or shape of an orthodontic appliance, in particular an abnormal position or shape of the orthodontic appliance, such as detachment of a ring or detachment of an orthodontic aligner, and / or deployment of the position or shape of the orthodontic appliance, and / or the speed of deployment of the position or shape of the orthodontic appliance for, in particular, - Optimizing the appointment date with an orthodontist or dentist; and / or, - Evaluating the effectiveness of an active orthodontic treatment; and / or, - Measuring the activity of an active orthodontic appliance; and / or, - Measuring the loss of effect of a passive orthodontic appliance; and / or, - Odontology; and / or, - Measuring the development of the shape of a patient's teeth between two dates, for example between two dates before treatment, in particular between two dates separated by an event that is likely to have changed the position and / or shape of at least one tooth, for example between two dates separated by the occurrence of an impact on the tooth, or between two dates separated by the use of a dental device that can have an adverse effect, such as a dental device intended for the treatment of sleep apnea, or between two dates separated by the occurrence of a graft in the patient's oral cavity, in particular a periodontal graft, in particular a gingival graft can be used for.

[0034] Steps A) and B), and / or steps a) and c), and / or steps 1) to 3) (excluding step b) are preferably carried out by a computer. The present invention also A computer program comprising program code instructions for implementing steps A) and B), and / or steps a) and c), and / or steps 1) to 3) (excluding step b); A computer medium, such as a memory or a CD-ROM, having recorded thereon such a program; and, A computer into which such a program has been loaded relating thereto.

[0035] Definition

[0036] The term "patient" is understood to mean any person on whom the method according to the invention is carried out, whether or not this person is ill or is undergoing orthodontic treatment.

[0037] "Orthodontic treatment" is all or part of a treatment aimed at changing the configuration of the dental arch (active orthodontic treatment) or at maintaining the configuration of the dental arch, in particular that carried out after the end of active orthodontic treatment (passive orthodontic treatment).

[0038] "Orthodontic appliance" is an appliance worn by or intended to be worn by a patient. An orthodontic appliance can be intended not only for therapeutic or prophylactic treatment, but also for aesthetic treatment. In particular, an orthodontic appliance can be an arch and tooth alignment appliance, or a tooth alignment aligner, or an auxiliary appliance of the Carriere Motion type. Such an aligner extends along the successive teeth of the dental arch where it is fixed. It generally defines a "U" shaped channel. The configuration of the orthodontic appliance is determined not only to ensure its attachment to the teeth, but also as a function of the desired target position of the teeth. More specifically, the shape is determined such that, in the use position, the orthodontic appliance exerts a force tending to move the treated teeth towards their target position (active orthodontic appliance), or a force tending to hold the teeth in this target position (passive orthodontic appliance, or "retainer").

[0039] The "dental situation" defines a set of features of the patient's dental arch at a given moment, for example, the position of the teeth, their shape, the position of the orthodontic appliance, etc. at this moment. These features can also relate to the general shape of the dental arch and / or its arrangement relative to other dental arches of the patient, in particular the arrangement in the position of "closed mouth".

[0040] The term "arcade" or "dental arcade" is understood to mean all or a part of the dental arcade. Thus, the term "image of the dental arcade" is understood to mean a two-dimensional depiction of all or a part of the dental arcade.

[0041] According to the international convention of the International Dental Federation, each tooth of the dental arcade has a predetermined number. The numbers of the teeth defined in this convention are shown in Figure 7.

[0042] A "scene" is formed by a set of elements that can be observed simultaneously. A "dental scene" is a scene that includes at least one tooth in the patient's oral cavity.

[0043] A "noteworthy point" is a point of the dental scene where, for example, the apex or tip of a tooth, the interproximal contact point, i.e., the tooth with an adjacent tooth, for example, the midpoint or distal point of the incisal edge of a tooth, or the center point of the tooth crown, i.e., the "barycenter", can be identified.

[0044] The term "dental body" is understood to mean an element that can be identified within the oral cavity, such as a tooth, a set of multiple teeth, for example a pair of adjacent teeth or a set of three, a gum or a device intended to be supported by the alveolar arch, in particular an orthodontic appliance, a crown, an implant, a bridge or a facet. The dental body can also be a subset of the aforementioned elements, such as a tooth with a determined number, or a set of two, three or more adjacent teeth, where the set of one of the teeth has a determined number.

[0045] The position of the dental body is "abnormal" if it does not conform to therapeutic or aesthetic criteria.

[0046] The term "image" is understood to mean a two-dimensional digital depiction, such as a photograph or an image taken from a film. An image is composed of a plurality of pixels.

[0047] "Angulation" is the orientation of the optical axis of the image acquisition device relative to the patient when acquiring an image. By extension, when an image is acquired at this angle, the angle is said to "indicate" or "have" or be "associated with" this angle.

[0048] The "tooth zone" of an image is the part of the image that depicts only the tooth, i.e., the part that follows the profile of this tooth on this image. In other words, the depiction of the tooth on this image depicts substantially 100% of the tooth zone.

[0049] The term "model" is understood to mean a digital three-dimensional model called a "3D" model. A model is composed of a set of voxels. A "tooth model" is a three-dimensional digital model of a single tooth. The terms "image of the alveolar arch" or "model of the alveolar arch" are understood to mean the two-dimensional or three-dimensional depiction of all or a part of the alveolar arch respectively.

[0050] A 3D scanner or "scanner" is a well-known device for obtaining a model of a tooth or dental arch. It can form a 3D model by conventionally using structured light and based on matching a plurality of different images and specific points on these plurality of images.

[0051] The method according to the present invention is carried out by a computer, preferably exclusively by a computer, except for the acquisition of images. The term "computer" is understood to mean any electronic device that includes a set of a plurality of machines having computer processing capabilities. The computer can be a server remote from the user. For example, the computer can be "cloud" - based. Preferably, the computer is a mobile phone.

[0052] Conventionally, a computer in particular includes a processor, a memory, a human - machine interface (conventionally including a screen), and a module for communicating via the Internet, via Wi - Fi, via Bluetooth 登録商標 via or via a telephone network. Software configured to implement the method of the present invention is loaded into the memory of the computer. The computer can also be connected to a printer.

[0053] "First", "second", "updated", "historical", "front", "rear", "static", "dynamic" are used for clarity.

[0054] " before of" and " after of" are consecutive instants over time refer to .

[0055] The "updated" patient is the patient whose dental situation is intended to be evaluated. The "historical" patient is the patient having a corresponding historical record.

[0056] "Spatial attribute" is a generic term that designates the structure of an item of spatial information. It defines an ordered sequence of variables of a three-dimensional reference frame, for example, an orthonormal reference frame, for example, for the 14th tooth: (abscissa of the barycenter; ordinate of the barycenter; applicate of the barycenter of the barycenter). The three-dimensional reference frame is determined relative to the patient under consideration, for example, relative to the center of the patient's oral cavity. The three-dimensional reference frame is preferably fixed relative to the patient under consideration or relative to a part of the patient under consideration. In particular, the origin of the reference frame can be the center of the oral cavity.

[0057] In particular, the three-dimensional reference frame is independent of the position and orientation of the image acquisition device when acquiring an image.

[0058] One or more values of an item of spatial information can always be zero. For example, if the spatial attribute is used to determine the abscissa of a point of interest in a dental scene along the X-axis of the three-dimensional reference frame, only the value of this abscissa is non-zero. Alternatively, the value is not always zero.

[0059] An item of spatial information cannot be inferred solely from the observation of an image where the acquisition conditions are unknown. In particular, for example, it cannot be inferred solely from the observation of a single acquired image by an image acquisition device whose orientation and distance relative to the dental body under consideration are unknown, such as a mobile phone not fixed on a support at a predetermined distance from the patient, for example, a mobile phone supported by the patient, or a mobile phone fixed on such a support but whose orientation can be changed.

[0060] In this way, the image provides information on the "surface" in the plane of the image, for example, the position of specific points of teeth drawn on the image in the two-dimensional reference frame of the image. An item of spatial information provides depth information relative to the plane of the image. In the example of the position of specific points of teeth drawn on the image, the spatial information provides the coordinates of this point that enable this point to be positioned not only on the image but also in the depth direction perpendicular to the plane of the image.

[0061] The historical spatial information can be relative, for example, when the historical spatial information represents the distance between two points to be noted for one tooth, or the distance between a point to be noted for one tooth and a point to be noted for another tooth, for example, a point to be noted for another adjacent tooth.

[0062] An item of spatial information is the occurrence of a spatial attribute. For example, (2; 3; 1.5) can define the position of the center of gravity of tooth No. 14 of patient "Mr Martin". This is referred to as "updated" or "historical" according to whether it is associated with the latest patient or the history.

[0063] “ before of” and “ after of” are temporally consecutive instants.

[0064] "Tooth entity" is a generic term and designates, for example, tooth No. 14. "Latest tooth entity" and "historical tooth entity" are respectively the occurrences of the tooth entity in the latest patient and the historical patient. For example, tooth No. 14 of patient "Mr Martin" can be the latest tooth entity.

[0065] "Vertical", "horizontal", "right", "left", "horizontal", "in front of" or "frontal", "rear", "up", "down" are refers to a patient standing upright in the vertical direction .

[0066] The information is one latestIt is "static" or "dynamic" according to whether it is the analysis result at a moment or the analysis results at a plurality of consecutive latest moments.

[0067] "Comprising", "including", or "having" should be understood in a non-limiting manner unless otherwise specified.

[0068] Further features and advantages of the present invention will become more clearly apparent upon reading the following detailed description of the invention and referring to the accompanying drawings.

Brief Description of the Drawings

[0069] [Figure 1] FIG. 1 schematically shows a method of repeatedly performing a method for training a neural network according to the present invention, each time obtaining a set of neural networks specialized for items of dental body and / or spatial information. [Figure 2] FIG. 2 schematically shows a method of repeatedly performing an analysis method according to the present invention with different specialized neural networks each time. [Figure 3] FIG. 3 schematically shows various steps of a method for determining the amount of movement of the latest dental body. [Figure 4] FIG. 4 schematically shows that a method for determining the amount of movement according to the present invention is repeatedly performed at two different moments using different specialized neural networks each time to obtain complex and accurate dynamic information. [Figure 5] FIG. 5 is an example of a set of latest images after performing a process for separating the dental region. [Figure 6] FIG. 6 is a graph showing the accuracy of the latest item of spatial information obtained according to the present invention. [Figure 7] FIG. 7 shows the numbers of teeth used in the dental field. [Figure 8] FIG. 8 is a graphical representation of the conformity index.

[0070] Throughout the various figures, the same reference numerals are used to denote similar or identical objects.

DETAILED DESCRIPTION OF THE INVENTION

[0071] A training method according to the present invention is shown in FIG. 1.

[0072] neural network In step a), the neural network is preferably specialized for image classification.

[0073] Preferably, the neural network is a convolutional neural network (CNN), and is preferably selected from the following neural networks: - AlexNet (2012); - ZF Net (2013); - VGG Net (2014); - GoogleNet (2015); - Microsoft ResNet (2015); - Caffe: BAIR Reference CaffeNet, BAIR AlexNet; - Torch; - VGG_CNN_S, VGG_CNN_M, VGG_CNN_M_2048, VGG_CNN_M_1024, VGG_CNN_M_128, VGG_CNN_F, VGG ILSVRC-2014 16-layer, VGG ILSVRC-2014 19-layer, Network-in-Network (Imagenet & CIFAR-10); - Google: Inception (V3, V4).

[0074] Preferably, a "squeeze-and-excitation" (SE) processing block described by Jie Hu et al., "Squeeze-and-Excitation Networks", arXiv:1709.01507v4 [cs.CV], 16 May 2019 is added to the CNN convolutional operator. More preferably, the neural network is of the VGG type having an SE block.

[0075] For operation, conventionally, an orientation neural network has to be trained using a learning process called "deep learning" based on a historical learning database adapted to the desired function.

[0076] historical learning database Training a neural network is a process well-known to those skilled in the art. It includes presenting a historical learning database containing a plurality of historical records each including input data and output data to the neural network.

[0077] Accordingly, the neural network learns to "match", i.e., connect, the input data and the output data with each other.

[0078] To learn to evaluate, from a set of latest images, the latest items of spatial information regarding the latest dental bodies depicted on these images, e.g., the spatial information regarding the tooth configuration depicted on these images, the historical learning database preferably consists of a set of historical records each including: - A set of "historical" images depicting the "historical" dental bodies of a "historical" patient, preferably photographs depicting at least one tooth of this patient; - A historical descriptor of the set of historical images, where the descriptor includes an item of historical spatial information regarding the historical dental bodies depicted in the historical images.

[0079] Preferably, the history learning database includes 1,000 hyper , 5,000 hyper , preferably 10,000 hyper , preferably 30,000 hyper , preferably 50,000 hyper , preferably 100,000 hyper history records. The larger the number of records, the better the analysis ability of the neural network. The number of history records is conventionally less than 10,000,000 or less than 1,000,000.

[0080] The history records are each associated with an individual historical patient.

[0081] historical image Set of historical images all of , regardless of the historical records considered, the same number of historical images include are included. This number is preferably 1, 2, 4, 5 larger than , and / or less than 100, 50, 20, 15 or 10, preferably 5 - 15.

[0082] In one embodiment, this number is 3, where the three historical images have different angles.

[0083] The historical image is an image acquisition device, preferably selected from a mobile phone, a "connected" camera, a "smart" watch, a tablet, or a personal computer, a fixed computer or a portable computer, and is acquired by the above image acquisition device equipped with an image acquisition system, such as a web camera or a camera.

[0084] The historical image is preferably a photo taken with a mobile phone.

[0085] Preferably, each historical image is a photograph or an image extracted from a film. Preferably, it is in color, preferably true color. Preferably, it depicts a dental scene substantially as seen by the operator of the image acquisition device, and in particular in the same color.

[0086] The historical image is preferably an extraoral image, i.e., the device for acquiring these images is not introduced into the oral cavity of the historical patient.

[0087] More preferably, the device for acquiring the historical image is 5 cm hyper , 8 cm hyper or 10 cm hyper away from the mouth of the historical patient, thereby avoiding condensation of water vapor on the optical system of the image acquisition device and facilitating focusing. Moreover, preferably, the image acquisition device, in particular a mobile phone, is not equipped with any specific optical system for acquiring the historical image, which is particularly possible due to the separation between the image acquisition device and the mouth of the historical patient during acquisition.

[0088] In one embodiment, the historical patient wears a dental retractor to better expose the teeth of the historical patient. The retractor can have the characteristics of a conventional retractor. Preferably, it is provided with a rim extending around the retractor opening and is arranged such that the lip of the historical patient can rest thereon while exposing the teeth of the historical patient through the retractor opening. Preferably, the retractor is provided with brackets for separating the cheeks such that the device for acquiring the historical image can acquire a photograph of the vestibular surfaces of the teeth, such as molars, located at the bottom of the oral cavity through the retractor opening.

[0089] In a preferred embodiment, a dental retractor is not used. In fact, tests show that photographs taken without a retractor are generally sufficient to carry out the method according to the invention. Of course, if necessary, the patient in question may have to move the cheek or lip away, for example with a finger or a spoon.

[0090] All historical images of the historical record depict at least partially the same dental body of the historical patient associated with the record, for example the incisors of the historical patient.

[0091] Historical images can in particular depict one or more teeth. Preferably, the historical images can depict several teeth and at least part of the gums, or the lip or nose of the historical patient.

[0092] The historical images of the historical record all depict the same historical dental body, but preferably at different angles, i.e. the historical images are taken at different orientations of the image acquisition device relative to the oral cavity of the historical patient. For example, the historical record can include six historical images each depicting the same tooth as seen, for example, as a "front view", "right front view", "right view", "left front view", "left view", and "bottom view".

[0093] The angles of the historical images of the historical record are preferably substantially the same for all historical records relating to the same historical dental body, for example the same type of tooth, for example all historical records relating to incisors, for example for the same tooth number, for example all historical records relating to the upper right incisor, regardless of the historical record being considered.

[0094] The historical record can include one or more historical images at the same angle.

[0095] When obtaining the historical image and the latest image of the method according to the present invention, it can be defined with an accuracy level where the angle is not restricted. However, the tests have shown that the angle does not need to be defined very precisely. Therefore, advantageously, the acquisition of these images does not require prior training of the operator of the image acquisition device. The historical image, like the latest image, can be acquired with minimal care, for example, using a simple mobile phone.

[0096] In a preferred embodiment, the angle is defined very generally. For example, the angle can be selected from one group of potential angles consisting of the "front view", "right side view", "right front view", "left side view", "left front view", "bottom view", "front bottom view", "top view", "front top view" of the angle.

[0097] The angle can be defined relative to a "natural" reference frame, that is, as a function of the way the patient perceives the image acquisition device. In this reference frame, the angle is selected from one group of potential angles consisting of, for example, the following angles. On the occlusal surface, - The case where the optical axis of the image acquisition device substantially coincides with a straight line at the intersection of the occlusal surface and the median sagittal plane is the "front view"; - The "right side view" when the optical axis of the image acquisition device is substantially within the occlusal surface and perpendicular to the median sagittal plane and the image acquisition device is on the right side of the patient; - The "left side view" when the optical axis of the image acquisition device is substantially within the occlusal surface and perpendicular to the median sagittal plane and the image acquisition device is on the left side of the patient; In the median sagittal plane, - The "top view" when the optical axis of the image acquisition device is substantially in the median sagittal plane and perpendicular to the occlusal surface and the image acquisition device is above the patient; - The "bottom view" when the optical axis of the image acquisition device is substantially in the median sagittal plane and perpendicular to the occlusal surface and the image acquisition device is below the patient.

[0098] The angle can be more precisely defined. In particular, for each of the above angles, the optical axis of the image acquisition device is in the occlusal plane or the median sagittal plane. For example, it is possible to add the following angles: - In one of the two planes of "right front" and "left front" that are inclined at 45° relative to the median sagittal plane and include the straight line at the intersection of the occlusal plane and the median sagittal plane, right front and right view, right front and left view, left front and right view, and left front and left view; or, - In one of the "upper occlusal surface" and "lower occlusal surface" that are inclined at 45° relative to the occlusal plane and include the straight line passing through the center of the oral cavity at the intersection of the occlusal plane and the plane parallel to the frontal plane, upper occlusal surface and front view; or, - When the optical axis is at the intersection of a first plane selected from one of the two planes of "right front" and "left front" and a second plane selected from one of the two planes of "upper occlusal surface" and "lower occlusal surface".

[0099] Preferably, the angle is selected from a group of potential angles consisting of the angles listed above.

[0100] Preferably, the angle is nevertheless determined as a function of the dental body. For example, if the dental body is one tooth or a group of teeth, the angle is preferably fixed as a function of the number of the one tooth or the group of teeth. The configuration of the mouth does not necessarily allow the same angle to be used for all teeth.

[0101] However, it can be the case that the same angle is used for two different dental bodies, for example incisors and canines.

[0102] Precise positioning of the acquisition device when acquiring historical images is not necessary. Thus, the historical images can be acquired at different distances from the mouth. Tests show that historical dental entities, such as teeth, can be depicted at different scales without substantially affecting in a significant manner the performance capabilities of a neural network that is trained depending on the historical images considered and / or depending on the historical records considered.

[0103] However, preferably, the image acquisition device is fixed on a support that is placed and supported on the body of the historical patient, and preferably introduced into the mouth of the historical patient. If the support is a rigid body, it advantageously imposes a predetermined distance between the image acquisition device and the mouth of the historical patient. Thereby, the performance capabilities of the neural network are improved.

[0104] Preferably, the support supports a conventional dental retractor. Such a dental retractor is conventionally provided with a rim extending around the retractor opening, and the lip of the historical patient is arranged so as to be able to rest thereon while exposing the patient's teeth through the retractor opening.

[0105] Preferably, an image capture device as described in European Patent Application No. 17,306,361.1 filed on October 10, 2017 is used.

[0106] Furthermore, preferably, the historical image is cropped before being incorporated into the historical record. "Cropping" or "re-cutting" is a conventional operation that includes trimming the image to separate the relevant part therefrom and then normalizing the dimensions. Preferably, the historical image is trimmed to separate the historical tooth body, i.e., to depict substantially only the historical tooth body. Re-cutting can be performed manually or, as described below, by a computer, in particular by a neural network trained for this purpose. Re-cutting the historical image significantly improves the performance capabilities of the trained neural network.

[0107] The test also shows that, as indicated above, the angle does not need to be set precisely.

[0108] historical dental body and specialization The tooth body can in particular be a single tooth, or a set of multiple teeth, or a dental arch. Preferably, the tooth body is selected to limit the diversity of its shape among various historical records. The tooth body is preferably a tooth of a specific type of tooth or a tooth with a specific number.

[0109] The "historical" tooth body refers to the tooth body in a specific case of the historical patient under consideration. In other words, in the historical learning database, all historical tooth bodies are specific occurrence examples of the tooth bodies associated with the learning database.

[0110] Preferably, the learning database, and thus the neural network, is specialized only for teeth with a specific number. For example, they are specialized for the upper right incisors. Then, all historical images of a certain historical record all depict the same historical tooth, the historical spatial information of this historical record is related to this historical tooth, and all historical teeth in the learning database use the same number. In this way, specializing the neural network for the tooth number significantly improves its efficiency.

[0111] In this way, several neural networks specialized in this manner are preferably trained, where each neural network is trained using a historical learning database dedicated to the type of tooth or tooth number. Thus, advantageously, the analysis of the dental situation can implement a neural network specialized for each tooth of the latest patient and for the corresponding tooth type or tooth number.

[0112] historical descriptor A set of historical descriptions of the historical images includes a set of items of historical space information, i.e., a set of values for variables of spatial attributes, where these values relate to the historical dental entity depicted in the historical image. By definition, the spatial attributes include at least three variables corresponding to the three dimensions of space. Therefore, the spatial information is a set of at least three values for at least three individual variables of the spatial attributes.

[0113] The spatial attributes are the same not only for all the historical images of the set but also for all the historical records.

[0114] The spatial attributes in particular - one or more positions in the space of one or more dental entities and / or in the space of one or more parts of a dental entity; and / or - one or more orientation directions in the space of one or more dental entities and / or in the space of one or more parts of a dental entity; and / or - one or more orientation courses in the space of one or more dental entities and / or in the space of one or more parts of a dental entity can be defined as:

[0115] One such part of one dental entity is, for example, - one or more points, for example, when the dental entity is a tooth, the contact point with an adjacent tooth or the center of gravity of the tooth; and / or - one or more rows, for example, when the dental entity is a tooth, the separation row between an adjacent tooth or an edge of a cusp of the tooth; and / or - one or more surfaces of this tooth body can be.

[0116] In particular, the spatial attribute (x A , y A , z A , x B , y B and z B) where, when defined in an ordered manner as the positions for two points (x A , y A and z A) and (x B , y B and z B ), with the position of point A before the position of point B), it indirectly defines the azimuth direction, azimuth course, and distance. If it defines the positions for three points and these positions are ordered, it indirectly defines three azimuth directions, and thus the angles between pairs of these directions, the azimuth courses along each of these directions, and three distances.

[0117] For example, when defining the centroid position of the first tooth and the centroid position of the tooth immediately adjacent to the left of the first tooth, the spatial attribute not only directly defines the positions of these centroids but also indirectly defines the azimuth direction of the straight line connecting these points, the azimuth course from the first centroid to the second centroid, and the distance between these centroids.

[0118] The spatial attribute can define the absolute position in a three-dimensional reference frame that is relatively fixed to the patient, for example, where its origin is at the center of the patient's oral cavity under consideration, the horizontal axis is horizontal and oriented towards the front, the vertical axis is horizontal and oriented towards the right, and the applicate axis is vertical and oriented upwards. It can also define the vector between two points, that is, the relative position of one point relative to another point. For example, the spatial attribute is (x B - x A , y B - y A , z B - z A) That is, it is possible to provide the relative position of point B relative to point A.

[0119] Therefore, the spatial information can define an absolute position or a relative position, and / or an orientation direction and / or an orientation course, and / or a distance in a specific dental situation.

[0120] Determining the historical spatial information of the historical record causes no problems. It can be determined by any means, for example, manually or by computer, and in particular, it can be determined by taking measurements from the historical patient, or from a plaster model of the teeth of the historical patient, or from a digital three-dimensional model of the dental arch supporting the tooth under consideration.

[0121] Preferably, the spatial attributes define fewer than 30 variables, preferably fewer than 20 variables, preferably fewer than 10 variables, preferably fewer than 5 variables, preferably fewer than 4 variables. The effectiveness of the trained neural network is improved.

[0122] In one embodiment, the spatial attribute is of the dental body, preferably of a tooth having a specific number one , preferably 1 hyper , preferably 2 hyper , and / or less than 5, preferably less than 4, notable points defines the position in space for. By limiting the historical spatial information related to the teeth to a limited number of points, the efficiency of the neural network is considerably improved.

[0123] In step b), the neural network is trained using the historical learning database by sequentially presenting the historical record, more specifically a set of historical images as input and historical spatial information as output.

[0124] Therefore, it learns to provide, as an output, corresponding items of spatial information from a set of images similar to a set of historical images presented as an input. In particular, after being trained in this way, the neural network can provide "up-to-date" spatial information regarding the "up-to-date" dental structure of the "up-to-date" patient at the "up-to-date" moment. Therefore, the up-to-date spatial information can be used alone or in combination with other information to analyze the up-to-date dental situation of the up-to-date patient. The analysis method according to the present invention includes steps a) to c), as shown in FIG. 2.

[0125] Therefore, the neural network thus trained can determine the most up-to-date items of spatial information for a set of the most up-to-date images taken and presented as an input in any up-to-date patient, which are compatible with the set of historical images of the historical learning database.

[0126] Step a) includes performing the above steps A) and B).

[0127] In step b), a set of the most up-to-date images depicting the most up-to-date dental structure of the most up-to-date patient, preferably the most up-to-date teeth, is acquired by the image acquisition device at the most up-to-date moment.

[0128] The most up-to-date moment can be - Irrespective of any dental correction treatment, for example, enabling each patient to monitor their dental situation at any time with their mobile phone; - During active dental correction treatment; - After passive dental correction treatment, particularly during passive dental correction treatment and can be.

[0129] In particular, the analysis method can be carried out during active dental correction treatment to monitor the progress of the active dental correction treatment, where the most up-to-date moment is preferably less than 3 months, less than 2 months, and / or 1 week after wearing an active dental correction device intended to correct the position of the teeth of the most up-to-date patient, such as an orthodontic aligner (i.e., an "aligner") or an orthodontic arch.hyper , preferably two weeks hyper , later.

[0130] The analysis method can also be performed after orthodontic treatment to confirm that the position of the teeth does not progress unfavorably (“recur”). Next, the most recent moment is preferably less than three months, less than two months, and / or one week after the passive orthodontic appliance (referred to as a “retainer”) intended to hold the teeth in a certain position is worn after the completion of the active orthodontic treatment hyper , preferably two weeks hyper , later.

[0131] The most recent image is preferably extraoral.

[0132] Preferably, the most recent image is an image extracted from a photograph or film. They are preferably in color, preferably true color. Preferably, they substantially depict the dental arch as seen by the operator of the image acquisition device.

[0133] In one embodiment, the most recent patient wears a dental retractor to better expose the teeth. However, preferably, a dental retractor is not used. Of course, if necessary, the most recent patient may need to spatially separate the cheek or lip, for example, using a finger or a spoon or other appliance suitable for this purpose.

[0134] The most recent image can depict one or more teeth. Preferably, the most recent image depicts at least a portion of the patient's gums or even the lips or nose.

[0135] All of the most recent images must be compatible with the neural network, i.e., they must be compatible with it. In other words, all of the most recent images must have been what could have been used for the history record.

[0136] The number of the latest images is preferably the same as the number of the historical images in the historical records.

[0137] The latest image must depict the same dental body as the historical image, for example, tooth No. 14.

[0138] The angle of the latest image is preferably the same as or close to the angle of the historical images of any historical record.

[0139] Generally, the latest image must be similar to the historical images used to train the neural network. For example, if these historical images are extraoral photographs generally taken by the historical patient himself at a variable distance, for example, at a distance of 10 to 50 cm from his own mouth, at a general angle (for example, as a "front view" or as a "right view"), preferably, the latest image is also a photograph taken under these shooting conditions. If the historical image depicts a figure taken with a dental retractor, it is preferably the same for the latest image.

[0140] The latest image is acquired by an image acquisition device, and the image acquisition device can preferably be the same as or different from the image acquisition device used to acquire the historical image, and is preferably an image acquisition device of the same type.

[0141] It is preferably selected from a mobile phone, a "connected" camera, a "smart" watch, a tablet, or a personal computer, a fixed computer or a portable computer, including an image acquisition system, such as a web camera or a camera. Preferably, the image acquisition device is a mobile phone. Preferably, the image acquisition device, especially a mobile phone, is not equipped with any specific optical system for acquiring the latest image.

[0142] More preferably, in order to acquire the latest image, the device for acquiring the latest image is from 5 cm hyper 8 cm hyper or 10 cmhyper Even and / or less than 50 cm, separated. Advantageously, this distance need not be set precisely.

[0143] However, preferably, similar to the history image, the latest image is "cropped" (or "recut") before being incorporated into the set of the latest images that are input into a neural network trained using the history learning database. Preferably, the latest image is trimmed to separate the latest tooth body, i.e., to depict only the latest tooth body substantially. Recutting can be performed manually or, preferably, by a computer, particularly by a neural network trained for this purpose.

[0144] In particular, the neural network can be trained to identify the tooth body on the image, for example, to identify the tooth region. Such a neural network for "identifying the tooth body" is described below in this specification. To crop the latest image, it is only necessary for the latest tooth body to be identified by this neural network, and the smallest rectangle that can contain the tooth body identified on this image needs to be defined to retain only the interior of this rectangle to define the cut, and then, to define the latest image to be incorporated into the set of the latest images, the dimensions of the cut need to be normalized. If the aspect ratio of the rectangle is substantially always the same regardless of the latest image, it is likely that the conditions for acquiring the image are similar, which is advantageous. Normalizing the dimensions of the cut includes adjusting these dimensions so that all the latest images have the same dimensions. Preferably, the operation of recutting the history image is the same as that of the latest image, and thus, the dimensions and the number of pixels of these images are the same or substantially identical.

[0145] Recutting the history image and the latest image significantly improves the performance of the trained neural network.

[0146] The image acquisition device is preferably used by an operator who is the latest patient or a close relative of the latest patient, but this can be any other person, in particular a dentist or orthodontist or caregiver. Preferably, the latest image is acquired by the latest patient.

[0147] Preferably, the latest image is acquired without using a support and without fixing the image acquisition device, in particular without a tripod, being supported on the ground.

[0148] However, in one embodiment, the image acquisition device is attached to a support that is placed and supported on the body of the historical patient and is preferably partially introduced into the mouth of the historical patient. If the support is a rigid body, it advantageously imposes a predetermined distance between the image acquisition device and the mouth of the latest patient. Thereby, the performance capabilities of the neural network are improved.

[0149] Preferably, the support supports a conventional dental retractor. Such a dental retractor is conventionally provided with a rim extending around the retractor opening and is arranged such that the lip of the latest patient can rest thereon while exposing the teeth of the latest patient through the retractor opening.

[0150] Preferably, an image capture device as described in European Patent Application No. 17,306,361.1 filed on October 10, 2017 is used.

[0151] the composition of the set of the latest images being acquired More preferably, the operator is guided to orient the image acquisition device at a predetermined angle and / or preferably to take a predetermined number of images at various angles and / or to orient the image acquisition device at the required angle, preferably in real time, during step b).

[0152] Therefore, in order to provide on-board control during step b), i.e., to confirm that the number and / or angle and / or quality of the latest image is satisfactory, it is preferable that an application be loaded into the image acquisition device.

[0153] In particular, the application can implement fool-proofing means that facilitate the approximate positioning of the image acquisition device relative to the latest patient before acquiring the latest image.

[0154] The fool-proofing means can in particular include displaying on the screen of the image acquisition device and criteria, such as superposition criteria, that the operator must match with the portion of the latest patient displayed on this screen, for example, the profile of a tooth, gum, lip or face.

[0155] The criteria can be, for example, geometric shapes, such as one point, one or more lines, such as parallel lines, stars, circles, ellipses, regular polygons, in particular squares, rectangles or rhombuses, or any combination of one or more of these shapes. The one or more criteria are preferably "fixed" on the screen, i.e., they do not move on the screen when the image acquisition device is moving.

[0156] The criteria can consist, for example, of a horizontal line intended to match the general direction of the on-screen depiction of the horizontal joint between the upper and lower teeth when the teeth are clamped together by the latest patient, and / or a vertical line intended to match the depiction of the vertical joint between the two upper incisors. The criteria can be formed, for example, by two circles that should be placed above the on-screen depiction of the eyes of the latest patient. The criteria can be formed, for example, by an ellipse that should be placed around the on-screen depiction of the mouth or face of the latest patient.

[0157] "A part of the patient" can also be supported by a support worn by the latest patient, for example, by a dental retractor or by a part bitten by the latest patient.

[0158] The application can also assist the operator in changing the angle of the image acquisition device by announcing or displaying on the screen messages such as "Please take a photo of the right side", "Higher", "Lower", or by emitting a series of beeping sounds whose frequency increases as the orientation of the image acquisition device is improved. For this purpose, the image displayed on the screen of the image acquisition device must be analyzed in real time to determine, in particular, whether a dental body is depicted, and preferably, to check whether the angle is appropriate.

[0159] Algorithms for detecting objects in an image are well known to those skilled in the art and can be used to search for dental bodies in the displayed image. Preferably, for example, a neural network is used to identify dental bodies, preferably selected from among the following "Object Detection Networks": - R-CNN (2013); - SSD (Single Shot MultiBox Detector: Object Detection Network), Faster R-CNN (Faster Region-based Convolutional Network method: Object Detection Network); - Faster R-CNN (2015); - SSD (2015); - RCF (Richer Convolutional Features for Edge Detection) (2017).

[0160] Training a neural network to detect dental bodies in an image, for example, a tooth with a determined number, is not a problem for those skilled in the art. In particular, the neural network is provided with an image as input and, as output, information related to the presence or absence of dental bodies.

[0161] The following article particularly deals with detection and segmentation: https: / / arxiv.org / pdf / 1405.0312.pdf and https: / / arxiv.org / pdf / 1703.06870.pdf

[0162] In one embodiment, the neural network is trained using a training database consisting of a set of records, 1,000 hyper , preferably 10,000 hyper , each record comprising: - an image including a zone depicting a dental body, for example, at least one dental zone related to a tooth with a determined number; - a description of the image identifying the zone on the image where the dental body is depicted .

[0163] During training, each image is supplied to the neural network as input, while the associated descriptor is supplied as output from the neural network.

[0164] Thus, upon completion of the training, the neural network can determine the zone, for example, the dental zone, where the dental body is depicted in the image supplied to the neural network as input.

[0165] Figure 5 represents a set of up-to-date images taken at different angles after being processed by a neural network trained in this way.

[0166] This angle can also be identified by a neural network trained for this purpose. The neural network is preferably selected from among CNNs, where the final layer of the neural network operates a regression.

[0167] The neural network consists of a set of 1,000 records hyper preferably 10,000 records hyper and is trained using a learning database consisting of a set of records, where each record is - an image including a zone depicting a dental body, for example, including at least one tooth zone related to a tooth having a determined number; - a descriptor of the image identifying the angle on the image including.

[0168] During training, each image is supplied as input to the neural network, while the associated descriptor is supplied as output from the neural network.

[0169] Therefore, upon completion of the training, the neural network can determine the angle of the image supplied to the neural network as input.

[0170] Preferably, the application defines a set of predetermined angles and, for each predetermined angle, the number of latest images to be acquired. When the application is launched, in step b), the application preferably, in real time, - analyzes the image displayed on the screen of the image acquisition device at the current moment, i.e., the "current image", to determine whether a dental body is depicted, preferably the angle of the image acquisition device, i.e., the "current angle"; - when a dental body is depicted, preferably when the current angle meets the current requirements, i.e., when it is still necessary to acquire the latest image at the current angle at the current moment, trigger the acquisition by the operator or automatically; - Otherwise, preferably, notify the operator to correct the angle of the image acquisition device, or end in step b) when there is no need to acquire additional up-to-date images. Perform the steps of.

[0171] In a preferred embodiment, the acquisition is triggered automatically, i.e., without operator action, as soon as the displayed image depicts the dental body and the angle is approved by the image acquisition device.

[0172] To guide the operator, a written and / or voice message can be issued by the image acquisition device. For example, the image acquisition device can announce "Please take a front photo" and issue a signal to notify the operator that the orientation is acceptable or, conversely, that the image acquisition device needs to retake the photo.

[0173] The end of the acquisition process may be announced orally by the image acquisition device or by a display on the screen.

[0174] Guiding the operator when acquiring the up-to-date image advantageously enables a set of up-to-date images that are immediately suitable for a trained neural network to be formed.

[0175] the composition of the set of the latest images after acquisition The set of up-to-date images can also be formed partially or entirely by selecting a sufficient number of up-to-date images that depict the up-to-date dental body at the desired angle after acquiring the up-to-date images.

[0176] Determining whether the up-to-date image can belong to a set of up-to-date images specialized for the up-to-date dental body, e.g., for the tooth number, includes checking whether this up-to-date image depicts this up-to-date dental body, e.g., the tooth with this number, and preferably checking whether the angle is appropriate.

[0177] This selection can be made manually, especially when the desired angle is approximate. For example, it is easy to take a front view photo, a right side photo of the open mouth, a left side photo of the open mouth, a right side photo of the closed mouth, and a left side photo of the closed mouth.

[0178] The latest image can also be selected by a computer. A neural network, for example, the neural network described above for obtaining images, can be used. Identification of the dental body and / or determination of the angle can also be performed using conventional image analysis, but such analysis is time-consuming.

[0179] When the latest image depicts the dental body at a certain angle and a set of the latest images still requires additional latest images for this angle, the latest image is added to the set.

[0180] If the latest image depicts the dental body at the desired angle but it is redundant, for example, if it is of higher quality because it depicts more surface area of the dental body than the other latest images, it can replace other latest images present in the specialized set.

[0181] In step c), the set of the latest images formed in step b) is input into the neural network trained in step a).

[0182] In response, the neural network provides the latest spatial information regarding the latest dental body.

[0183] multiple executions with different neural networks The neural network is preferably specialized for a limited dental body, for example, teeth having a determined number, and the spatial attributes preferably include a limited number of variables. Next, preferably, the analysis method is executed multiple times at the latest instant by modifying the considered dental body and / or the considered spatial attributes each time.

[0184] Therefore, all of the determined up-to-date spatial information is referred to as "static information". This enables a detailed analysis of the up-to-date dental situation of a patient to be obtained by re-lying on a specialized neural network.

[0185] In a preferred embodiment, in step a), several specialized neural networks are trained, where each neural network is specialized for a dental body and preferably for an individual tooth type or tooth number. Preferably, then, in step b), a number of up-to-date images sufficient to form a set of "specialized" up-to-date images suitable for each of the specialized neural networks are acquired.

[0186] Preferably, all of the up-to-date images are acquired substantially simultaneously. Optionally, the batch of acquired up-to-date images is analyzed to identify one or more specialized sets to which they can belong.

[0187] The static information can be enhanced. For example, for each of the three sets of adjacent teeth consisting of a central tooth, a left tooth, and a right tooth, when the central tooth is between the left and right teeth, the analysis method can be implemented to determine the position of the center of gravity of the teeth in a reference frame fixed relative to the dental arch supporting these teeth. One set of these three positions is the static information. To enhance this static information, it is possible to determine the angle formed by two straight line segments passing through the centers of gravity of the left and right teeth, respectively, with the center of gravity of the central tooth as a common origin, from the three positions. On the other hand, the distance between the center of gravity of the central tooth and the center of gravity of the left tooth or the right tooth can also be determined.

[0188] The analysis method can be executed multiple times by changing the considered spatial attributes each time. For example, the analysis method can be implemented to determine the position of the contact point between a tooth and a first adjacent tooth and then to determine the position of the contact point between the tooth and a second adjacent tooth. One set of the coordinates defining these two positions is the static information.

[0189] The parsing method is preferably executed multiple times by modifying the considered tooth body or the considered spatial attribute each time. For example, three sets of adjacent teeth consisting of the central tooth, the left tooth, and the right tooth are considered, where the central tooth is between the left and right teeth, and the parsing method is in a reference frame fixed relative to the dental arch supporting these teeth, determining the position of the center of gravity of the left tooth, then the position of the center of gravity of the right tooth, and then successively implementing to determine the position of the contact point of the central tooth with the left tooth and then the position of the contact point of the central tooth with the right tooth.

[0190] To enhance the static information thus obtained, for example, the angle between two planes perpendicular to the straight line connecting the two centers of gravity of the left and right teeth and the straight lines connecting the two contact points of the central tooth with the left and right teeth respectively can be measured. Therefore, this angle provides information related to the orientation of the central tooth relative to the left and right teeth.

[0191] use of static information The static information can be used to evaluate the current dental situation of the patient.

[0192] In particular, it can be used to evaluate whether the current tooth body is in a position belonging to a predetermined spatial region, in particular, for example, whether it belongs to a region defining a set of predefined positions considered acceptable. For example, such a region can be defined around a tooth or a point to be noted on a tooth, and thus, when the tooth is even partially away from this region or when the point to be noted is away from this region, the dental situation is considered abnormal. Such a region can also be defined around an orthodontic appliance or a point to be noted on an orthodontic appliance, and thus, when the orthodontic appliance is even partially away from this region or when the point to be noted is away from this region, the dental situation is considered abnormal.

[0193] The static information can be used to evaluate whether the latest dental body belongs to a set of orientations belonging to a given orientation, in particular a set of orientations considered acceptable, of a given orientation. In particular, the orientation of a tooth can be defined by the angle formed by two straight lines passing through a point of interest of the tooth, for example the center of gravity of the tooth, where the first of the two straight lines passes through a point of interest, for example the center of gravity, of the right side of the tooth, preferably the tooth adjacent to the tooth, and the second of the two straight lines passes through a point of interest, for example the center of gravity, of the left side of the tooth, preferably the tooth adjacent to the tooth.

[0194] The static information can also be used to evaluate the distance between a point of interest of the latest dental body and another point of interest of the dental arch supporting the latest dental body, for example, when the distance between the center of gravity of a tooth and the center of gravity of the tooth adjacent to the tooth belongs to a given distance range, in particular a set of distances that can be considered acceptable.

[0195] The limitation of the region defined for an item of static information or the limitation of the tolerance range defined for an item of static information, i.e., the "static constraint", is preferably defined by a dental practitioner.

[0196] The static information can in particular be used to determine whether the latest dental situation of a patient is abnormal. For example, to detect recurrence, the static constraint can be adapted to the dental situation at the end of an orthodontic treatment and can have an optional tolerance margin.

[0197] The static information can be used to determine the position of the first dental arch of a patient with respect to the second dental arch of the patient, in particular to detect and / or evaluate the presence of a vertical or horizontal overhang, in particular when the overhang is abnormal.

[0198] When monitoring orthodontic treatment, the static constraints can correspond to the dental situation expected at the latest moment or at the end of the orthodontic treatment, and can optionally have a tolerance margin.

[0199] If no treatment is performed, the static constraints can define a set of dental situations considered normal.

[0200] In one embodiment, the static constraints do not depend on the latest patient, i.e., they are applicable to any patient in the patient group. Therefore, they form a standard, i.e., a "standard set-up".

[0201] Preferably, the standard is specific to a disease state and / or the type of orthodontic treatment and / or a group of patients sharing common characteristics, for example, a group of patients belonging to the same age group and / or the same gender. The standard can in particular determine the arcade shape.

[0202] The standard can in particular define the dental situation at the end of treatment or at the latest moment.

[0203] In particular, if the static constraints are not respected, for example, if the positions, orientations and / or distances determined from the static information are not allowed, a message can be sent to the latest patient and / or the dental healthcare provider to inform them.

[0204] The static information can be displayed in the form of a graph.

[0205] For example, the static information can be presented on a computer screen or on a mobile phone screen.

[0206] For example, the graph can represent the teeth in a color that depends on a conformity index representing the conformity of the position of each tooth with a predefined position, for example, the predefined position at the latest moment. For example, the darker the tooth, the farther its position is from the predefined position.

[0207] This static information is in particular - Detecting or evaluating the position or shape of the teeth, and / or the development of the position or shape of the teeth, and / or the speed of development of the position or shape of the teeth: In particular, - During the period before treatment, that is, the period prior to orthodontic treatment; - When no orthodontic treatment is being carried out, in particular, for monitoring tooth eruption, or for detecting the recurrence position or abnormal position of the teeth, or for detecting tooth wear, for example, tooth wear due to grinding, or for monitoring the opening and closing of the space between two teeth or three or more teeth, in particular, the space between two adjacent teeth, or for monitoring the stability or correction of occlusion; - In the context of orthodontic treatment, in particular, for monitoring the movement of the teeth to a predetermined position, in particular, to an improved position of the teeth, or for monitoring tooth eruption, or for creating a space suitable for attaching, for example, a dental implant, for monitoring the opening and closing of the space between two teeth or three or more teeth, in particular, the space between two adjacent teeth; and / or, - Detecting or evaluating the position or shape of the orthodontic appliance, in particular, the abnormal position or shape of the orthodontic appliance, for example, detachment of the ring or detachment of the dental aligner, and / or the development of the position or shape of the orthodontic appliance, and / or the speed of development of the position or shape of the orthodontic appliance For, in particular, - Optimizing the appointment date with the orthodontist or dentist; and / or, - Evaluating the effectiveness of active orthodontic treatment; and / or, - Measuring the activity of the active orthodontic appliance; and / or, - Measuring the loss of effect of the passive orthodontic appliance; and / or, - Odontology; and / or, - Measuring the development of the shape of a patient's teeth between two dates, for example between two dates before treatment, in particular between two dates separated by an event that is likely to have changed the position and / or shape of at least one tooth, for example between two dates separated by the occurrence of an impact on the tooth, or between two dates separated by the use of a dental device that can have an adverse effect, for example a dental device intended for the treatment of sleep apnea, or between two dates separated by the occurrence of a graft in the patient's oral cavity, in particular a periodontal graft, in particular a gingival graft. It can be used for this purpose.

[0208] When the static information is used to evaluate the development, it can be compared with the defined situation at the moment before the latest moment without relying on the method according to the invention. For example, when the static information is used to evaluate the development of the position or shape of a tooth, it can be compared with this predefined tooth position or shape without performing steps a) to c), where the position or shape is predefined, for example, at the start of treatment or at an intermediate moment of treatment.

[0209] multiple executions at different latest moments: amount of movement The analysis method according to the invention can also be carried out one or more times at the latest moments of different " before of" and " after of", as shown in Figure 3.

[0210] before The " previous " latest spatial information (i.e., static information) obtained at the latest moment of after is compared with the " after of" latest spatial information (i.e., together with the static information respectively) obtained at the latest moment of

[0211] The term "amount of movement" refers to information obtained directly or indirectly as a result of such a comparison.

[0212] The amount of movement can in particular be the movement of the point of interest between two latest instants, or, by dividing this movement by the time interval between these two latest instants, can be the average movement speed between these latest instants.

[0213] Accordingly, the present invention relates to a method for determining the amount of movement of the dental structure of a latest patient, comprising steps 1) to 3), and optionally step 4).

[0214] In step 1), an analysis method according to the present invention is before executed at the latest instant and spatial information regarding the latest dental structure is obtained. before The latest instant can be, for example, less than 3 months, less than 2 months, less than 1 month, less than 1 week, less than 2 days, after the wearing of an active dental orthosis or a passive dental orthosis, such as an aligner, an arch or a retainer.

[0215] The before latest instant can be, for example, less than 3 months, less than 2 months, less than 1 month, less than 1 week, less than 2 days, after the wearing of an active dental orthosis or a passive dental orthosis, such as an aligner, an arch or a retainer.

[0216] The analysis method according to the present invention can be carried out multiple times as a function of the desired before static information, as described above.

[0217] In step 2), the same analysis method according to the present invention is after executed at the latest instant and after spatial information is obtained. Therefore, the after spatial information is related to the same latest dental structure and the same spatial attributes as the analysis method carried out at the before instant.

[0218] The after latest instant is preferably 2 weeks before later, 1 month hyper later, 2 months hyper later or 6 months hyper later than the latest instant of thehyper and / or less than 5 years, less than 3 years or less than 1 year, which is slow.

[0219] In step 2), the same one or more neural networks as in step 1) are used. Therefore, step a) is not necessary.

[0220] When the analysis method according to the present invention is carried out multiple times in step 1), similarly in the case of step 2), the before static information equivalent to the static information of after items are obtained.

[0221] In step 3), before the spatial information of after and the spatial information of

[0222] before the spatial information of after and the spatial information of

[0223] For example, before the spatial information of after and the spatial information of before are the coordinates (x1, y1 and z1) and (x2, y2 and z2) of the center of gravity of the teeth at the latest moments t1 of after and the latest moment t2 of 2 +(y2 - y1) 2 +(z2 - z1) 2 the square root of is such that the distance covered by this center of gravity can be before evaluated between the latest moment t1 of after and the latest moment t2 of

[0224] The result of the comparison can be composed of values of 1 or more. For example, considering a situation where the spatial attribute is (x'; y'; z'; x''; y''; z''), where (x'; y'; z') and (x''; y''; z'') are respectively the positions of two points to be noted of one tooth in a three-dimensional reference frame, for example, an orthonormal reference frame, that is, (Ox; Oy; Oz), the origin is, for example, the center of the dental arch supporting this tooth. previous Spatial information and after If the spatial information of and are respectively expressed as (x1'; y1'; z1'; x1''; y1''; z1'') and (x2'; y2'; z2'; x2''; y2''; z2''), it is as follows: -x1', y1', z1' are before the values of the coordinates x, y, and z of point P' at the latest moment of (for example, in mm); -x2', y2', z2' are after the values of the coordinates x, y, and z of point P'' at the latest moment of (for example, in mm); -x1'', y1'', z1'' are before the values of the coordinates x, y, and z of point P'' at the latest moment of (for example, in mm); -x2'', y2'', z2'' are after the values of the coordinates x, y, and z of point P'' at the latest moment of (for example, in mm).

[0225] Next, the comparison result can be, for example, the distance covered by point P' and the distance covered by point P'' between the latest moment t1 of obtained above and before the latest moment t2 of (the comparison result is two values), or the arithmetic mean of these two distances (the comparison result is one value). after In step 4), preferably, the amount of movement exists, for example, on the screen of the latest personal computer of the patient and / or the dental healthcare provider or on the screen of a mobile phone.

[0226]

[0227] ​As shown in FIG. 4, steps 1) and 2) can be carried out multiple times by modifying the spatial attributes of the latest dental body and / or analysis method each time. For example, they can each be repeated for multiple teeth of the latest patient. In step 3), the latest before spatial information and the latest after spatial information can be compared. The latest before spatial information and after spatial information obtained for different sets of steps 1) and 2) can also be combined to obtain improved information.

[0228] The term "dynamic information" generally refers to all information directly or indirectly resulting from one or more executions of steps 1) to 3), where each execution is at the latest instant t1 of before and the latest instant t2 of after .

[0229] use of dynamic information This dynamic information can be used to evaluate the development of the latest patient's dental situation.

[0230] It can be used to evaluate whether the translational or rotational movement speed of the notable points of the latest dental body is within a range of values that define a set of predefined speeds, for example, a set of speeds that can be considered acceptable.

[0231] This dynamic information can be used to evaluate whether the dynamics of the active dental correction treatment conform to the expected dynamics, that is, whether the teeth move at a speed that follows the dental correction treatment.

[0232] The limits of the tolerance range defined for the dynamic information, that is, the "dynamic constraints", are preferably defined by dental practitioners.

[0233] This dynamic information can be displayed in the form of a graph.

[0234] For example, the dynamic information can be presented on the screen of a computer or on the screen of a mobile phone.

[0235] Preferably, the graph summarizes all the information directly or indirectly resulting from one or more executions of steps 1) to 3), where each execution before at the latest instant t1 of after and at the latest instant t2 of

[0236] For example, in FIG. 8, in the context of an orthodontic treatment experienced by the latest patient, the teeth will be colored as a function of a conformity index indicating that the speed of movement of each tooth conforms to a predetermined speed.

[0237] The use of the dynamic information is generally easier than the use of static information. For example, it is generally easier to determine whether a point of interest on a tooth has moved abnormally than to determine whether the position of this point in space is abnormal.

[0238] For example, to detect recurrence, the dynamic constraint can correspond to a permitted movement margin for the center of gravity of the tooth. To monitor an orthodontic treatment, the dynamic stress can correspond to the movement process of a tooth relative to another tooth (a decrease or an increase in the distance between these teeth) to verify that two teeth move towards or away from each other. The dynamic constraint can also correspond to a threshold for the speed of movement of the orthodontic appliance or for the speed of movement of the points of the teeth on which the orthodontic appliance acts to verify the activity of the orthodontic appliance.

[0239] The dynamic information can also be used to measure the development of the shape of a set of one or more teeth.

[0240] Therefore, the dynamic information is particularly - detecting or evaluating the position or the development of the shape of a tooth, and / or the speed of the position or the development of the shape of a tooth; in particular, - during the period before treatment, i.e., the period prior to orthodontic treatment; - when no orthodontic treatment is being performed, in particular, for monitoring tooth eruption, or for detecting the recurrence or abnormal position of teeth, or for detecting tooth wear, such as tooth wear due to bruxism, or for monitoring the opening and closing of the space between two teeth or three or more teeth, especially between two adjacent teeth, or for monitoring the stability or correction of occlusion; - in the context of orthodontic treatment, in particular, for monitoring the movement of teeth to a predetermined position, especially to an improved position of the teeth, or for monitoring tooth eruption, or for creating a space suitable for attaching, for example, a dental implant, for monitoring the opening and closing of the space between two teeth or three or more teeth, especially between two adjacent teeth; and / or, - detecting or evaluating the position or shape of an orthodontic appliance, in particular, an abnormal position or shape of the orthodontic appliance, such as detachment of a ring or detachment of an orthodontic aligner, and / or the deployment of the position or shape of the orthodontic appliance, and / or the speed of deployment of the position or shape of the orthodontic appliance for the purpose of, in particular, - optimizing the appointment date with an orthodontist or dentist; and / or, - evaluating the effectiveness of an active orthodontic treatment; and / or, - measuring the activity of an active orthodontic appliance; and / or, - measuring the loss of effect of a passive orthodontic appliance; and / or, - dentistry; and / or - measuring the development of the shape of a patient's teeth between two dates, for example, between two dates before treatment, in particular, between two dates separated by an event that is likely to have changed the position and / or shape of at least one tooth, such as between two dates separated by the occurrence of an impact on the tooth, or between two dates separated by the use of a dental device that can cause an adverse effect, such as a dental device intended for the treatment of sleep apnea, or between two dates separated by the occurrence of a graft in the patient's oral cavity, in particular, a periodontal graft, especially a gingival graft can be used for.

[0241] In particular, if the dynamic constraints are not respected, for example, if the amplitude and / or speed and / or direction of the movement of the points to be noted of the teeth determined from the dynamic information are not allowed, the message can be sent to the latest patient and / or the dental healthcare provider in order to inform the latest patient and / or the dental healthcare provider of the message.

[0242] In one embodiment, the static information and / or the dynamic information are used to evaluate whether the objective has been achieved and / or to measure the difference between the latest dental situation of the latest patient at the latest moment and the achievement of the objective.

[0243] The objective is preferably selected from among the following objectives: - The latest patient achieves a class 1 occlusion for the canines; - The latest patient achieves a class 1 occlusion for the molars; - The space in the anterior sector of the latest patient is closed; - The space resulting from the extraction of the teeth of the latest patient is closed; - The latest patient has a normal horizontal overhang, i.e., "overjet", preferably a normal horizontal overhang of 1 to 3 mm; - The latest patient has a normal vertical overhang, i.e., "overbite", preferably a normal vertical overhang of 1 to 3 mm; - The incisors inter-incisal sectors between the upper arcade and the lower arcade of the latest patient are not offset; - The latest patient has no lateral offset of the lower arcade and / or the upper arcade relative to the sagittal plane of the patient's head; - the latest patient does not have a lateral offset of the upper dental arch relative to the lower dental arch; - the orthodontic appliance worn by the latest patient, such as an orthodontic arch and / or an orthodontic aligner and / or an auxiliary appliance, is passive, i.e., no longer acts to change the position of the teeth of the latest patient; - no or only limited tooth movement of the latest patient has been detected between the last two checks of the upper and / or lower dental arches, preferably, no tooth movement of the latest patient has been detected; - all deciduous teeth of the latest patient have fallen out; - absence of lateral open bite (closure of lateral open bite); - absence of posterior open bite; - absence of anterior open bite; - anterior crossbite absence; - posterior crossbite absence; - improvement of crowding; - stabilization of concavity; - closed interproximal space ; - absence of mucosal irregularities.

[0244] Example

[0245] In one example, the teeth considered are a pair of two teeth having a determined type or number, for example, tooth number 13 (right upper canine) and the adjacent tooth number 14 (right upper first premolar). The spatial attribute is a triple of coordinates for the vector connecting the center of gravity of tooth number 13 and the center of gravity of tooth number 14, i.e., parameters (X, Y, Z). Therefore, the spatial information is formed by a triplet of the values of these coordinates.

[0246] The spatial information is measured relative to an orthogonal reference frame (Ox; Oy; Oz), the origin O of which is the center of the upper dental arch of the patient considered.

[0247] In step a), a historical learning database containing 100,000 historical records is created.

[0248] All of the historical images are extraoral photographs taken in true color, without a retractor, and then cropped.

[0249] The photographs are most often taken with an individual's camera, generally a mobile phone, and sometimes with a dentist's camera. They are taken from various distances from the historical patient and then cropped, preferably without regard to the historical images considered, so that the sizes of teeth 13 and 14 are substantially the same.

[0250] Any historical record includes a set of four historical images, and the set of four historical images depicts all of teeth 13 and 14 of the historical patient, and their angles are, respectively, (on the occlusal surface) "front view", "top view", "right side view", "right front view".

[0251] The historical space information of a historical record is composed of a vector (Xi, Yi, Zi). In other words, starting from the center of gravity of tooth 13, moving in the order of value Xi along axis Ox, value Yi along axis Oy, and then value Zi along axis Oz, the center of gravity of tooth 14 is reached. The historical space information of each record is manually determined from a three-dimensional model of the dental arch of the historical patient created by a 3D scanner.

[0252] In step b), a neural network CNN, for example GoogleNet (2015), is trained using a historical learning database so that it can determine the vector between the center of gravity of tooth 13 and the center of gravity of tooth 14 based on a set of the latest images similar to the set of historical images used for training.

[0253] In step b), the latest patient, for example, a patient who has no scheduled orthodontic treatment and is not wearing a retainer, beforeIt is considered at the latest moment t1. The latest patient has a mobile phone that has downloaded an application capable of performing steps b) and c). The latest patient wishes to check the dental conditions regarding the 13th and 14th teeth of the latest patient, and the patient launches this application.

[0254] The application activates the camera of the mobile phone and guides the latest patient to obtain four photos taken of these two teeth at the angles of the above-mentioned "front view", "top view", "right side view", and "right front view".

[0255] Next, the application crops these photos taken at various distances from the latest patient so that the sizes of the 13th and 14th teeth are substantially the same regardless of the latest image and are substantially the same as the sizes of these teeth in the historical images.

[0256] Next, the application inputs all four latest images into a trained neural network. This trained neural network can be integrated within the application or can be installed on a computer remote from the mobile phone. In the latter case, the mobile phone sends the four latest images to the remote computer to input the four latest images into the trained neural network.

[0257] In step c), based only on these four latest images, the trained neural network preferably provides the latest item of spatial information in less than 120 seconds, less than 60 seconds, less than 40 seconds, less than 20 seconds, less than 10 seconds, or less than 5 seconds. The latest item of spatial information is a vector (Xa, Ya, Za), and in a rectangular coordinate reference frame (Ox; Oy; Oz) whose origin O is at the center of the upper dental arch of the latest patient, it enables the center of gravity of the 13th tooth of the latest patient to be connected to the center of gravity of the 14th tooth of the latest patient.

[0258] The vector (Xa, Ya, Za) is a static item of information that can be compared with preset static constraints. For example, it is possible to check whether |Xa| < Sx, and / or |Ya| < Sy, and / or |Za| < Sz, where Sx, Sy, and Sz are threshold values, for example, 0.5 mm, 0.7 mm, and 0.3 mm. For example, it is also possible to check whether |Xa| + |Ya| + |Za| < S, where S is a threshold value. For example, if the constraint is not respected because |Xa| > Sx, a message is sent to notify the latest patient of them. For example, a message asking the latest patient to contact a dental healthcare worker is displayed on the screen of their mobile phone.

[0259] If the neural network is in a remote computer, the computer sends the latest spatial information and / or the message to the application.

[0260] The latest patient is, for example, before at the latest moment t2, for example before one month after the latest moment, the same process (launching of the application, taking of photos, and inputting them into the trained neural network) can be executed.

[0261] For this second implementation of the present invention, the application can after not only analyze the static information obtained at the latest moment, i.e., the vector (Xa', Ya', Za'), in the same way as at the latest moment of before but can also compare it with the static information obtained at the latest moment of before i.e., the vector (Xa, Ya, Za). It can, for example, determine the following amounts of movement: |Xa' - Xa|, |Ya' - Ya|, |Za' - Za|, |Xa' - Xa| + |Ya' - Ya| + |Za' - Za|, |Xa' - Xa| / (t2 - t1), |Ya' - Ya| / (t2 - t1), |Za' - Za| / (t2 - t1), or (|Xa' - Xa| + |Ya' - Ya| + |Za' - Za|) / (t2 - t1).

[0262] These amounts of movement form dynamic information that advantageously teaches the evolution over time of the dental situation regarding tooth 13 and tooth 14, more specifically, the relative movement of tooth 13 with respect to tooth 14. Therefore, these amounts of the above movement complement static information.

[0263] The dynamic information can be compared with predefined dynamic constraints. For example, it is possible to check whether |Xa'-Xa| / (t2 - t1) < Vx, |Ya'-Ya| / (t2 - t1) < Vy, |Za'-Za| / (t2 - t1) < Vz, or (|Xa'-Xa| + |Ya'-Ya| + |Za'-Za|) / (t2 - t1) < V, where Vx, Vy, Vz, and V represent threshold values of 0.1 mm / month, 0.2 mm / month, 0.1 mm / month, and 0.3 mm / month respectively. If the constraints are not respected, a message is sent to notify the latest patient. For example, the message is displayed on the screen of the latest patient's mobile phone, prompting the latest patient to contact the dental healthcare provider.

[0264] If the neural network is in a remote computer, the computer sends all or part of the latest spatial information and / or the dynamic information and / or the message to the application.

[0265] Preferably, the latest patient performs the operations described above for all pairs of teeth in their dental arch (tooth 1 and tooth 2, tooth 2 and tooth 3, etc.). For each pair of teeth, the application preferably executes a specialized procedure to provide sufficient photographs taken at different predefined angles for the considered pair of teeth. The photographs for the determined pair of teeth are input into a neural network specialized for this pair of teeth.

[0266] In one embodiment, the dynamic information is used to measure the effectiveness of the active dental orthosis, i.e., its "activity", i.e., its ability to act on the dental arch at the latest moment. For example, the speed of tooth movement can be used to determine whether the dental orthosis remains effective (if these speeds are greater than a threshold value, for example, 0.1 mm / month), and thus whether the dental orthosis needs to be changed or modified, and / or whether an appointment with a dental healthcare provider is necessary. Where appropriate, written or oral messages are sent to the latest patient and / or to the dental healthcare provider.

[0267] example Figure 6 illustrates an example implementation of the method according to the present invention. The date is on the horizontal axis. The vertical axis provides the accumulation of movement in all directions in millimeters for all the teeth of the patient's mandibular dental arch.

[0268] The solid curve represents the "actual" development. To determine this curve, a digital three-dimensional model of the patient's dental arch is first generated with a 3D scanner. Next, it is deformed to correspond to the tooth arrangements observed at different moments by implementing the method described in International Patent Application PCT / EP2015 / 074859. Each point on this curve requires several hours of computer processing.

[0269] The dashed curve represents the development determined according to the present invention. Each point on this curve requires only a few seconds of computer processing.

[0270] Very surprisingly, it can be seen that the dashed curve follows the solid curve very well. Therefore, it realistically represents the actual development despite being able to be calculated very quickly.

[0271] As is now apparent, the present invention provides a solution for determining the position, distance, orientation direction, or orientation course in the volume of the patient's oral cavity. This solution is fast, reliable, and requires only limited computing resources.

[0272] In particular, by downloading the application to a mobile phone, it can be implemented within seconds.

[0273] In addition, it provides accurate information typically with an accuracy of 0.3 mm or less.

[0274] Finally, it can be implemented from a simple extraoral photo taken by the patient himself / herself with his / her own mobile phone, especially without any special attention and without the need for any pre-created 3D models.

[0275] Therefore, the present invention enables the dental situation of any person to be evaluated not only during active or passive dental orthodontic treatment, but also when no dental orthodontic treatment is being performed, and even when the person has never met a dental healthcare provider before. Note that the present invention may be configured as follows [Item 1] A method of training a neural network intended to analyze the dental condition of a latest patient, comprising A) creating a historical learning database regarding the dental body and spatial attributes associated with the dental body, where the historical learning database includes more than 1,000 historical records, each historical record being related to an individual historical patient, and each of the historical records a set of historical images, where all of the historical images depict the dental body (referred to as the "historical dental body") in the historical patient, and one item of spatial information (referred to as the "historical spatial information") for the historical patient, including a set of values for the spatial attributes including B) Training the neural network by providing the neural network with the set of historical images as input and the historical space information as output, where the spatial attribute defines an ordered sequence of a plurality of variables in a three-dimensional reference frame. The method of training as described above, including the step of [Item 2] The method of training according to Item 1, where the neural network is a convolutional type. [Item 3] The spatial attribute is The position of one or more points to be noted of the tooth body in a three-dimensional reference frame; and / or One or more vectors between a plurality of points to be noted of the tooth body and / or between one point to be noted of the tooth body and another point The method of training according to Item 1 or 2, which defines [Item 4] The tooth body is a single tooth or a set of less than 5 teeth; and / or The spatial attribute includes less than 30 variables; and / or The set of historical images of any historical record includes more than 2 and less than 30 historical images; and / or Each historical image of the set of historical images of any historical record has an angle selected from one group of potential angles including more than 2 and less than 30 potential angles; and / or At least three historical images of any set of historical images have different angles. The method of training according to any one of Items 1 to 3. [Item 5] The method of training according to any one of Items 1 to 4, where the historical tooth body is a set of teeth including a tooth having a predetermined number or a tooth having a predetermined number and one or two teeth adjacent to the tooth. [Item 6] The method of training according to any one of Items 1 to 5, where the historical image is a true-color photograph, and / or does not illustrate a dental retractor, and / or is taken outside the oral cavity. [Item 7] A method for analyzing the dental condition of a patient at the latest moment (referred to as the "latest patient"), including a) Training a neural network according to the method of training described in any one of Items 1 to 6 before the latest moment; b) At the latest moment, obtaining a set of latest images that match the neural network and depict the tooth body (referred to as the "latest tooth body") of the latest patient by an image acquisition device; c) Analyzing the set of latest images by the neural network to obtain an item of spatial information (referred to as the "latest spatial information") for the latest patient, including a set of values for the spatial attribute. The method of performing the analysis, including the steps of [Item 8] The method of performing the analysis according to Item 7, wherein the latest image is a true-color photograph and / or is taken outside the oral cavity, and / or in step b), the image acquisition device is a mobile phone, and / or the latest patient is not wearing a dental retractor. [Item 9] The method of performing the analysis according to Item 7 or 8, wherein in step b), the latest patient is wearing a dental retractor. [Item 10] In step a), a plurality of neural networks are trained with individual historical learning databases that differ in terms of being related to different tooth bodies and / or different spatial attributes to obtain a plurality of specialized neural networks; In step b), a latest image is acquired, and for each of the plurality of specialized neural networks, a set of the latest images is created from the latest image; In step c), each of the plurality of sets of the latest images is analyzed by a corresponding specialized neural network to obtain a plurality of latest items of spatial information The method of performing the analysis according to any one of Items 7 to 9. [Item 11] The method of performing the analysis according to Item 10, wherein in step a), each historical learning database relates to a tooth or a group of a plurality of teeth having a number unique to the historical learning database, and the numbers of the plurality of teeth in the group are unique to the historical learning database. [Item 12] The latest spatial information is used to evaluate whether the objective has been achieved and / or to measure the difference between the latest dental condition of the latest patient at the latest moment and the achievement of the objective, where the objective is selected from the following objectives: The latest patient achieves a Class 1 occlusion for the canine teeth; The latest patient achieves a Class 1 occlusion for the molar teeth; The anterior space of the latest patient is closed; The space caused by tooth extraction of the latest patient is closed; The latest patient has a normal horizontal overjet; The latest patient has a normal vertical overjet; The incisor sector between the upper and lower dental arches of the latest patient is not offset; The latest patient has no lateral offset of the lower and / or upper dental arches relative to the sagittal plane of the patient's head; That the latest patient does not have a lateral offset of the upper dental arch relative to the lower dental arch; That the orthodontic appliance worn by the latest patient no longer acts to change the positions of the teeth of the latest patient; That no or only limited tooth movement of the latest patient has been detected between the last two checks of the upper and / or lower dental arches; That all deciduous teeth of the latest patient have fallen out; Absence of a lateral open bite; Absence of a posterior open bite; Absence of an anterior open bite; Absence of an anterior crossbite; Absence of a posterior crossbite; Improvement of crowding; Stabilization of depressions; Closed interproximal spaces; Absence of mucosal irregularities, The method of analysis according to any one of items 7 to 11. [Item 13] A method for determining the amount of movement between a previous latest moment and a subsequent latest moment after the previous latest moment, 1) Performing the analysis method according to any one of items 7 to 10 at the previous latest moment to obtain a previous latest item of spatial information; 2) Performing the analysis method according to any one of items 7 to 10 at the subsequent latest moment to obtain a subsequent latest item of spatial information; 3) Comparing the previous latest spatial information with the subsequent latest spatial information to obtain the amount of movement between the previous latest moment and the subsequent latest moment; 4) Optionally, presenting the amount of movement to the latest patient and / or dental healthcare provider The method as described above, including the steps. [Item 14] The method of determination according to item 13, wherein the amount of movement defines the amplitude and / or velocity of translational and / or rotational movement between the previous latest moment and the subsequent latest moment of one or more points of the latest tooth body, and / or one or more vectors connecting a plurality of points of the latest tooth body, and / or one or more vectors connecting one or more points of the latest tooth body and one or more other points of the oral cavity of the latest patient. [Item 15] In step 3), the amount of movement is compared with a threshold value, and depending on the difference between the amount of movement and the threshold value, The effectiveness index of the dental orthosis worn by the latest patient; and / or The fitness index of the dental condition of the latest patient having a situation predetermined by the dental orthodontic treatment performed by the latest patient, or a situation resulting as a result of the dental orthodontic treatment performed by the latest patient, or a situation defined by a dental healthcare provider independently of the dental orthodontic treatment The method of determination according to item 13 or 14, wherein is determined. [Item 16] The method according to any one of items 13 to 15, wherein the effective index and / or the conformity index are presented in the form of a graph. [Item 17] Detecting or evaluating the position or shape of a tooth, and / or the development of the position or shape of a tooth, and / or the rate of development of the position or shape of a tooth; and / or, Detecting or evaluating the position or shape of an orthodontic appliance, and / or the development of the position or shape of an orthodontic appliance, and / or the rate of development of the position or shape of an orthodontic appliance; and / or Measuring the development of the shape of a patient's teeth between two dates; and / or Odontology The method of use of the analysis method according to any one of items 7 to 12 or the determination method according to any one of items 13 to 16 for the purpose of. [Item 18] Monitoring the eruption of teeth; and / or, Detecting the recurrence position or abnormal position of teeth; and / or, Detecting the wear of teeth; and / or, Monitoring the opening and closing of the space between two or more teeth; and / or, Monitoring the stability or correction of tooth occlusion; and / or, Monitoring the movement of teeth to a predetermined position; and / or, Detecting or evaluating the peeling of a ring or the peeling of an orthodontic aligner; Optimizing the appointment date with an orthodontist or dentist; and / or, Evaluating the effectiveness of an active orthodontic treatment; and / or, Measuring the activity of an active orthodontic appliance; and / or, Measuring the loss of effect of a passive orthodontic appliance; and / or, Measuring the development of the shape of a patient's teeth between two dates separated by the occurrence of an impact on the teeth, or between two dates separated by the use of a dental device intended for the treatment of sleep apnea syndrome, or between two dates separated by the occurrence of a graft in the patient's oral cavity The method of use according to item 17 for the purpose of.

Claims

**Claim 1** A method of training a neural network intended to analyze the dental condition of a current patient, comprising: A) creating a historical learning database regarding the dental entities and the spatial attributes associated with said dental entities, wherein said historical learning database includes more than 1,000 historical records, each historical record being related to an individual historical patient, and each said historical record includes a set of historical images, all of said historical images depicting said dental entity (referred to as "historical dental entity") in said historical patient, and one item of spatial information (referred to as "historical spatial information") for said historical patient, including a set of values for said spatial attributes ; B) training said neural network by providing said neural network with said set of historical images as input and said historical spatial information as output, wherein said spatial attributes define an ordered sequence of a plurality of variables in a three-dimensional reference frame . The method of training as described above. **Claim 2** The method of training according to claim 1, wherein said neural network is a convolutional type. **Claim 3** Said spatial attributes are the position of one or more remarkable points of said dental entity in a three-dimensional reference frame; and / or one or more vectors between a plurality of remarkable points of said dental entity and / or between one remarkable point of said dental entity and another point . The method of training according to claim 1 or 2. **Claim 4** Said dental entity is a single tooth or a set of less than 5 teeth; and / or said spatial attributes include less than 30 variables; and / or said set of historical images of any historical record includes more than 2 and less than 30 historical images; and / or each historical image of said set of historical images of any historical record has an angle selected from one group of potential angles including more than 2 and less than 30 potential angles; and / or at least 3 historical images of any set of historical images have different angles . The method of training according to any one of claims 1 to 3. **Claim 5** The method of training according to any one of claims 1 to 4, wherein said historical dental entity is a set of teeth including a tooth having a predetermined number or a tooth having a predetermined number and one or two teeth adjacent to said plurality of teeth. **Claim 6** The method of training according to any one of claims 1 to 5, wherein the history image is a true-color photograph and / or does not illustrate a dental retractor and / or is taken outside the oral cavity.

7. A method for analyzing the dental condition of a patient at the latest moment (referred to as the "latest patient"), comprising: a) training a neural network according to the method of training according to any one of claims 1 to 6 before the latest moment; b) at the latest moment, obtaining a set of latest images that are adapted to the neural network and depict the dental body (referred to as the "latest dental body") of the latest patient by means of an image acquisition device; c) analyzing the set of latest images by the neural network to obtain one item of spatial information (referred to as the "latest spatial information") for the latest patient, including a set of values for the spatial attributes. The method of analyzing as described above.

8. The method of analyzing according to claim 7, wherein the latest image is a true-color photograph and / or is taken outside the oral cavity and / or in step b), the image acquisition device is a mobile phone and / or the latest patient is not wearing a dental retractor.

9. The method of analyzing according to claim 7 or 8, wherein in step b), the latest patient is wearing a dental retractor.

10. In step a), a plurality of neural networks are trained in individual historical learning databases that differ in terms of being related to different dental bodies and / or different spatial attributes to obtain a plurality of specialized neural networks; In step b), a latest image is obtained, and for each of the plurality of specialized neural networks, a set of latest images is created from the latest image; In step c), each of the plurality of sets of latest images is analyzed by the corresponding specialized neural network to obtain a plurality of latest items of spatial information. The method of analyzing according to any one of claims 7 to 9.

11. The method of analyzing according to claim 10, wherein in step a), each historical learning database relates to a tooth or a group of teeth having a number unique to the historical learning database, and the numbers of the plurality of teeth in the group are unique to the historical learning database.

12. The latest spatial information is used to evaluate whether the objective has been achieved and / or to measure the difference between the latest dental condition of the latest patient at the latest moment and the achievement of the objective, where the objective is selected from among the following objectives: The latest patient achieves a Class 1 occlusion for the canine teeth; The latest patient achieves a Class 1 occlusion for the molar teeth; The space in the front part of the latest patient is closed; The space caused by tooth extraction of the latest patient's teeth is closed; The latest patient has a normal horizontal overjet; The latest patient has a normal vertical overbite; The incisal sector between the upper and lower dental arches of the latest patient is not offset; The latest patient has no lateral offset of the lower and / or upper dental arches relative to the sagittal plane of the patient's head; The latest patient has no lateral offset of the upper dental arch relative to the lower dental arch; The orthodontic appliance worn by the latest patient no longer acts to change the positions of the patient's teeth; No or only limited tooth movement of the latest patient has been detected between the last two checks of the upper and / or lower dental arches; All deciduous teeth of the latest patient have fallen out; Absence of a lateral open bite; Absence of a posterior open bite; Absence of an anterior open bite; Absence of an anterior crossbite; Absence of a posterior crossbite; Improvement of crowding; Stabilization of concavities; Closed interproximal spaces; Absence of mucosal irregularities, The method of analyzing according to any one of claims 7 to 11.

13. A method for determining the amount of movement between the previous latest moment and the subsequent latest moment after the previous latest moment, comprising: 1) Performing the analysis method according to any one of claims 7 to 10 at the previous latest moment to obtain the previous latest item of spatial information; 2) Performing the analysis method according to any one of claims 7 to 10 at the subsequent latest moment to obtain the subsequent latest item of spatial information; 3) Comparing the previous latest spatial information with the subsequent latest spatial information to obtain the amount of movement between the previous latest moment and the subsequent latest moment; 4) Optionally, presenting the amount of movement to the latest patient and / or to the dental healthcare professional The method comprising the step of. **Claim 14** The method of determining according to claim 13, wherein the amount of movement defines the amplitude and / or velocity of translational and / or rotational movement between the previous latest instant and the subsequent latest instant of one or more points of the latest dental body and / or one or more vectors connecting a plurality of points of the latest dental body or one or more vectors connecting one or more points of the latest dental body and one or more other points of the oral cavity of the latest patient. **Claim 15** In step 3), the amount of movement is compared with a threshold value, and depending on the difference between the amount of movement and the threshold value, The effectiveness index of the dental orthosis worn by the latest patient; and / or The fitness index of the dental condition of the latest patient, having a situation predetermined by the dental orthosis treatment performed by the latest patient, or a situation resulting as a result of the dental orthosis treatment performed by the latest patient, or a situation defined by a dental healthcare professional independently of the dental orthosis treatment Is determined, the method of determining according to claim 13 or 14. **Claim 16** The method of determining according to claim 15, wherein the effectiveness index and / or the fitness index are presented in the form of a graph. **Claim 17** Detecting or evaluating the position or shape of the teeth and / or the development of the position or shape of the teeth and / or the rate of development of the position or shape of the teeth; and / or, Detecting or evaluating the position or shape of the dental orthosis and / or the development of the position or shape of the dental orthosis and / or the rate of development of the position or shape of the dental orthosis; and / or Measuring the development of the shape of the patient's teeth between two dates; and / or Dentistry The method of use of the analysis method according to any one of claims 7 to 12 or the determination method according to any one of claims 13 to 16 for the purpose of. **Claim 18** Monitoring tooth eruption; and / or, Detecting the recurrence position or abnormal position of the teeth; and / or, Detecting tooth wear; and / or, Monitoring the opening and closing of the space between two or more teeth; and / or, Monitoring the stability or correction of tooth occlusion; and / or, Monitoring the movement of the teeth to a predetermined position; and / or, Detecting or evaluating the peeling of the ring or the peeling of the dental aligner; Optimizing the appointment date with the orthodontist or dentist; and / or, Evaluating the effectiveness of an active dental orthopedic treatment; and / or, Measuring the activity of an active dental orthopedic appliance; and / or, Measuring the loss of effectiveness of a passive dental orthopedic appliance; and / or, Measuring the development of the shape of a patient's teeth between two dates separated by the occurrence of an impact on the teeth, or between two dates separated by the use of a dental device intended for the treatment of sleep apnea syndrome, or between two dates separated by the occurrence of a graft in the patient's oral cavity The method of use according to claim 17 for such.

Images