Method and apparatus including tooth eruption prediction

Abstract

Devices and methods for evaluating dental x-ray images and predicting tooth eruption and / or loss. The device may use one or more trained neural networks to evaluate patient x-rays and dental measurements determined from the patient x-rays to predict about when permanent teeth can be germinated to replace deciduous teeth. The neural network may be trained using a data training set of x-ray images and accompanying dental measurement data.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This patent application claims priority to U.S. Provisional Patent Application Serial No. 63 / 478,495, filed on January 4, 2023, entitled “TOOTH ERUPTION PREDICTION,” and U.S. Provisional Patent Application Serial No. 63 / 581,278, filed on September 7, 2023, entitled “METHODS AND APPARATUSES INCLUDING TOOTH ERUPTIONPREDICTION,” the entire contents of each of which are incorporated herein by reference. Incorporation by reference

[0002] All publications and patent applications mentioned in this specification are herein incorporated by reference in their entirety to the same extent as if each individual publication or patent application was specifically and individually indicated to be incorporated by reference. Technical Field

[0003] The present disclosure relates generally to dental case evaluation and, more particularly, to predicting the eruption and / or loss of a patient's teeth. Background Art

[0004] Orthodontic and dental treatment using a series of patient-removable appliances (e.g., "appliances") is very useful for treating patients. Treatment planning is typically performed in conjunction with a dental professional (e.g., a dentist, orthodontist, dental technician, etc.) by generating a model of the final arrangement of the patient's teeth, and then breaking the treatment plan into multiple intermediate stages (steps) corresponding to individual appliances that are worn in sequence. This process can be interactive, with the phasing (scheduling) and, in some cases, the final target position being adjusted based on constraints on tooth movement and the preferences of the dental professional. Once the final treatment plan is complete, a series of appliances corresponding to the treatment plan can be manufactured.

[0005] Treatment planning for young patients (e.g., patients between the ages of 5–14 years old) can be complicated by the eruption of permanent teeth that replace the patient's precocious teeth. Tooth eruption can cause spaces to form within the dental arch, which affects tooth placement and movement. In some cases, an erupting tooth may cause the treatment plan to be revised, which costs money and may extend treatment time. If the timing of tooth eruption can be predicted, the clinician (dentist, orthodontist, etc.) can ensure that the treatment plan can accommodate the erupting tooth. In addition, predicting tooth eruption and loss can be difficult, particularly due to the high variability in tooth eruption timing across various patient demographics. In some cases, the only diagnostic data available may be a patient x-ray image. Predicting tooth eruption timing based on a patient x-ray image can be difficult due to the limited two-dimensional information.

[0006] Therefore, there is a need to predict tooth eruption and / or loss based on dental x-ray images. Summary of the Invention

[0007] Devices (e.g., systems and apparatus) and methods are described herein that can evaluate patient x-ray images and predict the eruption of teeth in the patient and / or the loss of teeth in the patient. In some examples, these devices and methods can include a machine learning agent (e.g., a neural network) trained to predict tooth eruption and / or tooth loss.

[0008] Generally, methods of evaluating patient x-ray images are described herein. Any of these methods may include receiving one or more patient x-ray images, determining one or more dental measurements associated with the one or more patient x-ray images, and predicting tooth eruption and / or tooth loss using a trained neural network. Any of these methods may include forming one or more (e.g., a series of) dental appliances configured to fit over a patient's dentition and including one or more accommodations for eruption / loss.

[0009] Typically, the patient x-ray images may include two-dimensional x-rays, such as panoramic x-rays or one or more bitewing x-rays. In some cases, the patient x-ray images may include three-dimensional x-rays, such as conventional computed tomography or cone-beam computed tomography. In some variations, the patient x-ray images may include both two-dimensional and three-dimensional images.

[0010] Typically, tooth eruption and tooth loss can be predicted using a trained neural network. Typically, the trained neural network can be trained using training data. The training data can include one or more x-ray images and dental measurements associated with each x-ray image. The dental measurements can describe various attributes, measurements, characteristics, etc. of a patient's dentition. For example, the dental measurements can include the distance from the crown of a tooth to the gum, the relative distance between the crown of an unerupted tooth and the mesial tooth, or other measurements.

[0011] Typically, a dental treatment plan can be developed or modified based on predicted tooth eruption or tooth loss. In some cases, a dental treatment plan can include a series of dental appliances. One or more dental appliances can be modified to accommodate the erupting permanent teeth.

[0012] A method for predicting changes in a patient's dentition is described herein. The method may include receiving dental measurements determined from one or more x-ray images of the patient, predicting changes in the patient's dentition based on the dental measurements using a trained neural network, wherein the trained neural network is trained using a plurality of training x-ray images and corresponding patient data, and generating a treatment plan based on the predicted changes in the patient's dentition. Notably, the changes in the patient's dentition may include tooth eruption, tooth loss, or a combination thereof.

[0013] In general, dental measurements can be associated with measurements of any feasible characteristic of a patient's dentition. In some cases, the measurements can include measurements from the crown of a tooth to a gingival feature. In some other cases, the measurements can include the relative distance between an unerupted tooth and a mesial tooth crown.

[0014] In any of the methods described herein, the corresponding patient data can include patient age, patient gender, patient weight, patient race, patient geographic location, or a combination thereof. The patient data can be used to train the neural network and, in some cases, can be used during execution of the neural network to predict tooth eruption and / or loss.

[0015] The corresponding patient data may also include facial measurements. The facial measurements may include the distance between any two or more facial reference points or landmarks that can be identified on the patient's face.

[0016] Typically, a treatment plan includes a plurality of dental appliances configured to accommodate the erupted teeth. In some variations, the dental appliances may include a cutout region proximate the predicted location of the erupted teeth. In some other variations, the dental appliances may include an enlarged pocket disposed in an area proximate the predicted location of the erupted teeth. In yet other variations, the dental appliances may include an undercut region to receive the erupted teeth.

[0017] In any of the methods described herein, the x-ray images of the patient may include bitewing x-rays, panoramic x-rays, or a combination thereof.

[0018] Generally, any of the methods described herein may further include receiving updated dental measurements determined from one or more subsequent x-ray images of the patient, and updating the prediction of changes in the patient's dentition based at least in part on the updated dental measurements. In some variations, the updated dental measurements may include measurements determined from an intraoral scan.

[0019] Also described herein are devices configured to perform any of these methods. For example, a device for predicting changes to a patient's dentition is described herein and may include a communication interface, one or more processors, and a memory storing instructions that, when executed by the one or more processors, cause the one or more processors to receive dental measurements determined from one or more x-ray images of the patient, predict changes to the patient's dentition based on the dental measurements using a trained neural network, wherein the trained neural network is trained using a plurality of training x-ray images and corresponding patient data, and generate a treatment plan based on the predicted changes to the patient's dentition.

[0020] Also described herein is a non-transitory computer-readable storage medium storing instructions that, when executed by one or more processors of a device, cause the device to receive dental measurements determined from one or more x-ray images of a patient, predict changes to the patient's dentition based on the dental measurements using a trained neural network, wherein the trained neural network is trained using a plurality of training x-ray images and corresponding patient data, and generate a treatment plan based on the predicted changes to the patient's dentition.

[0021] Also described herein is a method for predicting a change in a patient's dentition, comprising: receiving a reference 2D x-ray image and a corresponding reference 3D x-ray image, wherein the reference 2D x-ray image and the reference 3D x-ray image are based on an ideal dental arch describing ideal tooth positions; determining a transformation function to map the 2D x-ray image to a 3D x-ray image; receiving the 2D x-ray image of the patient, generating a predicted 2D model based on mapping the patient's 2D x-ray image to the reference 2D x-ray image; generating a predicted 3D model based on the transformation function; and predicting a change in the patient's dentition based on the predicted 3D model.

[0022] Typically, the method may further comprise generating a treatment plan based on predicted changes in the patient's dentition.Also typically, the changes in the patient's dentition may comprise tooth eruption, tooth loss, or a combination thereof.

[0023] In some variations, any of the methods may include predicting the geometry of the unerupted tooth based on statistical modeling. Typically, the 2D X-ray images of the patient may be a panoramic x-ray, a series of bitewing x-rays, a series of periapical x-rays, or a combination thereof.

[0024] Generally, in any of the methods described herein, an ideal dental arch can be selected based on the patient's demographics.

[0025] Typically, in any of the methods described herein, the 3D x-ray image can be a computed tomography (CT) or cone-beam computed tomography (CBCT) scan.

[0026] Also described herein is an apparatus for predicting changes in a patient's dentition, which may include a communication interface, one or more processors, and a memory storing instructions that, when executed by the one or more processors, cause the one or more processors to receive a reference 2D x-ray image and a corresponding reference 3D x-ray image, wherein the reference 2D x-ray image and the reference 3D x-ray image are based on an ideal dental arch describing ideal tooth positions, determine a transformation function to map the 2D x-ray image to a 3D x-ray image, receive a 2D x-ray image of the patient, generate a predicted 2D model based on mapping the patient's 2D x-ray image to the reference 2D x-ray image, generate a predicted 3D model based on the transformation function, and predict a change in the patient's dentition based on the predicted 3D model.

[0027] Also described herein is a non-transitory computer-readable storage medium storing instructions that, when executed by one or more processors of a device, cause the device to receive a reference 2D x-ray image and a corresponding reference 3D x-ray image, wherein the reference 2D x-ray image and the reference 3D x-ray image are based on an ideal dental arch describing ideal tooth positions, determine a transformation function to map the 2D x-ray image to a 3D x-ray image, receive a 2D x-ray image of a patient, generate a predicted 2D model based on mapping the patient's 2D x-ray image to the reference 2D x-ray image, generate a predicted 3D model based on the transformation function, and predict a change in the patient's dentition based on the predicted 3D model.

[0028] For example, methods are described herein that include extracting or receiving dental measurements from one or more x-ray images of a patient; predicting tooth eruption and / or shedding in the patient's dentition based on the dental measurements using a trained neural network, wherein the trained neural network is trained using a plurality of training x-ray images and corresponding patient data to generate a probability value indicating the likelihood of eruption and / or shedding of teeth at a particular location; and forming a dental appliance configured to be fitted over the patient's dentition and including an adaptation to accommodate the predicted tooth eruption and / or shedding, wherein the adaptation includes one or more of: an incision area configured to accommodate the predicted tooth eruption and / or shedding, a pocket configured to accommodate the predicted tooth eruption and / or shedding, and / or an undercut area configured to accommodate the predicted tooth eruption and / or shedding.

[0029] Any of these methods may include generating a treatment plan based on the probability values.

[0030] As described above, the dental measurement results may include the crown-to-gingival distance of at least one tooth. In some examples, the dental measurement results include the relative distance between the crown of an unerupted tooth and the mesial tooth. The corresponding patient data may include patient age, patient gender, patient weight, patient race, patient geographic location, or a combination thereof. The corresponding patient data may include facial measurements associated with two or more facial reference points. The patient's x-ray image may include a panoramic x-ray, a bitewing x-ray, or a combination thereof.

[0031] Any of these methods may include receiving updated dental measurements determined from one or more subsequent x-ray images of the patient; and updating the prediction of change to the patient's dentition based at least in part on the updated dental measurements. The updated dental measurements may include measurements determined from an intraoral scan.

[0032] Any of these methods may include outputting a probability value for a particular location.

[0033] Typically, forming a dental appliance can include forming a digital model of a dental appliance (or series of appliances) for upper and / or lower teeth. In some examples, forming a dental appliance can include manufacturing the dental appliance. For example, manufacturing the dental appliance can include manufacturing the dental appliance by a direct manufacturing process (e.g., 3D printing, etc.).

[0034] In any of these methods and apparatuses, the dental appliance may include a patient-removable appliance configured to fit over the patient's dentition.

[0035] Forming the dental appliance can include generating a digital model of the dental appliance and positioning an adaptor in an area on the dental appliance configured to be worn in a specific position. The dental appliance can include one or more tooth-receiving cavities configured to be worn on the patient's dentition (e.g., teeth).

[0036] Also described herein are apparatuses for performing any of these methods. For example, and these apparatuses may include: one or more processors; and memory storing instructions that, when executed by the one or more processors, cause the apparatus to perform a method comprising: extracting or receiving dental measurements from one or more x-ray images of a patient; predicting tooth eruption and / or shedding in the patient's dentition based on the dental measurements using a trained neural network, wherein the trained neural network is trained using a plurality of training x-ray images and corresponding patient data to generate probability values ​​indicating the likelihood of eruption and / or shedding of teeth at specific locations; and forming a dental appliance configured to accommodate the predicted tooth eruption and / or shedding, wherein the dental appliance includes one or more of: an incision region configured to accommodate the predicted tooth eruption and / or shedding, a pocket configured to accommodate the predicted tooth eruption and / or shedding, and / or an undercut region configured to accommodate the predicted tooth eruption and / or shedding.

[0037] In some examples, the method may include: positioning one or more maxillary arch splines relative to the patient's dental arch in a virtual model of the patient's dental arch; positioning one or more buccal-lingual planes in an area of ​​the patient's dental arch to receive erupted teeth; for each buccal-lingual plane, determining a two-dimensional (2D) tooth profile based on a projection of the one or more maxillary arch splines through the one or more buccal-lingual planes; and forming a dental appliance configured to be worn on the patient's dental arch, wherein the forming includes setting a void area of ​​the dental appliance based on the 2D tooth profile from the one or more buccal-lingual planes.

[0038] The one or more maxillary arch splines may pass through external points on the surface of the patient's existing teeth. The one or more maxillary arch splines may be associated with at least one of a buccal gingival point or a lingual gingival point on the surface of the patient's existing teeth. The one or more maxillary arch splines may be associated with at least one of a buccal cusp, a lingual cusp, or a groove arch point on the surface of the patient's existing teeth. The one or more maxillary arch splines may be described by a piecewise continuous polynomial that describes a curve passing through predetermined points of the patient's existing teeth. In some examples, the one or more buccal-lingual planes are orthogonal to the patient's dental arch. Determining a gap may include forming a volume based on a projection of the one or more maxillary arch splines through the one or more buccal-lingual planes. In some examples, determining the one or more maxillary arch splines includes adjusting the one or more maxillary arch splines by a clinician.

[0039] The adjustment can be performed by a clinician interacting with a graphical user interface.Any of these methods can include displaying the one or more jaw arch splines on a graphical user interface.Any of these methods can include displaying the dental appliance including the gap on a display.

[0040] Typically, any of these methods may involve modifying the dental appliance to accommodate teeth from a different arch.

[0041] All methods and apparatus described herein, in any combination, are contemplated herein and may be used to achieve the benefits as described herein. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] A better understanding of the features and advantages of the methods and apparatus described herein will be obtained by referring to the following detailed description and accompanying drawings which set forth illustrative embodiments, in which:

[0043] Figure 1 is a graph showing example probabilities of tooth loss versus age for young patients.

[0044] Figure 2 An example of a machine learning tooth eruption prediction device is schematically shown.

[0045] Figure 3 is a flow chart illustrating an example method for training a neural network to predict tooth eruption.

[0046] Figure 4 is shown for use Figure 3 A flow chart of an example of a method for using a machine learning approach to predict tooth eruption in a patient.

[0047] Figure 5 is a flow chart illustrating one example of a method for predicting the position and direction of tooth eruption in a patient.

[0048] Figure 6 is a flow chart illustrating an example method for training a neural network to predict tooth eruption based on facial features.

[0049] Figure 7 A diagram of a face with locations of possible facial fiducials is shown.

[0050] Figure 8 is shown for use Figure 6 A flow chart of an example of a method of a machine learning approach to predicting tooth eruption in a patient.

[0051] Figures 9A-9D An example dental appliance designed to accommodate erupting teeth is shown.

[0052] Figure 10 Shows that it can be Figure 2 A block diagram of an example apparatus of a machine learning tooth eruption prediction device.

[0053] Figure 11A is an x-ray scan image showing an example of prediction as described herein.

[0054] Figure 11B An example of a method for making predictions based on rescan times as described herein is shown.

[0055] Figure 12 An example of a workflow for predicting the position and orientation of erupted permanent teeth using x-ray images as described herein is schematically shown.

[0056] Figure 13 A simplified dental arch is shown illustrating another approach for appliance design to accommodate unerupted teeth.

[0057] Figure 14 An example of a buccal-lingual plane is shown.

[0058] Figure 15A and Figure 15B An exemplary use of buccal-lingual planes and mandibular arch splines is shown.

[0059] Figure 16 A portion of a dental arch is shown that may be displayed on a graphical user interface.

[0060] Figure 17 is a flow chart illustrating one example of a method for determining a gap in a dental appliance.

[0061] Figure 18 is a diagram illustrating an example of a computing environment that includes the tooth eruption prediction methods and apparatus (eg, modules) described herein. DETAILED DESCRIPTION

[0062] Images, such as dental images, are widely used to develop and monitor dental treatment plans. For example, some dental images can be used to determine a starting point for a dental treatment plan or, in some cases, to determine whether a patient is a viable candidate for any of a number of different dental treatment plans. The output of the apparatus and methods described herein can be tailored to the patient's age and the complexity of their treatment goals and can provide valuable insight to clinicians (doctors, dentists, orthodontists, etc.) in deciding whether to proceed or otherwise modify treatment.

[0063] Described herein are devices (e.g., systems and apparatus, including software) and methods for training and applying machine learning agents (e.g., neural networks) to evaluate patient dental images (including x-ray images) to predict the eruption and / or loss of a patient's teeth. In particular, these methods and devices can predict the timing of eruption, the location of eruption, the direction in which a tooth may erupt, and in some cases the geometry of the erupted tooth. These methods and devices may be particularly useful when used in conjunction with or as part of dental treatment planning, which may be divided into multiple steps (e.g., stages, segments, keyframes, etc.). The resulting predictions can be made in the context of the stages of a proposed or ongoing treatment plan. Thus, these methods and devices can use data from a patient (e.g., imaging data) to predict when the eruption of a tooth (or multiple teeth) will occur, including in some examples which stage or segment, and can further predict how long a tooth will take to erupt, where it will erupt, etc.

[0064] In any of these examples, the dental image can be an x-ray image of the patient's teeth, which can include one or more panoramic films, bitewing films, periapical films, or any other feasible x-ray images. In addition, measurement data determined from the patient's dental x-rays can be included. Exemplary measurement data can include the distance from the crown of a tooth to the gingiva, the relative distance between the crown of an unerupted tooth and the crown of a mesial tooth, etc. The processor or processing node can execute one or more neural networks that have been trained to predict tooth eruption and / or exfoliation from the patient's dental x-rays.

[0065] Also described herein are apparatus and methods for estimating tooth eruption and loss using a reference two-dimensional dental arch and a reference three-dimensional dental arch. A two-dimensional x-ray of a patient is mapped to a reference three-dimensional dental arch to predict movement of the patient's teeth.

[0066] After determining the estimated time of tooth eruption or loss, the clinician can change one or more aspects of the dental treatment plan. For example, based on the predicted eruption time, the clinician can decide to extract a young tooth. In another example, one or more dental appliances can be modified or adjusted to accommodate the erupting tooth.

[0067] Also described herein are methods comprising: extracting or receiving dental measurements from one or more x-ray images of a patient; predicting the eruption of teeth in the patient's dentition based on the dental measurements using a trained neural network, wherein the trained neural network is trained using a plurality of training x-ray images and corresponding patient data to generate a predicted shape of an erupted tooth at a specific location; and outputting the predicted shape of the erupted tooth at the specific location. These methods can be methods of determining the shape (e.g., morphology) of an erupted / unerupted tooth using the techniques described herein.

[0068] These methods can include any of the steps described herein. For example, any of the methods described herein can include (optionally) generating a treatment plan based on the predicted shape. In any of these methods, the method can include manufacturing (e.g., by a direct manufacturing process, etc.) one or more dental appliances, such as braces, from any of the treatment plans or otherwise.

[0069] Also described herein are apparatuses and methods for determining the presence, size, and / or location of a space (eg, a void) in a dental appliance for receiving an erupted tooth.

[0070] For example, a method is described herein that includes: positioning one or more maxillary arch splines relative to a patient's dental arch in a virtual model of the patient's dental arch; positioning one or more buccal-lingual planes in an area of ​​the patient's dental arch to receive erupted teeth; for each buccal-lingual plane, determining a two-dimensional (2D) tooth contour based on a projection of the one or more maxillary arch splines through the one or more buccal-lingual planes; and setting a void area of ​​a dental appliance based on the 2D tooth contours from the one or more buccal-lingual planes.

[0071] One or more maxillary arch splines can be defined based on the characteristics of the patient's existing teeth. These maxillary arch splines can be used to determine the outline of the gap. In some examples, one or more maxillary arch splines can be projected onto one or more buccal-lingual planes that are (virtually) located in the area of ​​the patient's dental arch that has or will have erupted teeth. The projection can be used to determine curves on the buccal-lingual planes. By using several curves, the outline or shape of the gap can be determined.

[0072] As described herein, a method for determining a gap in a dental appliance to receive an erupted tooth may include: determining one or more mandibular arch splines of a patient's dental arch; determining a position of one or more buccal-lingual planes, wherein the buccal-lingual planes are positioned in areas of the patient's dental arch for receiving the erupted tooth; determining a two-dimensional (2D) tooth profile for each buccal-lingual plane based on projections of the one or more mandibular arch splines through the one or more buccal-lingual planes; and determining a gap for the dental appliance based on the 2D tooth profiles from the one or more buccal-lingual planes.

[0073] In any of the methods described herein, the one or more jaw arch splines can pass through external points on the surface of the patient's existing teeth. Generally, the external points can be selected from any feasible points on the existing teeth. The external points can be associated with external dental features that affect the external contour or shape of the teeth. Thus, the jaw arch splines can be associated with any external dental feature.

[0074] For example, in any of the methods described herein, the one or more jaw arch splines can be associated with at least one of a buccal gingival point or a lingual gingival point on the surface of the patient's existing teeth. In some other examples, in any of the methods described herein, the one or more jaw arch splines can be associated with at least one of a buccal cusp, a lingual cusp, or a groove arch point on the surface of the patient's existing teeth.

[0075] In general, a jaw arch spline can be a mathematical equation that describes a curve through selected external points on the patient's teeth. In general, a jaw arch spline can be a piecewise continuous polynomial that describes a curve through predetermined points of the patient's existing teeth.

[0076] As described herein, the buccal-lingual plane can be positioned in an area where teeth are or will be erupting. Typically, the buccal-lingual plane can be orthogonal to the dental arch.

[0077] In general, the volume or shape of the void can be based on the projection of the one or more mandibular arch splines through the one or more buccal-lingual planes. For example, a curve on the buccal-lingual plane can be based on the intersection of one or more mandibular arch splines with the buccal-lingual plane. Many curves on many buccal-lingual planes can be used to determine the contour of the void.

[0078] In any of the methods described herein, one or more jaw arch splines can be adjusted by a clinician. By adjusting the jaw arch splines, the user can change or modify the shape of the gap. In some examples, a graphical user interface can display one or more jaw arch splines and control points that allow the clinician to interact with and change any jaw arch spline. After changing or modifying the jaw arch spline, the changes to the gap can be displayed on the graphical user interface.

[0079] In any of the methods described herein, a dental appliance comprising a void can be modified to accommodate teeth from different dental arches. For example, space that may be occupied by opposing teeth can be removed from the void to reduce or eliminate interference between the dental appliance and the opposing teeth.

[0080] An apparatus (e.g., a system and device including software) for determining a gap in a dental appliance is described herein. The apparatus may include one or more processors and a memory storing instructions that, when executed by the one or more processors, cause the apparatus to determine one or more jaw arch splines of a patient's dental arch, determine a position of one or more buccal-lingual planes, wherein the buccal-lingual planes are positioned in an area of ​​the patient's dental arch to receive erupted teeth, determine a two-dimensional (2D) tooth profile for each buccal-lingual plane based on a projection of the one or more jaw arch splines through the one or more buccal-lingual planes, and determine a gap in the dental appliance based on the 2D tooth profiles from the one or more buccal-lingual planes.

[0081] Also described herein is a non-transitory computer-readable storage medium storing instructions that, when executed by one or more processors of a device, cause the device to determine one or more mandibular arch splines of a patient's dental arch, determine a position of one or more buccal-lingual planes, wherein the buccal-lingual planes are positioned in an area of ​​the patient's dental arch to receive erupted teeth, determine a two-dimensional (2D) tooth profile for each buccal-lingual plane based on a projection of the one or more mandibular arch splines through the one or more buccal-lingual planes, and determine a space for a dental appliance based on the 2D tooth profiles from the one or more buccal-lingual planes.

[0082] Figure 1 Graph 100 illustrates an example probability of tooth loss versus age for young patients. Tooth eruption and shedding can introduce complications in treatment planning for young patients. In some examples, treatment plans may need to be modified depending on when tooth eruption and / or shedding occurs. In some cases, clinicians may desire to extract primary teeth as part of treatment planning to help ensure a good treatment outcome. In other cases, dental appliances may need to be modified based on the timing of eruption and / or shedding.

[0083] As shown in graph 100, tooth eruption occurs most frequently around age 11, however, the exact age may vary between patients. Therefore, a method for predicting when tooth eruption and / or loss will occur can help clinicians develop efficient and effective treatment plans.

[0084] Typically, when training a neural network to determine shedding as described herein, the network can be trained on data from adolescents (e.g., children under 18 years old, between 6 and 18 years old, between 6 and 16 years old, etc.) using a patient dataset where all patients in the training dataset have undergone treatment with dental appliances. Surprisingly, using this treatment-matched patient data is particularly helpful for these patients because it appears that the teeth of patients treated with dental appliances may shed and erupt on a different schedule than in textbook data. In some cases, for example, when using a shell appliance system, teeth may erupt faster than would be expected based on textbook data compared to treatment with a wire and bracket system. This may be one reason for complaints about improperly fitted appliances.

[0085] Typically, a training database for the machine learning models described herein may include before and after scans of a patient, cross-referenced by patient age; a predictive distribution table can be generated from this data for each type of tooth.

[0086] While the methods and apparatus described herein may be particularly helpful for children, the methods and apparatus may also be used with adult patients. For example, a predictive model included as part of the apparatus and methods described herein can estimate the probability that an adult tooth will begin to erupt as a replacement, and whether the adult tooth will be fully erupted by the end of treatment. This can also be useful in how and when to apply and design eruption compensation features, or dynamically design to match eruption rates. For patients with partially erupted dentition, the methods and apparatus may include the extent to which a tooth is likely to erupt further on its own, expressed as a distance remaining or a percentage of completion. As needed, the clinician can further pull the tooth by adding attachments or leave it alone to alleviate the known problem of overuse of attachments.

[0087] The methods and devices described herein can be used to examine an existing patient's dentition and generate a patient-specific growth pattern timeline for all remaining developmental events. This can also include predicting remaining mandibular growth and development if the correct anatomical landmarks can be detected, and these devices and methods can cross-reference these tooth positions against available submitted x-rays.

[0088] The methods and devices (including software) described herein can generally be configured to predict the eruption of one or more teeth. These methods and devices can be methods for indicating the likelihood of treatment success based on an eruption prediction. In some cases, the methods and devices can provide as an output (or part of an output) a recommendation to proceed or not to proceed. For example, the methods and devices described herein can estimate the probability of exfoliation occurring after the end of treatment; in some examples, the methods or devices can provide a recommendation based on the probability of exfoliation occurring after the end of treatment; for example, if this probability is greater than a threshold (post-treatment eruption threshold), e.g., X%, then a recommendation can be made to not proceed with treatment or to interrupt treatment (if it has already begun). Any of these methods and devices can determine the probability of exfoliation occurring within the next N months. In some examples, if the probability of exfoliation occurring within the next N months is greater than a threshold (predicted exfoliation threshold), e.g., Y%, then extraction of the obstructing tooth is recommended. Any of these methods can determine the probability of exfoliation occurring during treatment; if the probability of exfoliation occurring during treatment is greater than a threshold (on-treatment eruption threshold), e.g., Z%, then a shorter treatment period is recommended.

[0089] In any of these methods and apparatus, a probabilistic analysis of any of these estimates (e.g., post-treatment eruption, pre-treatment eruption, and during-treatment eruption) can include: a strict probability of tooth loss by age X, a probability of tooth loss by age Y, and / or a probability of tooth loss based on current age and / or current dentition (conditional on dentition). The probability of tooth loss by age Y can be a conditional probability of tooth loss. For example, this can include renormalizing the residual probability of tooth loss assuming the patient has not lost a tooth by age X (e.g., if the cumulative probability of losing a tooth by age 11 is 0.6, but you have not lost it yet, one model for conditional probability is to renormalize the residual probability by stretching it to a range of 0-1).

[0090] Conditional probabilities can be determined using a patient database (e.g., frequency counts from a patient database), historical data, and / or literature surveys. In any of the methods and devices described herein, a hierarchical statistical model of Bayesian machine learning can be used. Alternatively or additionally, a black box model can be used, wherein a machine learning model is used to make predictions (e.g., random forests, decision trees, neural networks, etc.).

[0091] Figure 2An example of a machine learning tooth eruption prediction device 200 is schematically shown. Although described herein as a system, the machine learning tooth eruption prediction device 200 can be implemented using any feasible device, for example, an apparatus, a system, etc., including hardware, software and / or firmware. In some examples, the machine learning tooth eruption prediction device 200 may include a processing node 210, an application programming interface (API) 250, and a data storage module 240. As shown, the API 250 and the data storage module 240 may be respectively connected to the processing node 210. In some examples, all components of the machine learning tooth eruption prediction device 200 may be implemented as a single device (e.g., within a single housing). In some other examples, the components of the machine learning tooth eruption prediction device 200 may be distributed within separate devices. For example, the connection between any two or more devices, nodes (any of which may be referred to as a module herein) and / or data storage modules may be through a network including the Internet. In this manner, the machine learning tooth eruption prediction device 200 may be configured to operate as a cloud-based device, where some or all components of the machine learning tooth eruption prediction device 200 may be coupled together via any feasible wired or wireless network, including the Internet.

[0092] The machine learning tooth eruption prediction device 200 can predict the eruption of one or more teeth associated with a patient based on information associated with or determined from the patient's x-ray images. In some embodiments, the machine learning tooth eruption prediction device 200 can use an API 250 to facilitate the receipt or input of patient x-ray images 220 and the output of treatment data through a treatment planning interface 230. The machine learning tooth eruption prediction device 200 can also include a tooth eruption prediction engine 270, (one or more) treatment planning systems 280, and an instrumentation engine 285. The tooth eruption prediction engine 270 can predict the eruption of one or more patient teeth based on the patient's x-ray images. In some variations, the tooth eruption prediction engine 270 can perform machine learning, including executing one or more neural networks trained to predict tooth eruption using patient x-ray images. For example, the processing node 210 (and / or the machine learning agent 215) can use x-ray training data 260 to train one or more neural networks that can form all or part of the tooth eruption prediction engine 270. In some variations, the processing node 210 may provide the patient x-ray image 220 to a tooth eruption prediction engine 270. The tooth eruption prediction engine 270 may, in turn, predict tooth eruption based at least in part on tooth characteristics (e.g., tooth measurements) determined from the patient x-ray image 220. Figure 3 Describe the training of neural networks in more detail.

[0093] The treatment planning system(s) 280 can generate, create, and / or provide any available dental or orthodontic treatment plan for a patient. The treatment planning system(s) 280 can receive or obtain patient data and treatment preferences associated with a user. As described herein, the treatment planning interface 230 can transmit and / or receive patient data, user treatment preferences, and dental instrument data. Thus, using the patient data and user treatment preferences, the treatment planning system(s) 280 can generate or provide an associated treatment plan. The treatment planning system(s) 280 can implement automated and / or real-time treatment planning.

[0094] The treatment planning system(s) 280 may include one or more engines configured to generate and / or provide treatment plans. In various embodiments, the treatment planning system(s) 280 identify and / or calculate a treatment plan having instructions for treating a medical condition. The treatment plan may specify treatment goals, specific outcomes, intermediate outcomes, and / or recommended instruments for achieving the goals / outcomes. The treatment plan may also include a treatment length and / or milestones. In various embodiments, the treatment planning system(s) 280 calculate an orthodontic treatment plan for treating a malocclusion, a restorative treatment plan for a patient's dentition, a medical treatment plan, and the like. Treatment planning may include automated and / or real-time elements and may include the technology described in U.S. patent application Ser. No. 16 / 178,491, entitled "Automatic Treatment Planning." In various embodiments, the treatment planning system(s) 280 are managed by a treating technician. As described herein, the treatment plan may be adapted to the patient data based on the user's treatment preferences.

[0095] In some examples, the treatment planning system(s) 280 can generate instrument data based on the patient's treatment plan. In some examples, the patient's instrument data can describe a series of consecutive appliances that can be used to execute (perform or provide) the patient's treatment plan.

[0096] The treatment planning system(s) 280 may include an engine that allows a user to visualize, interact with, and / or manufacture instruments that implement the treatment plan. The treatment planning system(s) 280 may support UIs that display virtual representations of orthodontic instruments that move the patient's teeth from an initial position toward a final position to correct dental malocclusion. The treatment planning system(s) 280 may similarly include an engine that enables the display of representations of restorative instruments and / or other medical instruments. The treatment planning system(s) 280 may support the manufacture of instruments through, for example, an instrument manufacturing engine 285. The treatment planning system(s) 280 may also include an engine to support user interaction with the treatment plan. In some variations, the treatment template may include structured data, UI elements (tables, text boxes, UI buttons, selectable UI elements, etc.), and the like.

[0097] Treatment planning system(s) 280 may include one or more data stores configured to store treatment templates expressed according to treatment domain specific protocols. In some variations, treatment templates may be stored in data storage module 240.

[0098] The device manufacturing engine 285 can be configured to manufacture one or more devices, including dental and non-dental devices. Examples of dental devices include braces, other polymer dental devices, crowns, veneers, bridges, retainers, dental surgical guides, etc. Examples of non-dental devices include orthodontic devices, hearing aids, surgical guides, medical implants, etc.

[0099] The device manufacturing engine 285 may include a thermoforming system configured to indirectly and / or directly form the device. The device manufacturing engine 285 may implement instructions for indirectly manufacturing the device. For example, the device manufacturing engine 285 may be configured to thermoform the device on a male or female mold. The indirect manufacture of the dental device may include one or more of the following steps: producing a male or female mold of the patient's dentition in a target arrangement (e.g., by additive manufacturing, milling, etc.), thermoforming one or more sheets of material on the mold to generate the device housing, forming one or more structures in the housing (e.g., by cutting, etching, etc.), and / or coupling one or more components to the housing (e.g., by extrusion, additive manufacturing, spraying, thermoforming, adhesives, bonding, fasteners, etc.). Optionally, one or more auxiliary instrument components as described herein (e.g., elastics, wires, springs, rods, arch expanders, palatal expanders, bimaxillary blocks, bite blocks, bite ramps, mandibular advancement splints, bite plates, bridges, hooks, brackets, headgear tubes, palatal bars, frames, pin-and-tube devices, buccal shields, buccal arches, wire shields, lingual flanges and pads, lip pads or bumpers, protrusions, notches, etc.) are formed separately from the instrument housing after the housing has been manufactured and coupled to the instrument housing (e.g., via adhesives, bonds, fasteners, mounting features, etc.).

[0100] The device manufacturing engine 285 may include a direct manufacturing system configured to directly manufacture the device. For example, the device manufacturing engine 285 may include a system, device, or apparatus configured to use additive manufacturing techniques (also referred to herein as "3D printing") or subtractive manufacturing techniques (e.g., milling). In some embodiments, direct manufacturing involves forming an object (e.g., an orthodontic device or portion thereof) without using a physical template (e.g., a mold, mask, etc.) to define the object's geometry. Additive manufacturing techniques may include: (1) vat photocuring (e.g., stereolithography), in which an object is constructed layer by layer from a vat of liquid photopolymer resin; (2) material jetting, in which material is sprayed onto a build platform using a continuous or drop-on-demand (DOD) process; (3) binder jetting, in which alternating layers of a build material (e.g., a powder-based material) and a binder material (e.g., a liquid binder) are deposited by a print head; (4) fused deposition modeling (FDM), in which material is pumped through a nozzle, heated, and deposited layer by layer; (5) powder bed fusion, including but not limited to direct metal laser sintering (DMLS), electron beam melting (EBM), selective heat sintering (SHS), selective laser melting (SLM), and selective laser sintering (SLS); (6) sheet lamination, including but not limited to layered object manufacturing (LOM) and ultrasonic additive manufacturing (UAM); and (7) directed energy deposition, including but not limited to laser engineered mesh forming, directed light manufacturing, direct metal deposition, and 3D laser cladding. For example, stereolithography may be used to directly manufacture one or more devices herein. In some embodiments, stereolithography involves using light (e.g., ultraviolet light) to selectively polymerize a photosensitive resin (e.g., a photopolymer) according to a desired cross-sectional shape. The object geometry can be constructed layer by layer by sequentially polymerizing multiple object cross-sections. As another example, the device manufacturing engine 170 can be configured to directly manufacture the device using selective laser sintering. In some embodiments, selective laser sintering involves using a laser beam to selectively melt and fuse layers of powdered material according to a desired cross-sectional shape to construct the object geometry. As yet another example, the device manufacturing engine 285 can be configured to directly manufacture the device by fused deposition molding. In some embodiments, fused deposition molding involves melting and selectively depositing filaments of thermoplastic polymers in a layer-by-layer manner to form an object. In yet another example, the device manufacturing engine 285 can be configured to implement material spraying to directly manufacture the device. In some embodiments, material spraying involves spraying or extruding one or more materials onto a build surface to form continuous layers of the object geometry.

[0101] In some embodiments, the device manufacturing engine 285 may include a combination of direct and indirect manufacturing systems. In some embodiments, the device manufacturing system (not shown) can be configured to build the object geometry in a layer-by-layer manner, where successive layers are formed in discrete building steps. Alternatively, or in combination, the device manufacturing engine 285 can be configured to use a continuous construction of the geometry of the object, referred to herein as "continuous direct manufacturing." Various types of continuous direct manufacturing systems can be used. For example, in some embodiments, the device manufacturing engine 285 can use "continuous liquid interphase printing," where an object is continuously built from a reservoir of photopolymerizable resin by forming a gradient of partially cured resin between the build surface of the object and a "dead zone" that inhibits polymerization. In some embodiments, a semi-permeable membrane is used to control the delivery of a photopolymerization inhibitor (e.g., oxygen) into the dead zone to form a polymerization gradient. Examples of continuous liquid interphase printing systems are described in U.S. Patent Publication Nos. 2015 / 0097315, 2015 / 0097316, and 2015 / 0102532 (corresponding to U.S. Patent Nos. 9,205,601, 9,216,546, and 9,211,678), the disclosure of each of which is incorporated herein by reference in its entirety. As another example, the device manufacturing engine 285 can be configured to achieve continuous construction of the object geometry by continuous motion (e.g., along the vertical or Z direction) of the building platform during the irradiation phase, such that the hardening depth of the polymer irradiated with light is controlled by the speed of motion. Thus, continuous polymerization of the material on the building surface can be achieved. An example system is described in U.S. Patent No. 7,892,474, the disclosure of which is incorporated herein by reference in its entirety.

[0102] In another example, the device manufacturing engine 285 can be configured to extrude a composite material consisting of a curable liquid material surrounding a solid strand. The composite material can be extruded along a continuous 3D path to form an object. An example system is described in U.S. Patent Publication No. 2014 / 0061974 corresponding to U.S. Patent No. 9,511,543, the disclosure of which is incorporated herein by reference in its entirety.

[0103] In yet another example, the device fabrication engine 285 can implement a "spiral lithography" method, in which a liquid photopolymer is cured using focused radiation while the build platform is continuously rotated and raised. Thus, the object geometry can be continuously built along a spiral construction path. An example of such a system is described in U.S. Patent Publication No. 2014 / 0265034, corresponding to U.S. Patent No. 9,321,215, the disclosure of which is incorporated herein by reference in its entirety.

[0104] The data storage module 240 can be any feasible data storage unit, device, or structure, including random access memory, solid-state memory, disk-based memory, non-volatile memory, etc. The data storage module 240 can store image data, including patient x-ray images 220 received via the API 250. The data storage module 240 can also store device data from the device manufacturing engine 285.

[0105] The data storage module 240 and / or the processing node 210 may also include a non-transitory computer-readable storage medium that stores instructions that may be executed by the processing node 210. For example, the processing node 210 may include one or more processors (not shown) that execute instructions stored in the data storage module 240 to perform any number of operations, including operations for evaluating the patient x-ray image 220 and predicting tooth eruption. For example, the data storage module 240 may store one or more neural networks that may be trained and / or executed by the processing node 210. Alternatively, the processing node 210 may include one or more machine learning agents 215 (e.g., a trained neural network as described herein), such as Figure 1 shown.

[0106] In some examples, the data storage module 240 may include instructions for training one or more neural networks to evaluate the patient x-ray images 220. Figure 3 Further details regarding the training of the neural network are described below. Additionally or alternatively, the data storage module 240 may include instructions for executing one or more neural networks to evaluate the patient x-ray images 220. Further details regarding the execution of the neural network are described below.

[0107] Figure 3 is a flow chart illustrating an example method 300 for training a neural network to predict tooth eruption. Some examples may utilize additional operations, fewer operations, operations in a different order, operations in parallel, and operations that are different to perform the operations described herein. Figure 2 The method 300 is described using the machine learning tooth eruption prediction device 200, however, the method 300 can be performed by any other suitable device, system, or apparatus. In some examples, a neural network can be trained to predict tooth eruption from patient x-ray images and / or associated dental measurements.

[0108] Method 300 begins in block 302, where processing node 210 obtains x-ray training data 260. X-ray training data 260 may include dental x-ray images (e.g., tooth images) illustrating one or more aspects of a patient's dentition. The x-ray images may include multiple bitewing x-ray images and, in some cases, panoramic x-ray images. For example, x-ray training data 260 may include x-ray images of some or all of a person's teeth, soft tissue, bone structure, and the like. X-ray training data 260 may also include data associated with each x-ray image, particularly measurement data. For example, when x-ray training data 260 includes teeth that have not yet erupted, x-ray training data 260 may also include measurement data associated with those teeth. The measurement data may include the crown-to-gingival distance of any unerupted tooth and / or the relative distance between the crown of the unerupted tooth and the mesial tooth, as well as the number of teeth. X-ray training data 260 may also include patient demographic data, including, but not limited to, age, gender, weight, race, geographic location, and the like. Any of the x-ray training data 260 images described herein may include panoramic x-ray images, bitewing x-ray images, periapical x-ray images, or a combination thereof.

[0109] Next, in block 304, processing node 210 may train one or more neural networks based on the input x-ray images to predict tooth eruption. In some examples, the input x-ray images may also include patient data, such as patient age, gender, weight, ethnic background, geographic location, etc. In some examples, the input x-ray images may also include dental measurement data that can be determined from the input x-ray images. As an example, the input x-ray images may include crown-to-gingival distance measurements for unerupted teeth.

[0110] Processing node 210 can train various neurons to recognize various aspects of x-ray training data 260 and predict associated tooth eruption times, as well as other dental data, such as component changes in erupted tooth geometry. In some examples, processing node 210 can execute or perform any feasible supervised or unsupervised learning algorithm to train the neural network. For example, processing node 210 can execute a linear classifier, a support vector machine, a decision tree, or an algorithm to predict tooth eruption from all available data associated with the input x-ray image.

[0111] In some other examples, processing node 210 may execute or perform any feasible regression algorithm (including a linear regression model) to train a neural network to predict tooth eruption and / or determine any associated erupted tooth geometry.

[0112] In some variations, prior to training any neural network, processing node 210 may adjust the contrast or brightness associated with any image of x-ray training data 260. Adjusting the contrast or brightness may enable processing node 210 to more easily detect any dental features (e.g., teeth, gums, soft tissue, etc.) in x-ray training data 260. The trained neural network may be stored in data storage module 240.

[0113] Next, in block 306, processing node 210 may use the patient data to refine the training of the neural network. This step may be optional, as Figure 3 , as shown by the dashed line in . For example, using a neural network, a clinician can predict the eruption time of the first tooth (at time T1). Additional patient x-rays performed at a later time T2 (T2 being after T1) can be obtained. Measurements associated with the first tooth (and any other teeth that have erupted or fallen out) can be obtained from the additional patient x-rays. Using these measurements, the trained neural network can be updated to provide a more accurate prediction of the eruption of the patient's teeth. In some variations, the slope of the fit can be used to determine the eruption velocity of the first tooth. Subsequently, the neural network can be updated to use data associated with the eruption velocity of the first tooth to predict tooth eruption.

[0114] Figure 4 Is shown using Figure 3 A flowchart of an example of a method 400 for predicting tooth eruption of a patient using a machine learning method. Figure 2 The method 400 is described using the machine learning tooth eruption prediction apparatus 200; however, the method 400 may be performed by any other suitable apparatus, system, or device.

[0115] Method 400 begins at box 402, where processing node 210 obtains or determines dental measurements from patient x-ray images 220. For example, x-ray images 220 may include one or more panoramic x-ray images of the patient. In some other examples, x-ray images 220 may include one or more bitewing or periapical x-rays of the patient. In some variations, processing node 210 may adjust the contrast or brightness associated with any x-ray images 220 of the patient. Adjustment of the contrast or brightness may enable processing node 210 to more easily detect any dental features (e.g., teeth, gums, soft tissue, etc.) in the x-ray images of the patient. Dental measurements may be determined by measuring distances between features using x-ray images 220. In some examples, dental measurements may include the crown-to-gingival distance of any unerupted tooth, and the associated number of teeth.

[0116] Next, in block 404, the intraoral scan can be used to refine the dental measurements. For example, in some cases, a panoramic x-ray may show one or more erupted teeth; however, the panoramic x-ray may lack sufficient detail to obtain dental measurements. The corresponding intraoral scan can be used to determine dental measurements of any feasible teeth that can be used to estimate tooth eruption. This step can be optional; for example, if the panoramic x-ray includes sufficient detail, block 404 can be skipped.

[0117] Next, in block 406, processing node 210 predicts tooth eruption based on the dental measurements determined from the patient x-rays. For example, processing node 210 may execute one or more machine learning based programs and / or neural networks to predict tooth eruption. Some neural networks may be as described with respect to Figure 3 The predicted tooth eruption may provide a probability of tooth loss in weeks or months from a reference time. The reference time may be associated with the time at which the patient x-ray scan was captured.

[0118] Next, in block 408, processing node 210 may obtain measurements from subsequent x-ray scans of the patient. The measurements obtained in block 408 may be similar to those described in blocks 402 and 404. However, these measurements are based on later x-ray scans. Using two or more time-separated x-ray scans may allow for more accurate prediction of tooth eruption and / or exfoliation. In some embodiments, block 408 may be optional.

[0119] Next, in block 410, processing node 210 updates the eruption prediction based on the subsequent measurements determined in block 408. Figure 3 As described, the tooth eruption or loss prediction may be updated based on a second, more recent set of dental measurements.

[0120] Figure 5 is a flow chart illustrating one example of a method 500 for predicting the position and direction of tooth eruption in a patient. Figure 2 The method 500 is described using a machine learning tooth eruption prediction device 200, however, the method 500 can be performed by any other suitable device, system or apparatus. Generally, two-dimensional (2D) x-ray images are easier to obtain than 3D cone beam computed tomography (CBCT) or conventional computed tomography (CT) images in dental clinics. In addition, 2D x-ray images can also have the advantage of low x-ray dose. However, it may be difficult to determine 3D information, particularly the direction and position of tooth eruption, from only 2D x-ray images. A method for updating 2D x-ray information with 3D-based image data to improve tooth eruption and shedding prediction is described herein.

[0121] Method 500 begins in block 502, where processing node 210 obtains or receives a reference 2D image and an associated reference 3D image. For example, the reference 2D image and the reference 3D image may be of the same person or model. Thus, the teeth, gums, and other dental anatomical structures of the two reference images may share a one-to-one correspondence. Furthermore, the reference 2D image and the reference 3D image may be prepared or formed into an "ideal dental arch." The ideal dental arch may describe the ideal position of teeth within the dental arch. Different ideal dental arches may exist for different patients based on age, gender, weight, or any other applicable demographic information.

[0122] In some examples, the 2D image may be a panoramic x-ray image, a series of bitewing x-ray images, or any other feasible 2D x-ray image. The 3D image may be a CBCT scan, a CT scan, or any other feasible 3D x-ray image.

[0123] Next, in block 504, processing node 210 maps the reference 2D image to the reference 3D image. In other words, elements of the reference 2D image may be mapped or transformed into the same elements in the reference 3D image. The transformation (mapping) performed in block 504 may be used to establish a transformation function between the 2D image and the associated 3D image.

[0124] Next, in block 506, the processing node 210 receives (or obtains) a 2D x-ray image of the patient. The 2D x-ray image of the patient may be a panoramic x-ray, a series of bitewing x-rays, or any other feasible 2D x-ray image. The 2D x-ray image of the patient may represent the starting point of the dental treatment received by the patient.

[0125] Next, in block 508, processing node 210 generates a predicted 2D dental model. For example, processing node 210 may warp (map) the patient's 2D x-ray image (received in block 506) to a reference 2D x-ray image. In some embodiments, the warping may map teeth or other dental anatomical structures in the patient's 2D x-ray to similar teeth and dental anatomical structures in the reference 2D image.

[0126] Next, in block 510, processing node 210 generates a predicted 3D tooth model. The processing node transforms (maps) the 2D tooth prediction from block 508 into a predicted 3D model. In some embodiments, processing node 210 may use a similar transformation function as established in block 504 to transform the 2D model into a 3D model. Using the predicted 3D dental model, a clinician can predict tooth eruption and / or exfoliation for any patient. For example, the predicted 3D tooth model may provide the direction and position of erupted and / or exfoliated teeth.

[0127] In some cases, the erupted teeth in the panoramic 2D image can be mapped to an intraoral scan of the patient, such as an iTero scan, and this reference then used to process the unerupted teeth. In the absence of 2D panoramic or CBCT images, the geometry of the unerupted teeth can be predicted using statistical shape modeling, in which the tooth geometry is predicted based on the principal component variations of the tooth geometry of each individual tooth. The variation in tooth geometry for each tooth can be decomposed into major modes of variation (e.g., crown width, cusp distance, molar socket depth). Each mode corresponds to a certain percentage of the variation in tooth geometry in the population, so using this method, a tooth geometry that captures a large portion of the population can be calculated and used in the appliance.

[0128] When taking any X-ray image, the position of the patient's head and the scanning parameters can affect the accuracy of the predicted model. In some cases, the user can physically measure tooth dimensions (e.g., incisor width) and then scale the scanned image. Scanning should also follow a protocol, and for model accuracy, the same scan settings should be used each time.

[0129] In some variations, tooth eruption, shedding, and / or placement can be predicted based on one or more facial features of the patient. For example, a machine learning neural network can be trained to predict the morphology of erupted and shedding teeth. Figure 6-8 An example of this method is described.

[0130] Figure 6 6 is a flow chart illustrating an example method 600 for training a neural network to predict tooth eruption based on facial features. Although described herein using facial features, the neural network can be trained to predict tooth eruption based on any other feasible physical features or characteristics. Figure 2 The method 600 is described using the machine learning tooth eruption prediction apparatus 200; however, the method 600 may be performed by any other suitable apparatus, system, or device.

[0131] Method 600 begins at block 602, where processing node 210 obtains or receives training data. The training data may include facial feature data and x-ray image data. The facial feature data may include any feasible location and measurement information of various facial features or landmarks. For example, facial features may include jaw width, head size, jaw height, etc. The x-ray image data may include 2D and / or 3D x-ray images and any measurement data that may be associated with each x-ray image. The measurement data may include the crown-to-gingival distance of any unerupted teeth, the number of teeth, tooth orientation, etc.

[0132] Next, in block 604, processing node 210 may train one or more neural networks to predict tooth eruption and / or exfoliation based on facial features. As described above, facial features may include any feasible facial features, landmarks, and any associated or related measurements. For example, processing node 210 may construct one or more linear regression models, tree-based models, etc., to train one or more neural networks to predict tooth morphology. Tooth morphology may include the length, height, and width of a bounding box, as well as other tooth parameters such as crown height and circumference.

[0133] Next, in block 606, the processing node 210 may apply a stochastic technique to the results or predictions of the trained neural network. For example, the various prediction models may predict different tooth geometries. It is crucial that any prediction model account for individual differences among patients.

[0134] Figure 7 A diagram of a face 700 is shown with possible locations of facial fiducials. In some cases, measurements between these fiducials can be used as Figure 6 Although the facial feature data described in Figure 6 One example of facial reference points is shown in , but in other variations, any number of similar or different facial reference points may be used.

[0135] Figure 8 Is displayed using Figure 6 A flowchart of an example of a method 800 for predicting tooth eruption of a patient using a machine learning method. Figure 2 The method 800 is described with reference to the machine learning tooth eruption prediction apparatus 200. However, the method 800 may be performed by any other suitable apparatus, system, or device.

[0136] The method begins at block 802 with processing node 210 obtaining or determining facial measurements of a patient. The facial measurements may be input by a user or may be determined by a processor (such as processing node 210) analyzing one or more facial photographs.

[0137] Next, in block 804, processing node 210 uses a machine language model to predict tooth eruption based on the facial measurements. For example, processing node 210 may execute one or more machine learning-based programs and / or neural networks to predict tooth eruption. Some neural networks may be as described in relation to Figure 6 The training is performed as described. The predicted tooth eruption can provide the probability of tooth loss in weeks or months from a reference time.

[0138] After determining an estimate of tooth eruption and / or loss, the clinician can create or modify a treatment plan that takes into account the patient's changing dentition. Treatment planning timing

[0139] In some cases, the clinician may be able to change the treatment time to accommodate tooth eruption and / or shedding. For example, if the prognosis indicates that a young tooth may be shedding within a month, the clinician may decide to delay starting treatment by about a month. Extraction of deciduous teeth

[0140] In some cases, a clinician may decide to extract a primary tooth (baby tooth) before starting a treatment plan. Tooth eruption / loss predictions can enable a clinician to determine when tooth eruption / loss is likely to occur relative to the treatment plan. For example, if a tooth is predicted to erupt in the middle of a treatment plan, the clinician may decide to remove the tooth before starting treatment. Additionally, some studies may show that baby teeth may be lost more quickly when a patient is treated with dental appliances. Therefore, it may be advantageous to remove baby teeth before starting a treatment plan that includes wearing dental appliances. Appliance modification

[0141] Knowing the predicted tooth eruption / loss time can enable the clinician to modify the treatment plan accordingly. For example, one or more dental appliances can be designed to accommodate the erupting teeth. Figures 9A-9D An example dental appliance designed to accommodate erupting teeth is shown. Figure 9A Shown is a dental appliance 900 comprising a cutout region 905. The cutout region 905 is centered at the location of the young tooth that is predicted to be lost. In this way, the cutout region 905 can be adapted to the patient's dentition before and after the eruption of the permanent teeth.

[0142] Figure 9B A primary teeth dental appliance 910 is shown that can be designed based on the dentition geometry of the primary teeth, particularly before the permanent teeth erupt. Thus, the primary teeth dental appliance 910 includes a conventional pocket piece 915 for the unerupted teeth. Figure 9C An eruption compensating appliance 920 designed according to the predicted position of the erupting permanent tooth is shown. The eruption compensating appliance 930 may include an eruption compensating pocket 925, which may be larger than a conventional tooth pocket 915. The eruption compensating pocket 925 may minimize the risk of tooth impaction.

[0143] Figure 9D Another possible variation of a dental appliance 930 is shown to accommodate erupting teeth. In this example, the dental appliance 930 can include a pocket 935 (shown in outline) that includes an undercut region 937 that can accommodate movement associated with erupting teeth.

[0144] Figure 10A block diagram of an apparatus 1000 is shown, which may be Figure 2 Although described herein as a device, the functions of the device 1000 can be performed by any feasible device, system or method. The device 1000 may include a communication interface 1010, a processor 1030 and a memory 1040.

[0145] The communication interface 1010, which can be coupled to a network and processor 1030, can transmit and receive signals to and from other wired or wireless devices, including remote (e.g., cloud-based) storage devices, cameras, processors, computing nodes, processing nodes, computers, mobile devices (e.g., cell phones, tablet computers, etc.), and / or displays. For example, the communication interface 1010 can include a wired (e.g., serial, Ethernet, etc.) and / or wireless (Bluetooth, Wi-Fi, cellular, etc.) transceiver that can communicate with any other available device via any available network. In some examples, the communication interface 1010 can receive training data 1041 and / or patient data 1042.

[0146] The processor 1030 , which is also coupled to the memory 1040 , may be any suitable processor or processors capable of executing scripts or instructions of one or more software programs stored in the device 1000 , such as within the memory 1040 .

[0147] Memory 1040 may include training data 1041. Training data 1041 may include a plurality of 2D and / or 3D x-ray images including associated patient characteristics. The x-ray images may include panoramic x-rays, bitewing x-rays, CT scans, CBCT scans, and the like. As described above, patient characteristics may include any feasible measurements associated with any feasible dental anatomical structure. By way of example and not limitation, dental characteristics may include the crown-to-gingival distance of any tooth, and tooth numbers identifying individual teeth. In some variations, patient characteristics may include patient demographic data, including gender, age, weight, ethnic background, geographic location, and the like. In some examples, the training data may include facial features or benchmarks and measurements associated with those facial features.

[0148] Memory 1040 may also include patient data 1042. Patient data 1042 may include one or more patient x-rays to be evaluated by apparatus 1000 to determine a prediction of the patient's erupted teeth. In some examples, patient data 1042 may include panoramic or bitewing x-rays. In some further examples, patient data 1042 may also include an image of the patient's face with two or more facial reference points. Patient data 1042 may include measurements associated with the facial reference points.

[0149] The memory 1040 may also include a non-transitory computer-readable storage medium (e.g., one or more non-volatile memory elements, such as EPROM, EEPROM, flash memory, hard drive, etc.) that may store a neural network training software (SW) module 1043, a neural network SW module 1044, an API 1045, a treatment planning SW module 1046, and a device manufacturing SW module 1047. Each software module includes program instructions that, when executed by the processor 1030, cause the apparatus 1000 to perform corresponding function(s). Thus, the non-transitory computer-readable storage medium of the memory 1040 may include instructions for performing all or part of the operations described herein.

[0150] The processor 1030 may execute the neural network training SW module 1043 to train one or more neural networks to perform Figure 2-8 In some examples, execution of the neural network training SW module 1043 may cause the processor 1030 to collect or obtain training data (such as x-ray images, facial images, and associated patient characteristics within the training data 1041) and train a neural network using the training data 1041. The trained neural network may be stored as one or more neural networks in the neural network training SW module 1043.

[0151] The processor 1030 may execute one or more neural networks in the neural network SW module 1044 to evaluate patient x-ray images (which may be stored in the patient data 1042) to determine a prediction of tooth eruption or loss. For example, execution of the neural network may evaluate a patient x-ray and determine a prediction of tooth eruption. In another example, execution of the neural network may evaluate an image of a patient's face to determine an estimate of tooth eruption.

[0152] Processor 1030 may execute instructions in API 1045 to receive a patient x-ray image, which may be included with patient data 1042, and output a treatment recommendation, which may be determined by executing neural network SW module 1044. In some variations, API 1045 may also output treatment plan and / or dental instrument data.

[0153] The processor 1030 can execute the treatment planning SW module 1046 to generate and store a patient treatment plan. In some examples, execution of the treatment planning SW module 1046 can also modify treatment preferences, generate modified treatment parameters, and / or generate a treatment plan based on predicted tooth eruption and / or exfoliation. Execution of the treatment planning SW module 1046 can access any applicable data in the memory 1040 (e.g., x-ray images, facial images, patient data, etc.).

[0154] The processor 1030 may execute the appliance manufacturing SW module 1047 to generate appliance data, which may then be used to manufacture one or more appliances. For example, execution of the appliance manufacturing SW module 1047 may use the patient data and treatment preferences stored in the memory 1040 to generate appliance data for a series of appliances. Example

[0155] In some examples, the methods and apparatus described herein can estimate predictions of eruption compensation, which can include one or more of: eruption time, location, direction, and / or geometry. The predictions can be provided directly as output, and / or they can be incorporated into treatment planning, such as (but not limited to) orthodontic appliance design that can incorporate the predicted information for treatment improvement.

[0156] Typically, these methods and devices can include time points as part of the eruption timing prediction, such as: t0, the earliest point in time that extraction should be recommended; t1, the last moment before treatment based on primary teeth hinders tooth eruption; and t2, the time at which the erupted permanent tooth needs to be included in the treatment. The time tolerance used to adjust treatment is typically t0 to t1.

[0157] Figure 11A A method for predicting tooth eruption based on an X-ray scan 1100 is shown. This example shows the relative distances between the crowns of teeth 5 and 6 (on the upper jaw) and the crown of the next mesial tooth. The crown-to-gingival distances of teeth 28 and 29 are shown on the lower jaw.

[0158] Use something like Figure 11A For an image (e.g., an x-ray image) as shown, the method or apparatus can determine a prediction based on an image (e.g., an x-ray scan) by first obtaining the position / distance (e.g., in mm) from the x-ray scan. Alternatively, if the absolute distance is unclear in the image (e.g., an x-ray scan), the image can be registered with an intraoral scan, and then the tooth dimensions in the scan can be used to obtain the absolute dimensions in the image (e.g., x-ray scan). Based on this, the crown-to-gingival distance d can be estimated. g You can also estimate the distance from the crown to the next mesial tooth crown (relative distance) d r . The method or apparatus may follow the order of eruption and assume that the next possible tooth is erupting. Starting from this simple model, the method and / or apparatus may assume a linear relationship with distance and time and predict the t1 of the next erupting tooth from the measured distance. Subsequently, the method or apparatus may predict the next erupting tooth from the relative distance. Any of these methods and apparatus may utilize existing data to calibrate the relationship between distance and time.

[0159] Any of the methods and apparatus described herein may make one or more of these predictions based on rescan times.

[0160] For example, if the rescan occurs at t1 of an erupted tooth, a linear regression machine learning model can be trained and used to predict the next t1 with features such as gender, age, and the next likely erupted tooth. For example, the prediction of t0 and t2 based on t1 can include modifying d by a preset distance. g :

[0161] t0:d=dg-d0

[0162] t2:d=dg+d2

[0163] The method or apparatus can then use the fitted slope (e.g., velocity) to determine the time, e.g. Figure 11B For example, using the current time interval:

[0164] t0=t1–dt0

[0165] t2=t1+dt2

[0166] Also described herein are methods and apparatus for using x-ray images (scans) to predict erupted permanent teeth, including 3D position and orientation. Figure 12 In the dental clinic, x-ray images may be easier to obtain than CT images and may also have the advantage of lower x-ray dose. However, it is often difficult to obtain 3D information, especially the direction and position of tooth eruption. The methods and devices described herein can be used to map 2D images to a reference 3D model (such as a 3D model). Figure 12 The patient and reference 2D images can then be warped to display a 2D image of the patient. The same warping can be performed on the reference 3D to provide the orientation and position of the erupted teeth. Alternatively, the erupted teeth in the panoramic 2D image can be mapped to an intraoral scan of the patient, such as an iTero scan, and the reference used to process the unerupted teeth.

[0167] In the absence of 2D panoramic or CBCT images, the geometry of unerupted teeth can be predicted using statistical shape modeling, in which the tooth geometry is predicted based on the principal component variations of the tooth geometry of each individual tooth. For example, the variation in tooth geometry of each tooth can be decomposed into major variation modes (e.g., crown width, cusp distance, molar groove depth). Each mode corresponds to a certain percentage of the variation in tooth geometry in the population, so using this method, a tooth geometry that captures the largest portion of the population can be calculated and used in the orthodontic appliance. This analysis can be performed on tooth 3 from 10 random dentitions. The geometric pattern shown in the video represents 32% of all tooth geometry variations in the dataset. Utilizing this form of analysis allows tooth geometry predictions to be made based on tooth position and can be further stratified by patient demographic data to provide more accurate predictions.

[0168] The position of the patient's head during the X-ray and the scanning parameters can affect the accuracy of the model. Physical measurements of tooth dimensions (e.g., incisor width) can be made and the scanned image scaled. Scanning should also follow a protocol, and for model accuracy, the same scan settings should be used each time.

[0169] Figure 13 A simplified dental arch 1300 is shown, which illustrates another approach for accommodating unerupted teeth for appliance design. The appliance can include a pocket, void, or "bubble area" that can receive one or more teeth as they erupt and grow into the dental arch. The bubble area can be a generally open area to allow the teeth to erupt unimpeded. Therefore, precise or accurate predictions about tooth eruption (size, direction, placement) may not be necessary. As shown, the dental arch 1300 can be associated with the upper jaw; however, in other examples, the dental arch 1300 can be associated with the lower jaw.

[0170] The example dental arch 1300 includes a bubble area 1330 (sometimes referred to as a void). As described above, the bubble area 1330 can be a portion of the dental arch 1300 that can accommodate one or more erupting teeth. While the bubble area 1330 shown can accommodate multiple erupting teeth, in other embodiments, the bubble area 1330 can be sized to accommodate only a single tooth. The bubble area 1330 can provide sufficient space for teeth to erupt and grow without damaging or interfering with dental appliances.

[0171] Generally, the bubble region 1330 can be at least partially defined by the mandibular arch splines. As shown, the dental arch 1300 can include a buccal mandibular arch spline 1310 and a lingual mandibular arch spline 1320. Generally, the mandibular arch splines can be described by a piecewise continuous polynomial that describes a curve passing through any predetermined point associated with the patient's teeth. The bubble region 1330 can fit between the buccal and lingual mandibular arch splines 1310, 1320.

[0172] In some embodiments, the bubble region 1330 may include details (e.g., shape) about the possible tooth contours of the erupted teeth. In particular, dental data from one or more of the patient's actual teeth may be used to provide additional jaw spline information for the tooth contours in the bubble region 1330. This will be discussed below in conjunction with Figure 14-17 Describe in more detail.

[0173] Figure 14 An example buccal-lingual plane 1400 is shown. A bubble region (such as Figure 13 The bubble area 1330 can be defined or determined by one or more buccal-lingual planes. Each buccal-lingual plane 1400 can be orthogonal to the dental arch. In addition, each buccal-lingual plane 1400 can include a curve 1401, which can be determined by the contours of one or more existing teeth in the dental arch. In some embodiments, the curve 1401 can predict a possible two-dimensional (2D) tooth contour of a tooth that can be erupted into the bubble area 1330.

[0174] Curve 1401 may include five points that define the shape of a 2D tooth profile. In some variations, curve 1401 may include fewer than five points, and in other variations, curve 1401 may include more than five points. In one example, curve 1401 may include a buccal gingival point 1410, a buccal cusp point 1420, a groove point 1430, a lingual cusp point 1440, and a lingual gingival point 1450. Each point 1410, 1420, 1430, 1440, and 1450 may be associated with any feasible characteristic or feature of a patient's teeth.

[0175] In some examples, points on the curve 1401 can be determined from a jaw arch spline associated with a specific point on the patient's existing teeth. For example, the buccal gum point 1410 and the lingual gum point 1450 can be associated with a jaw arch spline associated with the area where the teeth meet the area of ​​the patient's gums.

[0176] The buccal cusp point 1420 and the lingual cusp point 1440 can be associated with a mandibular arch spline that passes through a relatively high point on the upper tooth surface. Thus, the buccal cusp point 1420 can be associated with a high point on the tooth surface that is positioned toward the buccal side of the patient. The lingual cusp point 1440 can be associated with a high point on the tooth surface that is positioned toward the lingual side of the patient. The groove point 1430 can be associated with a mandibular arch spline that passes through the local lowest point on the tooth surface between the buccal cusp point 1420 and the lingual cusp point 1440.

[0177] To determine points 1410, 1420, 1430, 1440, and 1450, the associated mandibular arch splines can be projected through the buccal-lingual plane 1400. At the locations where the mandibular arch splines intersect the buccal-lingual plane 1400, the associated points are defined. For example, the buccal gingival mandibular arch spline can intersect the buccal-lingual plane 1400 at point 1410. Other points can be determined by the intersection of other mandibular arch splines and the buccal-lingual plane 1400. Curve 1401 is determined by connecting points 1410, 1420, 1430, 1440, and 1450. In some embodiments, curve 1401 can be formed using any feasible curve fitting operation. In other embodiments, curve 1401 can be a segmented linear curve (not shown).

[0178] Figure 15A and Figure 15B An exemplary use of buccal-lingual planes and mandibular arch splines is shown. Figure 15A A dental arch 1500 is shown. As shown, the dental arch 1500 may be missing several teeth in region 1520. For example, the missing teeth may be unerupted teeth of a young patient. Therefore, a dental appliance for the dental arch 1500 may require a void or air bubble region in region 1520.

[0179] To determine the contour, shape, and / or volume of the bubble area, one or more buccal-lingual planes 1530 may be projected into the area 1520. The example dental arch 1500 shows four buccal-lingual planes 1530, but in other examples, any number of buccal-lingual planes 1530 may be used. The buccal-lingual planes 1530 are generally orthogonal to the dental arch 1500. The mandibular arch splines may be projected from the patient's existing teeth through one or more buccal-lingual planes 1530. An exemplary mandibular arch spline 1510 is shown. In some examples, there may be five mandibular arch splines corresponding to the buccal gingiva, lingual gingiva, buccal cusps, lingual cusps, and alveolar mandibular arch splines. Other examples may include more, fewer, and different mandibular arch splines. The mandibular arch splines may be piecewise defined polynomial equations that are determined to pass through similar points on each existing tooth.

[0180] Because the projected mandibular arch spline is associated with prominent external features of the patient's existing teeth, the intersection of the mandibular arch spline with the buccal-lingual plane 1530 can reflect or predict the likely external features of the erupting teeth within region 1520 .

[0181] Figure 15B A side view of a dental arch 1500 is shown. Buccal-lingual planes 1530 may be spaced apart within region 1520. The jaw arch splines 1510 are shown intersecting the buccal-lingual planes 1530. Figure 14 As depicted, a plurality of jaw arch splines can be used to define a curve 1401 on the buccal-lingual plane 1530. The curve 1401 can be used to define the outline or shape of a void or bubble region of a dental appliance.

[0182] Figure 16 A partial dental arch 1600 is shown that can be displayed on a graphical user interface. The partial dental arch 1600 can include a first mandibular arch spline 1610, a second mandibular arch spline 1620, a third mandibular arch spline 1630, and a bubble area 1640. In some cases, the clinician or other user may be dissatisfied with the shape of the bubble area 1640. (Recall that the bubble area 1640 can be at least partially determined by the mandibular arch splines projected through multiple buccal-lingual planes). In some embodiments, the clinician or other user can adjust the position or trajectory of one or more mandibular arch splines to modify the bubble area 1640. In some cases, multiple control points (depicted here as points) associated with different mandibular arch splines can be displayed on the graphical user interface. The control points can provide the clinician or user with "handles" to interactively move, reposition, or adjust any mandibular arch spline. Thus, by moving the control points via the graphical user interface, the clinician or user can modify the shape of the bubble area 1640. As the clinician adjusts any mandibular arch spline, the bubble area 1640 can be updated on the graphical user interface.

[0183] As shown, first jaw spline 1610 may include one or more control points 1611, second jaw spline 1620 may include one or more control points 1621, and third jaw spline 1630 may include one or more control points 1631. Control points 1611, 1621, and 1631 may be distributed across the associated jaw arch splines in any feasible manner. In some examples, the number of control points may be proportional to the proposed size of bubble region 1640. Thus, a relatively larger bubble region may have more control points, while a relatively smaller bubble region may have fewer control points.

[0184] Figure 17 is a flow chart illustrating one example of a method 1700 for determining a gap in a dental appliance. Figure 2The method 1700 is described with reference to the apparatus 200; however, the method 1700 may be performed by any other suitable device, system, or apparatus.

[0185] The method begins at box 1702 where the processing node 210 determines one or more mandibular arch splines for a dental arch. In some examples, the dental arch information may be received via the API 250 or may be stored in the data storage module 240. The dental arch may be derived or determined from the x-ray image 220 or from any other data, including two-dimensional or three-dimensional dental scan data. The mandibular arch splines may be determined to fit any particular dental arch. In some variations, the one or more mandibular arch splines may be associated with particular aspects or characteristics of the teeth. Some aspects or characteristics may include buccal gingival points, lingual gingival points, buccal cusp points, lingual cusp points, and / or groove points or teeth. The processing node 210 may determine a polynomial equation for the piecewise definition of each mandibular arch spline.

[0186] Next, in block 1704, processing node 210 determines the location of one or more buccal-lingual planes. The buccal-lingual planes may be located (virtually located) in the region of the dental arch where teeth have or will have erupted. The buccal-lingual planes are typically orthogonal to the dental arch. Processing node 210 may place (define) any feasible number of buccal-lingual planes. In some cases, the buccal-lingual planes may be evenly spaced. In some other examples, the buccal-lingual planes may be unevenly spaced.

[0187] Next, in block 1706, the processing node 210 projects the mandibular arch spline determined in block 1702 through the buccal-lingual plane determined in block 1704. Where the mandibular arch spline intersects the buccal-lingual plane, a tooth contour (curve) may be defined.

[0188] Next, in block 1708, the processing node 210 determines a bubble region (e.g., a void) of the dental appliance based on the jaw arch spline and the buccal-lingual plane. More specifically, the processing node 210 can determine the volume or shape of the void based on the tooth contours (curves) determined by the intersection of the jaw arch spline and the buccal-lingual plane. In some examples, the shape of the void can be determined by the geometric union of the various tooth contours from each of the buccal-lingual planes.

[0189] In block 1710, the clinician or other user may adjust one or more jaw arch splines. Figure 17Indicated by the dashed line, this operation is optional. Next, in some examples, processing node 210 can display the bubble area (gap) and one or more jaw arch splines that contribute to the shape of the bubble area on a graphical user interface. If the clinician wants to modify the shape of the bubble area, the clinician can adjust one or more jaw arch splines. In some examples, processing node 210 can display one or more control points on the jaw arch splines (via a graphical user interface) with which the clinician can interact to modify their trajectory.

[0190] Next, in block 1712, the clinician re-evaluates the bubble area. If the clinician is satisfied with the modified bubble area, the method 1700 ends. On the other hand, if the clinician is not satisfied with the modified bubble area, the method returns to block 1710.

[0191] In some cases, the bubble area can interact with or otherwise interfere with the patient's bite. For example, teeth from the opposing jaw (opposing dental arch) may contact or otherwise collide with the bubble area. This interaction may cause discomfort and damage to the appliance. In some variations, the shape of the teeth from the opposing jaw can be removed from the bubble area. Removing the appliance material in the bubble area can reduce or eliminate unintended interactions or collisions with the patient's teeth.

[0192] Generally, these methods and apparatus (systems, devices, etc., including software, hardware and / or firmware) for predicting tooth eruption include the following: Figure 2-8 and Figure 10 Any of the components described herein, as shown and described herein, which may be collectively referred to as a tooth eruption prediction module, may be used in one or more portions of a dental computing environment, including as part of an intraoral scanning system, a physician system, a treatment planning (e.g., technician) system, a patient system, and / or a manufacturing system. In particular, these methods and apparatus may be used as part of a treatment planning system that integrates a dual-arch passive appliance into orthodontic treatment planning.

[0193] Figure 18 is a diagram illustrating a variation of a computing environment 1800 that can generate one or more orthodontic / dental appliances and / or patient-specific treatment plans and, under the guidance of a dental professional, manufacture dental appliances that can complete the treatment plan to treat the patient. Figure 18The example computing environment 1800 shown in includes an intraoral scanning system 1810, a doctor system 1820, a treatment planning system 1830 (e.g., a technician system), a patient system 1840, an instrument manufacturing system 1850, and a computer-readable medium 1860. Each of these systems can be equivalently referred to as a subsystem of the entire system (e.g., the computing environment). Although shown as discrete systems, some or all of these systems can be integrated and / or combined. In some variations, the computing environment (dental computing system) 1800 may include only one or a subset of these systems (which may also be referred to as subsystems of the entire system 1800). As described above, one or more of these systems can be combined or integrated with one or more other systems (subsystems), such as, for example, the patient system and the doctor system can be part of a remote server that can be accessed by the doctor and / or patient interface. The computer-readable medium 1860 can be divided between all or some of the systems (subsystems); for example, the treatment planning system and the instrument manufacturing system can be part of the same subsystem and can be on the computer-readable medium 1860. Additionally, each of these systems may be further divided into subsystems or components, which may be physically distributed (eg, between local and remote processors, etc.) or may be integrated.

[0194] An intraoral scanning system may include an intraoral scanner and one or more processors for processing images. For example, the intraoral scanning system 1810 may include an optical device 1811 (e.g., one or more lenses, filters, mirrors, etc.), (one or more) processors 1812, a memory 1813, a scan capture module 1814, and a result simulation module 1815. Typically, the intraoral scanning system 1810 can capture one or more images of a patient's dentition. The intraoral scanning system 1810 can be used in a clinical environment (doctor's office, etc.) or in an environment of the patient's choice (e.g., the patient's home). In some cases, the operation of the intraoral scanning system 1810 can be performed by an intraoral scanner, a dental camera, a cell phone, or any other feasible device.

[0195] The optical components 1811 may include one or more lenses and optical sensors to capture reflected light, particularly from the patient's dentition. The scan capture module 1814 may include instructions (such as non-transitory computer-readable instructions) that may be stored in the memory 1813 and executed by the processor(s) 1812 to control the capture of any number of images of the patient's dentition.

[0196] For example, an outcome simulation module 1815, which may be part of the intraoral scanning system 1810, may include instructions for simulating tooth positions based on the treatment plan. Alternatively or additionally, in some examples, the outcome simulation module 1815 may import tooth number information from the 3D model onto the 2D image to help determine the outcome simulation.

[0197] Any component system or subsystem of the dental computing environment 1800 can access or use the 3D model of the patient's dentition. In addition, any of these subsystems can use or access the above-mentioned (one or more) tooth eruption prediction modules 1839. For example, the doctor system 1820 may include a treatment management module 1821 and an intraoral state capture module 1822, which can access or use the 3D model and / or (one or more) tooth eruption prediction modules 1839. The doctor system 1820 can provide a "doctor-facing" interface to the computing environment 1800. The treatment management module 1821 can perform any operations that enable a doctor or other clinician to manage the treatment of any patient. In some examples, the treatment management module 1821 can provide a visualization and / or simulation of the patient's dentition in relation to the treatment plan. The user interface can also include the display and / or control of the eruption prediction as described above.

[0198] The intraoral state capture module 1822 may provide an image of the patient's dentition to the clinician via the physician system 1820. The image may be captured by the intraoral scanning system 1810 and may also include an image of a simulation of tooth movement based on the treatment plan.

[0199] In some examples, the treatment management module 1821 can enable the physician to modify or revise the treatment plan, particularly when the images provided by the intraoral status capture module 1822 indicate that the movement of the patient's teeth may not conform to the treatment plan. The physician system 1820 can include one or more processors configured to execute any applicable non-transitory computer-readable instructions to perform any applicable operations described herein.

[0200] Alternatively or additionally, the treatment planning system 1830 may include any of the methods and devices described herein. The treatment planning system 1830 may include a scan processing / refinement module 1831, a segmentation module 1832, a phasing module 1833, a treatment monitoring module 1834, a tooth eruption prediction module 1839, and (one or more) treatment plan databases 1835. In general, the treatment planning system 1830 can determine a treatment plan for any feasible patient. The scan processing / refinement module 1831 can receive or obtain dental scans (such as scans from the intraoral scanning system 1810) and can process the scans to "clean" them by removing scanning errors and, in some cases, enhancing the details of the scanned images. The treatment planning system 1830 can perform segmentation. For example, the treatment planning system may include a segmentation module 1832 that can segment the dental model into separate parts including separate teeth, gums, jaws, etc. In some cases, the dental model can be based on scan data from the scan processing / refinement module 1831.

[0201] The staging module 1833 can determine the different stages of the treatment plan. Each stage can correspond to a different dental appliance. The staging module 1833 can also determine the final position of the patient's teeth based on the treatment plan. Thus, the staging module 1833 can determine some or all of the patient's orthodontic treatment plan. In some examples, the staging module 1833 can simulate the movement of the patient's teeth according to the different stages of the patient's treatment plan. The staging module can also be integrated with any tooth eruption prediction module 1839 to account for tooth eruption and / or tooth loss.

[0202] The treatment monitoring module 1834 can monitor the progress of the orthodontic treatment plan. In some examples, the treatment monitoring module 1834 can provide an analysis of the progress of the treatment plan to the clinician.

[0203] Although not shown here, the treatment planning system 1830 may include one or more processors configured to execute any applicable non-transitory computer-readable instructions to perform any applicable operations described herein.

[0204] The patient system 1840 may include a treatment visualization module 1841 and an intraoral state capture module 1842. Generally, the patient system 1840 may provide a "patient-facing" interface to the computing environment 1800. The treatment visualization module 1841 may enable the patient to visualize how the orthodontic treatment plan is progressing and also visualize the predicted outcomes (e.g., the final positions of the teeth).

[0205] In some examples, the patient system 1840 can capture a dentition scan of the treatment visualization module 1841 via the intraoral state capture module 1842. The intraoral state capture module can enable the patient to capture his or her own dentition via the intraoral scanning system 1810. Although not shown here, the patient system 1840 can include one or more processors configured to execute any applicable non-transitory computer-readable instructions to perform any applicable operations described herein.

[0206] The appliance manufacturing system 1850 may include an appliance manufacturing machine 1851, a processor(s) 1852, a memory 1853, and an appliance generation module 1854. Generally, the appliance manufacturing system 1850 may directly or indirectly manufacture appliances to implement an orthodontic treatment plan. In some examples, the orthodontic treatment plan may be stored in a treatment plan database(s) 1835.

[0207] The device manufacturing machine 1851 may include any feasible device or equipment that can manufacture any suitable dental appliance. The appliance generation module 1854 may include any non-transitory computer-readable instructions that, when executed by the processor(s) 1852, can direct the device manufacturing machine 1851 to produce one or more dental appliances. The memory 1853 may store data or instructions for use by the processor(s) 1852. In some examples, the memory 1853 may temporarily store treatment plans, dental models, or intraoral scans.

[0208] The computer-readable medium 1860 may include some or all of the elements described herein with respect to the computing environment 1800. The computer-readable medium 1860 may include non-transitory computer-readable instructions that, when executed by a processor, may provide the functionality of any device, machine, or module described herein.

[0209] It should be appreciated that all combinations of the foregoing concepts and additional concepts discussed in greater detail below (provided such concepts are not mutually inconsistent) are considered to be part of the inventive subject matter disclosed herein and can be used to achieve the benefits described herein.

[0210] The process parameters and step sequences described and / or illustrated herein are provided as examples only and may be changed as needed. For example, although the steps illustrated and / or described herein may be shown or discussed in a particular order, these steps do not necessarily need to be performed in the order shown or discussed. The various example methods described and / or illustrated herein may also omit one or more of the steps described or illustrated herein, or include additional steps in addition to those disclosed.

[0211] Any of the methods described herein (including user interfaces) may be implemented as software, hardware, or firmware and may be described as a non-transitory computer-readable storage medium storing a set of instructions executable by a processor (e.g., a computer, tablet, smartphone, etc.), which, when executed by the processor, causes the processor to control the execution of any of the steps, including but not limited to: displaying, communicating with a user, analyzing, modifying parameters (including timing, frequency, intensity, etc.), determining, alerting, etc. For example, any of the methods described herein may be performed at least in part by an apparatus including one or more processors having a memory storing a non-transitory computer-readable storage medium storing an instruction set for the process(es) of the method.

[0212] A processor may include hardware that runs computer program code. Specifically, the term "processor" may include a controller and may encompass not only computers with different architectures such as single / multi-processor architectures and sequential (von Neumann) / parallel architectures, but also specialized circuits such as field programmable gate arrays (FPGAs), application specific circuits (ASICs), signal processing devices, and other devices.

[0213] Although various embodiments are described and / or illustrated herein in the context of a fully functional computing system, one or more of these example embodiments may be distributed as a program product in various forms, regardless of the specific type of computer-readable medium used to actually perform the distribution. The embodiments disclosed herein may also be implemented using software modules that perform certain tasks. These software modules may include scripts, batch processes, or other executable files that may be stored in a computer-readable storage medium or a computing system. In some embodiments, these software modules may configure a computing system to perform one or more of the example embodiments disclosed herein.

[0214] As described herein, the computing devices and systems described and / or illustrated herein broadly represent any type or form of computing device or system capable of executing computer-readable instructions, such as those contained in the modules described herein. In their most basic configuration, these computing devices may each include at least one memory device and at least one physical processor.

[0215] As used herein, the term "memory" or "memory device" generally refers to any type or form of volatile or non-volatile storage device or medium capable of storing data and / or computer-readable instructions. In one example, a memory device can store, load, and / or maintain one or more modules described herein. Examples of memory devices include, but are not limited to, random access memory (RAM), read-only memory (ROM), flash memory, hard disk drive (HDD), solid-state drive (SSD), optical disk drive, cache, variations or combinations of one or more thereof, or any other suitable storage memory.

[0216] Furthermore, the term "processor" or "physical processor" as used herein generally refers to any type or form of hardware-implemented processing unit capable of interpreting and / or executing computer-readable instructions. In one example, a physical processor can access and / or modify one or more modules stored in the aforementioned memory device. Examples of physical processors include, but are not limited to, a microprocessor, a microcontroller, a central processing unit (CPU), a field programmable gate array (FPGA) implementing a soft-core processor, an application-specific integrated circuit (ASIC), multiple portions of one or more of them, variations or combinations of one or more of them, or any other suitable physical processor.

[0217] Although shown as separate elements, the method steps described and / or illustrated herein may represent portions of a single application. Furthermore, in some embodiments, one or more of these steps may represent or correspond to one or more software applications or programs that, when executed by a computing device, may cause the computing device to perform one or more tasks, such as method steps.

[0218] Furthermore, one or more devices described herein may transform data, physical devices, and / or representations of physical devices from one form to another. Additionally or alternatively, one or more modules described herein may transform a processor, volatile memory, non-volatile memory, and / or any other portion of a physical computing device from one form of computing device to another form of computing device by executing on the computing device, storing data on the computing device, and / or otherwise interacting with the computing device.

[0219] As used herein, the term "computer-readable medium" generally refers to any form of device, carrier, or medium capable of storing or carrying computer-readable instructions. Examples of computer-readable media include, but are not limited to, transmission-type media (such as carrier waves) and non-transitory media (such as magnetic storage media (e.g., hard drives, tape drives, and floppy disks), optical storage media (e.g., compact disks (CDs), digital video discs (DVDs), and Blu-ray discs), electronic storage media (e.g., solid-state drives and flash memory media), and other distribution systems.

[0220] Those skilled in the art will recognize that any process or method disclosed herein may be modified in a variety of ways. The process parameters and step sequences described and / or illustrated herein are provided as examples only and may be modified as needed. For example, although the steps illustrated and / or described herein may be shown or discussed in a particular order, the steps do not necessarily need to be performed in the order shown or discussed.

[0221] The various exemplary methods described and / or shown herein may also omit one or more steps described or shown herein, or may also include other steps in addition to the disclosed steps. Additionally, the steps of any method disclosed herein may be combined with any one or more steps of any other method disclosed herein.

[0222] A processor as described herein may be configured to perform one or more steps of any method disclosed herein. Alternatively or in combination, a processor may be configured to combine one or more steps of one or more methods as disclosed herein.

[0223] When a feature or element is referred to herein as being "on" another feature or element, it can be directly located on the other feature or element, or there may be intervening features or elements. In contrast, when a feature or element is referred to as being "directly on" another feature or element, there are no intervening features or elements. It should also be understood that when a feature or element is referred to as being "connected," "attached," or "coupled" to another feature or element, it can be directly connected, attached, or coupled to the other feature or element, or there may be intervening features or elements. In contrast, when a feature or element is referred to as being "directly connected," "directly attached," or "directly coupled" to another feature or element, there are no intervening features or elements. Although described or illustrated with reference to one embodiment, the features and elements so described or illustrated may apply to other embodiments. Those skilled in the art will also recognize that references to structures or features set "adjacent" another feature may have portions that overlap with or are below the adjacent features.

[0224] The terms used herein are only used for the purpose of describing specific embodiments and are not intended to limit the present invention. For example, unless the context clearly indicates otherwise, the singular forms "a", "an" and "said" as used herein are intended to include the plural forms as well. It should also be understood that the terms "include" and / or "comprise" when used in this specification specify the presence of stated features, steps, operations, elements and / or parts, but do not exclude the presence or addition of one or more other features, steps, operations, elements, components and / or their groups. As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items and can be abbreviated as " / ".

[0225] Although the terms "first" and "second" may be used herein to describe various features / elements (including steps), these features / elements should not be limited by these terms unless the context indicates otherwise. These terms may be used to distinguish one feature / element from another. Thus, a first feature / element discussed below may be referred to as a second feature / element, and similarly, a second feature / element discussed below may be referred to as a first feature / element, without departing from the teachings of the present invention.

[0226] Throughout this specification and the appended claims, unless the context requires otherwise, the word "comprise" and variations such as "include" and "comprising" mean that various components can be employed together in methods and articles of manufacture (e.g., compositions and apparatus including devices and methods). For example, the term "comprising" will be understood to imply the inclusion of any stated elements or steps but not the exclusion of any other elements or steps.

[0227] In general, the devices and methods described herein should be understood to be inclusive, but all or a subset of components and / or steps may alternatively be exclusive and may be expressed as "consisting of" or alternatively "consisting essentially of" the various components, steps, subcomponents or sub-steps.

[0228] As used herein in the specification and claims, including in the examples, unless otherwise expressly stated, all numbers can be read as if beginning with the word "about" or "approximately", even if the term does not appear explicitly. When describing amplitude and / or position to indicate that the value and / or position described is within the reasonable expected range of value and / or position, the phrase "about" or "approximately" can be used. For example, a numerical value can have a value that is + / -0.1% of a set value (or range of values), + / -1% of a set value (or range of values), + / -2% of a set value (or range of values), + / -5% of a set value (or range of values), + / -10% of a set value (or range of values), etc. Unless the context indicates otherwise, any numerical value given herein is also understood to include being approximately that value or approximately that value. For example, if the value "10" is disclosed, "approximately 10" is also disclosed. Any numerical range described herein is intended to include all subranges contained therein. It should also be understood that, as will be appropriately understood by one skilled in the art, when a value is disclosed as "less than or equal to" that value, "greater than or equal to that value" and possible ranges between values ​​are also disclosed. For example, if a value "X" is disclosed, then "less than or equal to X" and "greater than or equal to X" (e.g., where X is a numerical value) are also disclosed. It should also be understood that throughout this application, data is provided in a variety of different formats and that this data represents endpoints and starting points and ranges for any combination of data points. For example, if a specific data point "10" and a specific data point "15" are disclosed, it should be understood that greater than, greater than or equal to, less than, less than or equal to, and equal to 10 and 15 are considered to be disclosed as well as between 10 and 15. It should also be understood that every unit between two specific units is also disclosed. For example, if 10 and 15 are disclosed, then 11, 12, 13, and 14 are also disclosed.

[0229] Although various illustrative embodiments have been described above, any of a number of changes may be made to the various embodiments without departing from the scope of the present invention as described in the claims. For example, in alternative embodiments, the order in which the various described method steps are performed may generally be changed, and in other alternative embodiments, one or more method steps may be skipped altogether. Optional features of the various device and system embodiments may be included in some embodiments and not included in other embodiments. Therefore, the foregoing description is provided primarily for exemplary purposes and should not be construed as limiting the scope of the invention as set forth in the claims.

[0230] The examples and descriptions included herein show, by way of illustration and not limitation, specific embodiments in which the subject matter may be practiced. As described above, other embodiments may be utilized and derived therefrom so that structural and logical replacements and changes may be made without departing from the scope of the present disclosure. For convenience only, such embodiments of the subject matter of the invention may be referred to herein individually or collectively by the term "invention," and are not intended to actively limit the scope of the present application to any single invention or inventive concept, if in fact more than one is disclosed. Therefore, although specific embodiments have been illustrated and described herein, any arrangement calculated to achieve the same purpose may replace the specific embodiments shown. The present disclosure is intended to cover any and all modifications or variations of the various embodiments. Upon reviewing the above description, those skilled in the art will appreciate the combination of the above embodiments and other embodiments not specifically described herein.

Claims

1. A method for predicting changes in a patient's dentition, the method comprising: extracting or receiving dental measurements from one or more x-ray images of a patient; predicting the eruption and / or loss of teeth in the patient's dentition based on the dental measurements using a trained neural network, wherein the trained neural network is trained using a plurality of training x-ray images and corresponding patient data to generate a probability value indicating the likelihood of eruption and / or loss of teeth at specific locations; as well as Output the probability value of the specific position. 2 . The method of claim 1 , further comprising generating a treatment plan based on the probability value.

3. The method according to claim 1 or 2, wherein: The dental measurement includes a crown-to-gingival distance of at least one tooth.

4. The method according to claim 1 or 2, wherein: The dental measurements include the relative distance between the unerupted tooth and the crown of the mesial tooth.

5. The method according to any one of claims 1 to 3, wherein The corresponding patient data includes patient age, patient gender, patient weight, patient race, patient geographic location, or a combination thereof.

6. The method according to any one of claims 1 to 5, wherein The corresponding patient data includes facial measurements associated with two or more facial reference points.

7. The method according to claim 2, wherein: The treatment plan includes a dental appliance configured to accommodate the predicted erupted teeth.

8. The method according to claim 7, wherein: The dental appliance includes a cutout region disposed in an area proximate the predicted erupted tooth.

9. The method according to claim 7, wherein: The dental appliance includes an enlarged pocket positioned in an area proximate the predicted erupted tooth.

10. The method according to claim 7, wherein: The dental appliance includes an undercut region that receives an erupted tooth.

11. The method according to any one of claims 1 to 7, wherein The x-ray images of the patient include panoramic x-rays, bitewing x-rays, or a combination thereof.

12. The method according to any one of claims 1 to 7, further comprising: receiving updated dental measurements determined from one or more subsequent x-ray images of the patient; as well as A prediction of a change in the patient's dentition is updated based at least in part on the updated dental measurements.

13. The method according to any one of claims 1 to 7, wherein The updated dental measurements include measurements determined from the intraoral scan.

14. An apparatus for predicting changes in a patient's dentition, the apparatus comprising: Communication interface; one or more processors; as well as a memory storing instructions that, when executed by the one or more processors, cause the apparatus to: extracting or receiving dental measurements from one or more x-ray images of a patient; predicting the eruption and / or loss of teeth in the patient's dentition based on the dental measurements using a trained neural network, wherein the trained neural network is trained using a plurality of training x-ray images and corresponding patient data to generate a probability value indicating the likelihood of eruption and / or loss of teeth at specific locations; as well as Output the probability value of the specific position.

15. A non-transitory computer-readable storage medium storing instructions that, when executed by one or more processors of a device, cause the processors to perform a method comprising: extracting or receiving dental measurements from one or more x-ray images of a patient; predicting the eruption and / or loss of teeth in the patient's dentition based on the dental measurements using a trained neural network, wherein the trained neural network is trained using a plurality of training x-ray images and corresponding patient data to generate a probability value indicating the likelihood of eruption and / or loss of teeth at specific locations; as well as Output the probability value of the specific position.

16. A method for predicting changes in a patient's dentition, the method comprising: receiving a reference 2D x-ray image and a corresponding reference 3D x-ray image, wherein the reference 2D x-ray image and the reference 3D x-ray image are based on an ideal dental arch describing ideal tooth positions; determining a transformation function to map the 2D x-ray image to the 3D x-ray image; receiving a 2D x-ray image of the patient; generating a predicted 2D model based on mapping the patient's 2D x-ray image to the reference 2D x-ray image; generating a predicted 3D model based on the transformation function; and A change in the patient's dentition is predicted based on the predicted 3D model.

17. The method of claim 16, further comprising generating a treatment plan based on predicted changes in the patient's dentition.

18. The method according to claim 16, wherein The changes in the patient's dentition include tooth eruption, tooth loss, or a combination thereof.

19. The method of claim 16, further comprising predicting the geometry of the unerupted teeth based on statistical modeling.

20. The method according to claim 16, wherein The 2D x-ray images of the patient are a panoramic x-ray, a series of bitewing x-rays, a series of periapical x-rays, or a combination thereof.

21. The method according to claim 16, wherein The ideal dental arch is selected based on patient demographics.

22. The method according to claim 16, wherein The 3D x-ray image is a computed tomography (CT) scan or a cone beam computed tomography (CBCT) scan.

23. An apparatus for predicting changes in a patient's dentition, the apparatus comprising: Communication interface; one or more processors; as well as a memory storing instructions that, when executed by the one or more processors, cause the apparatus to: receiving a reference 2D x-ray image and a corresponding reference 3D x-ray image, wherein the reference 2D x-ray image and the reference 3D x-ray image are based on an ideal dental arch describing ideal tooth positions; determining a transformation function to map the 2D x-ray image to the 3D x-ray image; receiving a 2D x-ray image of the patient; generating a predicted 2D model based on mapping the patient's 2D x-ray image to the reference 2D x-ray image; generating a predicted 3D model based on the transformation function, and A change in the patient's dentition is predicted based on the predicted 3D model.

24. A non-transitory computer-readable storage medium storing instructions that, when executed by one or more processors of a device, cause the processors to perform a method comprising: receiving a reference 2D x-ray image and a corresponding reference 3D x-ray image, wherein the reference 2D x-ray image and the reference 3D x-ray image are based on an ideal dental arch describing ideal tooth positions; determining a transformation function to map the 2D x-ray image to the 3D x-ray image; receiving a 2D x-ray image of the patient; generating a predicted 2D model based on mapping the patient's 2D x-ray image to the reference 2D x-ray image; generating a predicted 3D model based on the transformation function; and A change in the patient's dentition is predicted based on the predicted 3D model.

25. A method comprising: extracting or receiving dental measurements from one or more x-ray images of a patient; predicting the eruption of teeth in the patient's dentition based on the dental measurements using a trained neural network, wherein the trained neural network is trained using a plurality of training x-ray images and corresponding patient data to generate predicted shapes of erupted teeth at specific locations; as well as The predicted shape of the erupted tooth at the specific position is output.

26. The method of claim 25, further comprising generating a treatment plan based on the predicted shape.

27. The method according to claim 25, wherein The dental measurement includes a crown-to-gingival distance of at least one tooth.

28. The method according to claim 25, wherein The dental measurements include the relative distance between the unerupted tooth and the crown of the mesial tooth.

29. The method according to claim 25, wherein The corresponding patient data includes patient age, patient gender, patient weight, patient race, patient geographic location, or a combination thereof.

30. The method of claim 25, wherein: The corresponding patient data includes facial measurements associated with two or more facial reference points.

31. The method of claim 26, wherein: The treatment plan includes a dental appliance configured to accommodate the predicted erupted teeth.

32. The method according to claim 31, wherein The dental appliance includes a cutout region disposed in an area proximate the predicted erupted tooth.

33. The method according to claim 31, wherein The dental appliance includes an enlarged pocket positioned in an area proximate the predicted erupted tooth.

34. The method according to claim 31, wherein The dental appliance includes an undercut region that receives an erupted tooth.

35. The method of claim 25, wherein: The x-ray images of the patient include panoramic x-rays, bitewing x-rays, or a combination thereof.

36. A method comprising: positioning one or more jaw arch splines relative to the patient's dental arch in a virtual model of the patient's dental arch; positioning one or more buccal-lingual planes in an area of ​​the patient's dental arch to receive erupted teeth; For each buccal-lingual plane, determining a two-dimensional (2D) tooth contour based on a projection of the one or more jaw arch splines through the one or more buccal-lingual planes; and A void area of ​​the dental appliance is set based on the 2D tooth contours from the one or more buccal-lingual planes.

37. The method according to claim 36, wherein The one or more jaw arch splines pass through external points on the surface of the patient's existing teeth.

38. The method of claim 36, wherein: The one or more jaw arch splines are associated with at least one buccal or lingual gingival point on the surface of the patient's existing teeth.

39. The method according to claim 36, wherein The one or more jaw arch splines are associated with at least one of a buccal cusp, a lingual cusp, or a groove arch point on a surface of the patient's existing teeth.

40. The method of claim 36, wherein The one or more jaw arch splines are described by a piecewise continuous polynomial that describes a curve passing through predetermined points of the patient's existing teeth.

41. The method of claim 36, wherein: The one or more buccal-lingual planes are orthogonal to the patient's dental arch.

42. The method of claim 36, wherein: Determining the void includes forming a volume based on projections of the one or more jaw arch splines through the one or more buccal-lingual planes.

43. The method of claim 36, wherein: Determining the one or more jaw arch splines includes adjusting the one or more jaw arch splines by a clinician.

44. The method according to claim 43, wherein The adjustments are performed by a clinician interacting with a graphical user interface.

45. The method of claim 43, further comprising displaying the one or more jaw arch splines on a graphical user interface.

46. ​​The method of claim 36, further comprising displaying the dental appliance including the void on a display.

47. The method of claim 36, further comprising modifying the dental appliance to accommodate teeth from a different dental arch.

48. An apparatus for determining a gap in a dental appliance, the apparatus comprising: one or more processors; as well as a memory storing instructions that, when executed by the one or more processors, cause the apparatus to: positioning one or more jaw arch splines relative to the patient's dental arch in a virtual model of the patient's dental arch; positioning one or more buccal-lingual planes in an area of ​​the patient's dental arch to receive erupted teeth; For each buccal-lingual plane, determining a two-dimensional (2D) tooth contour based on a projection of the one or more jaw arch splines through the one or more buccal-lingual planes; and A void area of ​​the dental appliance is set based on the 2D tooth contours from the one or more buccal-lingual planes.

49. A non-transitory computer-readable storage medium storing instructions that, when executed by one or more processors of a device, cause the processors to perform a method comprising: positioning one or more jaw arch splines relative to the patient's dental arch in a virtual model of the patient's dental arch; positioning one or more buccal-lingual planes in an area of ​​the patient's dental arch to receive erupted teeth; For each buccal-lingual plane, determining a two-dimensional (2D) tooth contour based on a projection of the one or more jaw arch splines through the one or more buccal-lingual planes; and A void area of ​​the dental appliance is set based on the 2D tooth contours from the one or more buccal-lingual planes.

50. A method comprising: extracting or receiving dental measurements from one or more x-ray images of a patient; predicting the eruption of teeth in the patient's dentition based on the dental measurements using a trained neural network, wherein the trained neural network is trained using a plurality of training x-ray images and corresponding patient data to generate probability values ​​indicating the likelihood of eruption and / or exfoliation of teeth at specific locations; and A dental appliance is formed, the dental appliance being configured to be fitted over the patient's dentition and including an adaptation device to accommodate the predicted tooth eruption and / or shedding, wherein the adaptation device includes one or more of: an incision area configured to accommodate the predicted tooth eruption and / or shedding, a pocket configured to accommodate the predicted tooth eruption and / or shedding, and / or an undercut area configured to accommodate the predicted tooth eruption and / or shedding.

51. The method of claim 50, further comprising generating a treatment plan based on the probability value.

52. The method of claim 50, wherein: The dental measurement includes a crown-to-gingival distance of at least one tooth.

53. The method of claim 50, wherein: The dental measurements include the relative distance between the unerupted tooth and the crown of the mesial tooth.

54. The method of claim 50, wherein: The corresponding patient data includes patient age, patient gender, patient weight, patient race, patient geographic location, or a combination thereof.

55. The method of claim 50, wherein: The corresponding patient data includes facial measurements associated with two or more facial reference points.

56. The method of claim 50, wherein: The x-ray images of the patient include panoramic x-rays, bitewing x-rays, or a combination thereof.

57. The method of claim 50, further comprising: receiving updated dental measurements determined from one or more subsequent x-ray images of the patient; as well as A prediction of a change in the patient's dentition is updated based at least in part on the updated dental measurements.

58. The method of claim 57, wherein The updated dental measurements include measurements determined from the intraoral scan.

59. The method of claim 50, further comprising outputting a probability value for the specific location.

60. The method of claim 50, wherein: Forming the dental appliance includes forming the dental appliance by a direct manufacturing process.

61. The method of claim 50, wherein: The dental appliance includes a patient-removable appliance configured to fit over the patient's dentition.

62. The method of claim 50, wherein: Forming the dental appliance includes generating a digital model of the dental appliance and positioning the adaptive device in an area of ​​the dental appliance configured to be worn in the specific position.

63. A device comprising: one or more processors; as well as a memory storing instructions that, when executed by the one or more processors, cause the apparatus to perform the method, comprising: extracting or receiving dental measurements from one or more x-ray images of a patient; predicting the eruption of teeth in the patient's dentition based on the dental measurements using a trained neural network, wherein the trained neural network is trained using a plurality of training x-ray images and corresponding patient data to generate probability values ​​indicating the likelihood of eruption and / or exfoliation of teeth at specific locations; A dental appliance is formed, the dental appliance being configured to accommodate the predicted tooth eruption and / or shedding, wherein the dental appliance includes one or more of: a cutout region configured to accommodate the predicted tooth eruption and / or shedding, a pocket configured to accommodate the predicted tooth eruption and / or shedding, and / or an undercut region configured to accommodate the predicted tooth eruption and / or shedding.

64. A method comprising: positioning one or more jaw arch splines relative to the patient's dental arch in a virtual model of the patient's dental arch; positioning one or more buccal-lingual planes in an area of ​​the patient's dental arch to receive erupted teeth; For each buccal-lingual plane, determining a two-dimensional (2D) tooth contour based on a projection of the one or more jaw arch splines through the one or more buccal-lingual planes; as well as A dental appliance is formed, the dental appliance configured to be worn on a dental arch of the patient, wherein forming includes defining a void area of ​​the dental appliance based on the 2D tooth contours from the one or more buccal-lingual planes.

65. The method of claim 64, wherein The one or more jaw arch splines pass through external points on the surface of the patient's existing teeth.

66. The method of claim 64, wherein The one or more jaw arch splines are associated with at least one of a buccal gingival point or a lingual gingival point on a surface of the patient's existing teeth.

67. The method of claim 64, wherein The one or more jaw arch splines are associated with at least one of a buccal cusp, a lingual cusp, or a groove arch point on a surface of the patient's existing teeth.

68. The method of claim 64, wherein The one or more jaw arch splines are described by a piecewise continuous polynomial that describes a curve passing through predetermined points of the patient's existing teeth.

69. The method of claim 64, wherein The one or more buccal-lingual planes are orthogonal to the patient's dental arch.

70. The method of claim 64, wherein Determining the void includes forming a volume based on projections of the one or more jaw arch splines through the one or more buccal-lingual planes.

71. The method of claim 64, wherein: Determining the one or more jaw arch splines includes adjusting the one or more jaw arch splines by a clinician.

72. The method of claim 71, wherein The adjustments are performed by a clinician interacting with a graphical user interface.

73. The method of claim 71 further comprising displaying the one or more jaw arch splines on a graphical user interface.

74. The method of claim 64, further comprising displaying the dental appliance including the void on a display.

75. The method of claim 64, further comprising modifying the dental appliance to accommodate teeth from a different dental arch.

76. The method of claim 64, wherein Forming the dental appliance includes forming the dental appliance by a direct manufacturing process.

Images