Restorative decision support for dental treatments

Abstract

A method of providing repair decision support for a dental patient includes: receiving image data of an interior of an oral cavity of the patient, the image data corresponding to one or more imaging modes; deriving a plurality of parameters from the image data; applying the decision model to the plurality of parameters; and generating a repair decision recommendation based on the output of the decision model.

Description

Technical Field

[0001] Embodiments of the present disclosure relate to the field of dental diagnostics, and in particular to systems and methods for improving the process of providing restorative decisions. Background Art

[0002] In a typical dental practice, patients visit the dentist at least twice a year for a cleaning and exam. During this visit, the dental office may or may not generate a set of x-ray images of the patient's teeth. Additionally, the dental hygienist cleans the patient's teeth and notes any potential problem areas, which are then communicated to the dentist. The dentist then reviews the patient's history, reviews new x-rays (if any have been generated), and spends a few minutes examining the patient's teeth and gums during the patient exam. During the patient exam, the dentist may follow a checklist of different areas. The exam may begin by inspecting the patient's teeth for cavities, then review any existing restorations, then examine the patient's gums, then examine the patient's head, neck, and mouth for lesions or tumors, then examine the jaw joint, then examine the bite and jaw position and / or other orthodontic issues, and finally review any x-rays. Based on this review, the dentist determines whether there are any dental conditions that require immediate attention, as well as any other dental conditions that are not urgent but should ultimately be addressed and / or monitored. The dentist then needs to explain the identified dental condition to the patient and inform the patient of the risks, benefits, potential treatment or restoration options, alternatives, and the consequences of not treating the condition, and encourage the patient to make an informed treatment decision for their own health.

[0003] The goal of dentistry is to preserve as many of a patient's natural teeth as possible. If a tooth is damaged or decayed, a restoration is necessary to prevent tooth fracture, which may lead to root canal treatment or tooth extraction. Restorations are broadly divided into two categories: (i) direct restorations and (ii) indirect restorations. A direct restoration involves drilling out the damaged or decayed natural tooth material and using a "filling" made of amalgam, composite material, gutta-percha, gold foil, temporary fillings, or other materials. An indirect restoration can be used when a tooth is damaged or decayed and there is no longer enough natural tooth material to support a direct restoration. Indirect restorations include crowns, bridges, inlays, onlays, and veneers. A full coverage restoration (crown or cap) is used when the structural integrity of the tooth's cusp is compromised because the tooth is more susceptible to fracture when the cusp is weakened. If the cusp / tooth is fractured, the tooth may require root canal treatment / core build-up / crown, or in the worst case, extraction.

[0004] Current practice is for each physician to rely on their clinical judgment to determine the type of restoration. However, patients are often skeptical of a physician's recommendation for a crown or other restoration and may believe the recommendation is primarily motivated by profit. Alternatively, patients may not understand the consequences of not receiving treatment because the physician is unable to effectively communicate the need for treatment. Consequently, patients may refuse treatment or restoration that is in their best interest, even when objective evidence supports the physician's recommendation. Summary of the Invention

[0005] Several example implementations are summarized. Many other implementations are also contemplated.

[0006] In a first embodiment, a method of providing restorative decision support for a dental patient includes receiving image data of an intraoral cavity of the patient, the image data corresponding to one or more imaging modalities; deriving a plurality of parameters from the image data; applying a decision model to the plurality of parameters; and generating a restorative decision recommendation based on an output of the decision model.

[0007] In at least one embodiment, the one or more imaging modalities include an imaging modality selected from an intraoral scan, a radiograph, or a cone beam computed tomography (CBCT) scan.

[0008] In at least one embodiment, the one or more imaging modalities include intraoral scanning, and wherein the image data includes one or more three-dimensional (3D) point clouds, and at least one of: a two-dimensional (2D) near-infrared (NIR) image, a 2D ultraviolet image, or a 2D color image.

[0009] In at least one embodiment, the one or more imaging modalities include radiographs, and wherein the image data includes one or more of panoramic radiographs, bitewing radiographs, or periapical radiographs.

[0010] In at least one embodiment, the image data corresponds to two or more of the imaging modalities.

[0011] In at least one embodiment, the plurality of parameters are each selected from a geometric parameter, a volume / area parameter, or a fracture classification parameter.

[0012] In at least one embodiment, the geometric parameter includes an intercusp width.

[0013] In at least one embodiment, the volume / area parameters include one or more of a restoration volume fraction, a caries volume fraction, or a restoration surface fraction.

[0014] In at least one embodiment, the fracture classification parameters include information describing the location of the tooth fracture and the depth of the tooth fracture.

[0015] In at least one embodiment, the decision model includes one or more of a decision tree or a neural network.

[0016] In at least one embodiment, the restorative decision recommendation includes one or more of: a direct restorative recommendation or an indirect restorative recommendation, or an indication of a dental condition and a severity level of the dental condition.

[0017] In at least one embodiment, the dental condition is selected from the group consisting of caries, gum recession, tooth wear, malocclusion, crowded teeth, spaces between teeth, plaque, stains, cracked teeth, cervical lesions, and chipped or fractured teeth.

[0018] In at least one embodiment, the method further includes presenting the repair decision recommendation for display in a graphical user interface (GUI).

[0019] In at least one embodiment, the method further comprises storing the restorative decision recommendation and the actual restorative decision made by the dentist who provided the restorative decision recommendation in a record associated with the dentist in a key performance indicator (KPI) database.

[0020] In a second embodiment, a method includes: identifying teeth with associated dental conditions based on a first set of parameters derived from current image data of the interior of the patient's mouth; presenting a 2D or 3D image of the interior of the patient's mouth and an indication of the teeth with the associated dental conditions in a user interface (UI) (e.g., a graphical user interface (GUI)); and presenting a restoration decision recommendation in the GUI based on an output of a decision model using the first set of parameters as input.

[0021] In at least one embodiment, a restoration decision recommendation is presented in a GUI in response to a user selection of a tooth in a 2D or 3D image, and wherein the indication includes one or more of a label on the tooth, an outline on the tooth, or a color of the tooth.

[0022] In at least one embodiment, identifying a tooth with an associated dental condition includes: comparing a first set of parameters with a second set of parameters, the second set of parameters being derived from previous image data of the interior of the oral cavity captured prior to the current image data; and identifying the tooth by determining that a difference between a parameter from the first set of parameters and a parameter from the second set of parameters satisfies a threshold condition.

[0023] In at least one embodiment, the current image data corresponds to a first imaging mode, and wherein the previous image data corresponds to a second imaging mode different from the first imaging mode.

[0024] In a third embodiment, a method includes receiving image data corresponding to an interior of a patient's mouth; applying a trained machine learning model to the image data to derive a plurality of parameters from the image data; and applying a decision model to the plurality of parameters to generate a restorative decision recommendation.

[0025] In at least one embodiment, the trained machine learning model is adapted to calculate or estimate the volume of restorative material present on or in the tooth in the image data, and wherein deriving the plurality of parameters includes calculating at least one of a restoration volume or a surface ratio based on the estimated volume of restorative material.

[0026] In at least one embodiment, the trained machine learning model is adapted to receive as input image data corresponding to different imaging modalities, and wherein the different imaging modalities are independently selected from intraoral scans, radiographs, or cone beam computed tomography (CBCT) scans.

[0027] In another embodiment, a dental diagnostic system includes a memory and a processing device, the processing device being configured to execute instructions from the memory to perform the method of any of the aforementioned embodiments. For example, in at least one embodiment, the method includes: receiving image data of an interior of a patient's oral cavity, the image data corresponding to one or more imaging modalities; deriving a plurality of parameters from the image data; applying a decision model to the plurality of parameters; and generating a restoration decision recommendation based on an output of the decision model.

[0028] In another embodiment, an intraoral scanning system includes an intraoral scanner and a computing device operably connected to the intraoral scanner, wherein the computing device is configured to perform the method of any one of the preceding embodiments.

[0029] In another embodiment, a computer-readable medium includes instructions that, when executed by a processing device, cause the processing device to perform the method of any one of the preceding embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The present disclosure is illustrated by way of example and not by way of limitation in the figures of the accompanying drawings.

[0031] Figure 1A A user interface of a dental diagnostic hub is shown, in accordance with at least one embodiment of the present disclosure.

[0032] Figure 1B A user interface of a dental diagnostic center showing a time-lapse feature is shown, in accordance with at least one embodiment of the present disclosure.

[0033] Figure 1CA user interface of a dental diagnostic center is shown after a diagnostic has been run on a patient's dental arch, in accordance with at least one embodiment of the present disclosure.

[0034] Figure 1D A user interface for navigating diagnosis results provided by a dental diagnostic center to a mobile device is shown, according to at least one embodiment of the present disclosure.

[0035] Figure 1E A user interface of a dental diagnostic center is shown after a diagnostic has been run on a patient's dental arch, in accordance with at least one embodiment of the present disclosure.

[0036] Figure 2 A user interface for caries analysis in a dental diagnostic center according to at least one embodiment of the present disclosure is shown.

[0037] Figure 3 A user interface for amalgam analysis in a dental diagnostic center, according to at least one embodiment of the present disclosure, is shown.

[0038] Figure 4 An intraoral scanning system in accordance with at least one embodiment of the present disclosure is shown.

[0039] Figure 5 A 3D model of a lower dental arch is shown comparing predictions to a marked model to illustrate the segmentation quality between segmented and marked restorative material, in accordance with at least one embodiment of the present disclosure.

[0040] Figure 6 Classification of tooth caries regions comparing predictions to labeled models is shown, in accordance with at least one embodiment of the present disclosure.

[0041] Figure 7 Shown is a segmentation of the identified restorative material and the occlusal surface of the tooth, in accordance with at least one embodiment of the present disclosure.

[0042] Figure 8 A user interface of a key performance indicator dashboard according to at least one embodiment of the present disclosure is shown.

[0043] Figure 9 A flowchart of a method for generating a repair decision recommendation according to at least one embodiment of the present disclosure is shown.

[0044] Figure 10 A flowchart illustrating a method for presenting a restorative decision recommendation in a dental diagnostic center according to at least one embodiment of the present disclosure is shown.

[0045] Figure 11A flowchart of a method for generating repair decision recommendations based on parameters derived from image data using a trained machine learning model according to at least one embodiment of the present disclosure is shown.

[0046] Figure 12 A block diagram of an example computing device is shown in accordance with at least one embodiment of the present disclosure. DETAILED DESCRIPTION

[0047] Embodiments of a method for providing restorative decision support are described herein, which can be implemented, for example, as part of a dental diagnostic center. In at least one embodiment, a dentist or physician (terms used interchangeably herein) and / or their technician can collect various information about a patient. This information can include image data corresponding to one or more imaging modalities, including but not limited to an intraoral 3D scan of the patient's dental arch, an X-ray of the patient's teeth (e.g., optionally including a bitewing X-ray of the patient's teeth, a panoramic X-ray of the patient's teeth, a periapical and occlusal X-ray, etc.), a cone-beam computed tomography (CBCT) scan of the patient's jaw, an infrared image of the patient's teeth, or a color 2D image of the patient's teeth and / or gums. Other information can include but is not limited to biopsy information, malocclusion information, observation records of the patient's teeth and / or gums, etc. The intraoral scan can be generated by an intraoral scanner, and at least some of the other data can be generated by one or more devices other than the intraoral scanner. In addition, different data can be collected at different times. Each of the different data points can be useful for determining whether the patient has one or more types of dental conditions. The methods described herein can be used to derive multiple parameters (e.g., geometric parameters, volume / area parameters, or fracture classification parameters, each of which will be discussed in more detail below) from image data, and apply a decision model to the multiple parameters to generate a restorative decision recommendation that a physician can provide to a patient. The recommendation can be presented within a dental diagnostic center, which can provide a user interface that presents a unified view of the restorative decision recommendation, each analyzed dental condition type, which types of dental conditions may be worthy of the patient's attention, and which types of dental conditions may not be worthy of the patient's attention.

[0048] Certain embodiments utilize one or more imaging modalities to support physicians in making clinical decisions regarding restorations. For example, imaging modalities may include images generated from an intraoral scanner, radiographs (bitewing, periapical, or panoramic), CBCT scans, and the like. Restorative decision support systems and methods may use parameters derived from this information to provide a customizable objective measure of the health of the tooth and provide recommendations for direct restorations (e.g., fillings) or specific indirect restorations (e.g., inlays, onlays, crowns, bridges, or veneers). By providing objective evidence of the need for a direct or indirect restoration, the restorative decision support system may beneficially assist physicians in discussing restorative options with their patients, which may help physicians and patients make informed decisions while reducing the likelihood of tooth fractures that may result in root canal treatment, post and core construction, crowns, or extractions. In at least one embodiment, a customizable decision tree model is implemented to support the clinical decision-making process to identify teeth that require restoration and to help identify the type of restoration.

[0049] In at least one embodiment, a restoration decision system can identify teeth with associated dental conditions based on a set of parameters derived from image data of the interior of the patient's oral cavity. A 2D or 3D image of the interior of the oral cavity can be presented in a graphical user interface (GUI) (e.g., a GUI of a diagnostic center) along with an indication of the associated dental condition. The GUI can also present restoration decision recommendations based on the output of a decision model for which the set of parameters is used as input. The GUI also provides a user (e.g., a physician) with the option of visualizing the recommendations in response to a user selection of a tooth in the 2D or 3D image. For example, a tooth can be marked, outlined, or colored to draw attention to the tooth.

[0050] In at least one embodiment, a restoration decision system can receive image data corresponding to the interior of a patient's oral cavity (e.g., from a plurality of different imaging modalities). A trained machine learning model can be applied to the image data to derive a set of parameters from the image data. For example, the trained machine learning model can be adapted to calculate or estimate the volume of restorative material present on or within a tooth in the image data (e.g., a 2D or 3D image). Based on these estimates, one or more parameters can be derived, such as a restoration volume or a surface ratio. Subsequently, a decision model (e.g., a decision tree, a neural network, etc.) can be applied to these parameters to generate a restoration decision recommendation.

[0051] Traditionally, the various types of collected information are fragmented, with each type of information being stored in a separate system and accessed by a separate dental-related application. In order for a dentist to comprehensively assess a patient's dental health, they typically need to separately load and review each different type of data within each of the different dental-related applications specific to that type of data. The standard process for reviewing the data and making a diagnosis involves many manual steps on the dentist's part and requires a significant amount of effort and time on the dentist's part. Consequently, the standard process for a dentist to perform a comprehensive analysis of a patient's dental health is highly inefficient. Furthermore, once such a comprehensive analysis has been performed, it is often difficult for the dentist to present the comprehensive analysis to the patient, which again is a highly manual process on the dentist's part. What is lacking in conventional systems and what is provided in the embodiments described herein is a method for systematically and quickly and consistently collecting patient dental information, storing the information in a central repository, and deriving from this information various parameters that are useful in generating treatment and dental restoration recommendations.

[0052] In at least one embodiment, the dental diagnostic center brings together all of the distinct types of information related to a patient's dental health. The dental diagnostic center further performs automated analyses on each of the different types of dental conditions. Summary results of the various automated analyses can then be displayed together in a GUI. The summary results of the various automated analyses can include a severity rating for each type of dental condition. The summary results can identify those dental conditions with higher severity ratings, bringing them to the dentist's attention. The dentist can then select any type of dental condition to have the dental diagnostic center provide the patient with more detailed information about the selected type of dental condition.

[0053] In at least one embodiment, the Dental Diagnostic Center significantly increases the speed and efficiency of diagnosing a patient's dental conditions. The Dental Diagnostic Center enables dentists to identify all dental conditions that a patient may be concerned about with just a glance at the Dental Diagnostic Center's GUI. This enables dentists to easily and quickly prioritize the dental conditions to be addressed. In addition, the Dental Diagnostic Center can compare different identified dental conditions to determine any correlations between the different identified dental conditions. As a result, the Dental Diagnostic Center can identify some dental conditions as symptoms of other potential root cause dental conditions. For example, the Dental Diagnostic Center can identify dental crowding and caries formation caused by dental crowding. In addition, in at least one embodiment, the Dental Diagnostic Center creates presentations of the dental conditions, what would happen if those dental conditions were not treated, the root causes of the patient's dental conditions, treatment plan options, and / or simulations of treatment outcomes. Such presentations can be shown to patients to educate them about the condition of their dentition and their options for treating and / or not treating the problem.

[0054] A dental practitioner (e.g., a dentist or dental technician) can use an intraoral scanner to perform an intraoral scan of a patient's oral cavity. An intraoral scanning application running on a computing device operably connected to the intraoral scanner can communicate with the scanner to perform the intraoral scan and receive intraoral scan data (also referred to as intraoral images and intraoral scans). The result of the intraoral scan can be a sequence of intraoral scans that have been generated discretely (e.g., by pressing a "Generate Scan" button on the scanner for each image) or automatically (e.g., by pressing a "Start Scan" button and moving the intraoral scanner around the oral cavity while generating multiple intraoral scans). The operator can begin performing the intraoral scan at a first location in the oral cavity and move the intraoral scanner to various additional locations within the oral cavity until an intraoral scan has been generated for the entirety of one or more dental arches, or until a particular dental site has been completely scanned. In at least one embodiment, recording of the intraoral scan can be automatically started upon detection of a tooth or insertion into the oral cavity, and can be automatically paused or stopped upon detection of removal of the intraoral scanner from the oral cavity.

[0055] According to an example, a user (e.g., a dental practitioner) can subject a patient to an intraoral scan. In doing so, the user can apply the intraoral scanner to one or more locations within the patient's mouth. The scan can be divided into one or more segments. As an example, these segments can include the patient's lower cheek area, the patient's lower tongue area, the patient's upper cheek area, the patient's upper tongue area, one or more preparation teeth of the patient (e.g., the patient's teeth to which a dental device such as a crown or orthodontic device is to be applied), one or more teeth that serve as contacts for the preparation teeth (e.g., teeth that are not themselves subjected to a dental device but are located near one or more such teeth or engage with one or more such teeth when the mouth is closed), and / or the patient's occlusion (e.g., a scan is performed with the patient's mouth closed, wherein the scan is directed toward the joining area of ​​the patient's upper and lower teeth). In at least one embodiment, these segments include an upper dental arch segment, a lower dental arch segment, and a patient's occlusion segment. Via this scanner application, the scanner can generate intraoral scan data. A computing device executing the intraoral scanning application can receive and store the intraoral scan data.

[0056] Intraoral scan data can correspond to different imaging modes, which can include two-dimensional (2D) intraoral images (e.g., color 2D images), three-dimensional intraoral scans (e.g., intraoral images with depth information such as monochrome height maps), intraoral images generated using infrared or near-infrared (NIRI) light, and / or intraoral images generated using ultraviolet light. 2D color images, 3D scans, NIRI and / or infrared images and / or ultraviolet images can be generated by an intraoral scanner capable of generating each of these types of intraoral scan data. Such intraoral scan data can be provided from the scanner to a computing device in the form of one or more points (e.g., one or more pixels and / or groups of pixels). For example, the scanner can provide such intraoral scan data as one or more point clouds.

[0057] In at least one embodiment, an intraoral scan of a patient's oral cavity may be performed during a visit to a dentist's office. The intraoral scan may be performed as part of, for example, a semi-annual or annual dental health checkup. The intraoral scan may be a full scan of both the upper and lower dental arches and may be performed to gather information for performing a dental diagnosis. The dental information generated from the intraoral scan may include 3D scan data, 2D color images, NIRI and / or infrared images, and / or ultraviolet images.

[0058] In addition to performing intraoral scans, a dental practitioner may generate one or more other types of relevant dental health information, such as x-rays of the patient's teeth (e.g., optionally including bitewing x-rays of the patient's teeth, panoramic x-rays of the patient's teeth, etc.), cone-beam computed tomography (CBCT) scans of the patient's jaws, infrared images of the patient's teeth, color 2D images of the patient's teeth and / or gums that were not generated by an intraoral scanner (e.g., from photos taken by a camera), biopsy information, malocclusion information, observations of the patient's teeth and / or gums, and the like. For example, in addition to the intraoral scans of the oral cavity generated by the dental practitioner during an annual or semi-annual dentist appointment, the dental practitioner may also generate one or more x-rays of the patient's oral cavity during the dentist appointment. Additional types of dental information may also be collected when the dentist deems it appropriate to generate such additional information. For example, the dentist may collect a biopsy sample and send it to a laboratory for testing and / or may generate a panoramic x-ray and / or CBCT scan of the patient's oral cavity.

[0059] The intraoral scanning application can generate a 3D model (e.g., a virtual 3D model) of the patient's upper and / or lower dental arches based on the intraoral scan data. To generate the 3D model of the dental arch, the intraoral scanning application can register and stitch together the intraoral scans generated from the intraoral scanning session. In at least one embodiment, performing image registration includes capturing 3D data for various points of the surface in multiple intraoral scans and registering the intraoral scans by calculating a transformation between the intraoral scans. The intraoral scans can then be integrated into a common reference frame by applying an appropriate transformation to the points of each registered intraoral scan.

[0060] In at least one embodiment, registration is performed on each pair of adjacent or overlapping intraoral scans. For example, a registration algorithm can be performed to register two adjacent intraoral scans, which essentially involves determining a transformation that aligns one intraoral scan with the other intraoral scan. Registration can involve identifying multiple points in each intraoral scan (e.g., a point cloud) in a pair of intraoral scans, performing surface fitting on the points of each intraoral scan, and matching the points of the two adjacent intraoral scans using a local search around the points. For example, the intraoral scanning application can match points, edges, curvature features, rotation point features, etc. of one intraoral scan with the closest points, edges, curvature features, rotation point features, etc. interpolated on the surface of the other intraoral scan, and iteratively minimize the distance between the matched points. Registration can be repeated for each adjacent and / or overlapping scan to obtain a transformation to a common reference frame (e.g., rotation about one to three axes and translation in one to three planes). Using the determined transformation, the intraoral scanning application can integrate the multiple intraoral scans into a first 3D model of the lower dental arch and a second 3D model of the upper dental arch. The intraoral scan data may also include one or more intraoral scans showing the relationship between the upper dental arch and the lower dental arch. These intraoral scans may be used to determine the patient's occlusion and / or determine the patient's occlusal contact information. The patient's occlusion may include the determined relationship between the teeth in the upper dental arch and the teeth in the lower dental arch.

[0061] The intraoral scanning application or other application can further register data from one or more other imaging modes to the 3D model generated based on the intraoral scan data. For example, the processing logic can register x-ray images, CBCT scan data, ultrasound images, panoramic x-ray images, 2D color images, NIRI images, etc. to the 3D model. Each imaging mode in the different imaging modes can contribute different information about the patient's dentition. For example, NIRI images and x-ray images can identify caries, and color images can be used to add accurate color data to the 3D model, which can be used to determine tooth staining. The registered intraoral data from multiple imaging modes can be presented together in the 3D model, and / or presented side by side with one or more imaging modes that reflect the magnification and / or highlight and / or orientation of the 3D model shown. Data from different imaging modes can be provided as different layers, each of which can be used for a specific imaging mode. This can enable the doctor to turn on or off specific layers to visualize the dental arch with or without information from those specific imaging modes.

[0062] Figure 1A A user interface for a dental diagnostic center according to at least one embodiment of the present disclosure is shown. A 3D model of the patient's upper dental arch 140 and a 3D model of the patient's lower dental arch 141 can be generated by an intraoral scanning application and input into the dental diagnostic center. In addition, occlusal data showing the relationship between the upper dental arch 140 and the lower dental arch 141 can be input into the dental diagnostic center. The occlusal relationship data can also include how the upper and lower jaws dynamically relate to each other during functional movements, rather than just in a static relationship relative to each other. This can be useful for diagnosing problems related to the jaw joint, such as temporomandibular (TMJ) disorders. In addition, an intraoral scan of one or more prepared teeth of the patient may have been generated, which can also be input into the dental diagnostic center. The dental diagnostic center can then present a view of the upper dental arch 140, the lower dental arch 141, the prepared teeth, and / or the relative positions of the upper dental arch 140 and the lower dental arch 141 in the user interface of the dental diagnostic center.

[0063] Via the user interface of the dental diagnostic center, a practitioner can view one or more of the upper dental arch 140, the lower dental arch 141, a specific prepared tooth, and / or the patient's bite, each of which can be considered a separate scan segment or mode. The practitioner can select one or more scan segments to view via the scan segment selector 102. In at least one embodiment, as shown, the scan segment selector 102 can include an upper dental arch segment selection 105, a lower dental arch segment selection 110, and an occlusal segment selection 115. As shown, the upper dental arch segment selection 105 and the lower dental arch segment selection 110 are active, such that a 3D model of the upper dental arch 140 and a 3D model of the lower dental arch 141 are displayed. The practitioner can use the GUI to rotate the 3D model and / or change the zoom setting of the view of the 3D model.

[0064] The GUI of the dental diagnostic center may also include a diagnostic command 101. Selection of the diagnostic command 101 may cause the dental diagnostic center to perform one or more different analyses on the patient's dental arches 140, 141 and / or bite. The analyses may include an analysis for identifying cracks in the teeth, an analysis for identifying gum recession, an analysis for identifying tooth wear, an analysis of the patient's occlusal contacts, an analysis for identifying crowding of teeth (and / or spaces between teeth) and / or other malocclusions, an analysis for identifying dental plaque, an analysis for identifying tooth staining, an analysis for identifying caries, and / or other analyses of the patient's dentition. Once the analysis is complete, a dental diagnostic summary may be generated and displayed in the GUI of the dental diagnostic center, such as Figure 1C shown.

[0065] Some of the analyses performed to assess a patient's dental health are dental condition progression analyses, which compare the patient's dental condition at multiple different time points. For example, a caries assessment analysis may include comparing caries at a first time point and a second time point to determine the change (if any) in the severity of caries between the two time points. Other time-based comparative analyses that may be performed include time-based comparisons of gum recession, time-based comparisons of tooth wear, time-based comparisons of tooth movement, time-based comparisons of tooth discoloration, and the like. In at least one embodiment, processing logic automatically selects data collected at different time points to perform such time-based analysis. Alternatively, a user can manually select data from one or more time points for performing such time-based analysis.

[0066] Figure 1BA user interface for a dental diagnostic center according to at least one embodiment of the present disclosure is shown, illustrating a time-lapse feature 142. In at least one embodiment, the time-lapse feature is automatically initiated when a user selects the diagnostic command 101 to provide the user with the option of selecting which dental information from which time point to analyze. In at least one embodiment, the time-lapse feature 142 illustrates each of different time points (i.e., different timestamps) along a timeline at which dental information was collected. The selectable dental information may include at least intraoral scan data (e.g., 3D models generated based on one or more intraoral scanning sessions). The selectable dental information may also include x-rays generated at various time points, CBCT scan data generated at various time points, and / or other dental information generated at various time points. Via the time-lapse feature 142, the user can select one or more past data points (e.g., for a previously generated 3D model of a dental arch) and / or one or more current or recent data points (e.g., for a current 3D model of a dental arch). The selected data points can then be used to perform one or more time-based analyses on the patient's dentition.

[0067] In at least one embodiment, a time-based analysis of a patient's dentition compares 3D models and / or one or more dental conditions of the patient over time, and based on the comparison, identifies a dental condition and / or determines a rate of progression of one or more dental conditions. For example, 3D models of dental arches from different points in time can be compared to each other to determine the rate of progression of tooth wear, caries development, gum recession, gum swelling, malocclusion, etc. The rate of progression can be compared to a rate of progression threshold. The rate of progression threshold can be set by the physician or set to a default value. The amount of change in the dental condition can also be determined and compared to a rate of change threshold. Those dental conditions whose rate of progression meets or exceeds the rate of progression threshold for the dental condition and / or whose amount of change meets or exceeds the amount of change threshold can be identified as clinically significant dental conditions and / or dental conditions for which a problem or issue has been identified.

[0068] Time-based analysis can project the rate of progression or change of one or more detected dental conditions into the future to predict the severity level of the dental condition at a future point in time. In at least one embodiment, the progression of one or more dental conditions can be projected into the future, and the predicted dental condition at each projected time point can be compared to one or more criteria (e.g., such as a severity threshold). The one or more criteria can be default criteria and / or can be criteria set by a physician (e.g., a user of a dental diagnostic center). The criteria can also be set by aggregated data from the same practice or from a network of practices with similar patient profiles. When one or more criteria are met, it indicates that a clinically significant dental condition has been identified. In at least one embodiment, a future time point at which the projected dental condition will meet one or more criteria (e.g., by a severity threshold) can be recorded and added to the patient's record. In at least one embodiment, if the future time point at which the projected dental condition will meet one or more criteria for the dental condition is within a threshold amount of time from the current date, the dental condition can be identified as clinically significant or potentially clinically significant.

[0069] Figure 1C A user interface of a dental diagnostic center according to at least one embodiment of the present disclosure is shown, showing a dental diagnostic summary 103 after a patient's dental arch has been diagnosed. In at least one embodiment, the dental diagnostic summary 103 includes a scan segment selector 102 that includes an upper dental arch segment selection 105, a lower dental arch segment selection 110, and / or an occlusal segment selection 115. In at least one embodiment, the dental diagnostic summary 103 also includes a view of a selected dental segment or mode (e.g., a 3D model of the patient's upper dental arch 140 and a 3D model of the patient's lower dental arch 141).

[0070] The dental diagnosis summary 103 provides a single view showing multiple different types of possible dental conditions and an assessment of the presence and / or severity of each type of dental condition. In at least one embodiment, each dental condition is assigned one of three severity levels, including "No Problem Found" 145, "Potential Problem Found" 150, and "Problem Found" 155. Each dental condition can be coded or labeled with a severity rating determined for that type of dental condition. In at least one embodiment, the dental conditions are color-coded to graphically illustrate the severity level. For example, dental conditions with detected problems can be coded red, those with detected potential problems can be coded yellow, and dental conditions with no detected problems can be coded green. Many other coding schemes are also possible. In at least one embodiment, each dental condition is assigned a numerical severity level. For example, on a scale of 1 to 100, each dental condition can be assigned a severity level between 1 and 100 to indicate the severity level of the dental condition. Dental conditions with a severity level below a first threshold severity level can be identified as no problem dental conditions. Those dental conditions having a severity level that is higher than a first threshold severity level but lower than a second threshold severity level can be identified as dental conditions that are potential problem conditions. Those dental conditions having a severity level that is higher than a second threshold severity level can be identified as dental conditions that are problem conditions. In at least one embodiment, a different severity level threshold can be set for each of the different dental conditions. Alternatively, the severity levels of the different dental conditions can be standardized across multiple types of dental conditions, and the same severity level threshold can be used for multiple dental conditions. In at least one embodiment, the dental conditions are graded based on the severity level of the dental condition and / or based on the difference between the severity level of the dental condition and the associated severity level threshold.

[0071] In at least one embodiment, a physician can set a severity level threshold for one or more dental conditions. The severity level threshold that the physician can set can be a time-point severity level threshold that determines the time-point severity level of the dental condition based on data from a single time point. Additionally or alternatively, the severity level threshold that the physician can set can be a time-dependent threshold, such as an amount-of-change threshold and a rate-of-change threshold. Alarms can be set to alert the physician when threshold levels approach specific criteria.

[0072] In the absence of such a selected severity level threshold, a default severity level threshold may be automatically set for one or more of the dental conditions. In an example, the doctor may set a caries size threshold, and any detected caries whose size meets or exceeds the set caries size threshold may be identified as a found problem. In another example, the doctor may set a gum recession amount threshold, and any identified gum recession having a value that meets or exceeds the gum recession amount threshold may be identified as a found problem. The doctor may also set a rate of change threshold for one or more dental conditions, and / or such a rate of change threshold may be automatically set to a default value. For example, if the rate of change of tooth wear exceeds a tooth wear rate of change threshold, the patient may be identified as having an identified tooth wear problem. The doctor may also set a change amount threshold. If the detected change amount is greater than the set change amount threshold for the dental condition, the doctor may be alerted. Severity level thresholds can be set using a variety of different units, such as distance units (e.g., microns, millimeters, fractions of an inch, etc.), size units (e.g., microns, millimeters, fractions of an inch, etc.), rate of change units (e.g., microns / month, millimeters per year, etc.), brightness units, volume units (e.g., mm, etc.), and so on. 3 ), area units (e.g. mm 2 ), ratios, percentages (e.g., percent change), etc.

[0073] In at least one embodiment, the severity level threshold can depend at least in part on the location of the identified dental condition. For example, different caries severity thresholds can be set for different locations. Caries close to the dentin may be more urgent because they are more likely to cause pain and / or require root canal treatment than caries far from the dentin. Thus, for example, caries close to the dentin can have a lower threshold than caries far from the dentin. In at least one embodiment, the distance between the caries and the patient's dentin can be determined based on x-ray data, CBCT scans, and / or NIRI imaging of the interior of the mouth.

[0074] In at least one embodiment, the different types of dental conditions for which analysis is performed and included in the dental diagnostic summary 103 include tooth cracks, gum recession, tooth wear, occlusal contacts, crowding and / or spaces between teeth and / or other malocclusions, dental plaque, tooth staining, and caries. Additional, fewer, and / or alternative dental conditions may also be analyzed and reported in the dental diagnostic summary 103. In at least one embodiment, multiple different types of analyses are performed to determine the presence and / or severity of one or more dental conditions. One type of analysis that may be performed is a time point analysis that identifies the presence and / or severity level of one or more dental conditions at a particular point in time based on data generated at that point in time. For example, a single 3D model of a dental arch may be analyzed to determine whether, at a particular point in time, the patient's dental arch includes any caries, gum recession, tooth wear, problematic occlusal contacts, crowding, spaces or spaces between teeth, dental plaque, tooth staining, and / or tooth cracks. Another type of analysis that may be performed is a time-based analysis that compares dental conditions at two or more points in time to determine a change in a dental condition, a progression of a dental condition, and / or a rate of change in a dental condition, such as with reference to a dental diagnosis profile. Figure 1B As discussed. For example, in at least one embodiment, a comparative analysis is performed to determine differences between 3D models of a dental arch acquired at different time points. These differences can be measured to determine an amount of change, and this amount of change can be used, along with the time at which the intraoral scans used to generate the 3D model were acquired, to determine a rate of change. This technique can be used, for example, to identify the amount and / or rate of change of tooth wear, discoloration, plaque, crowding, gaps, gum recession, caries development, tooth cracks, and the like.

[0075] In at least one embodiment, one or more trained models are used to perform at least some of the one or more dental condition analyses. For example, the trained models may include physical models and / or machine learning models. In at least one embodiment, a single model may be used to perform multiple different analyses (e.g., to identify any combination of tooth cracks, gum recession, tooth wear, occlusal contacts, tooth crowding and / or spaces and / or other malocclusions, dental plaque, tooth staining and / or caries). Additionally or alternatively, different models may be used to identify different dental conditions. For example, a first model may be used to identify tooth cracks, a second model may be used to identify tooth wear, a third model may be used to identify gum recession, a fourth model may be used to identify problematic occlusal contacts, a fifth model may be used to identify crowding and / or spaces and / or other malocclusions of teeth, a sixth model may be used to identify dental plaque, a sixth model may be used to identify tooth staining, and / or a seventh model may be used to identify caries.

[0076] In at least one embodiment, intraoral data from one or more time points is input into one or more trained machine learning models that have been trained to receive intraoral data as input and output a classification of one or more types of dental conditions. In at least one embodiment, the trained machine learning model(s) are trained to identify an area of ​​interest (AOI) based on the input intraoral data and classify the AOI based on the dental condition. The AOI can be or include an area associated with a particular dental condition. For example, the area can include nearby or adjacent pixels or points that meet certain criteria. The intraoral data input into the one or more trained machine learning models can include three-dimensional (3D) data and / or two-dimensional (2D) data. The intraoral data can include, for example, one or more 3D models of a dental arch, one or more projections of one or more 3D models of a dental arch onto one or more planes (optionally including height maps), one or more x-rays of the teeth, one or more CBCT scans, panoramic x-rays, near-infrared and / or infrared imaging data, color images, ultraviolet imaging data, intraoral scans, and the like. If data from multiple imaging modalities is used (e.g., 3D scan data, color images, and NIRI imaging data), the data can be registered and / or stitched together so that the data is in a common reference frame and objects in the data are correctly positioned and oriented relative to objects in the other data. One or more feature vectors can be input into the trained model, where the feature vector includes multi-channel information for each point or pixel of the image. The multi-channel information can include color channel information from the color image, depth channel information from the intraoral scan data, a 3D model or a projected 3D model, intensity channel information from an x-ray image, and the like.

[0077] The trained machine learning model(s) may output a probability map, wherein each point in the probability map corresponds to a point in the intraoral data (e.g., a pixel in an intraoral image or a point on a 3D surface) and indicates a probability that the point represents one or more dental classes. In at least one embodiment, a single model outputs probabilities associated with multiple different types of dental classes, including one or more dental condition classifications. In an example, the trained machine learning model may output a probability map with probability values ​​for a tooth dental class and a gingival dental class. The probability map may further include probability values ​​for tooth cracks, gingival recession, tooth wear, occlusal contacts, crowding and / or spacing of teeth and / or other malocclusions, dental plaque, tooth staining, healthy areas (e.g., healthy teeth and / or healthy gums), and / or caries. In cases where a single machine learning model can identify each of tooth cracks, gum recession, tooth wear, occlusal contact, crowding and / or spacing of teeth and / or other malocclusions, dental plaque, tooth staining, and caries, 11 valued labels can be generated for each pixel, one for each of tooth, gum, healthy area, tooth cracks, gum recession, tooth wear, occlusal contact, crowding and / or spacing of teeth and / or other malocclusions, dental plaque, tooth staining, and caries. The corresponding predictions are probabilistic in nature: for each pixel, there are multiple numbers that can sum to 1.0 and can be interpreted as the probability that the pixel corresponds to these categories. In at least one embodiment, the first two values ​​for tooth and gum sum to 1.0, and the remaining values ​​for healthy area, tooth cracks, gum recession, tooth wear, occlusal contact, crowding and / or spacing of teeth and / or other malocclusions, dental plaque, tooth staining, and / or caries sum to 1.0.

[0078] In some instances, multiple machine learning models are used, where each machine learning model identifies a subset of possible dental conditions. For example, a first trained machine learning model can be trained to output a probability map having three values, one for each of healthy teeth, gums, and caries. Alternatively, a first trained machine learning model can be trained to output a probability map having two values, one for each of healthy teeth and caries. A second trained machine learning model can be trained to output a probability map having three values ​​(one for each of healthy teeth, gums, and tooth cracks) or two values ​​(one for each of healthy teeth and tooth cracks). One or more additional trained machine learning models can each be trained to output a probability map associated with identifying a particular type of dental condition.

[0079] In the case where the ML model is trained to identify three classes, it is convenient to store the predictions for such dental classes in RGB format. For example, the first value for the first dental class can be stored as a red intensity value, the second value for the second dental class can be stored as a green intensity value, and the third value for the third dental class can be stored as a blue intensity value. This makes it very easy to visualize the probability maps. Often, high precision is not required, and characters can be used instead of floating points, i.e., there are 256 possible values ​​for each channel of a pixel. Further optimizations can be performed to reduce size and improve performance (e.g., using 16 values ​​for quantization instead of 256).

[0080] The output of one or more trained machine learning models can be used to update one or more versions of the 3D model of the patient's upper and / or lower dental arch. In at least one embodiment, a different layer is generated for each dental condition classification. The layer can be turned on to graphically illustrate areas of interest on the upper and / or lower dental arch that have been identified or marked as having a specific dental condition.

[0081] If a probability map is generated for one or more input 2D images (e.g., such as a height map where pixel intensity represents height or depth), the probability map output by the (one or more) ML model(s) can be projected onto points in the virtual 3D model. Thus, each point in the virtual 3D model can include probability information from the probability map of one or more different intraoral images mapped to that point. In at least one embodiment, the probability information from the probability map is projected onto the 3D model as a texture. The updated 3D model can then include multiple sets of probabilities for one or more points, vertices, or voxels of the 3D model (e.g., vertices on a 3D mesh representing the surface of the 3D model), where different sets of probabilities associated with probability maps generated for different input images or other intraoral data can have different probability values.

[0082] The processing logic may modify the virtual 3D model by determining, for each point in the virtual 3D model, one or more dental classifications for the point. This may include using a voting function to determine the dental classification for each point. For example, each set of probability values ​​from the intraoral image may indicate a particular dental classification. The processing logic may determine the number of votes for each dental classification for a point and may then classify the point as having the dental classification that received the most votes. In at least one embodiment, a point may be associated with multiple dental conditions classified as such.

[0083] In at least one embodiment, image processing and / or 3D data processing can be performed on a 3D model of a dental arch generated from an intraoral scan and / or on the output of one or more trained models. Such image processing and / or 3D data processing can be performed using one or more algorithms that can be generic for multiple types of dental conditions or can be specific to a particular dental condition. For example, a trained model can identify areas on a 3D model of a dental arch that include caries, and image processing can be performed to assess the size and / or severity of the identified caries. Image processing can include performing automatic measurements, such as size measurements, distance measurements, change measurements, rate of change measurements, ratios, percentages, etc. Thus, image processing and / or 3D data processing can be performed to determine the severity level of the dental condition identified by the trained model(s). Alternatively, the trained model can be trained to classify an area as caries and identify the severity and / or size of the caries.

[0084] The one or more trained machine learning models used to identify, classify, and / or determine the severity level of a dental condition can be a neural network, such as a deep neural network or a convolutional neural network. In at least one embodiment, supervised training can be used to train such a machine learning model.

[0085] Artificial neural networks (e.g., deep neural networks and convolutional neural networks) typically include a feature representation component with a classifier or regression layer that maps features to a desired output space. For example, a convolutional neural network (CNN) hosts multiple layers of convolutional filters. Pooling is performed and nonlinear problems are solved at lower layers, and a multilayer perceptron is typically attached above the lower layers to map the top features extracted by the convolutional layers to decisions (e.g., classification outputs). Deep learning is a class of machine learning algorithms that uses a cascade of multiple layers of nonlinear processing units for feature extraction and transformation. Each successive layer uses the output from the previous layer as input. Deep neural networks can learn in a supervised (e.g., classification) and / or unsupervised (e.g., pattern analysis) manner. Deep neural networks include a hierarchy of layers in which different layers learn different representation levels corresponding to different levels of abstraction. In deep learning, each level learns to transform its input data into a slightly more abstract and complex representation. For example, in an image recognition application, the raw input might be a matrix of pixels; the first representation layer might abstract the pixels and encode edges; the second layer might combine and encode the arrangement of edges; the third layer might encode higher-level shapes (e.g., teeth, lips, gums, etc.); and the fourth layer might recognize that an image contains a face or define a bounding box around teeth in the image. Notably, the deep learning process can autonomously learn which features are best placed in which level. The "depth" in "deep learning" refers to the number of layers through which the data is transformed. More precisely, deep learning systems have a tangible credit assignment path (CAP) depth. The CAP is the chain of transformations from input to output. The CAP describes the possible causal relationships between input and output. For feedforward neural networks, the CAP depth can be the depth of the network, which can be the number of hidden layers plus one. For recurrent neural networks (where a signal can propagate through a layer more than once), the CAP depth can be infinite.

[0086] In at least one embodiment, a U-net architecture is used. U-net is a type of deep neural network that combines an encoder and a decoder with appropriate concatenation between them to capture both local and global features. The encoder is a series of convolutional layers that increase the number of channels while reducing height and width when processing from input to output, while the decoder increases height and width and reduces the number of channels. Layers with the same image height and width from the encoder can be concatenated with the output from the decoder. Any or all convolutional layers from the encoder and decoder can use traditional or depth-wise separable convolutions.

[0087] In at least one embodiment, the machine learning model is a recurrent neural network (RNN). An RNN is a type of neural network that includes a memory that enables the neural network to capture time dependencies. An RNN is able to learn input-output mappings that depend on both current input and past input. The RNN will process past and future intraoral data (e.g., intraoral scans taken at different times) and make predictions based on information spanning multiple time periods and / or patient visits. The RNN can be trained using a training data set to generate a fixed number of outputs. One type of RNN that can be used is a long short-term memory (LSTM) neural network.

[0088] A common architecture used for this task is LSTM (Long Short-Term Memory). Unfortunately, LSTM is not well suited for images because it does not capture spatial information as well as convolutional networks. For this purpose, ConvLSTM—a variant of LSTM that contains convolution operations within the LSTM cell—can be utilized. ConvLSTM is a variant of LSTM (Long Short-Term Memory) that contains convolution operations within the LSTM cell. ConvLSTM replaces matrix multiplication with convolution operations at each gate in the LSTM cell. By doing so, it captures the underlying spatial features through convolution operations in multi-dimensional data. The main difference between ConvLSTM and LSTM is the number of input dimensions. Since LSTM input data is one-dimensional, it is not suitable for spatial sequence data such as video, satellite, radar image datasets, etc. ConvLSTM is designed to take 3D data as its input. In at least one embodiment, a CNN-LSTM machine learning model is used. CNN-LSTM is an integration of CNN (convolutional layer) with LSTM. First, the CNN part of the model processes the data, and the one-dimensional result is fed to the LSTM model. In at least one embodiment, the network architecture for excess material removal can look like this Figure 11 A- Figure 11 As shown in B, it includes a ConvLSTM machine learning model.

[0089] In at least one embodiment, a class of machine learning models called MobileNet is used. MobileNet is an efficient machine learning model based on a streamlined architecture that uses depthwise separable convolutions to build a lightweight deep neural network. MobileNet can be a convolutional neural network (CNN) that can perform convolutions in both the spatial and channel domains. MobileNet can include a stack of separable convolution modules consisting of depthwise convolution and pointwise convolution (conv 1x1). Separable convolutions perform convolutions independently in the spatial and channel domains. This factorization of convolution can significantly reduce the total computational cost from HWNK2M to HWNK2 (depthwise) plus HWNM (conv 1x1), HWN(K2+M), where N represents the number of input channels, K2 represents the size of the convolution kernel, M represents the number of output channels, and HxW represents the spatial size of the output feature map. This can reduce the bottleneck of the computational cost of conv 1x1.

[0090] In at least one embodiment, a generative adversarial network (GAN) is used. A GAN is a type of artificial intelligence system that uses two artificial neural networks that compete with each other in a zero-sum game framework. The GAN includes a first artificial neural network that generates candidates and a second artificial neural network that evaluates the generated candidates. The GAN learns to map from a latent space to a specific data distribution of interest (a data distribution of variations of input images that are indistinguishable from photographs to the human eye), while the discriminator network discriminates between instances from a training dataset and candidates generated by the generator. The training goal of the generator network is to improve the error rate of the discriminator network (for example, by generating new synthetic instances that appear to be from the training dataset to deceive the discriminator network). The generator network and the discriminator network are trained together, with the generator network learning to generate images that are increasingly difficult for the discriminator network to distinguish from real images (from the training dataset), while the discriminator network simultaneously learns to better distinguish synthetic images from images from the training dataset. The two GAN networks are trained once they reach equilibrium. The GAN can include a generator network that generates artificial intraoral images and a discriminator network that segments artificial intraoral images. In at least one embodiment, the discriminator network can be a MobileNet.

[0091] In at least one embodiment, the machine learning model is a conditional generative adversarial (cGAN) network, such as pix2pix. These networks not only learn a mapping from an input image to an output image, but also a loss function used to train this mapping. GANs are generative models that learn a mapping from a random noise vector z to an output image y, G: z→y. In contrast, conditional GANs learn a mapping from an observed image x and a random noise vector z to y, G: {x, z}→y. The generator G is trained to produce outputs that are indistinguishable from "real" images by an adversarially trained discriminator D, which is trained to detect the generator's "fakes" as well as possible. In at least one embodiment, the generator can include a U-net or encoder-decoder architecture. In at least one embodiment, the discriminator can include a MobileNet architecture. An example of a cGAN machine learning architecture that can be used is the pix2pix architecture described in Isola, Phillip, et al., "Image-to-image translation with conditional adversarial networks," arXiv preprint (2017).

[0092] Training of neural networks can be achieved in a supervised learning manner, which involves feeding a training dataset consisting of labeled inputs through the network, observing its output, defining an error (by measuring the difference between the output and the labeled value), and using techniques such as deep gradient descent and backpropagation to adjust the network's weights across all its layers and nodes so that the error is minimized. In many applications, repeating this process for many labeled inputs in the training dataset produces a network that can produce correct outputs when presented with inputs different from those present in the training dataset. In high-dimensional settings such as large images, this generalization is achieved when a sufficiently large and diverse training dataset is available.

[0093] In order to train one or more machine learning models, a training dataset containing hundreds, thousands, tens of thousands, hundreds of thousands or more images (or multiple training datasets, one for each machine learning model to be trained) should be used to form a training dataset. In at least one embodiment, up to millions of cases of patient dentitions including one or more labeled dental conditions (such as cracked teeth, tooth wear, caries, gum recession, gum swelling, tooth staining, healthy teeth, healthy gums, etc.) are used, where each case may include a final virtual 3D model of a dental arch (or other dental part such as a portion of a dental arch). The machine learning model can be trained to automatically classify and / or segment the intraoral scan after the intraoral scanning session, and the segmentation / classification can be used to automatically determine the presence and / or severity of the dental condition.

[0094] For each 3D model with a marked dental classification, a set of images (e.g., height maps) can be generated. Each image can be generated by projecting the 3D model (or a portion of the 3D model) onto a 2D surface or plane. In at least one embodiment, different images of the 3D model can be generated by projecting the 3D model onto different 2D surfaces or planes. For example, a first image of the 3D model can be generated by projecting the 3D model onto a 2D surface at a top-down perspective, a second image can be generated by projecting the 3D model onto a 2D surface at a first side perspective (e.g., a buccal perspective), a third image can be generated by projecting the 3D model onto a 2D surface at a second side perspective (e.g., a lingual perspective), and so on. Each image can include a height map that includes depth values ​​associated with each pixel of the image. For each image, a probability map or mask can be generated based on the marked dental classification in the 3D model and the 2D surface onto which the 3D model is projected. The size of the probability map or mask can be equal to the pixel size of the generated image. Each point or pixel in the probability map or mask can include a probability value indicating the probability that the point represents one or more dental classifications. For example, there can be three dental classifications, including a first dental classification representing caries, a second dental classification representing healthy teeth, and a third dental classification representing gums. For example, a point with the first dental classification can have a value of (1,0,0) (the probability of the first dental classification is 100%, and the probability of the second and third dental classifications is 0%), a point with the second dental classification can have a value of (0,1,0), and a point with the third dental classification can have a value of (0,0,1).

[0095] A training dataset can be collected, where each data item in the training dataset can include an image (e.g., an image including a height map) and an associated probability map. Additional data can also be included in the training data items. The accuracy of the segmentation can be improved with the help of additional classifications, inputs, and multi-view support. Multiple sources of information can be incorporated into the model input and used jointly for prediction. Multiple dental classifications can be predicted simultaneously from a single model. Multiple problems can be solved simultaneously: tooth / gum segmentation, dental condition classification, etc. The accuracy is higher than traditional image and signal processing methods.

[0096] The additional data may include color images. For example, for each image (which may be monochrome), there may also be a corresponding color image. Each data item may include depth information (e.g., a height map) as well as color information (e.g., from a color image). Two different types of color images may be available. One type of color image is a viewfinder image and the other type of color image is a scan texture. A scan texture may be a combination or blend of multiple different viewfinder images. Each intraoral scan may be associated with a corresponding viewfinder image generated at approximately the same time as the intraoral image was generated. If hybrid scanning is used, each scan texture may be based on a combination of viewfinder images that are associated with the original scans used to produce the particular hybrid scan.

[0097] Another type of additional data may include images generated under specific lighting conditions (e.g., images generated under ultraviolet, near infrared, or infrared lighting conditions). The additional data may be 2D or 3D images and may or may not include depth information (e.g., a height map).

[0098] The result of such training is a function that can predict dental classifications directly from intra-oral data (e.g., a height map of intra-oral objects). Specifically, the machine learning model can be trained to generate a probability map, wherein each point in the probability map corresponds to a pixel of an input image and / or other input intra-oral data and indicates one or more of the following: a first probability that the pixel represents a first dental classification, a second probability that the pixel represents a second dental classification, a third probability that the pixel represents a third dental classification, a fourth probability that the pixel represents a fourth dental classification, a fifth probability that the pixel represents a fifth dental classification, and so on.

[0099] Based on the dental diagnosis summary 103, the dentist can select any type of dental classification. For example, the dentist can select any one of tooth cracks 134, caries 120, gum recession 122, amalgam 124, occlusion 126, crowding / spaces 128, dental plaque 130, and / or tooth staining 132. As discussed, a variety of different types of dental conditions can be displayed, and for each type of dental condition, a severity level for that dental condition can be shown. In the example shown, caries 120, amalgam 124, and crowding / spaces 128 are shown as having been identified as issues 155. Therefore, as a result of performing the caries analysis, tooth wear analysis, and crowding and / or space analysis, the severity levels of tooth crowding and / or space 128, amalgam 124, and caries 120 exceed the corresponding severity level thresholds. In the example shown, tooth staining 132 and occlusion 126 (eg, poor occlusal contact) are shown as potential issues 150 that have been identified, and tooth cracks 134 , gum recession 122 , and dental plaque 130 are shown as no issues 145 that have been identified.

[0100] After a quick glance at the dental diagnostic summary 103, the dentist may determine that the patient has caries, clinically significant tooth wear, crowding / spaces, and / or other malocclusions 128. Accordingly, the dentist may select the caries 120 view option, the amalgam 124 view option, and / or the crowding / spaces 128 view option to quickly review areas on the patient's dental arch where caries, amalgam, and / or crowding (and / or spaces) have been detected. The dentist may determine not to review gum recession, tooth cracks, or plaque for the patient because these dental conditions are classified as no problems found. The dentist may or may not review tooth staining and occlusal information because these dental conditions have been classified as potential problems found. Each dental condition shown may be shown with an icon, button, link, or selectable option that the user can select via the graphical user interface of the dental diagnostic center. Clicking on or otherwise selecting a specific dental condition may enable one or more tools associated with that particular dental condition.

[0101] The tools available for evaluating a selected dental condition may depend on the selected dental condition. For example, the evaluation tools available for tooth staining 132 may be different from the evaluation tools available for caries 120. Typically, one of the available tools associated with the selected dental condition includes a simulation of a prognosis of the dental condition. Through this simulation, the physician can determine what the region of interest (or regions of interest) exhibiting the dental condition looked like in the past and what they are predicted to look like in the future.

[0102] In at least one embodiment, the Dental Diagnostic Center, and in particular, the Dental Diagnostic Summary 103, helps doctors quickly identify dental conditions and their respective severity levels, helping doctors make better decisions regarding the treatment of dental conditions, and further helping doctors communicate with patients about their dental conditions and possible treatments. This makes the process of identifying, diagnosing, and treating dental conditions simpler and more efficient. Doctors can select any dental condition to determine the current prognosis for that condition and its likely future progression. In addition, the Dental Diagnostic Center can provide treatment simulations that demonstrate how one or more treatments will affect or eliminate the dental condition.

[0103] In at least one embodiment, a doctor can customize dental conditions and / or areas of interest by adding emphasis or annotations to specific dental conditions and / or areas of interest. For example, a patient may complain of pain in a specific tooth. The doctor can highlight the specific tooth on the 3D model of the dental arch. The dental conditions that are found to be associated with the specific highlighted or selected tooth can then be shown in the dental diagnosis summary. In another example, the doctor can select a specific tooth (e.g., the lower left molar) and can update the dental diagnosis summary by modifying the severity results to be specific to the selected tooth. For example, if a problem is found for caries and a possible problem is found for tooth staining for the selected tooth, the dental diagnosis summary 103 will be updated to show that no problem was found for amalgam 124, occlusion 126, crowding / space 128, plaque 130, tooth cracks 134, and gum recession 122 to show that a potential problem was found for tooth staining 132 and a problem was found for caries 120. This can help the doctor quickly identify the possible root cause of the patient's complaint of pain for the specific tooth selected. The doctor can then select a different tooth to obtain a summary of the dental problems of that other tooth. Additionally, the doctor can select a dental arch, a quadrant of a dental arch, or a group of teeth, and the dental diagnosis summary 103 can be updated to show the dental conditions associated with the selected group of teeth, quadrant of a dental arch, and / or dental arch.

[0104] Figure 1DA user interface for navigating diagnostic results provided by a dental diagnostic center to a mobile device 158 according to at least one embodiment of the present disclosure is shown. A dental diagnostic summary 103 generated by the dental diagnostic center can be sent to a patient's device, which can be a mobile device 158 or a traditional stationary device. Examples of mobile devices include mobile phones, tablets, laptops, and the like. Examples of traditional stationary devices include desktop computers, server computers, smart TVs, set-top boxes, and the like. In at least one embodiment, a link to the dental diagnostic summary 103 can be sent to the patient's device, and the patient can activate the link (e.g., click on the link) to access the dental diagnostic summary 103. In at least one embodiment, by selecting one or more dental conditions in the dental diagnostic summary 103, the patient can also access the underlying information summarized in the dental diagnostic summary 103. For example, this can show the patient which teeth exhibit a specific dental condition. The patient version of the dental diagnostic summary 103 can also include or be associated with a scheduling option or function 160. The patient can click on or otherwise select the scheduling option or function 160 to schedule an appointment. This can cause the mobile phone to call the dentist's office, for example, or can cause the patient's device to navigate to a calendar view showing available appointment times. The patient can click on or otherwise select an available appointment time to schedule an appointment with their dentist.

[0105] Figure 1E A user interface for a dental diagnostic center in accordance with at least one embodiment of the present disclosure is shown showing a dental diagnostic summary 161 after a diagnostic has been run on a patient's dental arch. The dental diagnostic summary 161 presents dental information about the patient that is organized differently than shown in the dental diagnostic summary 103. With the dental diagnostic summary 103, summary information for many different types of dental conditions is shown together without grouping the summary information based on dental category. On the other hand, the dental diagnostic summary 161 groups dental conditions based on dental category and indicates the specific type of dental condition or problem within each dental category. The dental condition information may also be arranged and presented in many different ways than the few examples shown herein.

[0106] In at least one embodiment, the dental diagnostic summary 161 includes a plurality of high-level dental categories or groupings, including a restorative / prosthodontic category 162, a TMJ category 188, an orthodontic category 174, a periodontal category 164, and an endodontic category 182. All restorative and / or prosthodontic dental conditions may be displayed under the restorative / prosthodontic category 162, all dental conditions associated with or caused by problems with the TMJ may be displayed under the TMJ category 188, all orthodontic dental conditions may be displayed under the orthodontic category 174, all periodontal dental conditions may be displayed under the periodontal category 164, and all endodontic dental conditions may be displayed under the endodontic category 182. Each of the high-level dental categories may be coded (e.g., color-coded) or otherwise include an indicator to show whether a dental condition falling under those high-level categories has been detected and / or the severity level of such a dental condition.

[0107] In at least one embodiment, one or more of the high-level dental categories (e.g., the restorative / prosthetic category 162, the TMJ category 188, the orthodontic category 174, the periodontal category 164, and the endodontic category 182) include summary information for subcategories and / or specific dental conditions that fall within the corresponding high-level dental category. In at least one embodiment, the TMJ category 188 includes a crack dental condition 190, an occlusal dental condition 192, and a tooth wear dental condition 194, which may correspond to, respectively, Figure 1C 134, occlusion 126, and amalgam 124. For each of the crack dental condition 190, occlusion dental condition 192, and tooth wear dental condition 194, the dental diagnostic summary 161 may indicate whether the patient's teeth are affected by the corresponding dental condition, which teeth are affected by the corresponding dental condition, and / or the severity of the corresponding dental condition. Each dental condition within the TMJ category 188 may be coded (e.g., color-coded) or otherwise include an indicator to show whether the corresponding dental condition has been detected and / or the severity level of such dental condition.

[0108] In at least one embodiment, the orthodontic category 174 includes a crowded dental category 176, a spaced dental category 178, and a jaw discrepancies dental category 180. The crowded dental category 176 may correspond to Figure 1CCrowding 128. The spacing dental category 178 can provide information about spaces or gaps between the patient's teeth. The jaw discrepancy dental category 180 can include information about issues with the patient's jaws, such as how the jaws close, overbites, underbites, overjet, and the like. For each of the crowding dental category 176, the spacing dental category 178, and the jaw discrepancy dental category 180, the dental diagnostic summary 161 can indicate whether the patient's teeth are affected by the corresponding dental condition, which teeth are affected by the corresponding dental condition, and / or the severity of the corresponding dental condition. Each dental condition within the orthodontic category 174 can be coded (e.g., color-coded) or otherwise include an indicator to show whether the corresponding dental condition has been detected and / or the severity level of such dental condition. Selecting any of the spacing dental category 178, the jaw discrepancy dental category 180, or the crowding dental category 176 can launch a dental analysis tool that displays the corresponding dental condition selected on the patient's dentition. From any of these dental analysis tools, an orthodontic tool can be launched.

[0109] In at least one embodiment, the periodontal category 164 includes an inflammation dental category 166, a bone loss dental category 170, and a gum recession dental category 167. The gum recession dental category 167 may correspond to Figure 1C The inflammation dental category 166 may include information regarding gum swelling or inflammation and / or the degree of swelling of one or more teeth. The bone loss dental category 170 may include information regarding bone density loss in one or more areas of the patient's jaw. For each of the gum recession dental category 167, the inflammation dental category 166, and the bone loss dental category 170, the dental diagnostic summary 161 may indicate whether the patient's teeth are affected by the corresponding dental condition, which teeth (e.g., tooth nos. 168, 169, 172) are affected by the corresponding dental condition, and / or the severity of the corresponding dental condition. Each dental condition within the periodontal category 164 may be coded (e.g., color-coded) or otherwise include an indicator to show whether the corresponding dental condition has been detected and / or the severity level of such dental condition. Selection of any of the gum recession dental category 167, the inflammation dental category 166, or the bone loss dental category 170 may launch a dental analysis tool that displays the corresponding dental condition selected on the patient's dentition.

[0110] In at least one embodiment, the endodontic category 182 includes one or more types of endodontic problems. Endodontic problems can include problems related to the root of the tooth and the soft tissue within the tooth, such as the pulp in the tooth. The endodontic category 182 can include endodontic conditions for one or more problem types 184, such as a first problem type for endodontic problems and a second problem type for root problems. For each problem type 184, one or more affected tooth numbers 186 can be indicated. Each dental condition within the endodontic category 182 can be coded (e.g., color coded) or otherwise include an indicator to show whether the corresponding dental condition has been detected and / or the severity level of such dental condition. Selection of any of the problem types 184 can launch a dental analysis tool that shows the corresponding dental condition selected on the patient's dentition.

[0111] In at least one embodiment, the restorative / prosthetic restoration category 162 includes one or more types of restorative and / or prosthetic restoration conditions. The term prosthetic restoration procedure refers, in particular, to procedures involving the oral cavity and directed to the design, manufacture, or installation of a dental prosthesis at a dental site within the oral cavity, or a real or virtual model thereof, or directed to the design and preparation of a dental site to receive such a prosthesis. For example, a prosthesis can include any restoration, such as a crown, veneer, inlay, onlay, and bridge, as well as any other artificial partial or complete denture. Prosthetic dental conditions or problems can include failing, failed, or broken / cracked prostheses, worn prostheses, loose prostheses, ill-fitting prostheses, and the like. Prosthetic dental conditions can also include conditions that can be corrected with a prosthesis, such as missing teeth, an edentulous dental arch, and the like. The restorative / prosthetic restoration category 162 can include existing prosthetic restoration conditions, which can constitute a first problem type 163, and conditions that can be addressed with a prosthetic restoration, which can constitute a second problem type 163. For each problem type 163, one or more affected tooth numbers 165 may be indicated. Each dental condition within the restorative / prosthetic restoration category 162 may be coded (e.g., color-coded) or otherwise include an indicator to show whether the corresponding dental condition has been detected and / or the severity level of such dental condition. Selection of any of the problem types 163 may launch a dental analysis tool that displays the corresponding dental condition selected on the patient's dentition.

[0112] In at least one embodiment, the dental diagnostic summary 161 includes a scan segment selector 102 that includes an upper dental arch segment selection 105, a lower dental arch segment selection 110, and / or an occlusal segment selection 115. In at least one embodiment, the dental diagnostic summary 103 also includes a view of a selected dental segment or mode (e.g., a 3D model of the patient's upper dental arch 140 and a 3D model of the patient's lower dental arch 141).

[0113] The dental diagnostic summary 161 provides a single view that shows multiple different types of possible dental conditions at both a high level and a low level, as well as an assessment of the presence and / or severity of each type of dental condition. In at least one embodiment, each dental condition is assigned one of three severity levels, including "no problem found," "potential problem found," and "problem found." Each of the dental conditions and / or dental categories (e.g., a high-level category that can include multiple potential conditions) can be coded or labeled with a severity rating determined for that type of dental condition. In at least one embodiment, the dental conditions and / or categories are color-coded to graphically illustrate the severity level. For example, those dental conditions and / or categories in which problems have been found can be coded red, those dental conditions and / or categories in which potential problems have been found can be coded yellow, and those dental conditions and / or categories in which no problems have been found can be coded green. Many other coding schemes are also possible. In at least one embodiment, each of the dental conditions and / or categories is assigned a severity level represented by a number. For example, on a scale of 1 to 100, each dental condition and / or category can be assigned a severity level between 1 and 100 to indicate the severity level of the dental condition. Those dental conditions and / or categories with a severity level below a first threshold severity level can be identified as non-problematic dental conditions. Those dental conditions and / or categories with a severity level above the first threshold severity level but below a second threshold severity level can be identified as potentially problematic dental conditions / categories. Those dental conditions and / or categories with a severity level above a second threshold severity level can be identified as problematic dental conditions / categories.

[0114] Figure 2A user interface for caries analysis in a dental diagnostic center, according to at least one embodiment of the present disclosure, is shown. The user interface for caries analysis can be provided in response to a physician selecting caries 120 from a dental diagnostic summary 103. In at least one embodiment, the caries analysis user interface includes a scan segment selector 102, which includes an upper arch segment selection 105, a lower arch segment selection 110, and / or an occlusal segment selection 115. In at least one embodiment, each tooth with an area of ​​interest associated with the selected dental condition is highlighted or otherwise emphasized or marked in the scan segment selector 102. Thus, a quick glance at the scan segment selector 102 can show the physician where to look further to review the AOI associated with the selected dental condition. In at least one embodiment, individual teeth can be selected in the scan segment selector 102 to view only those selected teeth. For example, the physician can select one or several teeth with AOIs to display a 3D model of only those teeth.

[0115] In at least one embodiment, the user interface for caries analysis also includes a view of a selected dental segment or mode. In the example shown, the lower dental arch is selected, and a 3D model of the patient's lower dental arch 141 is shown. An overlay of areas of interest (AOIs) reflecting detected caries is shown on the 3D model of the lower dental arch. For example, areas of interest 206A-F representing detected caries are shown on the 3D model of the lower dental arch 141. The upper dental arch segment can be selected to view the AOIs representing caries in the upper dental arch. In addition, both the upper and lower dental arches can be selected to show caries on both the upper and lower dental arches.

[0116] The physician can change the view of one or more displayed 3D models (e.g., the 3D model of the lower dental arch 141) via the user interface to better view the identified AOI. Such changes to the view can include changing the zoom setting (e.g., by zooming in or out), rotating the 3D model, panning left, right, up, down, etc., and the like. The physician can additionally use a focus tool to move the focus window 204 to any position on the 3D model to focus on an area of ​​the 3D model of the dental arch. Additional information from one or more additional imaging modes can be shown for the area within the focus window 204. For example, NIRI data for the area can be shown in the NIRI window 208, and color data for the area can be shown in the color window 210. For both the NIRI window 208 and the color window 210, the physician can zoom in or out and / or change the view of the area.

[0117] The physician can select the time-based simulation function to initiate a time-based simulation for a selected dental condition (e.g., for caries). The time-based simulation can use information about the AOIs as they exist at different time points from the patient's recorded history (e.g., intraoral scans, NIRI images, color images, x-rays, etc. from different time points) to project the progression of the dental condition into the future and / or past. The time-based simulation can generate a video that shows the onset of the dental condition and the progression of the dental condition over time to the current state of the dental condition and into the future. The time-based simulation can further include one or more treatment options and can show what the area of ​​interest will look like in the future after performing one or more selected treatments.

[0118] The user interface for caries analysis may indicate a caries severity level for each of the detected caries 206A-F. The severity level may be based on the size of the caries, the location of the caries, and / or the distance between the caries and the patient's dentin and / or pulp.

[0119] In at least one embodiment, a secure sharing mode may be provided in which a physician may securely collaborate with other care providers and / or may securely communicate with a patient (or a patient's parent) via a remote connection.

[0120] The physician may select a learning mode option (not shown) to bring up educational information about the differences between healthy teeth and teeth with caries, as well as the differences between different levels of caries severity. The patient's current dentition with currently detected caries may be shown, and further tooth decay may be predicted. The educational information may show what happens when the caries reaches the patient's dentin and / or pulp, indicating the amount of pain the patient can expect at various stages of tooth decay. Educational information may be shown to the patient to show the patient the stage of tooth decay in their teeth and what will happen if they do not treat the tooth decay.

[0121] Once the physician completes review of the patient's caries information, the physician may select the dental diagnosis summary view icon or navigation option 202 to navigate back to the dental diagnosis summary 103 .

[0122] Figure 3A user interface for amalgam analysis in a dental diagnostic center, according to at least one embodiment of the present disclosure, is shown. The user interface for amalgam analysis can be provided in response to a physician selecting amalgam 124 from a dental diagnostic summary 103. In at least one embodiment, the amalgam analysis user interface includes a scan segment selector 102, which includes an upper arch segment selection 105, a lower arch segment selection 110, and / or an occlusal segment selection 115. In at least one embodiment, each tooth with an area of ​​interest associated with the selected dental condition is highlighted or otherwise emphasized or marked in the scan segment selector 102. Thus, a quick glance at the scan segment selector 102 can show the physician where to look further to review the AOI associated with the selected dental condition. In at least one embodiment, individual teeth can be selected in the scan segment selector 102 to view only those selected teeth. For example, the physician can select one or several teeth with AOIs to display a 3D model of only those teeth.

[0123] In at least one embodiment, the user interface for amalgam analysis also includes a view of a selected dental section or mode. In the example shown, the lower dental arch is selected, and a 3D model of the patient's lower dental arch 341 is shown. An overlay of areas of interest (AOIs) reflecting detected amalgam is shown on the 3D model of the lower dental arch. For example, areas of interest 302A-B representing areas where amalgam is detected are shown on the 3D model of the lower dental arch 341. The upper dental arch section can be selected to view the AOIs representing amalgam in the upper dental arch. In addition, both the upper and lower dental arches can be selected to show amalgam on both the upper and lower dental arches. The physician can change the view of one or more displayed 3D models (e.g., the 3D model of the lower dental arch 341) via the user interface to better view the identified AOIs. Such changes to the view can include changing the zoom setting (e.g., by zooming in or out), rotating the 3D model, translating left, right, up, down, etc., and the like. The physician can additionally use the focus tool to move the focus window 304 to any location on the 3D model to focus on an area of ​​the 3D model of the dental arch. Additional information from one or more additional imaging modalities can be shown for the area within the focus window 304. Once the physician has completed reviewing the patient's tooth wear information, the physician can select the dental diagnosis summary view icon or navigation option 202 to navigate back to the dental diagnosis summary 103.

[0124] Other types of user interfaces are also contemplated, such as user interfaces for tooth wear analysis, occlusal contact analysis, malocclusion analysis, tooth staining analysis, and teeth after bleaching. Such user interfaces may be similar to the user interface described in U.S. Patent Publication No. 2022 / 0202295, the disclosure of which is incorporated herein by reference in its entirety. Dentists may be shown views for gum swelling, plaque, tooth cracks, and / or gum recession similar to those shown for caries and tooth wear. In addition, dental condition analysis tools similar to those provided for caries and / or tooth wear may be provided for gum swelling, plaque, tooth cracks, and / or gum recession. For example, a gum swelling analysis tool may project the amount of gum swelling into the future and show inflammation of the gums, bleeding gums, and the like. Similarly, a gum recession analysis tool may project the amount of gum recession into the future, show exposed portions of the tooth roots, and the like.

[0125] Figure 4 One embodiment of a system 400 for performing intraoral scanning, generating a virtual three-dimensional model of a dental site, and / or performing dental diagnostics is shown. In at least one embodiment, the system 400 performs the following reference Figures 1A-3 and Figure 5-Figure 11 System 400 includes a computing device 405 that can be coupled to a scanner 450 and / or a data store 410 .

[0126] The computing device 405 may include a processing device, a memory, a secondary storage device, one or more input devices (e.g., such as a keyboard, a mouse, a tablet, etc.), one or more output devices (e.g., a display, a printer, etc.), and / or other hardware components. The computing device 405 may be connected to the data storage 410 directly or via a network. The network may be a local area network (LAN), a public wide area network (WAN) (e.g., the Internet), a private WAN (e.g., an intranet), or a combination thereof. In some embodiments, the computing device 405 may be integrated into the scanner 450 to improve performance and mobility.

[0127] Data storage 410 may be an internal data storage or an external data storage connected to computing device 405 directly or via a network. Examples of network data storage include storage area networks (SANs), network attached storage (NASs), and storage services provided by cloud computing service providers. Data storage 410 may include a file system, a database, or other data storage arrangements.

[0128] In at least one embodiment, a scanner 450 for obtaining three-dimensional (3D) data of a dental site in a patient's mouth is operatively connected to the computing device 405. The scanner 450 may include a probe (e.g., a handheld probe) for optically capturing a three-dimensional structure (e.g., by confocal focusing of an array of beams). An example of such a scanner 450 is manufactured by Align Technology, Inc. Intraoral digital scanner. Other examples of intraoral scanners include 8M TM True Definition Scanner and Apollo DI intraoral scanner and CEREC AC intraoral scanner manufactured by.

[0129] Scanner 450 can be used to perform intraoral scans of a patient's oral cavity. An intraoral scanning application 408 running on computing device 405 can communicate with scanner 450 to perform the intraoral scan. The result of the intraoral scan can be a series of generated intraoral images or scans. Each intraoral scan can include x, y, and z position information for one or more points on the surface of the scanned object. In at least one embodiment, each intraoral scan includes a height map of the surface of the scanned object. An operator can initiate a scanning operation with scanner 450 at a first position in the oral cavity, move scanner 450 within the oral cavity to a second position while performing the scan, and then stop recording the intraoral scan. In at least one embodiment, recording can automatically begin when the scanner identifies any tooth. Scanner 450 can transmit the intraoral scan to computing device 405. Computing device 405 can store current intraoral scan data 435 from the current scanning session in data storage 410. The data store 410 may additionally include past intraoral scan data 438, additional current dental data 445 generated during the current patient visit (e.g., x-ray images, CBCT scan data, panoramic x-ray images, ultrasound data, color photographs, etc.), additional past dental data generated during one or more previous patient visits (e.g., x-ray images, CBCT scan data, panoramic x-ray images, ultrasound data, color photographs, etc.), and / or reference data 450. Alternatively, the scanner 450 may be connected to another system that stores data in the data store 410. In such an embodiment, the scanner 450 may not be connected to the computing device 405.

[0130] According to one example, a user (e.g., a practitioner) can perform an intraoral scan on a patient. In doing so, the user can apply the scanner 450 to one or more locations within the patient's mouth. The scan can be divided into one or more segments (e.g., upper dental arch, lower dental arch, and bite). Via such scanner application, the scanner 450 can provide current intraoral scan data 435 to the computing device 405. The current intraoral scan data 435 and / or past intraoral scan data 438 can include 3D surface data (e.g., in the form of a 3D image or an image with height information), 2D or 3D color image data, NIRI image data, ultraviolet image data, and the like. Such scan data can be provided from the scanner to the computing device 405 in the form of one or more points (e.g., one or more pixels and / or pixel groups). For example, the scanner 450 can provide a 3D image as one or more point clouds.

[0131] In at least one embodiment, the intraoral scanning application 408 includes a model generation module 425. When the scanning session is complete (e.g., all images of the dental part have been captured), the model generation module 425 can generate a virtual 3D model of the scanned dental part. To generate the virtual model, the model generation module 425 can register and "stitch" together the intraoral scans generated from the intraoral scanning session. In at least one embodiment, performing the registration includes capturing 3D data for various points of the surface in multiple scans (views from the camera) and registering the scans by calculating the transformation between the images, as discussed above.

[0132] In at least one embodiment, the computing device 405 includes a dental diagnostic center 430, which may include a UI 432, one or more dental health analyzers 434, and a recommendation engine 433. The UI 432 may be a graphical user interface and may include icons, buttons, graphics, menus, windows, etc. for controlling and navigating the dental diagnostic center 439.

[0133] Each dental health analyzer can be responsible for performing an analysis associated with a different type of dental condition. For example, the dental health analyzers 434 can include separate dental health analyzers 434 for tooth cracks, gum recession, tooth wear, occlusal contacts, dental crowding and / or other malocclusions, dental plaque, tooth staining, and / or caries. In at least one embodiment, a single dental health analyzer 434 performs each different type of dental health analysis associated with each type of dental condition discussed herein. In at least one embodiment, there are multiple dental health analyzers, some of which perform dental health analyses for multiple different dental conditions. As described above, current intraoral scan data 435, past intraoral scan data 438, additional current dental data 445, additional past dental data 448, and / or reference data 452 can be used to perform one or more dental analyses. For example, data related to an at-hand patient can include x-rays, 2D intraoral images, 3D intraoral images, 2D models, and / or virtual 3D models corresponding to the patient visit during which the scan occurred. Data about the patient at hand may additionally include the patient's past x-rays, 2D intraoral images, 3D intraoral images, 2D models and / or virtual 3D models (e.g., corresponding to the patient's past visits and / or the patient's dental records).

[0134] Reference data 452 may include pooled patient data, which may include x-rays, 2D intraoral images, 3D intraoral images, 2D models, and / or virtual 3D models for a large number of patients. Such a large number of patients may or may not include patients at hand. In compliance with regional medical record privacy regulations (e.g., the Health Insurance Portability and Accountability Act (HIPAA)), pooled patient data may be anonymized and / or used. Pooled patient data may include data corresponding to the scans discussed herein and / or other data. Reference data may additionally or alternatively include teaching patient data, which may include x-rays, 2D intraoral images, 3D intraoral images, 2D models, virtual 3D models, and / or medical illustrations (e.g., medical illustration diagrams and / or other images) used in an educational environment.

[0135] One or more dental health analyzers 434 can use intraoral data (e.g., current intraoral scan data 435, past intraoral scan data 438, additional current dental data 445, additional past dental data 448, and / or reference data 452) to perform one or more types of dental condition analysis, as discussed above. As a result, the dental diagnostic center 430 can determine multiple different dental conditions and severity levels for each of those types of identified dental conditions. In at least one embodiment, the dental health analyzer 434 additionally uses information about multiple different types of identified dental conditions and / or associated severity levels to determine correlations and / or causal relationships between two or more of the identified dental conditions. Multiple dental conditions can be caused by the same underlying root cause. Furthermore, some dental conditions can be potential root causes of other dental conditions. Treatment of the underlying root cause dental condition can alleviate or halt the further development of the other dental condition. For example, malocclusion (e.g., tooth crowding and / or interdental spaces or spacing), tooth wear, and caries can all be identified for the same tooth or group of teeth. The dental diagnostic center 430 can analyze these identified dental conditions that have common, overlapping, or adjacent areas of interest and determine a correlation or causal relationship between one or more of these dental conditions. In an example, the dental diagnostic center 430 can determine that caries and tooth wear for a particular grouping of teeth is caused by crowding of the teeth in that grouping of teeth. By performing orthodontic treatment on that grouping of teeth, malocclusion can be corrected, which can prevent or reduce further caries progression and / or tooth wear for that grouping of teeth. In another example, dental plaque, tooth discoloration, and gum recession can be identified for areas of the dental arch. Tooth discoloration and gum recession can be symptoms of excessive dental plaque. The dental diagnostic center 430 can determine that dental plaque is a potential cause of tooth discoloration and / or gum recession.

[0136] In at least one embodiment, a currently identified dental condition can be used by the dental diagnostic center 430 to predict a future dental condition that is not currently indicated. For example, severe occlusal contacts can be evaluated to predict tooth wear and / or tooth cracks in areas associated with severe occlusal contacts. Such analysis can be performed by inputting intraoral data (e.g., current intraoral data and / or past intraoral data) and / or dental conditions identified from the intraoral data into a trained machine learning model that has been trained to predict future dental conditions based on the current dental condition and / or the current dentition (e.g., the current 3D surface of the dental arch). The machine learning model can be any of the types of machine learning models discussed elsewhere herein. The machine learning model can output a probability map indicating the predicted location of the dental condition and / or the type of dental condition. Alternatively, the machine learning model can output a prediction of one or more future dental conditions without identifying where those dental conditions are predicted to be located.

[0137] The recommendation engine 433 can be used to provide restorative decision recommendations based on image parameters 454 derived from current intraoral scan data 435, past intraoral scan data 438, and other information or image data. In at least one embodiment, the recommendation engine 433 uses a decision model that can generate as output a restorative decision recommendation based on several parameters derived (e.g., automatically measured or approximated) from image data obtained by one or more of the following modalities: intraoral scans using, for example, NIRI, UV, or white light imaging (standard color), radiographs (e.g., panoramic, bitewing, or periapical radiographs), or CBCT.

[0138] In at least one embodiment, parameters useful for generating restorative decision recommendations can be calculated, for example, by applying a trained machine learning model to images from various modalities. Exemplary parameters include, but are not limited to, the parameters described in Table 1 below. As shown, different imaging modalities can be used to obtain the same type of parameters. For example, the intercuspal width can potentially be estimated / calculated based on a 3D intraoral scan or alternatively based on a CBCT image or photograph / projection. Thus, in at least one embodiment, the restorative decisions generated by the recommendation engine 433 can be agnostic to the modality from which the parameters are derived, thereby allowing for flexible selection or combination of imaging modalities to obtain parameters useful for generating recommendations.

[0139] Table 1: Example parameters and their associated measurement techniques

[0140]

[0141]

[0142] In at least one embodiment, the recommendation engine 433 can utilize one or more of a decision tree, a neural network, or other classifiers to recommend a type of restoration. In at least one embodiment, a decision tree can be used. An exemplary decision tree is now described for illustrative purposes:

[0143] If (the inter-cusp width of the existing restoration > X%) and ((%RVP + %DVP) > Y%), then generate a warning for a crown / onlay

[0144] Otherwise, if (the inter-cusp width of the existing restoration > X%) or ((%RVP + %DVP) > Y%), then generate a caution for a crown / onlay

[0145] Otherwise, if (the inter-cusp width of the existing restoration < X%) and ((%RVP + %DVP) > Y%), then

[0146] 1. If (%DVP > 0), then recommend a direct restoration (filling)

[0147] 2. Otherwise, do not recommend a restoration

[0148] In at least one embodiment, X% and Y% can be doctor / clinician adjustable parameters. In at least one embodiment, the parameters X% and Y% are determined based on a trained machine learning model that uses past restoration decisions and parameters derived from intraoral scan data (e.g., %RVP, %DVP) as inputs. It is envisioned that other decision trees can be designed to recommend between, for example, an inlay, an onlay, and a crown as a restoration decision based on dental measurement parameters.

[0149] In at least one embodiment, the restoration volume percentage (RVP) and the restoration surface percentage (RSP) are parameters that can be derived from image data, for example, using a trained machine learning model. For example, starting from a 2D rendering of a 3D model (e.g., a 3D model generated by the model generation module 425 based on data obtained from the scanner 450), a machine learning model can be trained to segment the teeth and detect / identify restoration objects / materials present on the surface of each tooth. Then, the RVP and RSP can be determined based on the overlapping area, as discussed below, and used in the decision tree model together with an indication of caries or other structural damage detected on the teeth. Although only amalgam is discussed below for illustrative purposes, other types of restoration objects / materials are also envisioned.

[0150] Figure 5-Figure 7A method for detecting the presence of amalgam in a 3D dentition model representing the interior of a patient's oral cavity is described. In at least one embodiment, a rendered image of the 3D model is first generated (e.g., buccal, lingual, and occlusal views). The restorative material present in each view can then be labeled and used to train a machine learning model for restorative material segmentation. In at least one embodiment, the model is applied to a 2D rendering and then projected onto the corresponding 3D mesh from which the 2D rendering was generated for segmentation. Figure 5 A 3D model of the lower dental arch is shown comparing the predicted results 501A with the labeled model 501B to illustrate the segmentation quality between the segmented amalgams 502A-D and the labeled amalgams 504A-D. For example, larger amalgam areas may be more accurately segmented than smaller areas and therefore less likely to return a false positive result on an existing crown.

[0151] Figure 6 The classification of dental caries regions is shown comparing the predicted results 601A with the labeled model 601B, and the comparison of the segmented dental caries regions 602A-B with the labeled dental caries regions 604A-B is shown. Training a machine learning model to identify dental caries regions can beneficially prevent misclassification of dental caries regions as amalgam. Even in the event of such misclassification, the overall region may be small enough to minimize the impact on restoration decisions.

[0152] In at least one embodiment, RSP can be calculated as the ratio of the area represented by the restorative material to the area of ​​the occlusal surface of the tooth. Figure 7 7. Segmentation of the identified restorative material and the occlusal surface of the tooth is shown. A 2D rendering of the segmented lower dental arch 701 is shown to identify a restorative material region 702 (e.g., amalgam) and an occlusal surface region 704. For example, %RSP can be calculated as the ratio of the number of pixels within region 702 divided by the number of pixels contained within region 704. In at least one embodiment, one or more models can be used to determine whether a patient is a candidate for restorative treatment, such as a crown. For example, in a heuristic model, if the %RSP is greater than a threshold amount, a recommendation for a crown can be generated.

[0153] Figure 8A user interface for a key performance indicator (KPI) dashboard according to at least one embodiment of the present disclosure is shown. The KPI dashboard can be implemented as part of any dental diagnostic center embodiment described herein, or as part of a dental clinic management system. In at least one embodiment, the information presented and visualized in the KPI dashboard can be performance metrics at the individual physician, physician classification (specialty), clinic, or regional level for tracking the outcomes of their respective patients. The outcomes include restoration types (e.g., direct restorations, crowns, inlays, and onlays), restoration counts (including the number of restorations diagnosed, the number of planned treatments, and the number of completed / billed treatments), and information related to the outcomes relative to the target (e.g., the percentage of treatments diagnosed / planned relative to the target percentage, the percentage of treatments diagnosed / completed relative to the target percentage, and the percentage of treatments planned / completed relative to the target percentage). In addition, for each record, an associated restoration decision recommendation and an indication of whether the physician followed the recommendation can be provided. The KPI dashboard can then be used as a more objective measure of treatment outcomes, and to semi-quantitatively assess whether a particular physician's or clinic's treatment plan meets the philosophy or standards set by a dental support organization (DSO). In at least one embodiment, the KPI dashboard can be expanded to include additional services besides remediation decision results, such as revenue forecasting.

[0154] Figure 9 A- Figure 11 A flowchart of a method performed by a dental diagnostic center according to at least one embodiment of the present disclosure is shown. These methods may be performed by processing logic comprising hardware (e.g., circuitry, dedicated logic, programmable logic, microcode, etc.), software (such as instructions running on a processing device), or a combination thereof. In at least one embodiment, the processing logic corresponds to Figure 4 computing device 405 (eg, corresponding to computing device 405 executing intraoral scanning application 408 and / or dental diagnostic center 430).

[0155] Figure 9A flowchart of a method 900 for generating a restorative decision recommendation according to at least one embodiment of the present disclosure is shown. At box 905, processing logic receives current image data, latest image data, or previous image data of the interior of the patient's mouth. In at least one embodiment, the image data corresponds to one or more imaging modes, and the one or more imaging modes include, for example, an imaging mode selected from an intraoral scan, a radiograph, or a cone beam computed tomography (CBCT) scan. In at least one embodiment, the one or more imaging modes include an intraoral scan, and wherein the image data includes one or more three-dimensional (3D) point clouds, and at least one of the following: a two-dimensional (2D) near-infrared (NIR) image, a 2D ultraviolet image, or a 2D color image. In at least one embodiment, the one or more imaging modes include radiographs, and at least one of the following: a panoramic film, a bitewing film, or a periapical film. In at least one embodiment, the image data corresponds to two, three, or more imaging modes.

[0156] In at least one embodiment, the image data is received as a result of a scan of the inside of a patient's mouth by a doctor or dental practitioner. The scan data may include 3D scan data (e.g., color or monochrome 3D scan data, which may be received in the form of a point cloud, height map, image, or other data type) and / or one or more 3D models of the patient's dental arch generated based on the scan data. The scan data may also include NIRI images and / or color images. The 3D scan data, NIRI images, and / or color images 912 may all be generated by an intraoral scanner. In at least one embodiment, the doctor or dental practitioner may generate additional patient data. In at least one embodiment, the processing logic receives the additional patient data. The additional patient data may include x-ray images (e.g., bitewing x-ray images and / or panoramic x-ray images) and / or dental information (e.g., such as observations by a dental practitioner, biopsy results, CBCT scan data, ultrasound data, etc.). In at least one embodiment, the processing logic may import a patient record for the patient being scanned. Imported patient records may include historical patient data, such as historical NIRI images, color images, 3D scan data or 3D models generated from such 3D scan data, x-ray images, and / or other information.

[0157] In at least one embodiment, processing logic may generate a current or updated version of a 3D model of one or more dental arches of the patient using the current or latest scan data and / or additional current or latest dental data (e.g., using model generation module 425). Alternatively, a 3D model may have already been generated, and the received current or latest scan data may include the 3D model of the dental arch.

[0158] At block 910, processing logic derives a plurality of parameters from the image data. For example, a trained machine learning model may be used to segment regions of the image data, and the model may then be used to calculate various physical parameters. In at least one embodiment, each parameter is selected from a geometric parameter, a volume / area parameter, or a fracture classification parameter. For example, a geometric parameter may include an intercuspal width. As another example, a volume / area parameter may include one or more of a restoration volume fraction, a caries volume fraction, or a restoration surface fraction. As another example, a fracture classification parameter may include information describing the location of a tooth fracture and the depth of a tooth fracture.

[0159] At block 915, processing logic applies a decision model to the plurality of parameters. In at least one embodiment, the decision model comprises one or more of a decision tree or a neural network. In an exemplary embodiment, a decision tree may be utilized that includes adjustable parameters (e.g., heuristic parameters) that can be used to determine whether various threshold conditions are met. The threshold conditions can be used to determine which types of recommendations are generated.

[0160] At box 920, processing logic generates a restorative decision recommendation based on the output of the decision model. In at least one embodiment, the restorative decision recommendation includes a direct restorative recommendation (e.g., a filling) or an indirect restorative recommendation (e.g., an inlay, onlay, crown, bridge, or veneer). In at least one embodiment, the restorative decision recommendation is presented for display in a GUI (e.g., UI 432 of the dental diagnostic center 430). The information presented in the user interface can include qualitative results and / or quantitative results of the various analyses. In at least one embodiment, a dental diagnostic summary is shown that includes high-level results but does not include low-level details or detailed information underlying the high-level results. All of the analysis results can be presented together in a unified view that increases clinical efficiency and improves communication between the doctor and the patient regarding the patient's oral health and how to best treat the dental condition.

[0161] In at least one embodiment, the restoration decision recommendation includes an indication of a dental condition and a severity level of the dental condition. For example, dental conditions may include, but are not limited to, caries, gum recession, tooth wear, malocclusion, crowded teeth, spaces between teeth, plaque, tooth staining, tooth cracks, cervical defects (stress injuries), or chipped or fractured teeth. In at least one embodiment, processing logic determines the recommended treatment. Each type of dental condition may be associated with one or more standard treatments performed in dentistry and / or orthodontics for treating that type of dental condition. A treatment plan may be recommended based on the location of the identified AOIs, the dental conditions of the identified AOIs, the number of AOIs with the dental condition, and / or the severity level of the dental condition. The physician may review the treatment plan and / or adjust the treatment plan based on his or her practice and / or preferences. In at least one embodiment, the physician may customize the dental diagnostic center to give preference to certain types of treatment options over other types of treatment options based on the physician's preferences. A treatment may be determined for each identified dental condition that is determined to be clinically significant.

[0162] In at least one embodiment, processing logic generates a diagnostic result based on the results of the performed dental condition analysis and the previously generated recommendations. The processing logic can generate a caries result, a discoloration result, a malocclusion result, a tooth wear result, a gum recession result, a plaque result, a gum swelling result, a tooth crowding and / or space result, and / or a tooth crack result. The diagnostic result can include the detected AOI associated with each type of dental condition, and a severity level of the dental condition of the AOI. The diagnostic result can include qualitative measurements such as the size of the AOI, the amount of recession of the gum area, the amount of wear of the tooth area, and the amount of change (e.g., for caries, tooth wear, gum swelling, gum recession, tooth discoloration, etc.), the rate of change (e.g., for caries, tooth wear, gum swelling, gum recession, tooth discoloration, etc.), and the like. The diagnostic result can also include qualitative results, such as an indication of whether the dental condition at the AOI has improved, remained the same, or has worsened, an indication of the rapidity with which the dental condition has improved or worsened, an acceleration of the improvement or worsening of the dental condition, and the like. An expected rate of change may have been determined (e.g., automatically or with physician input), and the rate of change of the dental condition measured at the AOI may be compared to the expected rate of change. The difference between the expected rate of change and the measured rate of change may be recorded and included in the diagnosis. Each diagnosis may be automatically assigned a Code and Nomenclature (CDT) code for the dental procedure or a procedure code for other health and ancillary services provided in dentistry. Appropriate insurance codes and related financial information may be automatically assigned to each diagnosis.

[0163] In at least one embodiment, processing logic stores the restorative decision recommendation in a key performance indicator (KPI) database and the actual restorative decision made by the dentist who provided the restorative decision recommendation in a record associated with the dentist.

[0164] Figure 10 A flowchart of a method 1000 for presenting restorative decision recommendations in a GUI of a dental diagnostic center according to at least one embodiment of the present disclosure is shown. At block 1005, processing logic identifies teeth with associated dental conditions based on a first set of parameters derived from current image data of the interior of the patient's mouth. In at least one embodiment, the current image data is received in the same or similar manner as described above with respect to block 905 of method 900. In at least one embodiment, the processing logic identifies teeth with associated dental conditions by: (1) comparing the first set of parameters with a second set of parameters derived from prior image data of the interior of the mouth captured prior to the current image data; and (2) identifying the teeth by determining that the difference between the parameters from the first set of parameters and the parameters from the second set of parameters meets a threshold condition. For example, whenever a new scan or CBCT of a patient is available, that information is stored in a record associated with the patient (e.g., stored in past intraoral scan data 438). As new information becomes available, the processing logic can calculate or recalculate various parameters (e.g., caries volume fraction). If it is determined that the change in the calculated / recalculated parameter exceeds a threshold for a given tooth, the tooth can be identified in the GUI as potentially having a condition. In at least one embodiment, the current image data corresponds to a first imaging mode and the previous image data corresponds to a second imaging mode that is different from the first imaging mode (e.g., an advantage of the recommendation engine 433 is that it is independent of the type of mode used to derive the parameter).

[0165] At block 1010, processing logic presents a 2D or 3D image of the interior of the patient's mouth and an indication of the teeth with associated dental conditions in a GUI. For example, the indication may include one or more of a label on the tooth, an outline on the tooth, or the color of the tooth.

[0166] At block 1015, processing logic presents a restoration decision recommendation in a GUI based on the output of the decision model for the first set of parameters used as input. In at least one embodiment, the restoration decision recommendation is generated, for example, as described above with respect to method 900. In at least one embodiment, the restoration decision recommendation is presented in the GUI in response to a user selecting a tooth in a 2D or 3D image. The recommendation may also be visually provided next to the AOI associated with the tooth for which the recommended restoration is intended. For example, a physician may review the restoration decision recommendation and the associated AOI to self-assess the presence and / or severity of a dental condition. This may include zooming in or out of the AOI, panning, rotating the AOI view, viewing additional data about the AOI, such as NIRI imaging data, UV imaging data, color data, x-ray data, and the like.

[0167] Figure 11 A flow chart is shown of a method 1100 for generating restorative decision recommendations based on parameters derived from image data using a trained machine learning model, in accordance with at least one embodiment of the present disclosure. At block 1105, processing logic receives image data corresponding to the interior of a patient's oral cavity, which may be received in a manner similar to that described above with respect to block 905 of method 900.

[0168] At block 1110, processing logic applies the trained machine learning model to the image data to derive a plurality of parameters from the image data. In at least one embodiment, the trained machine learning model is adapted to calculate or estimate the volume of restorative material present on or within the teeth in the image data, and wherein deriving the plurality of parameters includes calculating at least one of a restoration volume or a surface ratio based on the estimated volume of restorative material. For example, the machine model may be trained to segment a 2D projection of a 3D model of a patient's dentition (e.g., as described above with respect to Figure 5-Figure 7 In at least one embodiment, the trained machine learning model is adapted to receive as input image data corresponding to different imaging modalities. In other embodiments, different machine learning models may be utilized, each adapted to segment images of a different modality. In at least one embodiment, the different imaging modalities are independently selected from intraoral scans, radiographs, or cone beam computed tomography (CBCT) scans.

[0169] At block 1115 , processing logic applies the decision model to the plurality of parameters to generate a repair decision recommendation. In at least one embodiment, the repair decision recommendation is generated, for example, as described above with respect to method 900 .

[0170] Figure 12A graphical representation of a machine in the example form of a computing device 1200 is shown within which a set of instructions for causing the machine to perform any one or more of the methodologies discussed herein may be executed. In alternative embodiments, the machine may be connected (e.g., networked) to other machines in a local area network (LAN), an intranet, an extranet, or the Internet. The machine may operate as a server or a client machine in a client-server network environment, or as a peer machine in a peer-to-peer (or distributed) network environment. The machine may be a personal computer (PC), a tablet computer, a set-top box (STB), a personal digital assistant (PDA), a cellular phone, a network appliance, a server, a network router, a switch or a bridge, or any machine capable of executing a set of instructions (sequential or otherwise) to specify actions to be taken by the machine. Furthermore, while a single machine is shown, the term "machine" shall also be taken to include any collection of machines (e.g., computers) that individually or collectively execute a set (or multiple sets) of instructions to perform any one or more of the methodologies discussed herein. In at least one embodiment, the computing device 1200 corresponds to Figure 4 computing device 405.

[0171] The example computing device 1200 includes a processing device 1202, a main memory 1204 (e.g., read-only memory (ROM), flash memory, dynamic random access memory (DRAM) such as synchronous DRAM (SDRAM)), etc.), a static memory 1206 (e.g., flash memory, static random access memory (SRAM), etc.), and a secondary memory (e.g., data storage device 1228), which communicate with each other via a bus 1208.

[0172] The processing device 1202 represents one or more general-purpose processors, such as a microprocessor, a central processing unit, or the like. More specifically, the processing device 1202 may be a complex instruction set computing (CISC) microprocessor, a reduced instruction set computing (RISC) microprocessor, a very long instruction word (VLIW) microprocessor, a processor implementing another instruction set, or a processor implementing a combination of instruction sets. The processing device 1202 may also be one or more special-purpose processing devices, such as an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a digital signal processor (DSP), a network processor, or the like. The processing device 1202 is configured to execute processing logic (instructions 1226) for performing the operations and steps discussed herein.

[0173] The computing device 1200 may also include a network interface device 1222 for communicating with a network 1264. The computing device 1200 may also include a video display unit 1210 (e.g., a liquid crystal display (LCD) or a cathode ray tube (CRT)), an alphanumeric input device 1212 (e.g., a keyboard), a cursor control device 1214 (e.g., a mouse), and a signal generating device 1220 (e.g., a speaker).

[0174] The data storage device 1228 may include a machine-readable storage medium (or, more specifically, a non-transitory computer-readable storage medium) 1224 on which is stored one or more sets of instructions 1226 embodying any one or more of the methodologies or functions described herein. A non-transitory storage medium refers to a storage medium other than a carrier wave. During execution of the instructions by the computer device 1200, the instructions 1226 may also reside, completely or at least partially, within the main memory 1204 and / or within the processing device 1202, with the main memory 1204 and the processing device 1202 also constituting computer-readable storage media.

[0175] The computer readable storage medium 1224 may also be used to store a recommendation engine 1250, which may correspond to Figure 4 's similarly named components. The computer-readable storage medium 1224 may also store a software library containing methods for the recommendation engine 1250. Although the computer-readable storage medium 1224 is shown as a single medium in the example embodiment, the term "computer-readable storage medium" should be taken to include a single medium or multiple media (e.g., a centralized or distributed database, and / or associated caches and servers) that store one or more sets of instructions. The term "computer-readable storage medium" should also be taken to include any non-transitory medium (e.g., a medium other than a carrier wave) that is capable of storing or encoding a set of instructions for execution by a machine and causing the machine to perform any one or more methods of the present disclosure. Accordingly, the term "computer-readable storage medium" should be taken to include, but is not limited to, solid-state memory, and optical and magnetic media.

[0176] The following exemplary embodiments are now described:

[0177] Example 1: A method for providing restorative decision support for a dental patient, the method comprising: receiving image data of the interior of the patient's mouth, the image data corresponding to one or more imaging modes; deriving multiple parameters from the image data; applying a decision model to the multiple parameters; and generating a restorative decision recommendation based on the output of the decision model.

[0178] Example 2: The method of Example 1, wherein the one or more imaging modes include an imaging mode selected from intraoral scans, radiographs, or cone beam computed tomography (CBCT) scans.

[0179] Example 3: The method of Example 2, wherein the one or more imaging modes include the intraoral scan, and wherein the image data includes one or more three-dimensional (3D) point clouds, and at least one of the following: a two-dimensional (2D) near-infrared (NIR) image, a 2D ultraviolet image, or a 2D color image.

[0180] Embodiment 4: The method of embodiment 2, wherein the one or more imaging modalities include radiographs, and wherein the image data includes one or more of panoramic radiographs, bitewing radiographs, or periapical radiographs.

[0181] Embodiment 5: The method of embodiment 2, wherein the image data corresponds to two or more of the imaging modes.

[0182] Embodiment 6: The method according to any one of the preceding embodiments, wherein each of the plurality of parameters is selected from geometric parameters, volume / area parameters, or fracture classification parameters.

[0183] Example 7: The method according to Example 6, wherein the geometric parameter includes the intercusp width.

[0184] Example 8: The method according to Example 6, wherein the volume / area parameters include one or more of the restoration volume ratio, the caries volume ratio or the restoration surface ratio.

[0185] Embodiment 9: The method according to embodiment 6, wherein the fracture classification parameters include information describing the tooth fracture location and tooth fracture depth.

[0186] Embodiment 10: The method according to any one of the preceding embodiments, wherein the decision model comprises one or more of a decision tree or a neural network.

[0187] Embodiment 11: The method according to any one of the preceding embodiments, wherein the restorative decision recommendation comprises one or more of: a direct restorative recommendation or an indirect restorative recommendation, or an indication of a dental condition and a severity level of the dental condition.

[0188] Example 12: The method of Example 11, wherein the dental condition is selected from the group consisting of caries, gum recession, tooth wear, malocclusion, tooth crowding, tooth spaces, dental plaque, tooth staining, tooth cracks, tooth neck defects, and chipped or fractured teeth.

[0189] Embodiment 13: The method according to any one of the preceding embodiments further comprises: presenting a repair decision recommendation for display in a graphical user interface (GUI).

[0190] Embodiment 14: The method according to any one of the preceding embodiments further includes: storing the restorative decision recommendation and the actual restorative decision made by the dentist who provided the restorative decision recommendation in a record associated with the dentist in a key performance indicator (KPI) database.

[0191] Example 15: A method comprising: identifying teeth with associated dental conditions based on a first set of parameters derived from current image data of the interior of a patient's mouth; presenting a 2D or 3D image of the interior of the patient's mouth and an indication of the teeth with the associated dental conditions in a graphical user interface (GUI); and presenting a restoration decision recommendation in the GUI based on the output of a decision model using the first set of parameters as input.

[0192] Example 16: The method of Example 15, wherein the restoration decision recommendation is presented in the GUI in response to a user selecting a tooth in the 2D or 3D image, and wherein the indication includes one or more of a label on the tooth, a contour on the tooth, or a color of the tooth.

[0193] Example 17: A method according to Example 15 or Example 16, wherein identifying the tooth having the associated dental condition includes: comparing the first set of parameters with a second set of parameters, the second set of parameters being derived from prior image data of the interior of the oral cavity captured before the current image data; and identifying the tooth by determining that the difference between parameters from the first set of parameters and parameters from the second set of parameters satisfies a threshold condition.

[0194] Embodiment 18: The method of embodiment 17, wherein the current image data corresponds to a first imaging mode, and wherein the previous image data corresponds to a second imaging mode different from the first imaging mode.

[0195] Example 19: A method comprising: receiving image data corresponding to the interior of a patient's mouth; applying a trained machine learning model to the image data to derive multiple parameters from the image data; and applying a decision model to the multiple parameters to generate a restorative decision recommendation.

[0196] Example 20: A method according to Example 19, wherein the trained machine learning model is adapted to calculate or estimate the volume of restorative material present on or in the tooth in the image data, and wherein deriving the multiple parameters includes calculating at least one of the restoration volume or surface ratio based on the estimated volume of restorative material.

[0197] Example 21: A method according to Example 19 or Example 20, wherein the trained machine learning model is adapted to receive image data corresponding to different imaging modes as input, and wherein the different imaging modes are independently selected from intraoral scans, radiographs, or cone beam computed tomography (CBCT) scans.

[0198] Example 22: A dental diagnostic system comprising: a memory; and a processing device for executing instructions from the memory to perform a method comprising: receiving image data of the interior of a patient's mouth, the image data corresponding to one or more imaging modes; deriving multiple parameters from the image data; applying a decision model to the multiple parameters; and generating a restoration decision recommendation based on the output of the decision model.

[0199] Example 23: An intraoral scanning system comprising: an intraoral scanner; and the dental diagnostic system according to Example 22.

[0200] Embodiment 24: A non-transitory computer-readable storage medium having encoded thereon instructions that, when executed by a computing device, cause the computing device to perform the method according to any one of embodiments 1-21.

[0201] Claim language or other language herein reciting "at least one of" a set and / or "one or more" a set means that one member of the set or multiple members of the set (in any combination) satisfies the claim. For example, claim language reciting "at least one of A and B" or "at least one of A or B" means A, B, or A and B. In another example, claim language reciting "at least one of A, B, and C" or "at least one of A, B, or C" means A, B, C, or A and B, or A and C, or B and C, or A, B, and C. Language reciting "at least one of" a set and / or "one or more" a set does not limit the set to the items listed in the set. For example, claim language reciting "at least one of A and B" or "at least one of A or B" can mean A, B, or A and B, and can additionally include items not listed in the set of A and B.

[0202] It should be understood that the above description is intended to be illustrative and not limiting. Many other embodiments will become apparent upon reading and understanding the above description. Although embodiments of the present disclosure have been described with reference to specific example embodiments, it will be appreciated that the present disclosure is not limited to the described embodiments, but may be implemented with modifications and alterations within the spirit and scope of the appended claims. Accordingly, the description and drawings should be regarded as illustrative and not restrictive. Therefore, the scope of the present disclosure should be determined with reference to the appended claims and the full scope of equivalents to which such claims are entitled.

Claims

1. A method for providing restorative decision support to a dental patient, the method comprising: receiving image data of the interior of a patient's oral cavity, the image data corresponding to one or more imaging modalities; deriving a plurality of parameters from the image data; applying a decision model to the plurality of parameters; as well as A repair decision recommendation is generated based on the output of the decision model.

2. The method according to claim 1, wherein The one or more imaging modalities include an imaging modality selected from an intraoral scan, a radiograph, or a cone beam computed tomography (CBCT) scan.

3. The method according to claim 2, wherein: The one or more imaging modalities include intraoral scanning, and wherein the image data includes one or more three-dimensional (3D) point clouds and at least one of: a two-dimensional (2D) near-infrared (NIR) image, a 2D ultraviolet (UV) image, or a 2D color image.

4. The method according to claim 2, wherein: The one or more imaging modalities include radiographs, and wherein the image data includes one or more of panoramic radiographs, bitewing radiographs, or periapical radiographs.

5. The method according to claim 2, wherein: The image data corresponds to two or more of the imaging modes.

6. The method according to claim 1, wherein Each of the plurality of parameters is selected from geometric parameters, volume / area parameters, or fracture classification parameters.

7. The method according to claim 6, wherein: The geometric parameters include the intercusp width.

8. The method according to claim 6, wherein: The volume / area parameters include one or more of the following: restoration volume proportion, caries volume proportion or restoration surface proportion.

9. The method according to claim 6, wherein: The fracture classification parameters include information describing the tooth fracture location and tooth fracture depth.

10. The method according to claim 1, wherein The decision model includes one or more of a decision tree or a neural network.

11. The method according to claim 1, wherein The repair decision recommendation includes one or more of the following: Direct repair recommendation or indirect repair recommendation, or An indication of a dental condition and a level of severity of said dental condition.

12. The method according to claim 11, wherein The dental condition is selected from the group consisting of caries, gum recession, tooth wear, malocclusion, tooth crowding, spaces between teeth, dental plaque, tooth staining, cracked teeth, cervical lesions, and broken or cracked teeth.

13. The method according to claim 1, further comprising: The repair decision recommendation is presented for display in a graphical user interface (GUI).

14. The method according to claim 1, further comprising: In a key performance indicator KPI database, the restoration decision recommendation and the actual restoration decision made by the dentist who provided the restoration decision recommendation are stored in a record associated with the dentist.

15. A method comprising: identifying teeth having an associated dental condition based on a first set of parameters derived from current image data of the interior of the patient's mouth; presenting in a graphical user interface (GUI) a 2D or 3D image of the interior of the patient's mouth and an indication of the teeth with the associated dental condition; as well as A repair decision recommendation is presented in the GUI based on the output of the decision model with the first set of parameters used as input.

16. The method according to claim 15, wherein The restoration decision recommendation is presented in the GUI in response to a user selection of the tooth in the 2D or 3D image, and wherein the indication includes one or more of a label on the tooth, an outline on the tooth, or a color of the tooth.

17. The method according to claim 15, wherein: Teeth identified as having associated dental conditions include: comparing the first set of parameters with a second set of parameters derived from previous image data of the interior of the oral cavity captured before the current image data; and The tooth is identified by determining that a difference between a parameter from the first set of parameters and a parameter from the second set of parameters satisfies a threshold condition.

18. The method according to claim 17, wherein: The current image data corresponds to a first imaging mode, and wherein the previous image data corresponds to a second imaging mode different from the first imaging mode.

19. A method comprising: receiving image data corresponding to the interior of the patient's oral cavity; applying a trained machine learning model to the image data to derive a plurality of parameters from the image data; as well as A decision model is applied to the plurality of parameters to generate a repair decision recommendation.

20. The method according to claim 19, wherein The trained machine learning model is adapted to calculate or estimate the volume of restorative material present on or in the tooth in the image data, and wherein deriving the plurality of parameters comprises calculating at least one of a restoration volume or a surface ratio based on the estimated volume of restorative material.

21. The method according to claim 19, wherein The trained machine learning model is adapted to receive as input image data corresponding to different imaging modalities, and wherein the different imaging modalities are independently selected from intraoral scans, radiographs, or cone beam computed tomography (CBCT) scans.

22. A dental diagnostic system comprising: a memory and a processing device for executing instructions from the memory to: receiving image data of the interior of a patient's oral cavity, the image data corresponding to one or more imaging modalities; deriving a plurality of parameters from the image data; applying a decision model to the plurality of parameters; as well as A repair decision recommendation is generated based on the output of the decision model.

23. An intraoral scanning system comprising an intraoral scanner and the dental diagnostic system according to claim 22.

24. A non-transitory computer-readable storage medium having encoded thereon instructions that, when executed by a computing device, cause the computing device to perform the method of claim 1.

Images