Assessing tooth wear in a subject

Abstract

There is provided a method for assessing tooth wear in a subject. The method comprises generating edge profile data describing a profile of an edge of a tooth of the subject. The method further comprises analysing the edge profile data to determine a tooth wear metric, wherein the tooth wear metric is indicative of a presence of tooth wear of the tooth and / or a severity of tooth wear of the tooth.

Description

[0001] ASSESSING TOOTH WEAR IN A SUBJECT

[0002] FIELD OF THE INVENTION

[0003] The disclosure relates to methods and apparatus for assessing tooth wear in a subject.

[0004] BACKGROUND OF THE INVENTION

[0005] Several factors can cause a tooth to deteriorate over time such as bruxism, incorrect toothbrush usage, chewing on a pen, chemicals, or diet.

[0006] There currently exist certain challenges in detecting and quantifying tooth wear of a tooth. It would be advantageous to automatically detect and quantify tooth wear of a tooth, given an input image of a tooth.

[0007] SUMMARY OF THE INVENTION

[0008] As noted above, there currently exist certain challenges. Embodiments of the present disclosure address these and other challenges.

[0009] According to a first aspect, there is provided a method for assessing tooth wear in a subject. The method comprises generating edge profile data describing a profile of an edge of a tooth of the subject, and analysing the edge profile data to determine a tooth wear metric, wherein the tooth wear metric is indicative of a presence of tooth wear of the tooth and / or a severity of tooth wear of the tooth.

[0010] An advantage of the first aspect is that tooth wear of a tooth may be detected and quantified.

[0011] The edge of the tooth may be an incisal edge of the tooth, or a cutting edge of the tooth. Analysing edge profile data of a well defined cutting edge of a tooth (such as an incisal edge) may enable tooth wear to be more accurately identified and quantified.

[0012] The method may further comprise obtaining an intraoral image of a tooth of the subject. In these embodiments, generating the edge profile data comprises analysing the intraoral image to generate the edge profile data.

[0013] This may advantageously enable tooth wear of a tooth to be detected and / or quantified, given an input image of a tooth. The intraoral images that are analysed may be those obtained in other tooth inspection methods (for example, plaque and / or cavity detection, gum health assessment) of the teeth of the subject. These intraoral images may be obtained in other illumination settings or image modalities (for example, QLF, white light, etc). In these embodiments, a tooth wear assessment may be performed based on these images, which have been obtained for the above tooth inspection methods, andthus the tooth wear assessment may be performed without requiring any additional effort from the subject.

[0014] Analysing the intraoral image to generate the edge profde data may comprise analysing the intraoral image to generate boundary data describing a boundary of the tooth, and generating the edge profde data based on the intraoral image and / or the boundary data, wherein the edge profde data is a part of the boundary data. This enables a parameterisation of the tooth edge to be obtained based on the intraoral images, which can then be analysed to determine if this parameterised edge is wom / jagged, which may be indicative of tooth wear.

[0015] Analysing the intraoral image to generate the boundary data may comprise processing the intraoral image with a tooth segmentation algorithm. This may advantageously enable the tooth edge to be identified and / or parameterized using image processing methods for tooth instance detection and / or identification.

[0016] Generating the edge profde data may comprise generating the edge profde data based on the boundary data, or processing the intraoral image with an image analysis algorithm. This may advantageously enable the tooth edge to be identified and / or parameterised using image processing methods for tooth instance detection and / or identification.

[0017] Analysing the edge profde data to determine a tooth wear metric may comprise determining edge curvature information based on the edge profde data, wherein the edge curvature information describes the curvature of the profde of the edge, analysing the edge curvature information to determine the tooth wear metric. This may enable the curvature (or shape) of the tooth edge to be obtained, which can then be analysed to determine if this edge is wom / jagged, which may be indicative of tooth wear.

[0018] Analyzing the edge profde data to determine a tooth wear metric may further comprise comparing the edge profde data with corresponding edge profde data, wherein the corresponding edge profde data describes a profde of the edge of another tooth of the subject of the same type as the tooth, and determining and / or adjusting the tooth wear metric based on the comparison.

[0019] Analysing the edge profde data to determine a tooth wear metric may further comprise comparing the edge profde data with reference edge profde data, wherein the reference edge profde data describes an expected profde of the edge of the tooth, and adjusting the tooth wear metric based on the comparison. This may enable the tooth wear to be assessed based on a to known, typical, and / or default profiles of the tooth being assessed, and may enable wear caused by, for example, chipping, bruxism, and / or accidental damage to be identified. It will be appreciated that the aforementioned wear will result in a significantly different edge profde to an expected edge profde of the tooth in these instances.

[0020] The method may further comprise determining a tooth number (for example, in the ISO 3950 notation) of the tooth, and obtaining the reference edge profde data corresponding to the determined tooth number. This advantageously enables the edge profde of the tooth to be compared with an expectededge profile of for that specific tooth number. As different tooth numbers will have different expected edge profiles, this comparison may enable the tooth wear metric to be determined more accurately.

[0021] Analysing the edge profile data to determine a tooth wear metric may further comprise comparing the edge curvature information with historical edge curvature information, describing a previous curvature of the profile of the edge of the tooth of the subject, and adjusting the tooth wear metric based on the comparison. This may enable the tooth edge of the subject to be monitored over time, and changes in tooth shape due to wear to be detected.

[0022] The method may further comprise classifying, based on the edge curvature information and / or the edge profile data, a cause of tooth wear of the tooth. This classification may enable improved oral care feedback to then be generated and provided to the subject, which is specific to the cause of the wear.

[0023] The method may further comprise generating oral care feedback based on the determined tooth wear metric, and providing the oral care feedback to the subject. This may enable the subject to perform one or more actions to address and / or prevent the identified tooth wear from worsening.

[0024] The oral care feedback may include at least one of oral care routine advice, and dental professional examination advice.

[0025] According to a second aspect, there is provided an apparatus for assessing tooth wear in a subject. The apparatus comprises one or more processors configured to cause the apparatus to generate edge profile data describing a profile of an edge of a tooth of the subject, and analyse the edge profile data to determine a tooth wear metric, wherein the tooth wear metric is indicative of a presence of tooth wear of the tooth and / or a severity of tooth wear of the tooth.

[0026] According to a third aspect, there is provided a computer program product comprising a computer readable medium. The computer readable medium has a computer readable code embodied therein. The computer readable code is configured such that, on execution by a suitable computer or processor, the computer or processor is caused to perform the method described earlier.

[0027] These and other aspects will be apparent from and elucidated with reference to the embodiment(s) described hereinafter.

[0028] BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Exemplary embodiments will now be described, by way of example only, with reference to the following drawings, in which:

[0030] Fig. 1 is a flow chart illustrating a method according to an embodiment;

[0031] Fig. 2 shows an example intraoral image of a tooth of the subject;

[0032] Fig. 3 shows an example of polyline contour around a segmented tooth instance in an intra-oral image;

[0033] Fig. 4 shows an example of a cutting edge poly-line;

[0034] Fig. 5 is a visualization of determined local derivatives of a cutting edge polyline;Fig. 6 shows teeth that have been chipped on the side;

[0035] Fig. 7a shows an example tooth contour with a high tooth fracture measure value;

[0036] Fig. 7b shows an example tooth contour with a low tooth fracture measure value;

[0037] Fig. 8 is a block diagram illustrating an example apparatus according to embodiments of the disclosure; and

[0038] Fig. 9 is a block diagram illustrating an example processor according to embodiments of the disclosure.

[0039] DETAILED DESCRIPTION OF EMBODIMENTS

[0040] There are provided herein techniques for assessing tooth wear in a subject.

[0041] A method of assessing tooth wear in a subject may be implemented, for example, by training a convolutional neural network on an annotated set of teeth images with different degrees and types of wear, and then providing teeth images to this trained neural network to assess tooth wear of teeth present in these images. Such a convolution neural network (CNN) may therefore process a tooth image to directly compute a tooth wear metric. However, such an approach would require the acquisition and annotation of thousands of images, of which only a small percent will show teeth with a worn and / or damaged edge, and may require these images to be annotated by a qualified dentist in order to train the neural network, which would involve significant effort and cost. Certain embodiments described herein instead propose a hybrid image processing solution, which significantly reduces the amount of collected and annotated data required to implement the solution.

[0042] Fig. 1 illustrates a method 100 for assessing tooth wear in a subject according to an embodiment. More specifically, Fig. 1 illustrates a method 100 of operating an apparatus, such as the apparatus 800 described later with reference to Fig. 8, for assessing tooth wear in a subject. The method 100 illustrated in Fig. 1 is a computer-implemented method. As described later with reference to Fig. 8, the apparatus 800 comprises one or more processors 802. The method 100 illustrated in Fig. 1 can generally be performed by or under the control of the one or more processors 802 of the apparatus 800 described later with reference to Fig. 8, such as a processor 900 described later with reference to Fig. 9.

[0043] With reference to Fig. 1, at block 102, the method 100 comprises generating edge profile data describing a profile of an edge of a tooth of the subject. Block 102 may be performed by a processor of the apparatus 800. The edge profile data describing a profile of an edge of a tooth of the subject may be generated based on any suitable data relating to the edge of the tooth, that may have been obtained using one of the following modalities: an intra-oral image of the tooth, a 3D scan, a Magnetic Resonance Imaging scan, data obtained from an Inertial Measurement Unit, data obtained from a pressure sensor to which the tooth is applying pressure).

[0044] At block 104 of Fig. 1, the method 100 comprises analyzing the edge profile data to determine a tooth wear metric, wherein the tooth wear metric is indicative of a presence of tooth wear of the tooth and / or a severity of tooth wear of the tooth. Block 104 may be performed by a processor of theapparatus 800. The tooth wear metric measures the degree / severity / level of the wear of a tooth. The tooth wear metric may be determined by computing how jagged / rough / uneven the tooth edge (as represented by the edge profile data), as will be explained in greater detail below.

[0045] The tooth wear metric may be determined for each tooth of a specific type (for example, an incisor), as explained in greater detail below.

[0046] The method of Fig. 1 is now discussed in more detail.

[0047] As previously discussed, the method at block 102 comprises generating edge profile data describing a profile of an edge of a tooth of the subject. The edge of the tooth may be an incisal edge of the tooth, or a cutting edge of the tooth. That is, the edge of the tooth may be the part of the tooth used for cutting or tearing food. The methods described herein may be particularly suitable for analyzing teeth without an occlusal surface and a well-defined cutting edge (for example, incisors and canines).

[0048] In some embodiments, the method 100 may further comprise obtaining an intraoral image of a tooth of the subject, and wherein generating the edge profile data comprises analyzing the intraoral image to generate the edge profile data.

[0049] The intraoral image may be obtained by an imaging device in an intraoral device (IOD), or an intraoral scanner (IOS). The imaging device may comprise an image sensorthat is sensitive to visible light. The imaging device may be a camera. An example camera is an “RGB camera”, which is a camera that uses a Color Array Filter (CFA) with a standard Bayer pattern. The CFA is sensitive to the primary colors of visible light: red, green, and blue. However, the imaging device may be any suitable device that can be embedded in an IOD and can capture images of an oral cavity.

[0050] The method may comprise obtaining an intraoral image of a tooth of the subject prior to block 102. As discussed in further detail below, if the method is performed by an IOD, obtaining an intraoral image of a tooth of the subject may comprise using the imaging device to capture the images. If the method is performed by a processing device that is external to the IOD, obtaining an intraoral image of a tooth of the subject may comprise receiving the image, directly or indirectly, from the IOD.

[0051] Fig. 2 shows an example intraoral image of a tooth of the subject, that may be obtained from, for example, from an intra-oral scanner. That is, a tooth image may be collected using an intra-oral scanner.

[0052] Embodiments for analyzing the intraoral image to generate edge profile data are now discussed.

[0053] Analyzing the intraoral image to generate the edge profile data may comprise analyzing the intraoral image to generate boundary data describing a boundary of the tooth, and generating the edge profile data based on the intra-oral image and / or the boundary data, wherein the edge profile data is a part of the boundary data.

[0054] As an example, analyzing the intraoral image to generate boundary data describing a boundary of the tooth may comprise processing the intraoral image with a tooth segmentation algorithm. In these embodiments, the tooth object in the intra-oral image will therefore be segmented. The tooth (orinstance) segmentation algorithm may comprise, for example, a MMDetection model toolbox from OpenMMLab. The tooth segmentation algorithm may be a ML model.

[0055] For the purposes of the present disclosure, the term “ML model” encompasses within its scope the following concepts: ML algorithms, comprising processes or instructions through which data may be used in a training process to generate a model artefact for performing a given task, or for representing a real world process or system; the model artefact that is created by such a training process, and which comprises the computational architecture that performs the task; and the process performed by the model artefact in order to complete the task. The terms ML model, Artificial Intelligence (Al) model, AI / ML (AIML) model, ML algorithm, and Al algorithm may be used interchangeably herein.

[0056] For example, analyzing the intraoral image to generate boundary data describing a boundary of the tooth may comprise an ML algorithm being applied to the images. The ML algorithm may be trained to implement instance segmentation of teeth in the images input into the algorithm. For example, the ML algorithm may have been trained using deep learning techniques.

[0057] The tooth segmentation algorithm may therefore be an ML algorithm. An example ML algorithm is an AIML model operable to implement instance segmentation and / or image classification. The AIML model may be a neural network. The AIML model may be trained on collected and, optionally, annotated data of teeth. The annotation may be performed using any suitable framework, such as Amazon Sagemaker, Roboflow or Labelbox. The model by be trained in any suitable framework, such as PyTorch or TensorFlow. The AIML model may be trained on around 1000 images per image modality. In some embodiments, the image set may be selected by means of active learning to select the “best” samples. In some embodiments, the AIML model may trained by augmenting the images by means of classical augmentation strategies, or by means of generative AL In some embodiments, the tooth instance detection may be succeeded with tooth number identification. In some embodiments, the tooth instance and tooth number may be identified by a single model.

[0058] It is noted that these methods of tooth instance detection and identification may be common image processing steps in a number of types of tooth evaluation, such as determining an amount of plaque on a tooth, determining the presence of a cavity on a tooth, determining the presence of gum inflammation, etc. Thus, the methods described herein, for assessing tooth wear, may be performed in conjunction with other types of tooth evaluation, which also process intra-oral images to perform tooth instance detection and / or identification.

[0059] Segmentation of the tooth instance, or tooth object, in the intra-oral image may enable a tooth contour, (for example, a polyline, or a curve), surrounding the tooth instance, to be generated. Fig. 3 shows an example of a polyline contour 302 around the segmented tooth instance 304 in an intra-oral image. In the context of the present disclosure, the tooth contour, or the polyline contour, may be considered the boundary data above.

[0060] Generating the edge profile data based on the intra-oral image and / or the boundary data, wherein the edge profile data is a part of the boundary data, may comprise determining which part of theboundary data corresponds to the edge of the tooth. For example, following segmentation of the tooth instance, and the generation of the tooth contour 304 surrounding the segmented tooth instance, the location of the tooth cutting edge of the tooth may be identified. The tooth cutting edge may be identified through determining the part of the tooth contour which corresponds to the tooth cutting edge.

[0061] In some embodiments, generating the edge profile data may comprise generating the edge profile data based on the boundary data, or processing the intraoral image with an image analysis algorithm.

[0062] For example, the tooth cutting edge may be identified, for example, using a local contrast algorithm such as PCA or a convolutional neural network. For example, in some embodiments, the tooth cutting edge may be identified using a local contrast measure. For a RGB intraoral image, the intraoral image may be first converted to grey scale using a (local) PCA decomposition of intra oral image colours, such that the monochrome image corresponds to the 1st PCA coefficient, therefore maximizing the contrast. In these embodiments, the part of the tooth contour with high contrast, and adjacent to the dark background, would correspond to the tooth cutting edge. In some embodiments, a segmentation model (for example, MMDetection) may be trained to segment the cutting edge region of the tooth, and then intersect the obtained tooth contour with the result of detection. It is noted that cutting edge segmentation alone in an intraoral image may be less accurate, as teeth boundaries appear similarly in different teeth, and as the intraoral image may comprise other high contrast objects.

[0063] Following identification of the tooth cutting edge, a fragment (or part) of the tooth contour which corresponds to the tooth cutting edge, may be determined. For example, where the tooth contour is a polyline contour around the tooth instance, the part of the tooth contour which corresponding to the tooth cutting edge will be a cutting edge polyline. The tooth cutting edge may therefore be approximated in 2D, with, for example, a polyline or a curve. The edge profile data may therefore, in some embodiments, comprise a cutting edge polyline, or a cutting edge curve.

[0064] Alternatively, a polyline may be fitted to the detected boundary to enable a low dimensional parametrization of the cutting edge shape to be computed.

[0065] Embodiments for analyzing the edge profile data to determine a tooth wear metric are now discussed.

[0066] As noted above, the tooth wear metric may be determined by computing how jagged / rough / uneven the tooth edge (as represented by the edge profile data) is. In some embodiments, the tooth wear metric may be determined by computing how jagged / rough / uneven the edge polyline or edge curve (that is, the 2D representation of the edge) is.

[0067] In some embodiments, analyzing the edge profile data to determine a tooth wear metric may comprise determining edge curvature information based on the edge profile data, wherein the edge curvature information describes the curvature of the profile of the edge, and analyzing the edge curvature information to determine the tooth wear metric.For example, where a low dimensional parametrization of the cutting edge shape (e.g. a cutting edge polyline) has been determined, this low dimensional parametrization may be regressed to determine the tooth wear metric.

[0068] For example, determining the edge curvature information may comprise, in embodiments where a cutting-edge polyline has been determined, re-parameterising the cutting-edge polyline to have an equidistant control points. Fig. 4 shows an example of a re-parametrized cutting edge polyline 402 with equidistant control points in grey, and the original control points in white.

[0069] This re -parametrization allows a more accurate computation of the tooth fracture measure (or tooth wear metric, or tooth integrity index), which in this embodiment is based on the local derivatives of the cutting edge polyline. In these embodiments, for three consecutive control points A, B, C on the cutting edge polyline, the discrete local derivative is defined as AB x BC.

[0070] Therefore, in some embodiments, the determined local derivatives may correspond to the edge curvature information described herein. However, it will be appreciated that, in embodiments where a cutting edge polyline has been determined, information describing the curvature of this polyline, or other similar information, may be derived in any suitable manner.

[0071] In the following determination of the local derivatives of the cutting edge polyline, it is assumed that the contour that was determined from the instance segmentation, and therefore the cutting edge contour (the polyline) has a counterclockwise orientation. If this is not the case, this orientation may be changed to counterclockwise in any suitable manner.

[0072] A visualisation of the determined local derivatives of the cutting edge polyline are shown in Fig. 5. These local derivatives are discrete second order derivatives. As shown in Fig. 5, the discrete second order derivatives are naturally positive of the teeth comers 502, 504, while on tooth defects 506, 508, they have negative values.

[0073] Thus, the tooth fracture measure (or tooth wear metric), in these embodiments, can be defined as the variance of a histogram of the negative discrete derivative for the cutting edge polyline. Advantageously, determination of this metric does not require the collection of specific training data. However, it is noted that this metric may, in an embodiment, adapted so as to not measure tooth wear in certain specific scenarios, such as when a tooth has been chipped on one side. Fig. 6 shows an example of teeth that have been chipped on the side.

[0074] In other embodiments, the edge curvature information may be analyzed to determine the tooth wear metric as follows. The tooth wear metric can be determined based on a sequence of angles between consecutive polyline segments of the tooth cutting edge polyline. For example, a dense neural network may be trained, such that it receives as an input, a sequence of angles between consecutive polyline segments of the tooth cutting edge polyline, and outputs a determined value of the tooth wear metric. In these embodiments, the cutting edge polyline should be reparametrized to a fixed number of control points.In some embodiments, analyzing the edge profile data to determine a tooth wear metric may further comprise comparing the edge profde data with reference edge profde data, wherein the reference edge profde data describes an expected profde of the edge of the tooth, and adjusting the tooth wear metric based on the comparison.

[0075] Fig. 7a shows an example tooth contour with a high tooth fracture measure value, and Fig. 7b shows an example tooth contour with a low tooth fracture measure value. Both Fig. 7a and 7b show a incisor, however, the incisor shown in Fig. 7a has experienced a degree of tooth wear erosion. In this example, this is illustrated by the “pointed” cutting edge shape illustrated in Fig. 7a, whereas the incisor shown in Fig. 7b has an expected profde for the edge of a incisor.

[0076] It will be appreciated that comparing the edge profde data with reference edge profde data, for the tooth shown in Fig. 7a, would indicate that the tooth is eroded, as the profde of the edge of the tooth shown in Fig. 7a is significantly different to the expected profde of the tooth. This information may be used, for example, to normalise and / or adjust the tooth wear metric accordingly, and indicate that the tooth is eroded.

[0077] Determining the tooth wear metric may comprise normalising the determined tooth wear metric (for example, as determined based on the local derivates or using a neural network, as described above), based on a reference value (for example, a reference value of the tooth wear metric) that is specific to the tooth number of the tooth. It will be appreciated that this reference value will be specific to the tooth number, as different tooth numbers have different expected profiles.

[0078] Therefore, in some embodiments, the method 100 may further comprise determining a tooth number of the tooth, and obtaining the reference edge profile data corresponding to the determined tooth number. For example, the edge profile data of the imaged tooth may be compared to known, typical, and / or default profiles of the corresponding tooth (that is, with the same tooth number) of a generic user. The tooth number may be determined in any suitable manner (for example, analyzing the intra-oral image showing the tooth, analyzing one or more intra-oral images making up a complete scan of the subject’s oral cavity, and / or using any other means and / or suitable sensor information obtained during an intra-oral scan to determine the tooth number of the tooth).

[0079] In some embodiments, analyzing the edge profile data to determine a tooth wear metric may further comprise comparing the edge profile data with corresponding edge profile data, wherein the corresponding edge profile data describes a profile of the edge of another tooth of the subject of the same type as the tooth, and determining and / or adjusting the tooth wear metric based on the comparison.

[0080] It will be appreciated that, between different subjects, there may be significant variation in the shape of teeth of each subject. However, for any particular subject, the teeth on the left side of the subject’s mouth, and the corresponding teeth on the right side of the subject’s mouth (for example, the lower right canine and the lower left canine), will be genetically very similar. It will be appreciated that several types of teeth wear are not symmetric. Therefore, by comparing corresponding teeth (that is, the teeth which are mirrored on either side of the subject’s upper or lower jaw) on the left and right sides ofthe subject’s mouth, tooth wear on one of these teeth can be detected, by comparing the shapes of these corresponding teeth. As these corresponding teeth are expected to have similar shapes (and therefore similar edge profile data), a difference in the shapes and of the edge profile data of these corresponding teeth may indicate tooth wear on one of the teeth.

[0081] In some embodiments, comparing the edge profile data with corresponding edge profile data may comprise aligning the edge profile data and the corresponding edge profile data, and identifying whether any points of the edge profile data are below (or above) corresponding points of the corresponding edge profile data. For example, where one or points of the edge profile data are below corresponding points of the corresponding edge profile data, these points may indicate a notch in the tooth, and thus indicate tooth wear. In some embodiments, analyzing the edge profile data to determine a tooth wear metric may further comprise comparing the edge curvature information with historical edge curvature information, describing a previous curvature of the profile of the edge of the tooth of the subject, and adjusting the tooth wear metric based on the comparison.

[0082] For example, the tooth within the intraoral image may be compared with images of the tooth of the subject that have been previously obtained (for example, during previous intra-oral scans of the subject). These previously obtained images may then be stored (e.g., in a database that may be internal or external to the apparatus and / or the IOD) with an association to an identifier of the subject. The edge curvature information of the tooth in the present intraoral image may be compared to the edge curvature information of the tooth in one or more of these historical images of the subject’s tooth. This comparison may enable any deterioration of the tooth over time to be determined. In some embodiments, to identify the same tooth over time, teeth embeddings and similarity metrics may be used for re-identification of the tooth (for example, in a similar manner to how face identification may be performed). For example, it will be appreciated that comparing the edge profile data with historical edge profile data (for example, by registering the two or multiple profiles (for example, using Procrustes or any other suitable registration technique), for the tooth shown in Fig. 7a (where the historical intraoral image is shown in Fig. 7b), would indicate that the tooth has eroded over time, as the profile of the edge of the tooth shown in Fig. 7a is different to the previous profile of the tooth shown in Fig. 7b. This information may be used, for example, to normalise and / or adjust the tooth wear metric accordingly, and indicate that the tooth has eroded over time.

[0083] In some embodiments, the contour shape (or boundary data) of the tooth may be tracked over time (for example, through the historical image comparison described above) to detect changes in the tooth shape due to wear.

[0084] The resulting tooth wear metric may then be compared to a predetermined threshold. This predetermined threshold may correspond to the type of tooth, or the tooth number, to enable tooth wear across various teeth (with varying expected cutting edges) to be accurately assessed. The tooth may be classified as worn, or tooth wear may be identified, if the determined tooth wear metric exceeds thisthreshold. The severity / degree / level of tooth wear may also be indicated by the tooth wear metric (for example, by an amount the tooth wear metric exceeds the predetermined threshold).

[0085] In some embodiments, the method 100 may further comprise classifying, based on the edge curvature information and / or the edge profde data, a cause of tooth wear of the tooth. The cause of tooth wear of the tooth may comprise one or more of: erosion, accidental damage, bruxism, and excessive brushing force. For example, the edge curvature information and / or the edge profde data of the teeth shown in Fig. 6 is indicative of accidental damage. In Fig. 7a, the edge curvature information and / or the edge profde data of the tooth is indicative of bruxism. In some embodiments, a ML model may be trained, such that it receives as an input, an intra-oral image of a tooth, and outputs a classification result indicating one or more causes of tooth wear of the tooth. Different types of non-smoothness of a tooth edge may therefore be classified. For example, the method may allow differentiation between erosion due to normal usage, and accidental fractures.

[0086] In some embodiments, the method 100 may further comprise classifying, based on the edge curvature information and / or the edge profde data, a severity of tooth wear of the tooth. This may enable the method to further differentiate between different degrees and / or levels of erosion. In some embodiments, the method 100 may determine the severity of tooth wear of the tooth using one or more of: a machine learning technique, a deep learning technique, or a neural network.

[0087] In some embodiments, the method 100 may further comprise generating oral care feedback based on the determined tooth wear metric, and providing the oral care feedback to the subject. For example, if the tooth wear metric is indicative of a presence of tooth wear of the tooth, oral care feedback may be generated and provided to the subject.

[0088] In some embodiments, the oral care feedback may include at least one of oral care routine advice, and dental professional examination advice. For example, if the tooth wear metric is indicative of severe tooth wear of the tooth, oral care feedback including dental professional examination advice (for example, recommending a dentist visit) may be generated and provided to the subject. In another example, if the tooth wear metric is indicative of severe tooth wear of the tooth, oral care feedback including a recommendation to wear a bite or braces for a particular period of time, may be generated and provided to the subject. Additionally or alternatively, where the tooth metric indicates the presence of wear on the tooth, oral care routine advice may be generated and provided to the subject indicating that reduced brushing pressure should be applied to the tooth. Where the method 100 is performed by an IOD that comprises an oral care element (for example, where the IOD is a toothbrush comprising a brush head), the IOD may, for example, adjust a pressure applied by the brush head to the tooth, based on the tooth wear metric. For example, where the tooth metric indicates the presence of wear on a particular tooth, the toothbrush may reduce the pressure (for example, by reducing the toothbrush power) applied by the brush head to that tooth when the IOD is being used by the subject.

[0089] In embodiments in which the methods herein are being executed by an apparatus, the apparatus may be in communication with an oral care device (such as a toothbrush). In theseembodiments, the oral care feedback and / or the tooth wear metric may be provided to the oral care device. The oral care device may be configured to adapt its operation based on the received information. For example, the oral care device may adapt its operation (for example, reducing power when in contact with the tooth), when the tooth wear metric indicates that the tooth is worn.

[0090] Certain methods described herein, for assessing tooth wear, comprise detecting a tooth instance, detecting and parametrizing the cutting-edge boundary, and converting the parametrization into an actionable ‘fracture’ metric. It is noted that neither the tooth instance detection, nor the detection of cutting edge, requires specific annotations by a dentist, which significantly simplifies the training data collection to implement these methods.

[0091] As previously mentioned, the apparatus may be a processing device external to the IOD. For example, the apparatus may be one or more of a cloud computing network, a wireless device, and any other suitable communication device. In such embodiments, the method of Fig. 1 may be performed by one or more processors of the cloud computing network, wireless device, and / or any other suitable processors.

[0092] As noted above, the method described above may be implemented by analysing intra-oral images of a subject’s teeth, that have been obtained by an IOD. The IOD may also be configured to perform plaque detection and / or cavity detection, or the intra-oral images generated by the IOD may be further used for this purpose. That is, intra-oral images that have been obtained for a subject for use in plaque detection and / or cavity detection, may also be used to perform the tooth wear assessment method described herein. Therefore, the tooth wear assessment does not require an extra effort from the user.

[0093] The methods described herein refer generally to 2D intraoral images which may be analyzed to generate the edge profile data. However, the methods described herein may be applied to intraoral images that have been obtained using any type of intraoral imaging modality (for example, 3D intraoral images that have been obtained from 3D intraoral scans of the subject). It will be appreciated that, in these embodiments, in which 3D intraoral images are analyzed, the generated edge profile data may comprise 2D edge surface data (for example, 2D cutting surface data). In these embodiments, the 2D edge surface data may may be analyzed (for example, a metric derived from the 2D curvature may be analyzed) to determine the tooth wear metric. In embodiments in which a 3D intraoral image is processed with a tooth segmentation algorithm to generate boundary data, a part of the 3D surface corresponding to the cutting edge may segmented (for example, based on shape descriptors).

[0094] The methods described herein can therefore be added as a feature of a teeth condition assessment using an imaging oral device, such as a scanner. The assessed tooth wear can determine whether a change in operation of a toothbrush is required, such as reducing toothbrush power when the toothbrush is in contact with damaged teeth, or that a softer power tooth brush head should be used. The measured tooth wear can be tracked over time, and in combination with other teeth metrics (for example, the detection of a cavity), can trigger a recommendation of a dentist visit to the subject.Fig. 8 is a schematic diagram illustrating an apparatus 800 for assessing tooth wear in a subject according to an embodiment. The apparatus 800 may be a wireless device (e.g. a mobile phone, a smart phone, a tablet, a laptop, or any other wireless device), or any other device.

[0095] As illustrated in Fig. 8, the apparatus 800 comprises one or more processors 802. The one or more processors 802 can be implemented in numerous ways, with software and / or hardware, to perform the various functions described herein. The one or more processors 802 can comprise a plurality of software and / or hardware modules, each configured to perform, or that are for performing, individual or multiple steps of the method described herein.

[0096] The one or more processors 802 may comprise, for example, one or more microprocessors, one or more multi -core processors and / or one or more digital signal processors (DSPs), one or more processing units, and / or one or more controllers (e.g. one or more microcontrollers) that may be configured or programmed (e.g. using software or computer program code) to perform the various functions described herein. The one or more processors 802 may be implemented as a combination of dedicated hardware (e.g. amplifiers, pre-amplifiers, analog -to-digital convertors (ADCs) and / or digital -to-analog convertors (DACs)) to perform some functions and one or more processors (e.g. one or more programmed microprocessors, DSPs and associated circuitry) to perform other functions.

[0097] The one or more processors 802 can be configured to perform the method described herein.

[0098] In particular, the apparatus 800 may be for assessing tooth wear in a subject according to an embodiment. In such embodiments, the apparatus may be a processing device that is external to an IOD, in which case the apparatus may be communicatively coupled to the IOD (e.g., via the communications interface 610, discussed in more detail below). In such embodiments, the apparatus 800 may be a wireless device (e.g. a mobile phone, a smart phone, a tablet, a laptop, or any other wireless device), a processor of a cloud computing network, or any other suitable communication device. In particular, the one or more processors 802 may be configured to cause the apparatus to perform the method described with reference to Fig. 1. For example, the one or more processors 802 may be configured to cause the apparatus 800 to perform the embodiments of the method described herein, such as the embodiments of the method described with reference to Fig. 1.

[0099] Alternatively, the apparatus 800 may be, or be incorporated in, an IOD according to an embodiment. In such embodiments, the one or more processors 802 may be configured to cause the apparatus to perform the method described with reference to Fig. 1. For example, the one or more processors 802 may be configured to cause the apparatus 800 to perform the embodiments of the method described herein, such as the embodiments of the method described with reference to Fig. 1.

[0100] It will be appreciated that (regardless of whether the IOD is, is incorporated in, or is communicatively coupled to the apparatus 800), the IOD may comprise an imaging device to obtain intraoral image of a tooth of the subject. The imaging device may comprise an image sensor that is sensitive to visible light. The imaging device may be a camera. An example camera is an RGB camera.However, the imaging device may be any suitable device that can be embedded in an IOD and can capture images of an oral cavity.

[0101] As illustrated in Fig. 8, the apparatus 800 may comprise at least one memory 806.

[0102] Alternatively or in addition, at least one memory 806 may be external to (e.g. separate to or remote from) the apparatus 800. For example, another apparatus may comprise at least one memory 806 according to some embodiments. A hospital database may comprise at least one memory 806, at least one memory 806 may be a cloud computing resource, or similar. The one or more processors 802 of the apparatus 800 may be configured to communicate with and / or connect to at least one memory 806. The at least one memory 806 may comprise any type of non-transitory machine-readable medium, such as cache or system memory including volatile and non-volatile computer memory such as random access memory (RAM), static RAM (SRAM), dynamic RAM (DRAM), read-only memory (ROM), programmable ROM (PROM), erasable PROM (EPROM), and electrically erasable PROM (EEPROM). At least one memory 806 can be configured to store program code that can be executed by the one or more processors 802 of the apparatus 800 to cause the apparatus 800 to operate in the manner described herein.

[0103] Alternatively or in addition, at least one memory 806 can be configured to store information required by or resulting from the method described herein. For example, at least one memory 806 may be configured to store edge profile data, a tooth wear metric, an intraoral image, or any other information, or any combination of information, required by or resulting from the method described herein. The one or more processors 802 of the apparatus 800 can be configured to control at least one memory 806 to store information required by or resulting from the method described herein.

[0104] As illustrated in Fig. 8, the apparatus 800 may comprise at least one user interface 808. Alternatively or in addition, at least one user interface 808 may be external to (e.g. separate to or remote from) the apparatus 800. The one or more processors 802 of the apparatus 800 may be configured to communicate with and / or connect to at least one user interface 808. One or more processors 802 of the apparatus 800 can be configured to control at least one user interface 808 to operate in the manner described herein.

[0105] A user interface 808 can be configured to render (or output, display, or provide) information required by or resulting from the method described herein. For example, one or more user interfaces 808 may be configured to render (or output, display, or provide) oral care feedback, or any other information, or any combination of information, required by or resulting from the method described herein. Alternatively or in addition, one or more user interfaces 808 can be configured to receive a user input. For example, one or more user interfaces 808 may allow a user to manually enter information or instructions, interact with and / or control the apparatus 800. Thus, one or more user interfaces 808 may be any one or more user interfaces that enable the rendering (or outputting, displaying, or providing) of information and / or enables a user to provide a user input.

[0106] The user interface 808 may comprise one or more components for this. For example, one or more user interfaces 808 may comprise one or more switches, one or more buttons, a keypad, akeyboard, a mouse, a display or display screen, a graphical user interface (GUI) such as a touch screen, an application (e.g. on a smart device such as a tablet, a smart phone, or any other smart device), or any other visual component, one or more speakers, one or more microphones or any other audio component, one or more lights (e.g. one or more light emitting diodes, LEDs), a component for providing tactile or haptic feedback (e.g. a vibration function, or any other tactile feedback component), a smart device (e.g. a smart mirror, a tablet, a smart phone, a smart watch, or any other smart device), or any other user interface, or combination of user interfaces. One or more user interfaces that are controlled to render information may be the same as one or more user interfaces that enable the user to provide a user input.

[0107] As illustrated in Fig. 8, the apparatus 800 may comprise at least one communications interface (or communications circuitry) 810. Alternatively or in addition, at least one communications interface 810 may be external to (e.g. separate to or remote from) the apparatus 800. A communications interface 810 can be for enabling the apparatus 800, or components of the apparatus 800 (e.g. one or more processors 802, one or more sensors 804, one or more memories 806, one or more user interfaces 808 and / or any other components of the apparatus 800), to communicate with and / or connect to each other and / or one or more other components. For example, one or more communications interfaces 810 can be for enabling one or more processors 802 of the apparatus 800 to communicate with and / or connect to one or more sensors 804, one or more memories 806, one or more user interfaces 808 and / or any other components of the apparatus 800.

[0108] A communications interface 810 may enable the apparatus 800, or components of the apparatus 800, to communicate and / or connect in any suitable way. For example, one or more communications interfaces 810 may enable the apparatus 800, or components of the apparatus 800, to communicate and / or connect wirelessly, via a wired connection, or via any other communication (or data transfer) mechanism. In some wireless embodiments, for example, one or more communications interfaces 810 may enable the apparatus 800, or components of the apparatus 800, to use radio frequency (RF), Bluetooth, or any other wireless communication technology to communicate and / or connect.

[0109] Fig. 9 is a block diagram illustrating an example processor 900 according to embodiments of the disclosure. Processor 900 may be used to implement one or more processors described herein, for example, processor 802 shown in Fig. 800. Processor 900 may be any suitable processor type including, but not limited to, a microprocessor, a microcontroller, a digital signal processor (DSP), a field programmable array (FPGA) where the FPGA has been programmed to form a processor, a graphical processing unit (GPU), an application specific circuit (ASIC) where the ASIC has been designed to form a processor, or a combination thereof.

[0110] The processor 900 may include one or more cores 902. The core 902 may include one or more arithmetic logic units (AEU) 904. In some embodiments, the core 902 may include a floating point logic unit (FPLU) 906 and / or a digital signal processing unit (DSPU) 908 in addition to or instead of the ALU 904.The processor 900 may include one or more registers 912 communicatively coupled to the core 902. The registers 912 may be implemented using dedicated logic gate circuits (e.g., flip-flops) and / or any memory technology. In some embodiments the registers 912 may be implemented using static memory. The register may provide data, instructions and addresses to the core 902. In some embodiments, processor 900 may include one or more levels of cache memory 910 communicatively coupled to the core 902. The cache memory 910 may provide computer-readable instructions to the core 902 for execution. The cache memory 910 may provide data for processing by the core 902. In some embodiments, the computer-readable instructions may have been provided to the cache memory 910 by a local memory, for example, local memory attached to the external bus 916. The cache memory 910 may be implemented with any suitable cache memory type, for example, metal -oxide semiconductor (MOS) memory such as static random access memory (SRAM), dynamic random access memory (DRAM), and / or any other suitable memory technology.

[0111] The processor 900 may include a controller 914, which may control input to the processor 900 from other processors and / or components included in a system and / or outputs from the processor 900 to other processors and / or components included in the system. Controller 914 may control the data paths in the ALU 904, FPLU 906 and / or DSPU 908. Controller 914 may be implemented as one or more state machines, data paths and / or dedicated control logic. The gates of controller 914 may be implemented as standalone gates, FPGA, ASIC, or any other suitable technology. The registers 912 and the cache 910 may communicate with controller 914 and core 902 via internal connections 920A, 920B, 920C and 920D. Internal connections may implemented as a bus, multiplexor, crossbar switch, and / or any other suitable connection technology.

[0112] Inputs and outputs for the processor 900 may be provided via a bus 916, which may include one or more conductive lines. The bus 916 may be communicatively coupled to one or more components of processor 900, for example the controller 914, cache 910, and / or register 912. The bus 916 may be coupled to one or more components of the system.

[0113] The bus 916 may be coupled to one or more external memories. The external memories may include Read Only Memory (ROM) 932. ROM 932 may be a masked ROM, Electronically Programmable Read Only Memory (EPROM) or any other suitable technology. The external memory may include Random Access Memory (RAM) 933. RAM 933 may be a static RAM, battery backed up static RAM, Dynamic RAM (DRAM) or any other suitable technology. The external memory may include Electrically Erasable Programmable Read Only Memory (EEPROM) 935. The external memory may include Flash memory 934. The External memory may include a magnetic storage device such as disc 936. In some embodiments, the external memories may be included in a system.

[0114] There is provided a computer program comprising instructions which, when executed by a processor (such as one or more processors 802 of the apparatus 800 or the processor 902), cause the processor to perform at least part of, or all of, the method described herein.There is provided a computer program product comprising a computer readable medium. The computer readable medium has a computer readable code embodied therein. The computer readable code is configured such that, on execution by a suitable computer or processor (such as one or more processors 802 of the apparatus 800 or the processor 902), the computer or processor is caused to perform the method described herein. The computer readable medium may be, for example, any entity or device capable of carrying the computer program product. For example, the computer readable medium may include a data storage, such as a ROM (such as a CD-ROM or a semiconductor ROM) or a magnetic recording medium (such as a hard disk). Furthermore, the computer readable medium may be a transmissible carrier, such as an electric or optical signal, which may be conveyed via electric or optical cable or by radio or other means. When the computer program product is embodied in such a signal, the computer readable medium may be constituted by such a cable or other device or means. Alternatively, the computer readable medium may be an integrated circuit in which the computer program product is embedded, the integrated circuit being adapted to perform, or used in the performance of, the method described herein.

[0115] There is thus provided herein an apparatus, method, and computer program product for assessing tooth wear in a subject, which address the limitations associated with the existing techniques.

[0116] It will be understood that at least some or all of the method steps described herein can be automated. That is, at least some or all of the method steps described herein can be performed automatically. The method described herein can be a computer-implemented method.

[0117] While the invention has been illustrated and described in detail in the drawings and foregoing description, such illustration and description are to be considered illustrative or exemplary and not restrictive; the invention is not limited to the disclosed embodiments.

[0118] Variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the principles and techniques described herein, from a study of the drawings, the disclosure and the appended claims. In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite article "a" or "an" does not exclude a plurality. A single processor or other unit may fulfil the functions of several items recited in the claims. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage. A computer program may be stored or distributed on a suitable medium, such as an optical storage medium or a solid-state medium supplied together with or as part of other hardware, but may also be distributed in other forms, such as via the Internet or other wired or wireless telecommunication systems. Any reference signs in the claims should not be construed as limiting the scope.

Claims

CLAIMS:

1. A method for assessing tooth wear in a subject, the method comprising:generating edge profde data describing a profde of an edge of a tooth of the subject; and analyzing the edge profde data to determine a tooth wear metric, wherein the tooth wear metric is indicative of a presence of tooth wear of the tooth and / or a severity of tooth wear of the tooth.

2. The method of claim 1, wherein the edge of the tooth is an incisal edge of the tooth, or a cutting edge of the tooth, wherein the method further comprises:obtaining an intraoral image of a tooth of the subject, and wherein generating the edge profde data comprises analyzing the intraoral image to generate the edge profde data3. The method of claim 2, wherein analyzing the intraoral image to generate the edge profde data comprises:analyzing the intraoral image to generate boundary data describing a boundary of the tooth; andgenerating the edge profde data based on the intra-oral image and / or the boundary data, wherein the edge profde data is a part of the boundary data.

4. The method of claim 3, wherein analyzing the intraoral image to generate the boundary data comprises processing the intraoral image with a tooth segmentation algorithm.

5. The method of claim 3 or 4, wherein generating the edge profde data comprises:generating the edge profde data based on the boundary data; orprocessing the intraoral image with an image analysis algorithm.

6. The method of any preceding claim, wherein analyzing the edge profde data to determine a tooth wear metric comprises:determining edge curvature information based on the edge profde data, wherein the edge curvature information describes the curvature of the profde of the edge; andanalyzing the edge curvature information to determine the tooth wear metric.

7. The method of claim 6, wherein analyzing the edge profde data to determine a tooth wear metric further comprises:comparing the edge profile data with corresponding edge profile data, wherein the corresponding edge profile data describes a profile of the edge of another tooth of the subject of the same type as the tooth, and determining and / or adjusting the tooth wear metric based on the comparison.

8. The method of claim 6 or 7, wherein analyzing the edge profile data to determine a tooth wear metric further comprises:comparing the edge profile data with reference edge profile data, wherein the reference edge profile data describes an expected profile of the edge of the tooth; andadjusting the tooth wear metric based on the comparison.

9. The method of claim 8, wherein the method further comprises:determining a tooth number of the tooth; andobtaining the reference edge profile data corresponding to the determined tooth number.

10. The method of any of claims 6-9, wherein analyzing the edge profile data to determine a tooth wear metric further comprises:comparing the edge curvature information with historical edge curvature information, describing a previous curvature of the profile of the edge of the tooth of the subject; andadjusting the tooth wear metric based on the comparison.

11. The method of any preceding claim, wherein the method further comprises:classifying, based on the edge curvature information and / or the edge profile data, a cause of tooth wear of the tooth.

12. The method of any preceding claim, the method further comprising:generating oral care feedback based on the determined tooth wear metric; and providing the oral care feedback to the subject.

13. The method of claim 12, wherein the oral care feedback includes at least one of oral care routine advice, and dental professional examination advice.

14. An apparatus for assessing tooth wear in a subject, the apparatus comprising one or more processors configured to cause the apparatus to:generate edge profile data describing a profile of an edge of a tooth of the subject; and analyze the edge profile data to determine a tooth wear metric, wherein the tooth wear metric is indicative of a presence of tooth wear of the tooth and / or a severity of tooth wear of the tooth.

15. A computer program product comprising a computer readable medium, the computer readable medium having a computer readable code embodied therein, the computer readable code being configured such that, on execution by a suitable computer or processor:the computer or processor performs the method as claimed in any one of claims 1 to 13.

Images