Gum recession assessment

Abstract

A method and processing device configured to assesses gum recession based on analyzing fluorescence image data of dental surfaces illuminated with fluorescence-stimulating light to compute one or more gum recession indicators.

Description

[0001] 2025PF00028

[0002] 29.09.2025

[0003] GUM RECESSION ASSESSMENT

[0004] FIELD OF THE INVENTION

[0005] The present invention relates to the field of personal oral health devices and in particular devices having functionality for assessing one or more oral health indicators.

[0006] BACKGROUND OF THE INVENTION

[0007] Receding gum is a condition characterized by the gum tissue around the teeth pulling back or wearing away, which exposes more of the tooth or even the tooth’s root. This process typically occurs gradually and is often linked to gum disease or improper oral care. The consequences of receding gums include an increased risk of cavities in exposed roots, and in severe instances, it can lead to tooth loss. Additionally, individuals with receding gums often experience tooth sensitivity to hot, cold, or sweet foods.

[0008] Awareness of receding gums is important. If the condition develops rapidly, it can be a sign of severe gum disease, such as periodontitis, which requires professional diagnosis and intervention. Receding gums can also result from brushing with a too high force, indicating overbrushing or that the gums are sensitive to toothbrush abrasion. Therefore, there is a need for individuals to be made aware of this condition to enable them to adapt their brushing behavior or seek professional guidance.

[0009] SUMMARY OF THE INVENTION

[0010] The invention is defined by the claims.

[0011] An aspect of the invention is a method for assessing gum recession of a subject, the method comprising: receiving image data representative of one or more dental surfaces within the oral cavity of the subject under illumination by a stimulation light for stimulating fluorescence of tooth enamel and / or dentin; determining at least one gum recession indicator based on image analysis applied to image data captured by the imaging device; and generating a data output based on the computed gum recession indicator.

[0012] The proposed method allows for the assessment of gum recession by utilizing fluorescence imaging, which makes the distinction between tooth tissue and gum tissue clearer, thereby facilitating the detection of receding gums. In particular, within captured images, exposed dental surfaces will appear as green areas while surrounding gum regions will appear as darker regions.

[0013] A further aspect of the invention is a computer program product comprising computer program code configured, when run on a processor, to cause the processor to perform a method in accordance with any embodiment described in this disclosure or in accordance with any claim.2025PF00028

[0014] 2 29.09.2025

[0015] A further aspect of the invention is a processing device comprising one or more processors configured to perform a method in accordance with any embodiment described in this disclosure or in accordance with any claim.

[0016] The method may further comprise recording the computed gum recession indicator in a log, for example stored on a memory (local or remote, e.g. in the cloud).

[0017] In some embodiments, the stimulation light may have a wavelength within the range 380 to 420 nm, for example approximately 405 nm. For example, the stimulation light may comprise at least one wavelength component within the range 380 to 420 nm, for example approximately 405 nm. For example, the stimulation light may comprise a peak wavelength component within the range 380 to 420 nm, for example approximately 405 nm. This corresponds for example to blue-violet, or near ultraviolet light. It is the wavelength of light used in Quantitative Light-induced Fluorescence (QLF) imaging. Using this specific wavelength range, typical for QLF imaging, effectively stimulates fluorescence of tooth enamel and / or dentin, which makes tooth tissue appear green and gum tissue black or very dark purple, making the boundary between tooth and gum more visible.

[0018] In some embodiments, the image analysis may comprise detecting a gum-tooth boundary line within a captured image. A gum -tooth boundary line means a boundary between the gum and an exposed dental surface of a dental structure. This includes, for example, the boundary between the gum and the tooth part of the dental structure and the boundary between the gum and any exposed root part of the dental structure.

[0019] Different methods may be applied to detect the gum -tooth boundary line. One example is to segment the whole visible tooth area (e.g., based on color or intensity), segment the visible gum area (e.g., based on color or spectral characteristics), and identify the boundary between the two by applying morphological dilation to the segmented areas and detecting the resulting line of overlap. Another approach is to directly extract the boundary contour of the segmented tooth area and to identify, for example by shape prediction or model-based analysis, the portion of this contour corresponding to the gum-tooth boundary.

[0020] In some embodiments, the image analysis may further comprise a shape classification operation for classifying a shape of a detected gum -tooth boundary line. The method may comprise classifying a gum recession state of a tooth associated with a detected gum-tooth boundary line based on the shape classification, and the gum recession indicator may be computed at least in part based on the classified gum recession state of one or more imaged teeth. For example, the method may comprise classifying a tooth as having a normal gumline type or a receding gumline type. A receding gumline is typically associated with a more triangular shape than a normal gumline which exhibits a more shallow curvature. Classifying the shape of the gum-tooth boundary line provides a clear and objective way to identify and quantify gum recession.

[0021] In some embodiments, the image analysis may comprise a segmentation operation configured to segment dental surface areas within a captured image, and a quantification operation2025PF00028

[0022] 3 29.09.2025

[0023] configured to estimate, based on the segmented dental surface area for a tooth, an indication of an area size of an exposed root portion of the dental surface area. The gum recession indicator may be computed at least in part based on the estimated root portion area size for one or more imaged teeth. Quantifying the exposed root portion area allows for a direct measurement of the extent of gum recession, providing a quantitative indicator for assessment and tracking.

[0024] In some embodiments, the quantification operation may comprise estimating a total area size of the segmented dental surface area of a tooth; applying a trained neural network configured to predict an area size of a crown portion of an imaged dental surface area of a tooth based on an image input representative of the tooth dental surface area; and estimating an area size of an exposed root portion of the dental surface based on subtraction of the estimated area of the crown portion from the estimated total area size.

[0025] In some embodiments, the method may further comprise recording a log of computed gum recession indicators; computing tracking information based on changes in recorded gum recession indicators over time; and generating a data output indicative of the tracking information. The log may, for example, be stored on a memory. The memory may be local to the device or may be separate to the device. It may be a remote memory, e.g. a cloud-based data store. Additionally, or alternatively, it may comprise a memory of a mobile computing device with which the oral sensing device is communicatively coupled.

[0026] In some embodiments, the image analysis may comprise application of one or more trained neural networks, for example one or more convolutional neural networks or transformer-based neural networks.

[0027] In some embodiments, the image data may be image data captured by an imaging device being sensitive at least to fluorescence light emitted by tooth enamel and / or dentin when stimulated by the stimulation light.

[0028] In some embodiments, the method may further comprise controlling emission of the stimulation light by a light emitter comprising a light source operable to emit the stimulation light suitable for stimulating fluorescence of tooth enamel and / or dentin; and controlling acquisition of the image data by means of an imaging device being sensitive at least to fluorescence light emitted by tooth enamel and / or dentin when stimulated by the stimulation light.

[0029] In some embodiments, the imaging device may be sensitive to fluorescence light in a wavelength range of 510 to 520 nm. Optionally, the imaging device may comprise an RGB camera.

[0030] In some embodiments, the light emitter comprises a further light source operable to emit white visible light, and the imaging device is further sensitive to the white visible light emitted by the further light source. In some embodiments, the method comprises selectively operating in a selected one of: a fluorescence imaging mode in which image data is acquired of oral surfaces illuminated using the stimulation light, and a non-fluorescence imaging mode in which image data is acquired of oral surfaces illuminated using the white visible light.2025PF00028

[0031] 4 29.09.2025

[0032] In some embodiments, the method may further comprise communicating with an oral care device operable in use to perform an oral care function, and controlling configuration of at least one operational characteristic of the oral care function in dependence upon the at least one gum recession indicator. In some embodiments, the oral care device may comprise a powered toothbrush device and wherein the at least one operational characteristic may comprise a brushing intensity level. In other words, the method may comprise triggering adaptation of at least one operational setting of the oral care device based on the at least one gum recession indicator.

[0033] By way of example, a powered toothbrush may be adapted to a softer brushing mode responsive to the gum recession indicator indicating receding gums. Automated adaptation of an oral care device, such as a powered toothbrush, based on detected gum recession helps prevent further damage and supports improved oral care habits, directly addressing the cause or symptom of recession without requiring explicit user action.

[0034] In some embodiments, the method may further comprise outputting, using a user interface device, user feedback indicative of gum condition based on the at least one gum recession indicator.

[0035] Additionally or alternatively, the method may comprise outputting, using a user interface device, user feedback indicative of recommended actions responsive to the gum recession indicator indicating receding gums. By way of one example, the recommended actions may include using a different brush head attachment type, selected from a set of available brush head attachment types. For example, responsive to detecting receding gums, a recommendation may be generated recommending that the user attach a softer toothbrush head.

[0036] As mentioned previously, an aspect of the invention is a processing device comprising one or more processors configured to perform a method in accordance with any embodiment described in this disclosure or in accordance with any claim.

[0037] A further aspect of the invention is a system, comprising: an oral sensing device comprising: a mouth-insertable portion; a light emitter carried by the mouth-insertable portion, the light emitter comprising a light source operable to emit stimulation light suitable for stimulating fluorescence of tooth enamel and / or dentin; an imaging device carried by the mouth-insertable portion for imaging oral surfaces illuminated by the light emitter, the imaging device being sensitive at least to fluorescence light emitted by tooth enamel and / or dentin when stimulated by the stimulation light; and a processing device as described above, configured to determine the at least one gum recession indicator based on image analysis applied to image data captured by the imaging device.

[0038] In some embodiments, the system may further comprise the light emitter comprising a further light source operable to emit white visible light; wherein the imaging device is sensitive to the white visible light emitted by the further light source; and wherein the processing device is operable to selectively control the oral sensing device in a selected one of: a fluorescence imaging mode in which image data is acquired of oral surfaces illuminated using the stimulation light, and a non-fluorescence imaging mode in which image data is acquired of oral surfaces illuminated using the white visible light.2025PF00028

[0039] 5 29.09.2025

[0040] The white visible light may be of a spectral composition which does not stimulate fluorescence. Providing multiple imaging modes (fluorescence and white light) within the same system allows for versatile oral examination, as white light can offer general visual context while fluorescence specifically highlights gum recession due to differential tissue appearance, enhancing diagnostic capabilities.

[0041] In some embodiments, the processing device may be selectively operable in at least one mode in which the processing device is configured to compute the gum recession indicator based on image analysis applied to a combination of image data acquired in both the fluorescence imaging mode and the non-fluorescence imaging mode. This multimodal approach may involve acquiring image data sequentially in a controlled multi-modal image capture sequence, wherein the device automatically switches between illumination modes to capture complementary image information of the same oral region. The processing device may implement fusion algorithms that combine analysis results from both imaging modes, or may utilize one or more neural networks configured to process both fluorescence and white light image data simultaneously suing multiple input channels. For example, the neural network architecture may include separate processing pathways for each image modality that converge at intermediate or final layers to generate integrated recession assessments.

[0042] These and other aspects of the invention will be apparent from and elucidated with reference to the embodiment(s) described hereinafter.

[0043] BRIEF DESCRIPTION OF THE DRAWINGS

[0044] For a better understanding of the invention, and to show more clearly how it may be carried into effect, reference will now be made, by way of example only, to the accompanying drawings, in which:

[0045] Fig. 1 outlines steps of an example method in accordance with one or more embodiments of the invention;

[0046] Fig. 2 is a block diagram of an example processing device and system in accordance with one or more embodiments of the invention;

[0047] Fig. 3 schematically illustrates an example oral sensing device as may be utilized in accordance with one or more embodiments of the invention;

[0048] Fig. 4 schematically illustrates detection of a gum-tooth boundary line within acquired image data in accordance with one or more embodiments; and

[0049] Fig. 5 schematically illustrates detection of an exposed root area portion of an imaged dental surface.

[0050] DETAILED DESCRIPTION OF THE EMBODIMENTS

[0051] The invention will be described with reference to the Figures.

[0052] It should be understood that the detailed description and specific examples, while indicating exemplary embodiments of the apparatus, systems and methods, are intended for purposes of2025PF00028

[0053] 6 29.09.2025

[0054] illustration only and are not intended to limit the scope of the invention. These and other features, aspects, and advantages of the apparatus, systems and methods of the present invention will become better understood from the following description, appended claims, and accompanying drawings. It should be understood that the Figures are merely schematic and are not drawn to scale. It should also be understood that the same reference numerals are used throughout the Figures to indicate the same or similar parts.

[0055] Fig. 1 outlines in block diagram form steps of an example method according to one or more embodiments. The steps will be recited in summary, before being explained further in the form of example embodiments.

[0056] The provided method 10 is for assessing gum recession of a subject.

[0057] The method 10 comprises receiving 12 image data representative of one or more dental surfaces within the oral cavity of the subject under illumination by a stimulation light for stimulating fluorescence of tooth enamel and / or dentin. Fluorescence is the emission of light by a substance after it has absorbed light or other electromagnetic radiation. In the context of teeth, the stimulation light causes the tooth enamel and / or dentin to emit light at a different wavelength. Optionally, in some embodiments, the stimulation light has a wavelength within the range of 380 to 420 nm. For example, the wavelength may be approximately 405 nm. This stimulation light may comprise at least one wavelength component within this range, or it may comprise a peak wavelength component within this range. This wavelength range corresponds to blue-violet, or near ultraviolet light. This is the wavelength of light used in Quantitative Light-induced Fluorescence (QLF) imaging. When the tooth is illuminated with QLF light, tooth tissue appears green, and gum tissue appears black or very dark purple. This contrast makes receding gum more visible compared to images taken under white light.

[0058] The method 10 further comprises determining 14 at least one gum recession indicator based on image analysis applied to image data captured by the imaging device. The imaging device is optionally sensitive at least to fluorescence light emitted by tooth enamel and / or dentin when stimulated by the stimulation light. In some embodiments, the imaging device is sensitive to fluorescence light in a wavelength range of 510 to 520 nm. This imaging device may optionally comprise an RGB camera.

[0059] The method 10 further comprises generating 16 a data output based on the computed gum recession indicator.

[0060] The method may further comprise controlling emission of the stimulation light by a light emitter comprising a light source operable to emit the stimulation light suitable for stimulating fluorescence of tooth enamel and / or dentin. The method may further comprise controlling acquisition of the image data by means of an imaging device being sensitive at least to fluorescence light emitted by tooth enamel and / or dentin when stimulated by the stimulation light.

[0061] As noted above, the method can also be embodied in hardware form, for example in the form of a processing device which is configured to carry out a method in accordance with any embodiment described in this disclosure, or in accordance with any claim.2025PF00028

[0062] 7 29.09.2025

[0063] To further aid understanding, Fig. 2 presents a schematic representation of an example processing device 32 configured to execute a method in accordance with one or more embodiments of the invention. The processing device is shown in the context of a system 30 which comprises the processing device. The processing device alone represents an aspect of the invention. The system 30 is another aspect of the invention. The provided system need not comprise all the illustrated hardware components and it may comprise only a subset.

[0064] The processing device 32 comprises one or more processors 36 configured to perform a method in accordance with that outlined above, or in accordance with any embodiment described in this disclosure, or in accordance with any claim. In the illustrated example, the processing device further comprises an input / output or communication interface 34.

[0065] In the illustrated example of Fig. 2, the system 30 further comprises an oral sensing device 40 comprising a mouth-insertable portion 42. These elements are indicated schematically in the block diagram of Fig. 2.

[0066] The system 30 further comprises a light emitter 60 carried by the mouth-insertable portion 42. The light emitter 60 comprises a light source operable to emit stimulation light suitable for stimulating fluorescence of tooth enamel and / or dentin. The system 30 further comprises an imaging device 70 carried by the mouth-insertable portion 42 for imaging oral surfaces illuminated by the light emitter. The imaging device is sensitive at least to fluorescence light emitted by tooth enamel and / or dentin when stimulated by the stimulation light.

[0067] The processing device 32 is configured to determine the at least one gum recession indicator based on image analysis applied to image data captured by the imaging device 70.

[0068] The system 30 may further comprise a memory 38 for storing computer program code (i.e. computer-executable code) which is configured for causing the one or more processors 36 of the processing unit 32 to perform the method as outlined above, or in accordance with any embodiment described in this disclosure, or in accordance with any claim.

[0069] As mentioned previously, the invention can also be embodied in software form. Thus another aspect of the invention is a computer program product comprising computer program code configured, when run on a processor, to cause the processor to perform a method in accordance with any embodiment described in this disclosure, or in accordance with any claim.

[0070] As discussed above, the method comprises computing at least one gum recession indicator.

[0071] Regarding the gum recession indicator, various options are possible, as will be explained in more detail later. In general terms, the gum recession indicator may comprise a determined metric or classification indicative of a state of gum recession for one or more dental surfaces.

[0072] A gum recession indicator may, by way of example, be computed pertaining to a single tooth, a group of teeth, an area of the mouth, or the whole mouth. In some embodiments, more than one gum recession indicator may be computed. In some embodiments, one or more gum recession indicators2025PF00028

[0073] 8 29.09.2025

[0074] may be computed for each of a plurality of teeth or a plurality of groups of teeth. In some embodiments, a composite or amalgamated gum recession indicator may be computed based on a combination of the gum recession indicators of individual teeth or groups of teeth.

[0075] By way of example, in some embodiments, the gum recession indicator comprises a classification of a gum recession state of a tooth. The classification may be selected from a discrete set of classifications. In some embodiments, the classification may be a binary classification, e.g. normal vs abnormal. Additionally or alternatively, the classification may be selected from a discrete set of three or more classifications, for example grades of gum recession. In some embodiments, the classification of the gum recession state may be determined based on a shape or geometry, e.g. a curvature, of the gum-tooth boundary line. In some embodiments, the classification of the gum recession state may be determined based on a length of the gum -tooth boundary line. In some embodiments, the classification of the gum recession state may be determined based on an area size of an exposed root portion of the dental surface of a tooth. In some embodiments, the classification of the gum recession state may be determined based on a detected change in any one of the above-mentioned metrics or measures between two time points.

[0076] In some embodiments, the gum recession indicator may comprise at least one quantitative gum recession indicator. For example, in some embodiments, the gum recession indicator may be computed as a function of an estimated area size of an exposed root portion. As will be explained below, this may be quantified by subtracting a predicted crown portion area from a total segmented dental surface area. In some embodiments, a quantitative gum recession indicator may be computed based on determining a quantitative gum-tooth boundary length metric or quantitative gum-tooth boundary shape metric.

[0077] In some embodiments, values of one or more gum recession indicators may be tracked or trended over time to generate trend information. This trend information may for example be computed from recorded logs of historically computed gum recession indicator values.

[0078] The trend analysis may enable early detection and prevention of gum recession by identifying subtle changes or patterns that precede clinically apparent recession. For example, the processing device may be configured to detect gradual changes in gum-tooth boundary characteristics, progressive increases in exposed tooth surface area, or evolving shape patterns that indicate the onset of recession processes even when current measurements remain within normal ranges. This predictive capability may allow for early intervention and preventive measures to be implemented before recession becomes clinically significant.

[0079] As mentioned previously, the method may comprise generating a data output based on the computed gum recession indicator. By way of example, the method may further comprise outputting, using a user interface device, user feedback indicative of gum condition based on the at least one gum recession indicator.

[0080] The user interface may comprise a graphical user interface including a display configured to render a representation of the one or more gum recession indicators or information derived therefrom.2025PF00028

[0081] 9 29.09.2025

[0082] In certain embodiments, the presentation may take the form of a mouth map visual in which different regions of the dentition are depicted and annotated with corresponding gum recession indicators. In some implementations, visual indicia may be displayed (such as colored overlays, highlighting, or warning symbols) identifying particular teeth or areas of the mouth for which the computed indicators are representative of pathology or of a condition requiring monitoring. The presentation may further comprise graphical elements such as color-coding, shading, or symbolic markers to convey the degree of recession or severity level in an intuitive manner. The user interface may be implemented on a mobile computing device, such as a smartphone or tablet, thereby enabling convenient access by patients or practitioners in clinical or non-clinical settings. In such embodiments, the communicated data may be dynamically updated as new measurements are obtained, permitting real-time visualization and tracking of gum health overtime.

[0083] In some embodiments, the method may comprise using the computed at least one gum recession indicator to configure operation of an associated oral care device. For example, in some embodiments, the method may further comprise communicating with an oral care device operable in use to perform an oral care function, and controlling configuration of at least one operational characteristic of the oral care function in dependence upon the at least one gum recession indicator. In other words, the method comprises triggering adaptation of at least one operational setting of the oral care device based on the at least one gum recession indicator.

[0084] By way of one non-limiting example, the oral care device may comprise a powered toothbrush device and wherein the at least one operational characteristic comprises a brushing intensity level.

[0085] By way of further non-limiting examples, the oral care device may comprise a powered irrigator device, wherein the at least one operational characteristic comprises a fluid pressure level or pulsing pattern applied during use. In another embodiment, the oral care device may comprise a lightbased or vibration-based therapy device, wherein the at least one operational characteristic comprises an energy output level, frequency, or treatment duration. In such embodiments, the computed gum recession indicator may thus be used to automatically adapt the oral care function to provide an intensity or treatment profde appropriate to the condition of the user’s gum tissue.

[0086] The oral care device may be associated with an oral care device controller which is configured to control the at least one operational characteristic. The method may comprise communicating with the oral care device controller to cause modification or configuration of the at least one operational characteristic.

[0087] The method may in some embodiments be performed at least in part by a processing device of a mobile communication device, wherein the mobile communication device is in communicative relationship with the oral care device and the oral sensing device. The processing device of the mobile communication device may be configured to execute a software application which is2025PF00028

[0088] 10 29.09.2025

[0089] configured to monitor gum recession indicator values measured by the oral sensing device and control the at least one operational characteristic of the oral care device in dependence thereon.

[0090] As noted above, in some embodiments, a powered toothbrush may be adapted to a softer brushing mode responsive to the gum recession indicator indicating receding gums. In some embodiments, the adaptation may be applied globally across the device so that all brushing actions are performed at a reduced intensity. In other embodiments, the adaptation may be applied selectively, such that softer brushing is effected only at one or more detected locations corresponding to areas of gum recession, while standard brushing intensity is maintained elsewhere.

[0091] In further embodiments, the oral care device may comprise a powered oral irrigator device, and wherein a fluid pressure level is adapted to a less powerful pressure level responsive to the gum recession indicator indicating receding gums.

[0092] Again, this may be applied globally across the mouth or locally. The reduction in fluid pressure may be particularly beneficial for areas with gum recession because recession can lead to exposed root dentin, which may result in dentin hypersensitivity. Individuals with dentin hypersensitivity often experience pain when high pressure is applied to exposed dentin locations, so detecting such areas and automatically reducing the irrigation pressure locally may help prevent discomfort during oral care routines. In some embodiments, the processing device may be configured to identify specific locations where root surfaces are exposed and automatically adjust the irrigator settings to provide gentler treatment in those regions while maintaining standard pressure levels in areas with healthy gum coverage.

[0093] Additionally or alternatively, the method may comprise outputting, using a user interface device, user feedback indicative of recommended actions responsive to the gum recession indicator indicating receding gums. By way of one example, the recommended actions may include using a different brush head attachment type for a powered toothbrush, selected from a set of available brush head attachment types. For example, responsive to detecting receding gums, a recommendation may be generated recommending that the user attach a softer toothbrush head.

[0094] In some embodiments, the communicated data output may additionally be employed to provide the user with feedback or informational content through an associated software application. The application may be configured to present knowledge articles, guidance materials, or recommendations relating to gum recession and measures that may be taken by the user to address or manage the condition. In some embodiments, the application may track the progression of the gum recession indicators over time, thereby enabling the user to monitor changes in gum health and to receive longitudinal insights or alerts based on trends detected in the computed indicators.

[0095] As mentioned above, the method comprises processing image data representative of one or more dental surfaces within the oral cavity of the subject under illumination by a stimulation light for stimulating fluorescence of tooth enamel and / or dentin.

[0096] Tooth structures, in particular enamel and dentin, contain organic and inorganic constituents that exhibit natural fluorescence when subjected to excitation by light of appropriate2025PF00028

[0097] 11 29.09.2025

[0098] wavelength. In healthy enamel, this phenomenon generally produces an emission in the blue to green portion of the visible spectrum, while dentin tends to fluoresce more strongly and with a spectral profile shifted toward the yellow-green range.

[0099] Gingival tissue (gum tissue) also exhibits fluorescence, though typically weaker and spectrally distinct from that of enamel and dentin. Owing to its higher content of proteins, hemoglobin, and other chromophores, gum tissue tends to fluoresce with broader emission shifted toward the red region of the spectrum. As a result, under illumination with stimulation light, tooth tissues and gum tissues present markedly different fluorescence responses. This spectral contrast permits clearer delineation of the gumline relative to the tooth surface, thereby enabling the boundary between gum and tooth to be identified more readily in image data. Such identification may then be used in the computation of gum recession indicators, since more accurate detection of the gingival margin relative to the tooth surface is achievable.

[0100] In general, the stimulation light may be provided in the ultraviolet to blue spectral range, for example within about 350 to 450 nm, such that fluorescence of enamel, dentin, and gingival tissue is effectively stimulated and emissions are generated in the longer wavelength visible range, typically between about 450 to 600 nm. In certain preferred embodiments, the stimulation light has a wavelength within the range of 380 to 420 nm, for example approximately 405 nm. For example, the stimulation light may comprise at least one wavelength component within the range of 380 to 420 nm, or may comprise a peak wavelength component within the range of 380 to 420 nm, for example approximately 405 nm. Light within this narrower band corresponds to blue-violet or near-ultraviolet light, and is the wavelength region commonly employed in Quantitative Light-induced Fluorescence (QLF) imaging. Use of this range of stimulation light has been found to provide strong and diagnostically useful fluorescence from tooth structures, while also enhancing contrast between teeth and surrounding gum tissue.

[0101] As noted above, the oral sensing device includes a light emitter 60 that is configured to emit stimulation light of a spectral composition suitable for stimulating fluorescence of tooth enamel and / or dentin.

[0102] By way of example, the light emitter may comprise one or more solid-state light sources, such as light emitting diodes (LEDs) or laser diodes, configured to emit stimulation light of the desired spectral composition. In certain embodiments, the light emitter may be integrated into, or mounted upon, a mouth-receivable portion 42 of the oral sensing device 40 so as to direct the stimulation light toward one or more dental surfaces within the oral cavity when the device is in use. The light emitter may be arranged in an array or distributed around a surface of the mouth-receivable portion to provide uniform illumination of the teeth and gum line. In some implementations, the light emitter may further comprise optical components such as one or more lenses, diffusers, or waveguides to shape, spread, or otherwise control the spatial distribution of the emitted light, thereby enhancing coverage and reducing glare or shadowing. The integration of the light emitter into the mouth-receivable portion allows for localized2025PF00028

[0103] 12 29.09.2025

[0104] delivery of the stimulation light near the target areas, thereby improving efficiency of fluorescence excitation and facilitating reliable image capture.

[0105] As discussed above, the method comprises receiving 12 image data representative of one or more dental surfaces within the oral cavity of the subject under illumination by a stimulation light for stimulating fluorescence of tooth enamel and / or dentin, and wherein the image data is image data captured by an imaging device 70 being sensitive at least to fluorescence light emitted by tooth enamel and / or dentin when stimulated by the stimulation light.

[0106] In some embodiments, the imaging device may be configured to detect fluorescence emissions in a broad wavelength range, for example between about 450 to 600 nm, thereby capturing the characteristic emission responses of both enamel and dentin. The imaging device may be sensitive at least to a narrower wavelength band within this range, for example approximately 510 to 520 nm, where the fluorescence signal of enamel is particularly strong and provides enhanced contrast relative to adjacent gum tissue.

[0107] In a practical implementation, the imaging device may comprise an RGB camera, which inherently provides sensitivity across the visible spectrum and can therefore be utilized in both reflective imaging and fluorescence imaging modes. For instance, when the stimulation light is inactive, the RGB camera may operate in reflective mode to capture conventional color images of the oral cavity. When the stimulation light is active, the same RGB camera may operate in fluorescence mode, capturing images in which exposed dental surfaces appear as bright or green-colored regions, while surrounding gum tissue appears darker due to its weaker fluorescence. This dual -mode functionality enables both conventional imaging and quantitative light-induced fluorescence (QLF) imaging to be achieved using a simple imaging device.

[0108] As discussed above, certain embodiments employ use of an oral sensing device 40 for acquiring the image data.

[0109] Fig. 3 schematically illustrates an example oral sensing device 40.

[0110] The oral sensing device 40 comprises a mouth-insertable portion 42 for being received in the oral cavity of a user. The device further comprises a stem portion 44 coupled with and extending from the mouth insertable portion. The stem portion for example provides a handle portion enabling the device to be held by the user while the mouth-insertable portion is inserted into the mouth for acquiring image data.

[0111] The device 40 further comprises a light emitter 60 and an imaging device 70 which may be provided in accordance with any of the descriptions presented previously in this disclosure.

[0112] The light emitter 60 is configured to emit stimulation light 62 of a spectral composition suitable for stimulating fluorescence of tooth enamel and / or dentin. The light emitter 60 may comprise dual-mode illumination functionality permitting selective operation with white light illumination or stimulation light for fluorescence imaging.2025PF00028

[0113] 13 29.09.2025

[0114] In some embodiments, the oral sensing device 40 may be configured as an intraoral scanner suitable for consumer use.

[0115] The imaging device 70 may comprise an RGB camera or similar color imaging sensor positioned on or within the mouth-insertable portion 42. The imaging device 70 may be configured to capture image data under different illumination conditions provided by the light emitter 60.

[0116] In fluorescence imaging mode, the stimulation light causes tooth tissue to exhibit fluorescent properties, typically appearing with a green coloration in captured images, while gum tissue appears substantially darker, such as black or dark purple coloration. This differential fluorescence response creates enhanced visual contrast between tooth structure and gum tissue compared to conventional white light imaging, thereby facilitating improved detection and analysis of gum recession conditions.

[0117] The device 40 may be configured to operate in multiple imaging modes, switching between white light reflective imaging and fluorescence imaging. The processing device may be configured to analyze image data captured under either or both illumination conditions to determine gum recession indicators and other oral health parameters.

[0118] Optionally, the imaging device 70 may further be configured with adjustable camera settings such as exposure time, shutter speed, gain, or aperture settings that may be optimized for each imaging mode. For example, fluorescence imaging may utilize longer exposure times or higher gain settings to capture the relatively weaker fluorescence signals, while white light imaging may employ standard exposure parameters optimized for conventional color photography.

[0119] As discussed above, the method 10 comprises determining 14 at least one gum recession indicator based on image analysis applied to image data captured by the imaging device.

[0120] With regards to the image analysis, a variety of approaches are possible.

[0121] One general approach is to detect by the image analysis a gum-tooth boundary line and compute a gum recession indicator based on geometric properties of this boundary line, e.g. shape, curvature, length etc. Another general approach is to detect by the image analysis an area corresponding to an exposed root portion of the visible tooth area and compute a gum recession indicator based on properties of this extracted area, e.g. area size, area shape, ratio of root area size to crown area size etc.

[0122] An example implementation in accordance with the first general approach will now be discussed. In this approach, the image analysis comprises detecting a gum-tooth boundary line within a captured image.

[0123] The gum-tooth boundary line refers to the boundary between gingival tissue and an exposed dental surface of a dental structure. The dental structure includes both the crown portion and the root portion of the tooth. The position of the gum-tooth boundary line varies depending upon the gum recession state associated with the tooth. In a healthy state, the gum-tooth boundary generally lies near the junction between the crown and the root, commonly referred to as the cervical line or cemento-enamel junction (CEJ). In cases of gum recession, the gum-tooth boundary line may he apically of the cervical2025PF00028

[0124] 14 29.09.2025

[0125] line, exposing part of the root surface of the tooth. The gum-tooth boundary line thus corresponds to the gingival margin, or free gingival margin, as visible in the image.

[0126] A variety of methods are possible for detecting the gum-tooth boundary line from captured fluorescence images.

[0127] In one example approach, the image analysis comprises segmenting the visible dental surface area within the captured image and subsequently extracting boundary information from the segmented region. In this approach, the processing device may apply image segmentation techniques to identify and isolate regions corresponding to fluorescent tooth structure, which appear characteristically bright green under stimulation light. The segmentation may be performed using methods such as colorbased thresholding, region-growing algorithms, or machine learning-based segmentation models trained to distinguish fluorescent dental tissue from surrounding non-fluorescent regions. Once the dental surface area has been segmented, the gum-tooth boundary line may be derived by extracting the contour of the segmented region. Boundary extraction may be achieved using edge-detection algorithms, contourdetection techniques, or morphological operations to determine the perimeter of the segmented tooth region. The gum-tooth boundary line corresponds to portions of this perimeter where the segmented fluorescent tooth region directly interfaces with adjacent darker gingival tissue. These portions may be identified, for example, by shape-prediction or model-based analysis to select the boundary segment consistent with the expected location and curvature of the gumline.

[0128] In another approach, the image analysis may comprise segmenting both the visible tooth area (e.g., based on color or intensity) and the visible gum area (e.g., based on color or spectral characteristics), and subsequently determining the interface between the two. In this case, the segmented tooth and gum regions may be subjected to morphological dilation until they overlap, with the resulting overlap line corresponding to the gum-tooth boundary. Alternatively, the boundary may be detected by directly computing the shared edge between the segmented tooth region and the segmented gum region. This approach enables the gum-tooth boundary to be identified explicitly as the transition between two separately segmented regions, which may improve robustness in cases where tooth and gum boundaries are irregular or partially obscured.

[0129] A further example approach comprises directly detecting the gum-tooth boundary line by exploiting the characteristic intensity transition that occurs at the interface between fluorescent tooth structure and gingival tissue. At the gum -tooth boundary, a characteristic darkening or intensity gradient is observed in the fluorescence signal due to the transition from highly fluorescent enamel or dentin to less fluorescent gingival tissue. This intensity transition may also be influenced by optical effects such as shadowing or light scattering at the tissue interface, as well as by the presence of subgingival plaque or calculus deposits that exhibit distinct fluorescence properties. The processing device may apply edgedetection algorithms, gradient-based methods, or trained neural networks to identify these characteristic intensity transitions directly. Such methods may include applying a Sobel or Canny edge detector, computing local intensity gradients, or training a machine learning model such as a convolutional neural2025PF00028

[0130] 15 29.09.2025

[0131] network (CNN) or object detection model to recognize the specific visual characteristics of gum-tooth boundaries in fluorescence images. For example, object detection frameworks may be employed to identify and localize gum-tooth boundary regions within the captured images, treating boundary detection as an object recognition task. Unlike the segmentation-based approaches described above, this method operates directly on pixel intensity transitions, thereby providing an alternative pathway for boundary detection.

[0132] In some embodiments, the processing device may be configured to detect multiple gumtooth boundary lines for individual teeth, particularly for multi -rooted teeth such as molars and premolars. Multi-rooted teeth may exhibit distinct gum-tooth boundary segments corresponding to different root surfaces, which may experience varying degrees of recession. The processing device may apply segmentation or boundary detection techniques to identify separate boundary line segments, for example by analyzing the connectivity and continuity of detected boundary regions. Each identified boundary segment may be analyzed independently to assess recession characteristics specific to that root surface, enabling more detailed and localized assessment of gum recession patterns within individual teeth.

[0133] Once a gum-tooth boundary line is extracted, one or more gum recession indicators may be computed based on its geometric or shape characteristics. In one example implementation, the image analysis may comprise extracting one or more geometric parameters of the gum -tooth boundary line. By way of non-limiting example, such parameters may include: a degree of curvature along the boundary, the depth of any apical displacement relative to a reference cervical line, the angularity or smoothness of transitions at boundary inflection points, and a contour symmetry between mesial and distal aspects of the tooth. These quantitative measures provide objective criteria for differentiating between a normal gumline, which typically follows a shallow and symmetric curved profile, and a receding gumline, which may exhibit increased apical displacement, sharper angular features, or a more triangular contour.

[0134] Optionally, the processing device may perform a shape classification operation using the extracted geometric parameters, for example by applying pattern recognition techniques, machine learning algorithms, or rule-based systems to categorize the boundary into one of a set of predefined shape classes.

[0135] Based on the extracted geometric parameters and / or the results of any shape classification, the method may comprise classifying a gum recession state of the tooth associated with the gum-tooth boundary line. For multi -rooted teeth with multiple boundary segments, this classification may be performed segment by segment and then aggregated to determine an overall recession state for the tooth. The gum recession state may be mapped to corresponding gum health levels, for example classifying the tooth as exhibiting a normal gumline type or a receding gumline type. A receding gumline may be characterized by a more triangular or elongated boundary profile, whereas a normal gumline typically exhibits a shallower, smoothly curved contour that follows the natural cervical line of the tooth. The classification may further include intermediate categories or severity grades, such as mild recession, moderate recession, or severe recession states.2025PF00028

[0136] 16 29.09.2025

[0137] In some embodiments, the classification may be performed using a rules-based approach, for example by applying predefined threshold criteria to one or more geometric parameters. As one example, a tooth may be classified as having mild recession if a measured displacement of the gumline relative to the cervical line exceeds a first threshold but remains below a second threshold, and as having moderate or severe recession if higher thresholds are exceeded. In other embodiments, the classification may be performed by mapping the extracted geometric parameters into recession states using a regression model or statistical classifier. In further embodiments, the classification may be performed using a machine learning algorithm trained on datasets of clinically validated gum recession cases, such as a support vector machine, decision tree, or neural network model, configured to receive shape descriptors or feature vectors as inputs and to output a predicted recession state. In implementations where the shape classification step is performed separately, the processing device may map from the determined shape class (for example triangular vs. shallow curved) to a corresponding gum recession state using a lookup or rule-based mapping. For multi-rooted teeth, the processing device may apply aggregation rules to combine individual root surface assessments, such as taking the maximum recession severity, computing an average recession score, or identifying the most clinically significant recession pattern among the multiple root surfaces.

[0138] In some embodiments, the classification may further be aligned with established clinical classification systems used in dentistry. For example, the computed gum recession state may be mapped to categories corresponding to Miller’s Classification (Class I-IV) or Cairo’s Classification (RT1-RT3), or to other clinically recognized systems of grading gingival recession. This alignment enables the automated analysis to provide results consistent with accepted clinical practice, thereby facilitating interpretation by practitioners and integration with existing diagnostic workflows.

[0139] By way of schematic illustration, Fig. 4 illustrates a dental structure 80 as may be captured within an image obtained using stimulation light of a wavelength suitable for stimulating fluorescence of the dental surface area. In an image of the dental structure captured under stimulation light, the dental surface area would appear green, while background areas, including a gum region 86, would appear darker. The schematic figure illustrates an outline boundary of the exposed surface area of the dental structure as may be detected in accordance with at least one set of embodiments of the invention. The tooth-gum boundary line 92 is indicated by an arrow. By way of reference, the location of the tooth cervical line 94 (the interface between the root 82 and crown 84 part of the tooth is also shown. In the example depicted, the relevant tooth is exhibiting severe gum recession.

[0140] As mentioned above, a second general approach for assessing gum recession is based on estimating a dental surface area size of the exposed root portion and using one or more metrics associated therewith to provide gum recession indicators. This area based approach may be used alone or in combination with gum -tooth boundary detection approach discussed above.

[0141] In some embodiments, the image analysis may comprise (i) a segmentation operation configured to segment dental surface areas within a captured image, followed by (ii) a quantification2025PF00028

[0142] 17 29.09.2025

[0143] operation configured to estimate, based on the segmented dental surface area for a tooth, an indication of an area size of an exposed root portion of the dental surface area. The gum recession indicator may be computed at least in part based on the estimated root portion area size for one or more imaged teeth.

[0144] One implementation of this area-based approach comprises segmenting a total visible dental surface area of a tooth, estimating a crown area portion, then determining the exposed root area through subtraction.

[0145] The estimation of the crown area portion may comprise applying a trained neural network pre-trained to predict an area size of a crown portion of an imaged dental surface area of a tooth based on an image input representative of the tooth dental surface area. This neural network may be trained on datasets of teeth with known crown dimensions and may account for factors such as tooth type, viewing angle, and individual anatomical variations to provide accurate crown area predictions.

[0146] The processing device may estimate an area size of an exposed root portion of the dental surface based on subtraction of the estimated area of the crown portion from the estimated total area size. This subtraction-based approach leverages the principle that any visible dental surface area beyond the expected crown boundaries likely represents exposed root surface due to gum recession.

[0147] The trained neural network may be configured to predict normal crown boundaries and expected crown surface area based on tooth morphology, enabling the identification of areas that extend beyond these normal anatomical limits. Additional metrics may be derived from the estimated root area, such as the ratio of exposed root area to total tooth area, the geometric distribution of exposed root regions, or morphological characteristics such as the curvature or elongation of the exposed areas, which may provide insights into recession patterns and severity. In some embodiments, the neural network may be further configured to receive additional input data beyond the image data, such as demographic information about the subject, for example including age, gender, and / or other relevant characteristics.

[0148] The neural network for predicting crown area may comprise a convolutional neural network (CNN) architecture, such as a U-Net or ResNet-based model, or a transformer-based architecture such as a Vision Transformer, configured to perform semantic segmentation or regression tasks on dental images. The model may include multiple convolutional layers with pooling operations for feature extraction, followed by fully connected layers or upsampling layers depending on whether the output is a segmented crown mask or a direct area measurement. In embodiments where demographic information is incorporated, the network architecture may include additional input pathways or fusion layers configured to combine image features with demographic data, for example through concatenation of feature vectors or multi-modal attention mechanisms. Training may be performed using supervised learning on a dataset comprising fluorescence images of teeth with corresponding ground truth annotations indicating normal crown boundaries, which may be obtained from clinical dental records, 3D dental scans, or expert manual annotations by dental professionals. The training data may include images of various tooth types (incisors, canines, premolars, molars) captured under different viewing angles and lighting conditions, with labels indicating the expected crown surface area and / or crown boundary contours for each tooth, and may2025PF00028

[0149] 18 29.09.2025

[0150] optionally include associated demographic information for each subject to enable training of demographic-aware models. The model may be trained using standard optimization techniques such as stochastic gradient descent or Adam optimizer, with loss functions such as mean squared error for area regression tasks or Dice loss for segmentation tasks, and may incorporate data augmentation techniques such as rotation, scaling, and color variation to improve generalization across different imaging conditions and patient populations.

[0151] Another approach is to estimate the area of the exposed root portion of the tooth directly. For example, in some embodiments, the processing device may be configured to directly identify and segment exposed root portions based on their distinct fluorescence characteristics under stimulation light. When tooth roots become exposed due to gum recession, the root surface exhibits different fluorescence properties compared to the enamel-covered crown portion of the tooth. Specifically, exposed root surfaces, which are covered by cementum rather than enamel, typically appear darker or exhibit reduced fluorescence intensity compared to the brighter green fluorescence of enamel under stimulation light. In some embodiments, this fluorescence differential may be exploited to perform colorbased or intensity-based segmentation of the exposed root regions. For example, the processing device may be configured to apply thresholding techniques to identify regions within the tooth structure that exhibit fluorescence intensities within a predetermined range, or may use color space analysis to distinguish between the brighter green regions corresponding to enamel and the darker green or less fluorescent regions corresponding to exposed cementum.

[0152] This segmentation approach may enable direct quantification of exposed root area. The processing device may be configured to compute one or more metrics such as the total area of exposed root surface, the ratio of exposed root area to total visible tooth area, or the linear extent of root exposure along the tooth's long axis. Additionally, the presence of exposed root regions creates a characteristic "stretched" or elongated appearance of the overall tooth shape in fluorescence images, as the visible tooth structure extends beyond the normal crown boundaries into the previously subgingival root area. The processing device may be configured to analyze this shape elongation as an additional or alternative indicator of gum recession, for example by comparing the aspect ratio or length-to-width ratio of the segmented tooth region against expected normal values.

[0153] By way of schematic illustration, Fig. 5 illustrates a dental structure 80 as may be captured within an image obtained using stimulation light of a wavelength suitable for stimulating fluorescence of the dental surface area. In an image of the dental structure captured under stimulation light, the dental surface area would appear green, while background areas, including a gum region 86, would appear darker. In this example, a crown area portion 84 and a root area portion 82 have been detected, for example in accordance with one of the example methods discussed above. The tooth-gum boundary line 92 is indicated by an arrow.

[0154] The at least one gum recession indicator may be computed at least in part based on the classified gum recession state of one or more imaged teeth. The computation may involve aggregating2025PF00028

[0155] 19 29.09.2025

[0156] classification results across multiple teeth to generate overall oral health metrics or recession severity scores. For example, the processing device may calculate the percentage of teeth exhibiting recession, determine the average recession severity across the dentition, or identify specific regions of the mouth showing concerning recession patterns. The gum recession indicator may also incorporate temporal analysis when multiple imaging sessions are available, tracking changes in recession classification over time to assess progression or improvement. In some implementations, the indicator computation may weight different teeth differently based on their clinical significance or may focus on specific tooth types that are more susceptible to recession.

[0157] In some embodiments, the method may further comprise recording a log of computed gum recession indicators to enable longitudinal tracking and analysis of gum health over time. The processing device may be configured to store computed gum recession indicators along with associated metadata such as timestamps, tooth identifiers, imaging conditions, and / or other measurement parameters. This logging functionality enables the accumulation of historical data that can be analyzed to identify trends, assess treatment effectiveness, or detect early signs of recession progression. The log may be structured as a database or data file containing entries for each assessment session, with each entry including the computed recession indicators for individual teeth, overall oral health metrics, and / or contextual information that may be relevant for subsequent analysis.

[0158] The log may be stored on various types of memory systems. The memory may be local to the oral sensing device itself, such as internal flash memory or removable storage media. Alternatively, the memory may be separate from the device, such as a remote memory system or cloud-based data store. In some embodiments, the memory may comprise a memory of a mobile computing device, such as a smartphone or tablet, with which the oral sensing device is communicatively coupled.

[0159] The processing device may be configured to compute tracking information based on changes in recorded gum recession indicators over time by analyzing temporal patterns and trends in the logged data. This analysis may involve comparing current measurements with historical baselines, calculating rates of change in recession indicators, identifying accelerating or decelerating trends, or detecting significant deviations from expected patterns. For example, the tracking information may include metrics such as the rate of recession progression for individual teeth, the overall trend in oral health scores, or the identification of teeth showing changes that may require professional attention. The processing device may apply statistical analysis techniques, trend fitting algorithms, or machine learning models.

[0160] The method may further comprise generating a data output indicative of the tracking information. The data output may include graphical representations such as trend charts showing recession progression overtime, comparative visualizations highlighting changes between assessment sessions, or alert notifications when significant changes are detected. The output may also include textual summaries, numerical scores, or recommendations for oral care adjustments based on the observed trends. This tracking information may be integrated with oral care device control systems to automatically adjust2025PF00028

[0161] 20 29.09.2025

[0162] treatment parameters, or may be formatted for sharing with dental professionals to support clinical decision-making and treatment planning.

[0163] Optionally, in some embodiments, the light emitter 70 comprises a further light source operable to emit white visible light, and wherein the imaging device 60 is further sensitive to the white visible light emitted by the further light source. In some embodiments, the processing device 32 may be operable to selectively control the oral sensing device in a selected one of (i) a fluorescence imaging mode in which image data is acquired of oral surfaces illuminated using the stimulation light, and (ii) a non-fluorescence imaging mode in which image data is acquired of oral surfaces illuminated using the white visible light. The white visible light is of a spectral composition which does not stimulate fluorescence.

[0164] The processing device 32 may be configured to implement a control function that enables selective operation between the two imaging modes based on user input or automatic mode selection criteria. For example, the oral sensing device may include user interface elements such as buttons, touchscreen controls, or voice commands that allow a user to manually select the desired imaging mode. Alternatively, the processing device may automatically select the imaging mode based on factors such as the specific analysis objectives, ambient lighting conditions, or predetermined assessment protocols. The control function may also enable sequential operation in both modes during a single assessment session to capture complementary image data for comprehensive analysis.

[0165] In each imaging mode, the processing device is configured to control the light emitter 60 to emit the appropriate illumination type while coordinating image capture timing and camera settings. In fluorescence mode, the stimulation light is activated and image data is acquired. Optionally one or more imaging device parameters may be optimized for capturing the fluorescence signals, for example including adjusted exposure times, gain settings, and / or spectral filtering. In non-fluorescence mode, the white light source is activated and image data is acquired. Optionally, one or more camera settings may be configured for conventional color imaging or other imaging modalities, for example with standard RGB color balance and exposure parameters, or other example settings.

[0166] In some embodiments, the processing device may be selectively operable in at least one mode in which the processing device is configured to compute the gum recession indicator based on image analysis applied to a combination of image data acquired in both the fluorescence imaging mode and the non-fluorescence imaging mode. This multimodal approach may involve acquiring image data sequentially in a controlled multi-modal image capture sequence, where the device automatically switches between illumination modes to capture complementary image information of the same oral region. The processing device may implement fusion algorithms that combine analysis results from both imaging modes, or may utilize multimodal neural networks configured to process both fluorescence and white light image data simultaneously as input channels. For example, the neural network architecture may include separate processing pathways for each image modality that converge at intermediate or final layers to generate integrated recession assessments.2025PF00028

[0167] 21 29.09.2025

[0168] Embodiments of the invention described above employ a processing device. The processing device may in general comprise a single processor or a plurality of processors. It may be located in a single containing device, structure or unit, or it may be distributed between a plurality of different devices, structures or units. Reference therefore to the processing device being adapted or configured to perform a particular step or task may correspond to that step or task being performed by any one or more of a plurality of processing components, either alone or in combination. The skilled person will understand how such a distributed processing device can be implemented. The processing device includes a communication module or input / output for receiving data and outputting data to further components.

[0169] The one or more processors of the processing device can be implemented in numerous ways, with software and / or hardware, to perform the various functions required. A processor typically employs one or more microprocessors that may be programmed using software (e.g., microcode) to perform the required functions. The processor may be implemented as a combination of dedicated hardware to perform some functions and one or more programmed microprocessors and associated circuitry to perform other functions.

[0170] Examples of circuitry that may be employed in various embodiments of the present disclosure include, but are not limited to, conventional microprocessors, application specific integrated circuits (ASICs), and field-programmable gate arrays (FPGAs).

[0171] In various implementations, the processor may be associated with one or more storage media such as volatile and non-volatile computer memory such as RAM, PROM, EPROM, and EEPROM. The storage media may be encoded with one or more programs that, when executed on one or more processors and / or controllers, perform the required functions. Various storage media may be fixed within a processor or controller or may be transportable, such that the one or more programs stored thereon can be loaded into a processor.

[0172] Variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, 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.

[0173] A single processor or other unit may fulfill the functions of several items recited in the claims.

[0174] 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.

[0175] A computer program may be stored / 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.2025PF00028

[0176] 22 29.09.2025

[0177] If the term "adapted to" is used in the claims or description, it is noted the term "adapted to" is intended to be equivalent to the term "configured to".

[0178] Any reference signs in the claims should not be construed as limiting the scope.

Claims

2025PF0002823 29.09.2025CLAIMS:

1. A processing device (32) comprising one or more processors (36) configured to perform a method (10) for assessing gum recession of a subject, the method comprising:receiving (12) image data representative of one or more dental surfaces within the oral cavity of the subject under illumination by a stimulation light for stimulating fluorescence of tooth enamel and / or dentin;determining (14) at least one gum recession indicator based on image analysis applied to image data captured by the imaging device; andgenerating (16) a data output based on the computed gum recession indicator.

2. The processing device of claim 1, wherein the stimulation light has a wavelength within the range 380-420 nm, for example approximately 405 nm.

3. The processing device of claim 1 or 2, wherein the image analysis comprises detecting a gum -tooth boundary line within a captured image.

4. The processing device of claim 3,wherein the image analysis further comprises a shape classification operation for classifying a shape of a detected gum-tooth boundary line;wherein the method comprises classifying a gum recession state of a tooth associated with a detected gum-tooth boundary line based on the shape classification; andwherein the gum recession indicator is computed at least in part based on the classified gum recession state of one or more imaged teeth.

5. The processing device of any preceding claim,wherein the image analysis comprisesa segmentation operation configured to segment dental surface areas within a captured image; anda quantification operation configured to estimate, based on the segmented dental surface area for a tooth, an indication of an area size of an exposed root portion of the dental surface area; andwherein the gum recession indicator is computed at least in part based on the estimated root portion area size for one or more imaged teeth.2025PF0002824 29.09.20256. The processing device of claim 5, wherein the quantification operation comprises:estimating a total area size of the segmented dental surface area of a tooth; applying a trained neural network configured to predict an area size of a crown portion of an imaged dental surface area of a tooth based on an image input representative of the tooth dental surface area; andestimating an area size of an exposed root portion of the dental surface based on subtraction of the estimated area of the crown portion from the estimated total area size.

7. The processing device of any preceding claim, wherein the method further comprises:recording a log of computed gum recession indicators;computing tracking information based on changes in recorded gum recession indicators over time; andgenerating a data output indicative of the tracking information.

8. The processing device of any preceding claim, wherein the method further comprises:controlling emission of the stimulation light by a light emitter (60) comprising a light source operable to emit the stimulation light suitable for stimulating fluorescence of tooth enamel and / or dentin; andcontrolling acquisition of the image data by means of an imaging device (70) being sensitive at least to fluorescence light emitted by tooth enamel and / or dentin when stimulated by the stimulation light.

9. The processing device of claim 8,wherein the imaging device is sensitive to fluorescence light in a wavelength range of 510-520 nm; andoptionally wherein the imaging device comprises an RGB camera.

10. The processing device of claim 8 or 9, whereinthe light emitter comprises a further light source operable to emit white visible light, and the imaging device is further sensitive to the white visible light emitted by the further light source; and wherein the method is selectively operable in a selected one of:a fluorescence imaging mode in which image data is acquired of oral surfaces illuminated using the stimulation light, anda non-fluorescence imaging mode in which image data is acquired of oral surfaces illuminated using the white visible light.2025PF0002825 29.09.202511. The processing device of any preceding claim,wherein the method further comprises communicating with an oral care device operable in use to perform an oral care function, and controlling configuration of at least one operational characteristic of the oral care function in dependence upon the at least one gum recession indicator, and optionally wherein the oral care device comprises a powered toothbrush device and wherein the at least one operational characteristic comprises a brushing intensity level.

12. A system (30), comprising:an oral sensing device (40) comprising:a mouth-insertable portion (42);a light emitter (60) carried by the mouth-insertable portion, the light emitter comprising a light source operable to emit stimulation light suitable for stimulating fluorescence of tooth enamel and / or dentin;an imaging device (70) carried by the mouth-insertable portion for imaging oral surfaces illuminated by the light emitter, the imaging device being sensitive at least to fluorescence light emitted by tooth enamel and / or dentin when stimulated by the stimulation light; anda processing device (32) in accordance with any of claims 1-11, configured to determine the at least one gum recession indicator based on image analysis applied to image data captured by the imaging device.

13. The system (30) of claim 12,wherein the light emitter comprises a further light source operable to emit white visible light;wherein the imaging device is further sensitive to the white visible light emitted by the further light source; andwherein the processing device is operable to selectively control the oral sensing device in a selected one of:a fluorescence imaging mode in which image data is acquired of oral surfaces illuminated using the stimulation light, anda non-fluorescence imaging mode in which image data is acquired of oral surfaces illuminated using the white visible light.

14. A method for assessing gum recession of a subject, the method comprising:receiving image data representative of one or more dental surfaces within the oral cavity of the subject under illumination by a stimulation light for stimulating fluorescence of tooth enamel and / or dentin;2025PF0002826 29.09.2025determining at least one gum recession indicator based on image analysis applied to image data captured by the imaging device; andgenerating a data output based on the computed gum recession indicator.

15. A computer program product comprising computer program code configured, when run on a processor, to cause the processor to perform a method in accordance with claims 14.

Images