Oral care device using tooth movement detection

The oral care device uses sensors to objectively assess tooth mobility and predict progression, addressing the limitations of existing clinical methods by providing standardized metrics and predictive insights.

JP2024541431AInactive Publication Date: 2024-11-08KONINKLIJKE PHILIPS NV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
JP2024529748
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Priority Date
2022-04-28
Filing Date
2022-11-17
Publication Date
2024-11-08
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Existing clinical methods for assessing tooth mobility are subjective, time-consuming, and lack standardization, preventing direct comparison of measurements over time and failing to provide predictive insights into tooth movement trends.

Method used

An oral care device with integrated sensors and processing apparatus that measures force and movement during cleaning sessions to derive standardized tooth mobility metrics, using machine learning algorithms to predict tooth movement progression.

Benefits of technology

Enables objective, standardized assessment of tooth mobility for direct comparison and long-term trend analysis, facilitating early detection of pathological conditions and personalized treatment plans.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024541431000001_ABST
    Figure 2024541431000001_ABST
Patent Text Reader

Abstract

Means are provided for deriving tooth mobility metrics based on force or movement data obtained from sensors incorporated into the oral care device and acquired during an oral cleaning session. By processing the waveform of the sensed signal to detect one or more signal characteristics and further acquiring one or more characteristics of the mechanical engagement of the teeth with the device, a standardized tooth mobility metric can be derived.
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present application relates to oral care devices, in particular oral cleaning devices. [Background technology]

[0002] Tooth mobility is a useful clinical marker in oral care.

[0003] For example, knowing tooth mobility (TM) is important for monitoring periodontitis disease progression, for bone loss, and for monitoring periodontal or orthodontic treatment response and efficacy. Increased tooth mobility is a common symptom of periodontal disease and one of the main reasons for visits to the dentist (DP).

[0004] The teeth are fixed in the jaw socket by the periodontal ligament, which not only provides the teeth with a firm attachment support but also allows slight movement of the teeth during chewing. This physiological mobility allows the teeth to withstand the considerable forces exerted during chewing.

[0005] Normal (physiological) tooth mobility varies between different teeth and ranges, for example, from 40 μm (for molars) to 120 μm (for incisors). However, certain pathological conditions of the bone or periodontal ligament can adversely affect tooth mobility. The increase in tooth mobility is a gradual and irreversible process. This increase can also cause tooth loss if not corrected in time.

[0006] For example, 50% of Americans over the age of 30 (65 million people) have periodontitis. Studies have shown that 10-20% of people with periodontitis have some degree of tooth mobility in multiple tooth sites. Increased tooth mobility can also be caused by orthodontic treatment, malocclusion problems, and teeth grinding.

[0007] The clinical decision is based on the severity or degree of tooth mobility detected as to whether the tooth mobility requires a treatment response.

[0008] Dentists usually test tooth mobility by applying pressure to the teeth with a probe. Clinically, TM is usually classified into various grades, such as mild, moderate, and severe, based on the degree of displacement in response to pressure (i.e., the degree of mobility). Such classification scales have been shown to be very useful in monitoring trends in tooth mobility, and these classification scales are used as predictors for disease progression. However, these existing clinical methods for TM assessment are subjective, time-consuming, and prone to error.

[0009] Researchers have proposed other methods for TM assessment, such as the use of a laser beam or the Periotest® (a handheld device that applies a rapid tapping motion to the tooth or implant to assess stability based on the measurement of contact time). The output of the Periotest is a tooth mobility metric in the form of a graduated scale, ranging from -8 to +50 Periotest Value (PTV). The lower the Periotest value, the more damping the test object (tooth or implant) exhibits, and therefore the more stable the object. Another clinical device for tooth mobility assessment is the Penguin RFA Implant Stability Monitor, which measures the resonant frequency of a temporarily attached dental screw by a non-contact technique. The frequency is displayed as an ISQ value (Implant Stability Index) between 1 and 99. The ISQ value correlates strongly with implant micromobility.

[0010] However, currently there is no at-home method for rapid and objective screening of tooth mobility or stability / implant mobility or stability. Having such a method would be beneficial for early detection of tooth mobility or for detecting the stability of dental implants.

[0011] Furthermore, tooth mobility assessments are currently performed at wide time intervals by visiting a dentist. Such ad-hoc analyses are not sufficient to properly analyze the progression of tooth mobility in different classes, such as physiological vs. pathological, acute vs. chronic tooth mobility. Long-term assessment of tooth mobility is also important to study the suitability of dental implant stability or changes over time. Furthermore, existing methods do not provide predictive capabilities. Such predictions may be useful for scheduling future dental consultations or for modifying treatment regimes. Summary of the Invention [Problem to be solved by the invention]

[0012] In summary, therefore, existing clinical methods for measuring tooth mobility have certain disadvantages associated with them, including the following: Subjectivity in assessing tooth mobility or dental implant stability. Non-standardization of agitation assessment, meaning that measurements taken at different times cannot be directly compared. The need for specialized hardware, limiting at-home applicability. It is time consuming for the clinician, which can lead to the common practice of only assessing tooth mobility when the underlying gingival disease is already at an advanced stage. Lack of information regarding possible future trends in tooth mobility or the stability of dental implants.

[0013] A key challenge in providing tooth mobility assessment is the provision of an objective, standardized metric for tooth mobility that allows measurements taken at different times on different teeth to be compared and that allows clinical decisions to be made based on individual absolute measurements (e.g., rather than relying on trends over time).

[0014] One objective of the embodiments proposed herein is to provide an objective and standardized metric of tooth mobility (TM) that allows for direct comparison of different tooth mobility, for clinical assessment, and for derivation of longitudinal trends in TM. [Means for solving the problem]

[0015] The invention is defined by the claims.

[0016] According to an example according to one aspect of the present invention, there is provided a processing apparatus for an oral care device, comprising: An input / output section; One or more processors The processing device includes: receiving at the input / output sensor signals indicative of a load force or pressure applied to the at least one cleaning element or a support to which the cleaning element is coupled and a movement of the at least one cleaning element or a support to which the cleaning element is coupled, the sensor signals spanning a time period corresponding to a duration of mechanical engagement of the at least one cleaning element with the at least one tooth during an oral cleaning session; Processing the waveform of the signal over a period of time to extract one or more characteristics of the signal waveform; obtaining measurements of at least one further characteristic which is a characteristic of the teeth and / or a characteristic of the mechanical engagement of the at least one cleaning element with the teeth; determining a tooth mobility metric for the at least one tooth as a function of the one or more signal characteristics and the at least one further characteristic; A processing device is provided for generating a data signal indicative of the determined metric and for providing the data signal to the input / output.

[0017] Therefore, embodiments propose to derive tooth mobility metrics based on force, pressure or movement information on cleaning elements in a home oral care device. For example, tooth mobility can be assessed from sensor measurements acquired during an oral cleaning session with an oral cleaning device such as a toothbrush or mouthpiece device. In this way, tooth mobility can be assessed at home in a convenient manner.

[0018] In addition to the cleaning element movement characteristic data, embodiments propose to use at least one further characteristic of the sensed tooth or of the mechanical engagement of at least one cleaning element with the tooth in order to standardize or normalize the obtained metric and to make it objectively comparable with metrics obtained for other teeth under different measurement conditions, e.g. in later or previous cleaning sessions. The further characteristic increases the specificity of the metric.

[0019] The cleaning elements, in some examples, are bristles.

[0020] In some embodiments, the obtained tooth mobility metric defines a gradual level of tooth mobility.

[0021] In some embodiments, the processing device further accesses a data store that records historical tooth mobility metric data for the user, and stores the determined tooth mobility metric in the data store by outputting a data signal to the data store.

[0022] In some embodiments, the processing device further accesses a data store that records historical tooth mobility metric data for the user, and determines a long-term trend of the tooth mobility metric for one or more teeth of the user based on the historical tooth mobility metric data and the determined tooth mobility metric during the current cleaning session.

[0023] In some embodiments, the processing device further generates a prediction of tooth mobility progression based on long-term trends.

[0024] The processor generates a data signal indicative of the long-term trend for output from the device, for example to a data store (the same or a different data store) or to a user interface.

[0025] In some embodiments, the processing device further obtains an identification of the at least one tooth. The above-referenced data signal further includes an identification of the at least one tooth.

[0026] The processor further obtains an indication of a tooth type for the at least one tooth. The at least one further characteristic referred to above includes tooth type. Different teeth have different external morphologies and different root structures (e.g., one root, two roots, three roots) that affect normal tooth mobility. The classification of tooth mobility takes into account tooth type.

[0027] Information regarding tooth type can for example be obtained from position sensors integrated into the oral cleaning device or can be obtained from other methods such as camera images by using standard image processing methods.

[0028] In some examples, determining the tooth mobility metric comprises applying a machine learning algorithm. In some examples, determining the tooth mobility metric comprises applying a multi-parametric linear regression algorithm.

[0029] In some embodiments, the received sensed signal includes signal segments corresponding to respective periods of successive engagement of the at least one cleaning element with each of the plurality of teeth. The processor determines a respective tooth mobility metric for each of the plurality of teeth. The processor further obtains an identification of each of the teeth.

[0030] The sensed signal spans a period of successive engagement of the at least one cleaning element with a plurality of teeth, and the processor accordingly divides the signal into a plurality of signal segments, each segment spanning a respective period of engagement with a respective tooth.

[0031] Dividing the signal into segments includes identifying the beginning and end of each signal segment by detecting a predetermined signature pattern indicative of the beginning and end of each tooth segment, or by using a positioning signal for at least one cleaning element received from a positioning signal source. The division of the signal may otherwise be based on user input.

[0032] There are various options regarding the at least one further characteristic of the teeth and / or the at least one further characteristic of the mechanical engagement of the at least one cleaning element with the teeth.

[0033] In some embodiments, the at least one further characteristic includes a baseline user load parameter indicative of a baseline or average load force applied by a user to the cleaning element during use.

[0034] In some embodiments, the processing device performs a calibration operation that includes determining a baseline or average load force applied to at least one cleaning element during a cleaning session using the sensed signal.

[0035] The baseline or average force indicates the load applied by the user.

[0036] In some embodiments, the at least one further characteristic includes a brushing stroke speed parameter indicative of the speed of a brushing stroke applied by a user during a cleaning operation.

[0037] This may be determined and stored in a calibration operation, or it may be determined dynamically during a cleaning session, for example using an inertial measurement unit (IMU).

[0038] In some embodiments, the at least one further characteristic comprises one or more structural characteristics of the at least one cleaning element, such as one or more of a protrusion angle of the cleaning element and a stiffness metric of the cleaning element.

[0039] Thus, certain properties regarding the cleaning element (eg, bristle type, bristle stiffness, bristle coefficient of friction, bristle wear, bristle shape, bristle arrangement, etc.) may be used as further properties.

[0040] In some embodiments, the processing device further processes the obtained tooth mobility metrics to derive a clinical classification of tooth mobility for at least one tooth. This embodiment therefore proposes to categorize the tooth mobility metrics according to clinically relevant categories to clinically assess tooth mobility, e.g. disease extent or disease predictor, taking into account clinical interpretation and cross-correlation between teeth and cleaning sessions.

[0041] In other words, clinical classification puts tooth mobility into a clinical context, i.e. indicates the extent to which a measured level of tooth mobility is normal or abnormal, i.e. a pathological level of mobility.

[0042] Deriving the clinical classification comprises comparing the obtained tooth mobility metric to one or more selected thresholds.

[0043] In some embodiments, the one or more thresholds are selected based on at least one of tooth type and a demographic classification of the user.

[0044] In some embodiments, the sensed signal includes signal segments corresponding to a duration of engagement of the at least one cleaning element with the plurality of teeth, and the processor determines a respective tooth mobility metric for each of the plurality of teeth, and deriving a clinical classification with respect to tooth mobility for the first tooth further comprises comparing the tooth mobility metric obtained for the first tooth with a further tooth to detect a deviation level. The further tooth is preferably of the same tooth type. The further tooth is preferably on an opposite side of the dental arch to the first tooth. The deviation level is preferably further compared to a deviation threshold to obtain the clinical classification.

[0045] In some embodiments, the processing device further accesses a data store recording historical tooth mobility metric data for the user and compares the obtained tooth mobility metric for at least one tooth to historical values ​​for the same tooth. The clinical classification is further based on said comparison.

[0046] In an advantageous set of embodiments, the tooth mobility metric is used to derive predictive data regarding predicted tooth mobility progression or clinical classification.

[0047] For example, in some embodiments, the processing device generates a prediction of tooth mobility progression based on application of machine learning algorithms.

[0048] In some embodiments, a machine learning algorithm is trained to receive as input the derived tooth mobility metrics for a single cleaning session and a demographic classification of the user, and to generate as output a prediction of tooth mobility progression, the algorithm being trained based on training data for a population of patients in different demographic classifications, the training data including the tooth mobility metrics at two different time points for each of the multiple patients, and optionally a demographic classification for the patient.

[0049] Additionally or alternatively, in some embodiments a machine learning algorithm is provided that is trained to receive as input derived tooth mobility metrics for a series of cleaning sessions across a series of time points, and to generate as output a prediction of tooth mobility progression, the algorithm being trained on training data for a population of patients, the training data including tooth mobility metrics across a series of different time points for each of a plurality of patients.

[0050] Determining the tooth mobility metric involves first determining a provisional indication of the degree of tooth mobility, which in some instances includes performing a spectral analysis of the sensed signals.

[0051] In particular, the one or more characteristics of the signal waveform referred to above include the amplitude of one or more selected frequency components of the signal and / or a bandwidth associated with a dominant (i.e., resonant or harmonic) peak in the frequency spectrum of the sensed signal, or a quality factor associated with a dominant peak in the frequency spectrum of the sensed signal.

[0052] In particular, the processor calculates a frequency spectrum for the sensed signal. The processor detects the amplitude of one or more frequency components of the sensed signal. The processor detects the amplitude of a center frequency of a dominant peak of the frequency spectrum. The processor detects a bandwidth associated with the dominant peak of the frequency spectrum and / or a quality factor associated with the dominant peak of the frequency spectrum.

[0053] In this disclosure, a dominant peak refers to the peak in a frequency spectrum that has the highest amplitude at its center frequency, otherwise known as the main harmonic peak or resonant peak.

[0054] Determining the tooth mobility metric may comprise comparing one or more of the above-mentioned characteristics of the frequency spectrum to a baseline criterion. Alternatively, determining the tooth mobility metric may comprise processing one or more of the above-mentioned characteristics using a transfer function or a look-up table to derive the tooth mobility metric.

[0055] The oral care device comprises an actuation means for driving a vibratory motion of at least one cleaning element, which, in a reference state, exhibits a reference frequency spectrum having at least one dominant (i.e., main harmonic or resonant) peak having a centre frequency.

[0056] In some examples, determining the degree of tooth mobility includes determining a correlation between a reference frequency spectrum and a detected frequency spectrum. the amplitude of the center frequency component of the aforementioned dominant peak of the frequency spectrum, the bandwidth associated with the dominant peak in the frequency spectrum, and / or The quality factor associated with the dominant peak in the frequency spectrum of, and the desire for change.

[0057] In other words, the vibrational movement has a spectral signature, and determining the tooth mobility metric comprises detecting changes in said spectral signature. However, it is not essential to compare the measured frequency spectrum characteristics with such a reference spectrum. As mentioned above, in other cases, a transfer function such as a multi-parametric linear regression function may be used to derive the tooth mobility metric from one or more characteristics of the frequency spectrum and one or more further characteristics.

[0058] The reference frequency spectrum referred to above may, for example, correspond to a predicted operating frequency spectrum for a controlled calibration condition, e.g. for a reference load, and for a particular reference tooth (e.g. a non-mobile tooth), for example in a brushing session after a dental examination, or against a well-defined defined surface using a standardized method.

[0059] The reference frequency spectrum may be stored in the memory of the processing device as calibration data.

[0060] A number of different reference frequency spectra may be stored, corresponding to various possible applied user loads.

[0061] Another aspect of the present invention is a method for producing a A support; a plurality of cleaning elements for engagement with an oral surface, the cleaning elements protruding from a surface of the support; - sensor means for sensing measurements indicative of a load force applied to at least one of the cleaning elements or to the support or indicative of a movement of at least one of the cleaning elements or to the support; A processing device according to any embodiment described in this disclosure or according to any claim of this application; An oral care device comprising:

[0062] In one non-limiting set of examples, the oral care device is a power toothbrush. In a further non-limiting set of examples, the oral care device is a brushing mouthpiece device.

[0063] These and other aspects of the invention will be apparent from and elucidated with reference to the embodiment(s) described hereinafter.

[0064] For a better understanding of the present invention and to show more clearly how the same may be carried into effect, reference will now be made, by way of example only, to the accompanying drawings in which: [Brief description of the drawings]

[0065] [Figure 1] 1 illustrates an exemplary oral care device according to one or more embodiments. [Diagram 2] FIG. 1 illustrates a block diagram of components of an exemplary oral care device according to one or more embodiments. [Diagram 3] 1 shows the relationship between detection signal response and the degree of tooth mobility. [Figure 4] 13 shows simulation results illustrating how the degree of tooth mobility can be derived from a spectral analysis of the sensed signal. [Diagram 5] 13 shows simulation results illustrating how the degree of tooth mobility can be derived from a spectral analysis of the sensed signal. [Figure 6] 13 shows simulation results illustrating how the degree of tooth mobility can be derived from a spectral analysis of the sensed signal. [Figure 7] 13 shows simulation results illustrating how an applied user load affects the measured sensed signal response for a given degree of tooth movement. [Figure 8] 13 shows simulation results illustrating how an applied user load affects the measured sensed signal response for a given degree of tooth movement. [Figure 9] 1 illustrates how tooth movement and user load are correlated. [Figure 10] 1 shows how clinical metrics are derived from tooth mobility metrics. [Figure 11] 1 shows an exemplary decision tree for classifying tooth mobility. [Figure 12] 1 illustrates the configuration and workflow of an exemplary machine learning model. [Figure 13] This shows that future tooth mobility values ​​are predicted. [Figure 14] 1 shows an overview of an exemplary embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0066] The present invention will now be described with reference to the drawings.

[0067] 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 of 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, the appended claims, and the accompanying drawings. It should be understood that the drawings are merely schematic and are not drawn to scale. It should also be understood that the same reference numerals are used throughout the drawings to denote the same or similar parts.

[0068] The present invention provides a means for deriving a tooth mobility metric based on force or movement data obtained from a sensor means incorporated in the oral care device and acquired during an oral cleaning session. By processing the waveform of the sensed signal to detect one or more signal characteristics, and by further acquiring one or more characteristics of the mechanical engagement of the teeth with the device, a standardized tooth mobility metric can be derived.

[0069] 1 and 2 show an oral care device 10 according to at least one set of embodiments of the present invention. The device 10 comprises a support 12 from at least one surface of which a plurality of cleaning elements 14 project for mechanical engagement with an oral surface 42. The oral surface 42 is a tooth 40 or a surface of, for example, a dental implant. The device further comprises sensor means 30 for sensing measurements indicative of a load force applied to at least one of the cleaning elements 14 or the support 12 or indicative of a movement of at least one of them.

[0070] The device further comprises a processing unit 20 .

[0071] The processing device includes an input / output 22 and one or more processors 24 and receives a sensed signal 32 at the input / output 22 indicative of a load force applied to or indicative of the movement of the at least one cleaning element 14 of the oral care device. The signal spans a time period corresponding to a duration of mechanical engagement of the at least one cleaning element with the at least one tooth 40 during an oral cleaning session.

[0072] The processing device processes the signal waveform 60 over a period of time to extract one or more characteristics of the signal waveform.

[0073] The processing device further obtains measurements of at least one further property that is a property of the teeth and / or a property of the mechanical engagement of the at least one cleaning element with the teeth.

[0074] The processing unit further determines a tooth mobility metric 26 for at least one tooth in response to the one or more signal characteristics and the at least one further characteristic, and generates a data signal 25 indicative of the determined metric and provides the data signal to the input / output unit 22.

[0075] The device further comprises a user interface that provides a user output indicative of the tooth mobility metrics.

[0076] Therefore, embodiments propose a method to provide standardized tooth mobility metrics for one or more teeth, obtained during one or more cleaning sessions.

[0077] The input data includes data from sensing means capable of sensing force, pressure and / or movement signals (such as acceleration), for example integrated into the support (eg the brush head of a toothbrush or the device handle).

[0078] The input data further comprises at least one further characteristic of the teeth and / or at least one further characteristic of the mechanical engagement of the at least one cleaning element with the teeth. By way of non-limiting examples, the input data may comprise brushing speed (stroke speed), brushing duration, brushing angle, or vibration frequency (or derivatives of these parameters).

[0079] In some embodiments, the input data further includes positioning information to allow identification of specific teeth with matching tooth mobility metrics. Positioning data may be obtained from a position / orientation sensor incorporated in the oral care device in some instances. Positioning data may also be obtained from a motion sensor.

[0080] In some embodiments, the processing device further generates a data message for transmission to a clinician device to provide an indication if a particular level of tooth movement or an increase in tooth movement is detected.

[0081] With reference to Figure 1, the oral care device in the illustrated embodiment is shown in the form of a toothbrush with a brush head 16 coupled to a body 18. The body forms a handle for the device. Support 12 is provided by the platen of the brush head. Cleaning elements 14 include bristles. The bristles are arranged in tufts. References above to at least one cleaning element may be references to at least one bristle tuft in this set of embodiments.

[0082] Although bristles are mentioned as an example, one skilled in the art will recognize that a variety of other cleaning elements are also possible, such as elastomeric cleaning elements (thicker than typical individual bristles) or cleaning structures that protrude from a support having an expanded shape but are also designed to engage oral surfaces to provide a mechanical cleaning action.

[0083] Although a toothbrush is shown in FIG. 1, in further examples, the oral care device may take different forms, such as a brushing mouthpiece device, an oral cleaning device, or an oral flossing device.

[0084] The block diagram structure of FIG. 2 is applicable to any particular type of oral care device, not just toothbrushes.

[0085] With further reference to FIG. 1, preferably, the oral care device comprises an actuation means 19 for driving the vibrational motion of the at least one cleaning element 14. This induced vibrational motion pattern of the at least one cleaning element exhibits a frequency spectrum. The device records in the memory (not shown) of the processing unit 20 one or more reference frequency spectra corresponding to the frequency spectrum of the driven vibrational motion pattern in a particular reference or calibration state, such as in the presence of a particular reference load force, and corresponding to a particular tooth having a reference level of motion (e.g. zero motion). The reference state can be any desired state, for example in the absence of an externally applied load force on the at least one cleaning element and / or for a tooth having a non-zero motion. Two or more reference spectra can be stored for different conditions, for example different applied user loads (which will be discussed further below). In all cases, the reference frequency spectrum has at least one dominant peak (which substantially corresponds to the main harmonic or main resonant component of the vibration). For example, there are multiple dominant peaks, each at each harmonic of the main frequency component of the vibration. A dominant peak in this context means the peak that has the highest amplitude at its center frequency. Reference to a dominant peak (singular) means the peak that has the highest amplitude of all the peaks in the frequency spectrum.

[0086] The dominant peak of the frequency spectrum has a bandwidth and a quality factor associated with it.By comparing characteristics of the frequency spectrum of the detected vibratory motion pattern with characteristics of a reference vibratory motion pattern, the level of tooth mobility can be detected.

[0087] The sensor means 30 referred to above may take various forms, for example a pressure sensor, a force sensor, an inertial sensor (for example an accelerometer) or a load cell. The sensor means detects changes in acceleration, pressure and / or force on the support or the cleaning element, for example in one or more directions perpendicular to the support surface from which the cleaning element projects, or preferably in three orthogonal axes.

[0088] The processing unit converts changes in the sensed load signal into a determination of tooth mobility level or grade.

[0089] Referring to FIG. 2, optionally in some embodiments, the processing device 20 further accesses a data store 28 that records historical tooth mobility metric data for the user, and stores the determined tooth mobility metric in the data store by outputting a data signal 25 to the data store.

[0090] In some embodiments, the one or more processors 24 access a data store 28 that records historical tooth mobility metric data for a user and determine a long-term trend of the tooth mobility metric for one or more teeth of the user based on the historical tooth mobility metric data and the determined tooth mobility metric for the current cleaning session. The one or more processors further generate a prediction of tooth mobility progression based on the long-term trend.

[0091] The processor generates a digital signal indicative of the long-term trend for output from the device, for example to a data store (the same or a different data store) or to a user interface.

[0092] An identification of the particular tooth to which each mobility metric corresponds is also obtained (eg, using positioning means) and this identification is stored in association with the tooth mobility metric in the recorded data.

[0093] Referring to FIG. 2, another aspect of the present invention provides a system 8 including an oral care device 10 according to any embodiment described in this disclosure or according to any claim of the present application, and further including a data store 28 pre-storing determined tooth mobility metrics for one or more subjects.

[0094] Referring again to FIG. 2, optionally, in some embodiments the system may further include a secondary device 112 with a processing unit 104 that accesses the data store 28 and retrieves multiple tooth movement data records for a given subject across one or more different time points.

[0095] Data records spanning multiple different time points are retrieved, each record including tooth mobility metrics for one or more teeth. The processor 104 of the secondary device 112 processes the retrieved data records to generate trend information indicative of one or more tooth mobility metrics as a function of time. The processor 104 of the secondary device 112 controls the user interface 106 of the secondary device to display the trend information.

[0096] Secondary devices include, for example, mobile computing devices such as smartphones.

[0097] Another aspect of the present invention provides only the processing device 20 without the oral care device 10 or other components of the system 8 .

[0098] Deriving a tooth mobility metric is a process based on two main parts. The first step is to derive a measurement from the sensed signal waveform that is indicative of (i.e., correlates with) the relative level of tooth mobility. The second step is to convert the relative level of tooth mobility into an objective metric based on the further use of at least one further characteristic related to the mechanical engagement of at least one cleaning element with or on the teeth. Alternatively, these two parts are performed by a single algorithm or function in some examples, for example using a multi-parametric regression function.

[0099] Each of these two parts of the processing operation will now be discussed further.

[0100] As stated, the first requirement is to obtain a measurement that indicates the level of tooth mobility, which measurement is a relative level, i.e., not standardized and therefore not directly comparable between teeth or cleaning sessions, in other words, a measurement that correlates with tooth mobility but does not necessarily represent an objective absolute measurement.

[0101] Thus, in some embodiments, the processing device determines the level of tooth movement by comparing one or more characteristics of the sensed signal waveform to corresponding characteristics for a stored baseline or reference sensed signal, for example corresponding to a signal for a particular reference tooth (e.g., a healthy permanent tooth that may be indicated by a dental practitioner).

[0102] In some embodiments, an averaged sensed signal for multiple cleaning sessions is calculated by storing the sensed signal for each session in memory, then retrieving the stored signals and calculating the average. This averaged signal can be used to determine the level of tooth movement instead of (or in addition to) the signal for only a single cleaning session.

[0103] In some embodiments, a measure of tooth mobility may be derived for a given tooth by comparing one or more characteristics of the signal waveform for the given tooth with characteristics of one or more other teeth acquired within the same cleaning session.

[0104] Typically, sensing means 30 incorporated in the oral cleaning device can be used to record the applied pressure (user load) and resistance from the teeth by analysing the reaction force (or pressure / acceleration) applied to the support 12. This information can be used to derive tooth mobility by analysing the change in a characteristic parameter such as vibration amplitude in one or more frequency components, or vibration damping derived from normal / tangential reaction forces, pressures or accelerations applied to the support and / or cleaning elements 14.

[0105] For example, Figure 3 shows a derived (linear) relationship between the tooth movement displacement amplitude (wobbling amplitude) and the corresponding support normal reaction force (wobbling of the cleaning element and thus the tooth hitting the support). The force is relatively large, e.g., 0.7-2.5 N in the example shown, and can be measured, e.g., using a Hall sensor.

[0106] In at least one set of embodiments, it is proposed to determine tooth mobility using a spectral analysis of the sensed signal. For example, determining tooth mobility includes evaluating in the frequency domain of the sensed signal the change in signal amplitude at one or more frequency points and / or the attenuation of the main harmonic components of the vibration pattern over one or more cleaning sessions. Changes are monitored using differential measurements between different teeth, over multiple cleaning sessions or within the same cleaning session.

[0107] The sensing principle is illustrated with reference to Figures 4 and 5. Figure 4 illustrates the mechanical engagement of the cleaning element 14 of the exemplary oral care device 10 with a tooth 40. When a user cleans his or her teeth, the user exerts a certain user-loaded force on the cleaning element, which includes a component perpendicular to the tooth surface. As shown in Figure 4, when the cleaning element engages a tooth with a high level of mobility (a "loose tooth"), the tooth exhibits movement. The movement of the highlighted tooth 40a is primarily about the y-axis, but also exhibits some smaller degree of movement about the x-axis.

[0108] 4 to 6, a computer simulation was performed to test and demonstrate the feasibility of detecting tooth mobility using an in-silico tooth-jaw model.

[0109] 4-7 relate to simulations that tested three different levels of tooth movement of the tooth 40a of interest: low tooth movement (maximum tooth displacement of 0.05 mm, representing normal physiological tooth movement), medium tooth movement (maximum tooth displacement of 0.44 mm), and high level tooth movement (maximum tooth displacement of 1.5 mm, which is typical of tooth movement for teeth that will be substantially lost).

[0110] The in-silico model consists of a jaw with soft gums around a reference tooth 40a, which is free to rotate about a mesial-distal (Y) axis and is connected to the root by a torsion spring.

[0111] The level of motion of the reference tooth is adjusted in the model by adjusting the stiffness coefficient of the torsion spring attached to the model. Brushing conditions were simulated based on a Philips A3 brush head with a 90° angle of the cleaning element relative to the tooth surface (90° brushing angle) and a constant user load of 1.88 N (average load across users of Philips toothbrushes). The motion pattern of the cleaning element was modeled as a large amplitude mesial-distal stroke (applied by the user) superimposed with a smaller amplitude oscillatory motion pattern induced by the actuation / drive mechanism of the power toothbrush.

[0112] Simulated sensor signals representing the reaction force on the brush head platen 12 as a function of time and simulated tooth displacements were calculated from the model for the three different perturbation conditions described above.

[0113] FIG. 4 (right side) shows a representation in the time domain of the resulting simulated sensed signal waveforms 60 for three different perturbation conditions.

[0114] The received sensed signal waveform 60 typically includes signal segments corresponding to respective periods of engagement of the at least one cleaning element with each of the plurality of teeth. The processing device 20 determines respective tooth mobility metrics for each of the plurality of teeth. An identification of each of the teeth is further obtained in some instances.

[0115] In the simulated sensed signal results of FIG. 4, a signal segment 62 is shown that corresponds to the modeled tooth 40a of interest.

[0116] According to at least one set of embodiments, determining the degree of tooth mobility includes performing a spectral analysis of the sensed signals.

[0117] This is further illustrated with reference to FIG. 5(a).

[0118] To calculate a measurement indicative of the degree of tooth mobility, the processing device according to at least one set of embodiments calculates a frequency spectrum 70, 72 for the sensed signal. This can be done, for example, using a Fourier transform, such as a Fast Fourier Transform (FFT).

[0119] The plot in Fig. 5(a) shows the frequency spectrum for each of three different tooth movement levels, labelled A, B and C, respectively. Level A corresponds to a movement amplitude of 0.05 mm. Level B corresponds to a movement amplitude of 0.44 mm. Level C corresponds to a movement amplitude of 1.5 mm. Fig. 5(a) shows the frequency spectrum 70 calculated for the normal reaction force component (x-direction in the figure) of the sensed signal at each of the three different tooth movement levels A, B and C.

[0120] As described above, in this example the oral care device is an oral care device comprising an actuation means 19 for driving a vibratory motion of at least one cleaning element 14, which vibratory motion exhibits a baseline / reference vibration frequency spectrum having a central frequency component.

[0121] Observation of the frequency domain plots 72, 74 for the reaction forces reveals a large increase in the reaction force amplitude levels for frequency components around harmonics of the dominant frequency (approximately 250 Hz) of the reference frequency spectrum of the actuation means. There is also an increase in the attenuation levels between frequency components around these harmonic frequencies. In particular, the amplitude of the frequency components corresponding to the dominant harmonic component of the vibratory driving force applied by the actuation means (at approximately 250 Hz) shows an amplitude increase of approximately 4 dB (in the normal direction) and 9.6 dB (in the mesial-distal direction) from the lowest to the highest levels of simulated tooth movement.

[0122] The change in damping around the driving vibration harmonics is qualitatively observable by the change in bandwidth and can be determined qualitatively, for example, by calculating the quality factor (Q-factor), a dimensionless parameter defined as the ratio of the resonant center frequency to its bandwidth. By resonant center frequency we mean the center frequency component of the dominant peak in the frequency spectrum. For example, for a -3 dB bandwidth, the Q-factor for the dominant peak (main harmonic peak) ranges from 30 for low levels of perturbation (case A) to 25 for high perturbation (case C) for normal forces, and similarly decreases from 31 to 25.7 for mesial-distal forces.

[0123] 5(b) shows the resulting relationship between the amplitude of the normal reaction force at the center frequency of the dominant peak of the frequency spectrum of the actuation means (e.g., 250 Hz in this particular example) and the wobbling amplitude (i.e., tooth mobility level) of the tooth of interest. The reaction force is measured by the sensing means 30 in the oral care device.

[0124] FIG. 6 shows the calculated relationship between tooth mobility level and normal reaction force amplitude for different signal segments 62 spanning different time periods.

[0125] FIG. 6(a) shows the same simulated time-domain sensed signal as in FIG.

[0126] Figure 6(b) shows the calculated frequency spectra for three different tooth movement levels A, B, C over a smaller time signal segment 62. The signal segment 62 in Figure 6 corresponds to a time window from 0.075 s to 0.2 s. The signal segment in Figure 5 corresponds to a time window from 0.16 s to 0.2 s.

[0127] 6(c) shows the resulting relationship between the amplitude of the normal reaction force at the central vibration frequency of the actuation means (e.g., 250 Hz in this particular example) and the wobbling amplitude of the tooth of interest (i.e., the level of tooth movement). The reaction force is measured by the sensing means 30 in the oral care device. It can thus be seen that different levels of tooth movement can be associated with different measured force levels, no matter what time window is selected to analyze the measurement signal.

[0128] Thus, in summary, following this approach, and according to at least one set of embodiments, determining the degree of tooth movement comprises detecting the amplitude of one or more frequency components of the detection signal, and more particularly based on changes in the amplitude of the central frequency of a dominant peak in the vibration frequency spectrum.

[0129] In some examples, determining the degree of tooth movement is performed by determining the change between a frequency spectrum of the vibrations applied by the actuation means and a sensed frequency spectrum.

[0130] Additionally or alternatively, in some examples, determining the degree of tooth movement is performed by determining the change in a bandwidth associated with a dominant peak in the frequency spectrum, or a quality factor associated with a dominant peak in the frequency spectrum, between the vibration frequency spectrum of the vibration applied by the actuation means and the detected frequency spectrum.

[0131] Based on these results, it can be seen that the measured normal reaction force amplitude (herein called Fn) at the center frequency of the dominant (or main harmonic) peak of the vibration frequency spectrum is directly proportional to the level of tooth mobility (TM). Therefore, the reaction force amplitude at the dominant frequency can be used as a direct proxy parameter for tooth mobility (TM). This can be expressed in the following mathematical form: Reaction force amplitude (Fn) = f(TM) ... (1) where TM represents the level of tooth mobility.

[0132] In some embodiments, the processing device further obtains an identification of the at least one tooth, and the data signal further comprises the identification of the at least one tooth.

[0133] The identification is obtained using positioning data for the at least one cleaning element, for example obtained using an orientation sensor or a motion sensor (e.g., an inertial measurement unit (IMU)) integrated into the oral care device.

[0134] Additionally or alternatively, tooth identification may be derived by analysis of the sensed signals.

[0135] For example, as described above, the received detection signal includes signal segments 62 corresponding to respective periods of engagement of the at least one cleaning element with each of the plurality of teeth, and the processing device determines a respective tooth mobility metric for each of the plurality of teeth and obtains an identification of each of the teeth.

[0136] The beginning and end of each signal segment 62 can be identified by detecting a predetermined signature pattern indicative of the beginning and end of each tooth segment, or by use of a positioning signal for at least one cleaning element received from a positioning signal source. In this manner, it can be determined which portions of the sensed signal relate to various teeth.

[0137] The processor further, in some examples, identifies the type of tooth to which each signal segment corresponds based on user input indicating the type of tooth.

[0138] As described above, deriving the tooth mobility metric may further include obtaining a measurement of at least one further characteristic, the further characteristic being a characteristic of the teeth and / or a characteristic of the mechanical engagement of at least one cleaning element with the teeth.

[0139] Various factors such as cleaning pressure (user load), angle of the cleaning element relative to the teeth, the characteristics with which the teeth are brushed, material properties of the cleaning element, and geometrical arrangement of the cleaning element tend to affect the measured dynamic sensing signal (e.g., reaction force), making it difficult to directly compare characteristic parameter values ​​obtained from different teeth or different brushing sessions without taking such additional factors into account when determining the tooth mobility metric.

[0140] The present invention investigated the above-mentioned factors using simulation experiments to determine the effect of each of them on the sensor signal response. For each simulation experiment, one parameter was varied at a time while the other parameters were held constant.

[0141] Various options for the additional characteristic or characteristics to be used in determining the tooth mobility metric are now described.

[0142] According to at least one set of embodiments, the at least one further characteristic includes a user baseline load parameter indicative of a baseline or average load force applied by a user to the cleaning element during use.

[0143] The processing device determines a baseline or average load force applied to the at least one cleaning element during a cleaning session using the sensing signal from the sensing means, in some instances as part of a calibration operation.

[0144] To further explain, the proposed embodiment involves receiving a sensing signal indicative of a force, pressure or movement during use of the oral care device in a cleaning session. In such a context, a user exerts a load on the cleaning elements 14 and the support 12 when cleaning their teeth. To increase the specificity of the measurement, the at least one further characteristic includes a user baseline load parameter indicative of a baseline or average load force exerted by the user on the cleaning elements during use.

[0145] Referring to Figures 7 and 8, a new set of simulations was used to investigate the effect of various applied user loads (1.88N, 1.44N and 1N) on the measured reaction force on the support 12 for simulated teeth having the same level of tooth mobility.

[0146] The normal force amplitude at the main harmonic component of the driving vibration applied to the cleaning element is the user load (U L ) was observed to have a monotonic and approximately linear relationship with

[0147] FIG. 7(a) shows the waveform of the resulting normal reaction force sense signal 60 for each of three different applied user loads (1N, 1.4N and 1.88N).

[0148] 7(b) shows the frequency spectrum calculated for signal segment 62 corresponding to the time window from 0.075 s to 0.2 s. This may be calculated, for example, using a Fourier transform. The frequency spectra for each of the three different applied user loads are labelled A, B and C, where A corresponds to the IN load, B corresponds to the 1.4N load and C corresponds to the 1.88N load.

[0149] FIG. 8 shows the derived approximately linear relationship between normal force amplitude at the dominant driving vibration frequency component (eg, 250 Hz in the simulated example) and the applied user load.

[0150] The amplitude of tooth deflection and the applied load force i.e. U L It is noted that similar linear relationships between and have been reported in the literature, see for example the following article: M. Goellner, J. Schmitt, S. Holst, A. Petschelt, M. Wichmann, and C. Berthold, "Correlations between tooth mobility and the Periotest method in periodontally involved teeth," Quintessence Int, Vol. 44, No. 4, pp. 307-316, April 2013.

[0151] Thus, the sensed signal, e.g. the sensed reaction force on the support 12, is a function of the tooth movement, and similarly the reaction force is a function of the user load (U L ) is a function of the user load (U), as shown diagrammatically in FIG. L ) can be expressed as a function of

[0152] It is further noted that the user load is functionally at least partially related to the structural properties of the cleaning element 14, since the cleaning elements behave differently under various applied user loads. Different loads result in different bending states of the cleaning element, which affects the stiffness of the force transmission by the cleaning element between the mobile teeth and the associated sensor means.

[0153] Under low user loads, the force transmission characteristics of the cleaning element become less stiff and less sensitive to tooth displacements, so there is a minimum user load above which "loose tooth" dynamic behavior can be detected.

[0154] It has been found that using only the user load as an additional feature used in the tooth mobility calculation is sufficient to provide an objective tooth mobility metric that allows meaningful comparisons between mobility measurements at different times.

[0155] In practice, the level of applied user load may be sensed in a dedicated calibration operation or may be calculated based on sensed signals obtained in one or more normal cleaning sessions. The user load may be detected as a baseline of the sensed force or pressure signal from the sensor means 30, and the applied user load may be expected to be a dominant component of the baseline, and thus a relative change in the baseline corresponds to a change in the user load. The average of the user load may in some instances be obtained over multiple cleaning sessions.

[0156] With regard to determining a specific relationship between user load and sensor signal response for a given level of tooth movement, this may be based on a predefined (population average) relationship. Alternatively, a customer relationship may be derived for a particular user. For example, this may be achieved in a dedicated calibration procedure. In other words, in some embodiments, the processing device 20 may have performed a calibration operation that comprises sensing a baseline or average load force applied to at least one cleaning element during a cleaning session, and the degree of tooth movement is determined as a function of the baseline or average load force.

[0157] For example, the user is instructed to apply the cleaning element 14 to a single tooth and to vary the load applied by the cleaning element. The resulting measurement signal for the calibration procedure corresponds to a single tooth having a single level of motion. Variations in the load applied by the user can be detected, for example, by extracting a baseline of the force or pressure sensed signal, and the applied user load can be expected to be the dominant component of the baseline, and thus a relative change in the baseline corresponds to a change in user load. At the same time, the amplitude of the dominant harmonic component of vibration can be monitored from the sensed signal over the calibration operation period as a function of time over this period. A plot of amplitude as a function of (relative) applied user load can then be derived by temporally registering the extracted user load signal and the extracted amplitude signal with respect to one another, from which a linear correlation (i.e., a linear mapping) between the extracted user load signal and the extracted amplitude signal is derived.

[0158] This correlation can then be applied during operation to calibrate the sensing signal according to varying user loads during a cleaning session, or according to an average user load derived from data acquired over one or more cleaning sessions.

[0159] Another factor that influences the relationship between the detection signal and tooth mobility is the morphology of the tooth being cleaned. It has been found that single-rooted and multi-rooted teeth can have significantly different mobility patterns. For example, the maximum mobility amplitude for molars is typically around 40 μm, whereas the maximum mobility amplitude for incisors is typically around 120 μm. See, for example, M. Perlitsh, "A systematic approach to the interpretation of tooth mobility and its clinical implications," Dental Clinics of North America, Vol. 24, pp. 177-93, May 1980.

[0160] Therefore, according to some embodiments, tooth morphological characteristics are taken into account when calculating tooth mobility metrics.

[0161] As mentioned above, an indication of tooth type can be obtained based on obtaining a tooth identification, which may be derived based on a position sensor incorporated in the device, or based on analysis of the sensed signal waveform itself, or based on further sensing components, such as a camera incorporated in the device and positioned to capture image data of the teeth being cleaned. Standard image processing methods can be used to determine the tooth type from the image data.

[0162] For each tooth type, its mechanical properties are pre-determined and functional parameters are stored in the memory of the processor for use in calculating the mobility metric. For example, a valid tooth stiffness parameter k may be pre-derived for each of a number of different possible tooth types, e.g., from experimental or population-based data. The value of k for each tooth type may be stored in the memory. In operation, once a tooth type is detected, the corresponding stiffness parameter is retrieved. Calculation of the tooth mobility metric involves application of a function that includes k as an independent parameter / variable.

[0163] Additionally or alternatively, the at least one further characteristic may include one or more parameters relating to structural or mechanical properties of the at least one cleaning element.

[0164] In particular, the stiffness of the cleaning element, the degree of wear of the cleaning element, and the protrusion angle (or brushing angle (b A )) cleaning elements and supports (e.g., brush heads b H ) are the relevant parameters.

[0165] In particular, the protrusion angle of the cleaning element and the stiffness metric of the cleaning element are each fixed parameters of at least a portion of the oral care device and can therefore be easily determined. In some examples, the support and the cleaning element are part of a cleaning unit (e.g., a brush head) that is removably coupled to the body of the device to allow replacement. In some examples, an electronically readable tag (e.g., an RFID tag) that is readable by a sensor in the body portion may be incorporated into the cleaning unit portion, the tag including information about the structural characteristics of the cleaning element and / or the support. In some examples, the information may instead be input by a user (e.g., by an app loaded on a secondary device that is communicatively connected to the processing device).

[0166] In some embodiments, the at least one further characteristic includes a brushing stroke speed parameter indicative of the speed of a brushing stroke applied by a user during a cleaning operation.

[0167] This can be determined and stored in a calibration operation, or it can be determined dynamically during a cleaning session, for example using an IMU built into the device.

[0168] Thus, physical properties specific to the cleaning unit support and cleaning elements (type of cleaning element, stiffness of cleaning element, coefficient of friction of cleaning element, wear / deformation of cleaning element, cleaning element geometry, etc.) may be included among the at least one further characteristic.

[0169] In a preferred set of embodiments, the tooth mobility metric is derived using a combination of the different further characteristics mentioned above. This can be achieved, for example, by defining a tooth mobility metric calculation function that is a function of the different input parameters. For example, it is a linear function of the various further characteristics. The linear function may be derived using a multi-parametric linear regression technique, where the coefficients of the linear function are pre-derived by a regression fitting operation based on experimental data. In the final device, the tooth mobility metric calculation function is stored in the memory of the processing device and is retrieved based on the measured waveform characteristics and the at least one further characteristic during the computation of the tooth mobility metric calculation.

[0170] In some examples, a machine learning model is used to calculate the tooth mobility metrics.

[0171] In some examples, a multi-parametric linear regression model is used to calculate the tooth mobility metrics.

[0172] To explain further, in general, the calculation function f is determined in advance, and the calculation function f is a) a signal waveform characteristic derived from a sensed signal indicative of a measured force applied to at least one cleaning element or indicative of movement of at least one cleaning element; b) at least one further characteristic of the teeth and / or at least one further characteristic of the mechanical engagement of the at least one cleaning element with the teeth; and c) Measurement criteria for tooth mobility is defined to provide a mapping from

[0173] In the particular example described above, the signal waveform characteristics derived from the sensed signal are F n and F n In this context, refers to the reaction force amplitude at the central frequency of the dominant peak in the frequency spectrum (i.e., the force amplitude at the principal harmonic or resonant frequency of vibration).

[0174] Regarding (b), examples of further properties as mentioned above are as follows: k = tooth hardness metric, U L = user baseline load, b H = one or more structural characteristics of the cleaning element, such as stiffness and / or wear level, and b A = the angle of the cleaning element relative to the tooth surface while cleaning the teeth (i.e., the brushing angle).

[0175] Therefore, deriving a tooth mobility (TM) metric is TM = f(F n ,k,U L ,b H ,b A ) The process is carried out based on a processing device that calculates where f is a predefined calculation function as described above. As described above, the tooth mobility metric is calculated by subtracting F from the reference frequency spectrum. n The function f may be constructed based on a reference frequency spectrum to provide an appropriate mapping to the TM.

[0176] Of course, the above is only one example, and f may alternatively be a function of one or more different combinations of further characteristics of the teeth and / or further characteristics of the mechanical engagement of the at least one cleaning element with the teeth. n Alternatively, different characteristics of the sense signal waveform may be used.

[0177] Once a standardized metric of tooth mobility is obtained, in some embodiments this metric is further processed to derive a clinical classification of tooth mobility for at least one tooth.

[0178] This set of embodiments therefore proposes to clinically assess tooth mobility by categorizing tooth mobility metrics according to clinically relevant categories, allowing for clinical interpretation and cross-correlation between teeth and cleaning sessions.

[0179] Deriving the clinical classification comprises comparing the obtained tooth mobility metric to one or more selected thresholds.

[0180] Optionally, the threshold or thresholds against which the tooth mobility metric is compared may be selected based on at least one of tooth type and a demographic classification of the user.

[0181] For example, during a brushing session, a sensing signal is received from the sensing means (e.g., reaction force measurements). The sensing signal includes signal segments corresponding to a series of different dental positions, each spanning a particular time window of the signal. The particular tooth to which each signal segment corresponds can be identified based on a location method, such as using a position or orientation sensor means incorporated in the device, as described above, or any other suitable technique.

[0182] The measured sensing signal segments 62 for each tooth may then be processed in combination with one or more of the additional characteristics described above to calculate a tooth mobility metric for the corresponding tooth, for example, using a linear regression model or other suitable equation, function or algorithm.

[0183] The derived TM values ​​are compared to predefined thresholds specific to a particular tooth to derive a clinical classification for each tooth. In this way, teeth exhibiting pathological tooth mobility are identified. In some instances, a graded classification indicating the level of pathology can be determined.

[0184] The classification is one of a set of individual classifications, e.g., no disease, low level disease, moderate disease, severe disease. The individual classification is derived by comparing the derived tooth mobility metric to one or more tooth-specific thresholds. In some examples, the classification is a graded score along a pathological scale, e.g., from +1 to +10.

[0185] Because different types of teeth have different normal levels of movement, for each of a number of different tooth types, a look-up table is used to store appropriate tooth movement thresholds for the various classifications. These thresholds can be derived from literature or can be derived experimentally using experimental data.

[0186] FIG. 10 illustrates diagrammatically how a pre-defined tooth mobility threshold may be applied to derive a clinical classification.

[0187] FIG. 11 outlines in block diagram form the decision flow for deriving a clinical classification.

[0188] Following the procedures outlined above, sensing signals are acquired (202) over the course of a cleaning session and tooth mobility metrics are derived for each tooth (204).

[0189] The tooth mobility metric is compared 206 to a physiological threshold, which, as described above, is specific to a particular tooth.

[0190] If the tooth mobility exceeds the physiological threshold, a conclusion is reached that the tooth mobility is pathological (i.e. abnormal and indicative of some form of disease) (208), if not, a conclusion is reached that the tooth mobility is non-pathological and this indication is output in the form of a data signal.

[0191] The first step in deriving a clinical classification for tooth mobility for a given tooth then comprises comparing (210) the tooth mobility metric obtained for the tooth with a tooth mobility metric of a further tooth to detect a level of deviation between these two tooth mobility metrics. The further tooth is preferably of the same tooth type and on the opposite side of the dental arch to the first tooth (e.g., the upper right first molar compared to the upper left first molar). Additionally or alternatively, tooth mobility metrics may be obtained for multiple segments of each tooth and metrics for matching segments on opposite sides of the mouth may be compared.

[0192] The level of deviation between these two tooth mobility metrics is then compared to a deviation threshold (212).

[0193] If the deviation level exceeds the threshold, the disease is classified as localized disease (214). If the deviation level does not exceed the threshold, the disease level is classified as systemic disease (216).

[0194] The tooth mobility metric derived for each tooth (or tooth segment) is then further compared to previous tooth mobility metrics derived for the same tooth or tooth segment to ascertain whether there has been deterioration (218, 230). For example, the method may include accessing a data store that records historical tooth mobility metric data for a user and comparing the resulting tooth mobility metric for at least one tooth to historical values ​​for the same tooth, wherein clinical separation is further based on the comparison.

[0195] If the difference between the current tooth mobility metric and the previous value(s) for a given tooth is greater than a predetermined threshold (220, 232), the disease is classified as chronic (224, 234), otherwise the disease is classified as acute (226, 238). In an alternative example, the reverse logic may instead be applied, i.e., persistently high tooth mobility indicates chronic disease, whereas a sudden increase indicates acute disease.

[0196] In particular, significant deviations in tooth mobility metrics may indicate, for example, progressive disease. This information is then communicated to the user or dentist, for example, using a user interface component.

[0197] In some embodiments, the processing device further generates an alert and / or a recommendation to schedule a visit to a dentist depending on the derived clinical classification, which in some examples depends on a trend in tooth mobility values ​​or clinical classification, such as a deterioration in mobility or disease.

[0198] Optionally, the derived tooth mobility metric values ​​and / or clinical classifications for each tooth may be stored in memory for further reference.

[0199] As a further example, FIG. 12 shows a logical configuration for deriving a clinical classification based on tooth mobility. As an example, it is proposed to derive a measurement of tooth mobility (represented by Ωtooth in FIG. 12) for one or more teeth, which measurement is derived from sensor data from an oral care device, e.g., force or pressure measurement. It is further proposed to pre-create a rule base, stored in a data store, that records a set of thresholds to which the derived measurement of tooth mobility for a user is compared in order to obtain a clinical classification. The thresholds in the example include Ωnormal and Ωwobbly. As shown in FIG. 12, in this example, if Ωtooth<=Ωnormal, no action is required. If Ωwobbly>Ωtooth>Ωnormal, the clinical classification is an early warning classification and the user is advised to remain vigilant about the progression of tooth mobility. If Ωtooth>Ωwobbly, the clinical classification is that the teeth have abnormal mobility and the user is advised to seek advice from a dentist.

[0200] With regard to the thresholds stored in the rule base, these are derived based on empirical data, e.g., based on a population database including tooth mobility measurements and corresponding diagnoses / classifications from dentists. Thus, the thresholds are set based on a combination of empirical data (tooth mobility measurements) and data from dentists. The thresholds are further personalized to the user based on past mobility measurements for the user and / or based on user-specific input from the dentist, e.g., the dentist adjusts or sets the thresholds based on a physical examination of the user.

[0201] According to some embodiments, the processing device derives a prediction of tooth mobility progression based on, for example, the use of machine learning algorithms trained on a database of historical data for a population of users.

[0202] For example, in at least one set of embodiments, a machine learning algorithm is trained to receive as input the derived tooth mobility metrics for a single cleaning session and a demographic classification of the user, and to generate as output a prediction of tooth mobility progression.

[0203] The machine learning algorithm is pre-trained based on training data for a population of patients in different demographic categories, the training data including tooth mobility metrics across at least two different time points for each of the patients and demographic categories for the patients, the training data being data extracted from a population database.

[0204] Additionally or alternatively, a machine learning model is used that is trained to receive as input the derived tooth mobility metrics for a series of cleaning sessions over a series of time points and to generate as output a prediction of tooth mobility progression, the algorithm being trained on training data for a population of patients, the training data including tooth mobility metrics over a series of different time points for each of a plurality of patients, again the training data being data extracted from a population database.

[0205] For example, a population database may be used that includes recorded sensor signal data for multiple cleaning sessions for multiple users, and also records other parameters related to the cleaning elements or related to the mechanical engagement of the cleaning elements with the teeth, such as brushing angle, tooth type, etc. The database also records corresponding tooth mobility measurements and clinical classifications for each tooth included in the sensor signal data for each user. The data may be longitudinal, i.e., includes data for each individual over a long period of time. Other data may also be included, such as demographic data, dental data, and data related to brushing behavior for each user, that allows for customized predictions (i.e., based on a group level).

[0206] To further explain, a population database includes relevant data for a large number of individuals. It includes, for each individual, longitudinal data regarding tooth movement over an extended period of time (e.g., from the first onset of movement onward). It also includes other data, such as demographic data, for each patient.

[0207] The data obtained about the user of the device may include very limited longitudinal data regarding tooth movement, for example the data includes tooth movement data that only covers a relatively short amount of time.

[0208] The processing unit 20 uses the population database to generate a prediction for the user regarding the future progression of tooth mobility over time.

[0209] To do this, a machine learning algorithm can be applied, which has been trained on a population database.

[0210] According to a further example, a "nearest neighbor approach" is instead applied to generate the prediction. A subset of the population database is selected based on a calculated similarity metric between the data of the selected subset and the data for the current user, for example based on the similarity between the longitudinal movement data and demographic factors of the user with the longitudinal movement data and demographic data of the population database subset. A prediction of the future progression of tooth movement for the user is then derived based on the tooth movement progression for individuals in the selected subset of the population database.

[0211] Since different teeth have different mobility profiles, different machine learning models can be pre-trained and applied for different tooth types. Furthermore, different machine learning models can be trained and used for different types or classifications of diseases. Disease types or classifications (e.g., acute vs. chronic, localized vs. generalized) are related to tooth mobility progression, and therefore models can be modified or weights adapted for different diseases.

[0212] The use of machine learning models for prediction represents just one example: other types of algorithms, functions, or models may alternatively be used.

[0213] In algorithmic form, tooth mobility prediction can be considered as a time series prediction, where future tooth mobility values ​​are predicted based on previously observed values. For this purpose, various mathematical models can be used, such as moving average or autoregressive average linear (ARIMA).

[0214] FIG. 13 shows an exemplary result of tooth mobility prediction using an ARIMA model based on previous tooth mobility values. Previous values ​​are in the light grey area. Predicted values ​​forming a forward predicted trend are in the darker grey area. A severe disease threshold line is plotted on the chart. The predicted future trend allows for the predicted future time point when the tooth mobility value will cross the threshold.

[0215] If the predicted tooth movement value exceeds a certain predefined threshold, this predicted tooth movement value may be displayed to the user, for example by generating a sensory output, for example by audiovisual means, this information can be used by the user, for example for scheduling dental visits.

[0216] Optionally, the population database may include data on both users who have undergone TM treatment and those who have not, which could potentially allow for the generation of two predictions (predictions for both scenarios, i.e., with and without TM treatment), such that these insights can be used to prompt users into treatment.

[0217] By way of overview, a flow chart outlining various features implemented in accordance with various embodiments of the present invention is shown in FIG.

[0218] The parameters shown in area B are needed only during training of the algorithms used to calculate the various parameters. Although this exemplary chart focuses on parameters for toothbrushes, the same principles can be applied to other types of oral care devices, such as brushing mouthpiece devices.

[0219] In summary, force, pressure or motion sensing data, along with at least one further parameter related to the teeth or the engagement characteristics of the teeth with the device, are used as inputs to derive a tooth mobility (TM) metric. As mentioned in FIG. 14, this may be based on the application of a machine learning (ML) model trained with experimental data or population TM data. The derived tooth mobility metric is then used to clinically classify each tooth's mobility based on one or more thresholds. A prediction of future tooth mobility trends for one or more teeth is further derived.

[0220] Although FIG. 14 refers to machine learning (ML) algorithms as examples, any other suitable type of algorithm or function may be used instead.

[0221] The above-described embodiments of the invention use a processing device. The processing device generally includes a single processor or multiple processors. The processing device may be located in a single containing device, structure or unit, or may be distributed among multiple different devices, structures or units. Thus, a reference to a processing device performing or configured to perform a particular step or task corresponds to the step or task being performed by any one or more of multiple processing components, either alone or in combination. A person skilled in the art will understand how such a distributed processing device can be implemented. The processing device includes a communication module or input / output for receiving and outputting data to further components.

[0222] The processor or processors of the processing device can be implemented in numerous ways using software and / or hardware to perform the various functions required. A processor typically uses one or more microprocessors that are programmed using software (microcode) to perform the functions required. A processor may also 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.

[0223] Examples of circuitry used 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).

[0224] In various implementations, the processor is associated with one or more storage media, such as volatile and non-volatile computer memories, such as RAM, PROM, EPROM, and EEPROM. The storage media are encoded with one or more programs that, when executed on the one or more processors and / or controllers, perform the required functions. The various storage media may be attached within the processor or controller, or may be transportable such that one or more programs stored on the storage media can be loaded into the processor.

[0225] 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 word "a" or "an" does not exclude a plurality of elements.

[0226] The functions of several items recited in the claims may be carried out by a single processor or other unit.

[0227] 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.

[0228] The 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 telecommunications systems.

[0229] It is noted that when the term "adapted to" is used in the claims or the specification, it is intended to be equivalent to the term "configured to."

[0230] Any reference signs in the claims shall not be construed as limiting the scope.

Claims

1. 1. A processing device for an oral care device, comprising: An input / output section; one or more processors; The processing device includes: receiving at the input / output a sensing signal indicative of a load force applied to and movement of at least one cleaning element of the oral care device, the sensing signal spanning a time period corresponding to a duration of mechanical engagement of at least one tooth with the at least one cleaning element during an oral cleaning session; processing the sensed signal waveform over the time period to extract one or more characteristics of the signal waveform; obtaining measurements of at least one further characteristic which is a characteristic of the teeth and / or a characteristic of the mechanical engagement of the at least one cleaning element with the teeth; determining a tooth mobility metric for the at least one tooth in response to the one or more signal characteristics and the at least one further characteristic; A processing unit generating a data signal indicative of the determined metric and supplying said data signal to said input / output.

2. The processing device of claim 1 , wherein the tooth mobility metric defines a gradation level of tooth mobility.

3. The processing device further comprises: accessing a data store that records historical tooth mobility metric data for users; 3. A processing device according to claim 1 or 2, further comprising: a processor for storing the determined tooth mobility metric in the data store by outputting the data signal to the data store.

4. the one or more processors access the data store recording the historical tooth mobility metric data for the user; The one or more processors may further determine a long-term trend of tooth mobility metrics for one or more teeth of the user based on the historical tooth mobility metric data and the determined tooth mobility metric. The processing device of claim 3 , wherein the one or more processors are further configured to generate a prediction of tooth mobility progression based on the long-term trends.

5. 5. A processing device according to claim 1, further comprising: a processor for deriving an identification of the at least one tooth; and the data signal further comprising said identification of the at least one tooth.

6. 6. The processing device of claim 5, wherein the received detection signal includes signal segments corresponding to respective durations of engagement of the at least one cleaning element with each of a plurality of teeth, and the processing device determines respective tooth mobility metrics for each of the plurality of teeth to obtain an identification of each of the teeth.

7. The start and end of each said signal segment, by detecting a predetermined signature pattern indicating the beginning and end of each tooth segment; or The processing device of claim 6 , further comprising determining by use of a positioning signal for the at least one cleaning element received from a positioning signal source.

8. and wherein the at least one further characteristic includes a user baseline load parameter indicative of a baseline or average load force applied by the user to the cleaning element during use; Optionally, the processing device performs a calibration operation comprising determining the baseline or average load force applied to the at least one cleaning element during a cleaning session using the sensed signal.

9. The at least one further characteristic is a brushing stroke speed parameter indicative of the speed of brushing strokes applied by a user during said cleaning action; and / or One or more parameters relating to structural characteristics of the at least one cleaning element, the protrusion angle of the cleaning element; and Stiffness metric of the cleaning element One or more parameters, including one or more of: The processing device according to claim 1 , comprising:

10. The processing device of claim 1 , further comprising: a processor for processing the obtained tooth mobility metric to derive a clinical classification of the tooth mobility for the at least one tooth.

11. 11. The processing device of claim 10, wherein deriving the clinical classification comprises comparing the obtained tooth mobility metric to one or more selected thresholds, optionally the one or more thresholds being selected based on at least one of tooth type and a demographic classification of the user.

12. the sensed signal includes signal segments corresponding to a duration of engagement of the at least one cleaning element with a plurality of teeth, and the processor determines a respective tooth mobility metric for each of the plurality of teeth; 12. The processing device of claim 10 or 11, wherein deriving the clinical classification for the tooth mobility for a first tooth comprises comparing the tooth mobility metric obtained for the first tooth with a further tooth, the further tooth being of the same tooth type and on an opposite side of the dental arch to the first tooth, to detect a deviation level, and wherein the deriving comprises comparing the deviation level to a deviation threshold.

13. The processor further comprises: accessing a data store that records historical tooth mobility metric data for users; comparing the resulting tooth mobility metric for the at least one tooth to historical values ​​for the same tooth; 13. The processing structure of any one of claims 10 to 12, wherein said clinical classification is further based on said comparison.

14. the processor generates a prediction of the tooth mobility progression based on application of a machine learning algorithm; 14. The processing device of claim 1, wherein the machine learning algorithm is trained to receive as input the derived tooth mobility metrics for a single cleaning session and a demographic classification of the user, and to generate as output a prediction of the tooth mobility progression, the algorithm being trained on training data for a population of patients in different demographic classifications, the training data including tooth mobility metrics at two different time points for each of the plurality of patients, and a demographic classification for the patient.

15. A support; a plurality of cleaning elements for engagement with an oral surface, the cleaning elements protruding from a surface of the support; sensor means for sensing measurements indicative of a load or pressure applied to at least one of said cleaning elements or indicative of movement of said at least one of said cleaning elements; A processing device according to any one of claims 1 to 14; An oral care device comprising: