Occlusion vertical distance detection method and device and storage medium

Abstract

The invention provides an occlusion vertical distance detection method and device and a storage medium. The occlusion vertical distance detection method comprises the following steps: in a continuous tooth clicking process of a target main body, acquiring tooth clicking data of the target main body; according to the tooth clicking data, determining an resting vertical distance of the target main body; and determining the occlusion vertical distance of the target main body according to the rest vertical distance and a preset rest gap.

Description

Technical Field

[0001] This disclosure relates to the field of information processing technology, and in particular to a method, apparatus and storage medium for detecting bite vertical distance. Background Technology

[0002] Complete denture restoration is a routine treatment for edentulous patients. When performing occlusal reconstruction for edentulous patients, determining the occlusal vertical dimension (OVD) is crucial to ensuring the quality and efficiency of the restoration.

[0003] OVD (Orthogonal Volume Depth) refers to the distance from the base of the nose to the base of the chin when the natural teeth are in a centric occlusal position, which is equivalent to the height of the lower third of the face. Current OVD testing methods are highly dependent on the experience and skills of clinicians. If the OVD determined by the doctor is too low or too high, it will significantly reduce the effectiveness of complete denture restoration. Summary of the Invention

[0004] The inventors noted that existing OVD (occlusal vertical distance) testing methods are highly dependent on the experience and skills of clinicians and are easily affected by factors such as patient positioning and psychological state; they are also objectively susceptible to soft tissue deformation. To improve the quality of complete denture restorations, there is an urgent need to establish an efficient, accurate, and objective method for detecting occlusal vertical distance.

[0005] Accordingly, this disclosure provides a method for detecting occlusal vertical distance, which can improve the accuracy of occlusal vertical distance detection results, thereby achieving accurate OVD detection.

[0006] According to a first aspect of the present disclosure, a method for detecting occlusal vertical distance is provided, comprising: acquiring occlusal data of a target subject during continuous occlusal movements; determining a resting vertical distance of the target subject based on the occlusal data; and determining the occlusal vertical distance of the target subject based on the resting vertical distance and a preset resting gap.

[0007] In some embodiments, determining the resting vertical distance of the target subject based on the tapping data includes: processing the tapping data using numerical analysis methods to determine the resting vertical distance of the target subject; or processing the tapping data using a machine learning model to determine the resting vertical distance of the target subject.

[0008] In some embodiments, processing the percussion data using numerical analysis methods to determine the resting vertical distance of the target subject includes: processing the naso-mental distance of N sampling points in the percussion data to obtain the velocity, acceleration, and jerk of the N sampling points in the percussion data, where N is the total number of sampling points in the percussion data; dividing the percussion data into multiple data segments based on the velocity of the N sampling points; determining the optimal phase corresponding to the resting jaw position based on the multiple data segments; determining the representative value of each data segment in the multiple data segments based on the optimal phase; and determining the resting vertical distance of the target subject based on the representative values ​​of all data segments.

[0009] In some embodiments, determining the optimal phase corresponding to the resting jaw position includes: extracting the nasomental-mental distance from the sampling points corresponding to the preset initial phase in each of the plurality of data segments to obtain a set of nasomental-mental distances; determining the deviation of the set of nasomental-mental distances; iteratively optimizing the preset initial phase based on the deviation of the set of nasomental-mental distances to obtain the optimal phase, wherein the optimal phase minimizes the deviation of the set of nasomental-mental distances.

[0010] In some embodiments, determining the representative value of each data segment among the plurality of data segments based on the optimal phase includes: determining a first phase window based on the optimal phase and a redundancy value; calculating a first velocity threshold and a first sag threshold for each data segment; determining candidate sampling points for each data segment based on the first phase window, the first velocity threshold, and the first sag threshold; determining representative sampling points for each data segment from the candidate sampling points; and determining the representative value of each data segment based on the representative sampling points.

[0011] In some embodiments, calculating the first velocity threshold and the first judder threshold of each data segment includes: calculating the first velocity quantile of the absolute value of the velocity of all sampling points in each data segment to obtain the first velocity threshold of each data segment; and calculating the first judder quantile of the absolute value of the judder of all sampling points in each data segment to obtain the first judder threshold of each data segment.

[0012] In some embodiments, determining the candidate sampling point for each data segment includes: if the i-th sampling point in the k-th data segment is located within the first phase window, then determining whether the absolute value of the velocity of the i-th sampling point is greater than or equal to a first velocity threshold of the k-th data segment, wherein... K is the total number of data segments. Let I be the total number of sampling points in the k-th data segment; if the absolute value of the velocity of the i-th sampling point is greater than or equal to the first velocity threshold of the k-th data segment, then determine whether the absolute value of the jump of the i-th sampling point is greater than or equal to the first jump threshold of the k-th data segment; if the absolute value of the jump of the i-th sampling point is greater than or equal to the first jump threshold of the k-th data segment, then determine whether the absolute value of the jump of the i-th sampling point is the maximum value in the neighborhood of the i-th sampling point; if the absolute value of the jump of the i-th sampling point is the maximum value in the neighborhood of the i-th sampling point, then the i-th sampling point is selected as a candidate sampling point.

[0013] In some embodiments, determining the candidate sampling points for each data segment further includes: determining the candidate sampling points for the k-th data segment based on the first phase window, a first velocity threshold for the k-th data segment, and a first sag threshold for the k-th data segment, wherein... K is the total number of data segments; if the number of candidate sampling points of the kth data segment is 0, then adjust at least one of the phase window, velocity threshold and jog threshold until the number of candidate sampling points of the kth data segment is greater than 0.

[0014] In some embodiments, adjusting at least one of the phase window, velocity threshold, and judder threshold includes: increasing the first phase window to obtain a second phase window; decreasing the first velocity threshold to obtain a second velocity threshold; and decreasing the first judder threshold to obtain a second judder threshold.

[0015] In some embodiments, determining a representative sampling point for each data segment from the candidate sampling points of each data segment includes: determining a score for each candidate sampling point based on the velocity, jerk, and phase of each candidate sampling point in the k-th data segment, wherein... K is the total number of data segments; the candidate sampling point with the highest score among the candidate sampling points of the k-th data segment is taken as the representative sampling point of the k-th data segment.

[0016] In some embodiments, determining the score of each candidate sampling point includes: calculating the ratio of the absolute value of the velocity of each candidate sampling point to a first velocity threshold of the k-th data segment to obtain the relative velocity of each candidate sampling point; calculating the ratio of the absolute value of the judder of each candidate sampling point to a first judder threshold of the k-th data segment to obtain the relative judder of each candidate sampling point; calculating the difference between the phase of each candidate sampling point and the optimal phase to obtain the phase deviation of each candidate sampling point; and obtaining the score of each candidate sampling point based on the relative velocity, relative judder, and phase deviation of each candidate sampling point.

[0017] In some embodiments, dividing the tapping data into multiple data segments based on the velocities of the N sampling points includes: determining whether the velocity direction of the nth sampling point is the same as the velocity direction of the (n-1)th sampling point, wherein... If the velocity direction of the nth sampling point is not the same as the velocity direction of the (n-1)th sampling point, then it is determined that the velocity direction from the nth sampling point to the (n-1)th sampling point is... Are the velocity directions of the sampling points the same? The hysteresis length is the distance from the nth sampling point to the nth sampling point. If the velocity directions of all sampling points are the same, then the nth sampling point is taken as the segment boundary; according to the segment boundary, the tapping data is divided to obtain the multiple data segments.

[0018] In some embodiments, dividing the tapping data according to the segmentation boundary to obtain the plurality of data segments includes: dividing the tapping data into a plurality of pre-selected segments according to the segmentation boundary; calculating the amplitude ratio of the amplitude of each pre-selected segment to a reference amplitude; calculating the target sampling point percentage of each pre-selected segment; and determining the plurality of data segments from the plurality of pre-selected segments according to at least one of the amplitude ratio of the amplitude of each pre-selected segment to the reference amplitude and the target sampling point percentage of each pre-selected segment.

[0019] In some embodiments, calculating the amplitude ratio of each preselected segment to a reference amplitude in the plurality of preselected segments includes: calculating the amplitude of the nasal-mental distance of all sampling points in each preselected segment to obtain the amplitude of each preselected segment; taking the median of the amplitudes of all preselected segments as the reference amplitude; and calculating the ratio of the amplitude of each preselected segment to the reference amplitude to obtain the amplitude ratio of each preselected segment to the reference amplitude.

[0020] In some embodiments, calculating the target sampling point ratio of each preselected segment includes: determining a job reference value based on the absolute value of the job of N sampling points in the tapping data; if the absolute value of the job of the nth sampling point is greater than the job reference value, then the nth sampling point is taken as the target sampling point; calculating the ratio of the number of target sampling points in each preselected segment to the total number of sampling points in each preselected segment to obtain the target sampling point ratio of each preselected segment.

[0021] In some embodiments, determining multiple data segments from the plurality of preselected segments includes: if the amplitude ratio of the preselected segment to the reference amplitude is within a preset amplitude ratio range, then the preselected segment is determined as a data segment; if the proportion of target sampling points in the preselected segment is less than or equal to a preset proportion threshold, then the preselected segment is determined as a data segment.

[0022] In some embodiments, the method for training the machine learning model includes: acquiring percussion data samples for each subject during continuous percussion; acquiring intermittent percussion data for each subject during intermittent percussion under the guidance of a doctor; determining a resting vertical distance label corresponding to the percussion data sample for each subject based on the intermittent percussion data; constructing a dataset using the percussion data samples of all subjects and the corresponding resting vertical distance labels; and training the machine learning model using the dataset.

[0023] According to a second aspect of the present disclosure, an occlusal vertical distance detection device is provided, comprising: a memory; and a processor coupled to the memory, the processor being configured to execute instructions stored in the memory to implement the occlusal vertical distance detection method as described in any of the above embodiments.

[0024] According to a third aspect of the present disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions that, when executed by a processor, implement the bite vertical distance detection method as described in any of the above embodiments.

[0025] According to a fourth aspect of the present disclosure, a computer program product is provided, including computer instructions, wherein when executed by a processor, the computer instructions implement the bite vertical distance detection method as described in any of the above embodiments.

[0026] Other features and advantages of this disclosure will become clear from the following detailed description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description

[0027] To more clearly illustrate the technical solutions in the embodiments of this disclosure or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0028] Figure 1 This is a schematic flowchart of a bite vertical distance detection method according to an embodiment of the present disclosure;

[0029] Figure 2 This is a flowchart illustrating a numerical analysis method according to an embodiment of the present disclosure;

[0030] Figure 3 This is a schematic flowchart of a model training method according to an embodiment of the present disclosure;

[0031] Figure 4 This is a schematic diagram of the marker pasting position according to an embodiment of the present disclosure;

[0032] Figure 5 This is a schematic diagram of marker recognition according to an embodiment of the present disclosure;

[0033] Figure 6 This is a schematic diagram of intermittent tapping data according to an embodiment of the present disclosure;

[0034] Figure 7 This is a schematic diagram of tapping data according to an embodiment of the present disclosure;

[0035] Figure 8 This is a schematic diagram of tapping data for a single cycle according to an embodiment of the present disclosure;

[0036] Figure 9 This is a schematic diagram of tapping data according to another embodiment of the present disclosure;

[0037] Figure 10 This is a schematic diagram of occlusal data according to yet another embodiment of this disclosure;

[0038] Figure 11 This is a schematic diagram of the structure of a bite vertical distance detection device according to an embodiment of the present disclosure. Detailed Implementation

[0039] The technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, and not all embodiments. The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit this disclosure or its application or use. All other embodiments obtained by those skilled in the art based on the embodiments of this disclosure without creative effort are within the scope of protection of this disclosure.

[0040] Unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of this disclosure.

[0041] At the same time, it should be understood that, for ease of description, the dimensions of the various parts shown in the accompanying drawings are not drawn according to actual scale.

[0042] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0043] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0044] It should be noted that similar labels and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be discussed further in subsequent figures.

[0045] The inventors noted that existing OVD detection methods include mechanical and physiological methods. Mechanical methods include pre-extraction recording, while physiological methods include resting jaw position reference, swallowing method, facial landmark method, finger measurement method, X-ray projection measurement method, and speech method. All of these methods have certain limitations.

[0046] The resting jaw position reference method is a widely accepted and applicable method for OVD (Occupational Void Distance) detection in clinical practice. Its principle is that when the head is upright and the mouth is not chewing, swallowing, or speaking, the mandible is in its habitual resting position (RJP), with a resting jaw gap of approximately 2-3 mm between the upper and lower dental arches. At this time, the distance from the base of the nose to the base of the chin is the resting vertical dimension (RVD). Subtracting the resting jaw gap gives the OVD. While this method is relatively convenient and provides a certain quantitative standard, it still has the following drawbacks:

[0047] 1) Measurement results are easily affected by subjective factors such as the patient's posture and psychological state;

[0048] 2) Different doctors have different criteria for judging the resting jaw position;

[0049] 3) Measuring tools (such as a vertical distance ruler) and techniques can affect the accuracy of measurement results;

[0050] 4) The technology is highly sensitive, subjective, and has poor objective stability, which increases patients' consultation time and costs.

[0051] Other methods for detecting oral vena cava (OVD) also have their limitations. For example, the pre-extraction recording method is only suitable for the fabrication of the first denture. The swallowing method is highly subjective. The facial landmark method is susceptible to errors due to soft tissue contraction. In X-ray projection measurement, current research has not found stable landmarks strongly correlated with OVD, and the results may be affected by race and gender. The speech method requires too much clinical experience from the physician.

[0052] The limitations of the aforementioned OVD (Occlusal Vertical Distance) testing methods lead to inaccurate OVD results, resulting in poor outcomes for complete denture restorations and increased treatment time and costs. Specifically, excessively low OVD results can cause patients to present with an aged appearance, characterized by a prominent chin and reduced lip fullness. Excessively high OVD results can lead to condylar displacement, increasing the risk of articular disc and ligament damage, weakening the movement of the jaw-lifting muscles, and reducing electromyographic activity. Furthermore, improper OVD recovery can result in decreased chewing efficiency, causing temporomandibular joint pain, headaches, and other discomfort. To improve the quality of complete denture restorations, there is an urgent need to establish an efficient, accurate, and objective method for measuring the occlusal vertical distance.

[0053] Accordingly, this disclosure provides a method for detecting occlusal vertical distance, which can improve the accuracy of occlusal vertical distance detection results, thereby achieving accurate OVD detection.

[0054] Figure 1 This is a schematic flowchart of a method for detecting the vertical occlusal distance according to an embodiment of the present disclosure. In some embodiments, the following method for detecting the vertical occlusal distance is performed by a vertical occlusal distance detection device, including steps 11-13.

[0055] In step 11, during the continuous teeth tapping process of the target subject, the teeth tapping data of the target subject is acquired.

[0056] For example, the target subject sits upright, and markers are placed on the base of their nose and chin. The subject then performs continuous teeth tapping at a comfortable frequency and force. A binocular high-speed camera is used to record the continuous teeth tapping process, resulting in a teeth tapping video. The sampling frame rate of the binocular high-speed camera is greater than or equal to 500 frames per second. Accompanying software is used to identify the nasal and chin markers in the teeth tapping video, and distance parameter calibration and marker tracking analysis are performed to obtain the subject's teeth tapping data.

[0057] In some embodiments, the tapping data of the target subject includes the nasal-mental distance of N sampling points, where N is the total number of sampling points for the tapping data, and the nasal-mental distance is the distance between the detected nasal floor landmark and the mental floor landmark.

[0058] For example, the distance between the nasal base and chin base at the nth sampling point is .

[0059] In some embodiments, the tapping data is filtered to remove drift and noise from the tapping data.

[0060] For example, the distance between the nasal base and chin base at the nth sampling point in the teeth tapping data. It can be decomposed into effective signal, drift and noise, as shown in Equation (1).

[0061] (1)

[0062] in, For drifting, For a valid signal, It is noise.

[0063] For example, suppose the drift extracted using a large-window Savitzky-Golay filter is The distance between the base of the nose and the base of the chin is Then remove the drifted nasal-mental distance. As shown in formula (2).

[0064] (2)

[0065] For example, a Butterworth filter can be used to remove drift in the nasomental-mental distance. After processing, the distance between the base of the nose and the base of the chin after noise removal is obtained as follows: .

[0066] In step 12, the resting vertical distance of the target body is determined based on the tapping data.

[0067] In some embodiments, numerical analysis methods are used to process the tapping data to determine the resting vertical distance of the target subject. The numerical analysis methods include... Figure 2 As shown.

[0068] In some embodiments, a machine learning model is used to process the tapping data to determine the resting vertical distance of the target subject. The machine learning model utilizes, for example... Figure 3 The model training method shown is used to obtain the model.

[0069] For example, using numerical analysis methods or machine learning models, the resting vertical distance of the target subject can be obtained as follows: .

[0070] In step 13, the bite vertical distance of the target body is determined based on the resting vertical distance and the preset resting gap.

[0071] For example, assuming the resting vertical distance is The preset rest interval is The vertical distance of the bite of the target body As shown in formula (3).

[0072] (3)

[0073] It should be noted here that the preset rest interval... This can be determined based on clinical experience. For example, a pre-set rest interval. The value range is 2-4mm.

[0074] The bite vertical distance detection method described in the above embodiments can improve the accuracy of bite vertical distance detection results, thereby achieving accurate OVD detection.

[0075] In some embodiments, the method for determining the resting vertical distance of the target subject using numerical analysis methods is as follows: Figure 2 As shown.

[0076] Figure 2 This is a flowchart illustrating a numerical analysis method according to an embodiment of the present disclosure, including steps 21-25.

[0077] In step 21, the nasal-mental distances of the N sampling points in the teeth tapping data are processed to obtain the velocity, acceleration, and jerk of the N sampling points in the teeth tapping data. Here, N is the total number of sampling points in the teeth tapping data.

[0078] For example, assuming the nasomental distance for noise removal is... Then the velocity of the nth sampling point in the tapping data As shown in formula (4), the acceleration at the nth sampling point As shown in formula (5), the jerk j[n] of the nth sampling point is shown in formula (6).

[0079] (4)

[0080] (5)

[0081] (6)

[0082] in, Distance between the base of the nose and the base of the chin The differential, The timestamp corresponding to the sampling point. For timestamps The differential, For speed The differential, For acceleration The differential.

[0083] In step 22, based on the velocity of N sampling points, the tapping data is divided into multiple data segments.

[0084] In some embodiments, the method for dividing the tapping data into multiple data segments based on the velocity of N sampling points includes the following steps S101-S104.

[0085] In step S101, it is determined whether the velocity direction of the nth sampling point is the same as the velocity direction of the (n-1)th sampling point, where .

[0086] In step S102, if the velocity direction of the nth sampling point is not the same as the velocity direction of the (n-1)th sampling point, then it is determined that the velocity direction from the nth sampling point to the (n-1)th sampling point is... Are the velocity directions of the sampling points the same? This is the hysteresis length.

[0087] For example, hysteresis length The value range is 2-5.

[0088] In step S103, if from the nth sampling point to the nth sampling point... If the velocity directions of all sampling points are the same, then the nth sampling point is used as the segment boundary.

[0089] It should be noted here that if from the nth sampling point to the... If the velocity directions of the sampling points are different, then the velocity reversal at the nth sampling point is considered to be due to jitter. The method described in the above embodiment can determine the boundary of half a cycle of mandibular movement by judging whether the velocity direction changes. Simultaneously, the addition of hysteresis length as a constraint can prevent false boundaries from being generated under noise due to velocity direction reversal.

[0090] In step S104, the tapping data is divided according to the segmentation boundaries to obtain multiple data segments.

[0091] In some embodiments, the method of dividing the tapping data according to the segmentation boundary to obtain multiple data segments includes the following steps S201-S204.

[0092] In step S201, the tapping data is divided into multiple pre-selected segments according to the segmentation boundaries.

[0093] In step S202, the amplitude ratio of each preselected segment to the reference amplitude is calculated.

[0094] In some embodiments, the amplitude of the nasomental distance at all sampling points in each preselected segment is calculated to obtain the amplitude of each preselected segment. The median of the amplitudes of all preselected segments is used as the reference amplitude. The ratio of the amplitude of each preselected segment to the reference amplitude is calculated to obtain the amplitude ratio of each preselected segment relative to the reference amplitude.

[0095] For example, suppose the amplitude of the m-th pre-selected segment is ,in M is the total number of pre-selected segments, then the reference amplitude As shown in formula (7).

[0096] (7)

[0097] in, This is a median function.

[0098] For example, suppose the amplitude of the m-th pre-selected segment is... The reference amplitude is Then the amplitude ratio of the m-th pre-selected segment is... As shown in formula (8).

[0099] (8)

[0100] In step S203, the proportion of target sampling points for each pre-selected segment is calculated.

[0101] In some embodiments, a jerk reference value is determined based on the absolute value of the jerk of N sampling points in the tapping data. If the absolute value of the jerk of the nth sampling point is greater than the jerk reference value, then the nth sampling point is taken as the target sampling point. The ratio of the number of target sampling points in each pre-selected segment to the total number of sampling points in each pre-selected segment is calculated to obtain the proportion of target sampling points in each pre-selected segment.

[0102] In some embodiments, the judder reference value is the 95th percentile of the absolute value of the judder of N sampling points.

[0103] For example, suppose the judder of the nth sampling point is Then the jump reference value As shown in formula (9).

[0104] (9)

[0105] in, Let be the absolute value of the jerk at the nth sampling point. It is the 95th percentile function.

[0106] In step S204, multiple data segments are determined from multiple pre-selected segments based on at least one of the amplitude ratio of each pre-selected segment to the reference amplitude and the proportion of target sampling points of each pre-selected segment.

[0107] In some embodiments, if the amplitude ratio of the preselected segment to the reference amplitude is within a preset amplitude ratio range, the preselected segment is determined as a data segment.

[0108] In some embodiments, the preset amplitude ratio range is (0.8, 1.2).

[0109] For example, if the amplitude of the m-th pre-selected segment is greater than... Then the m-th pre-selected segment is determined as the data segment.

[0110] It should be noted that if the amplitude ratio of a pre-selected segment exceeds the preset amplitude ratio range, the data fluctuation of that pre-selected segment is considered too large. This may be due to excessive noise or incomplete tapping motion. Therefore, the pre-selected segment is considered an unstable segment.

[0111] In some embodiments, if the proportion of target sampling points of a preselected segment is less than or equal to a preset proportion threshold, the preselected segment is determined as a data segment.

[0112] In some embodiments, the preset percentage threshold is 0.2.

[0113] It should be noted that if the proportion of the target sampling points of the m-th pre-selected segment is less than or equal to 0.2, then the m-th pre-selected segment will be determined as a data segment.

[0114] It should be noted that the higher the proportion of target sampling points in the pre-selected segment, the greater the data jitter of the pre-selected segment. Therefore, this pre-selected segment is considered an unstable segment.

[0115] The method described in the above embodiments determines multiple data segments from multiple pre-selected segments based on the amplitude ratio of the pre-selected segments and / or the proportion of target sampling points, which can improve the stability of the data segments and thus improve the accuracy of the bite vertical distance detection results.

[0116] In step 23, the optimal phase corresponding to the resting jaw position is determined based on multiple data segments.

[0117] It should be noted that when a person's head is upright and the mouth is not chewing, swallowing, or speaking, the lower jaw is in a habitual resting position, which is called the resting jaw position.

[0118] For example, the kth data segment is ,in K represents the total number of data segments; the first segment in the tapping data... Each sampling point represents the k-th data segment. The starting sampling point, where ; The first in the data of teeth tapping Each sampling point represents the k-th data segment. The end sampling point, where That is to say, the k-th data segment .

[0119] It's important to note here that phase refers to the relative position of a sampling point within the k-th data segment. For example, the starting sampling point of the k-th data segment. The phase is 0, and the end sampling point of the k-th data segment. The phase is 1, and the phase of the sampling point at the very center of the k-th data segment is 0.5.

[0120] In some embodiments, determining the optimal phase corresponding to the resting jaw position includes the following steps S301-S303.

[0121] In step S301, the nasal-mental distance of the sampling point corresponding to the preset initial phase in each of the multiple data segments is extracted to obtain the nasal-mental distance set.

[0122] For example, assuming a preset initial phase If the value is 0.5, then the k-th data segment will have a phase with the preset initial phase. corresponding sampling points As shown in formula (10).

[0123] (10)

[0124] For example, in the k-th data segment, the initial phase is... corresponding sampling points Nasal base to chin base distance As shown in formula (11).

[0125] (11)

[0126] For example, the nasomental-mental distance of the sampling point corresponding to the preset initial phase in each of the K data segments is calculated to obtain the set of nasomental-mental distances. .

[0127] In step S302, the deviation of the nasal base-mental base distance set is determined.

[0128] In some embodiments, the difference between each naso-chin-base distance in the set of naso-chin-base distances and the median of the set of naso-chin-base distances is calculated to obtain the median difference of each naso-chin-base distance. The median of the median differences of each naso-chin-base distance in the set of naso-chin-base distances is calculated to obtain the deviation of the set of naso-chin-base distances.

[0129] For example, suppose the set of distances between the nasal base and chin base is... The deviation of the distance between the nasal base and chin base from the set As shown in formula (12).

[0130] (12)

[0131] It should be noted here that the deviation of the distance between the nasal base and chin base is significant. The smaller the value, the better the K data segments are. The more concentrated the readings at different locations, the better the data consistency.

[0132] In step S303, the preset initial phase is iteratively optimized based on the deviation of the nasal base-chin base distance set to obtain the optimal phase, wherein the optimal phase minimizes the deviation of the nasal base-chin base distance set.

[0133] It should be noted here that by iteratively optimizing the preset initial phase, the deviation of the nasal base-mental base distance set is reduced. Minimize to obtain the optimal phase. In some embodiments, a low-velocity or high-jump penalty term may also be superimposed on the deviation of the nasal-mental distance set.

[0134] In the method described in the above embodiments, the phase is used to normalize the sampling distance of each data segment, scan the phase in stable data segments, calculate the robust dispersion as the stability loss of the data segment, and thus obtain the individualized optimal phase that minimizes the loss.

[0135] In step 24, a representative value for each of the multiple data segments is determined based on the optimal phase.

[0136] In some embodiments, a method for determining a representative value for each of a plurality of data segments based on an optimal phase includes the following steps S401-S405.

[0137] In step S401, the first phase window is determined based on the optimal phase and the redundancy value.

[0138] In some embodiments, the lower limit of the first phase window is the difference between the optimal phase and the redundancy value. The upper limit of the first phase window is the sum of the optimal phase and the redundancy value.

[0139] In some embodiments, the redundancy value ranges from [0.08, 0.12].

[0140] For example, suppose the optimal phase is Redundancy value Then the first phase window As shown in formula (13).

[0141] (13)

[0142] For example, optimal phase It can be 0.5.

[0143] It should be noted that the phase window is used to avoid the optimal phase falling in a region that is almost still or fluctuates greatly.

[0144] In step S402, a first velocity threshold and a first slap threshold for each data segment are calculated.

[0145] In some embodiments, a first velocity quantile is calculated to determine the absolute value of the velocities of all sampling points in each data segment, thereby obtaining a first velocity threshold for each data segment.

[0146] In some embodiments, the first velocity quantile is the 80th percentile;

[0147] For example, suppose the velocity of all sampling points in the k-th data segment is If the first velocity quantile is the 80th percentile, then the first velocity threshold for the k-th data segment is... As shown in formula (14).

[0148] (14)

[0149] in, It is the 80% quantile function.

[0150] In some embodiments, the first hop quantile is calculated as the absolute value of the hops of all sampling points in each data segment to obtain the first hop threshold for each data segment.

[0151] In some embodiments, the first quantile is the 95th percentile.

[0152] For example, suppose the judder of all sampling points in the k-th data segment is If the first hop quantile is the 95th percentile, then the first hop threshold for the k-th data segment is... As shown in formula (15).

[0153] (15)

[0154] in, It is the 95th percentile function.

[0155] In step S403, candidate sampling points for each data segment are determined based on the first phase window, the first velocity threshold for each data segment, and the first sag threshold for each data segment.

[0156] In some embodiments, the method for determining candidate sampling points for each data segment includes the following steps S501-S504.

[0157] In step S501, if the i-th sampling point in the k-th data segment is located within the first phase window, then it is determined whether the absolute value of the velocity of the i-th sampling point is greater than or equal to the first velocity threshold of the k-th data segment, where... K is the total number of data segments. I represents the total number of sampling points in the k-th data segment.

[0158] In step S502, if the absolute value of the velocity of the i-th sampling point is greater than or equal to the first velocity threshold of the k-th data segment, then it is determined whether the absolute value of the jump of the i-th sampling point is greater than or equal to the first jump threshold of the k-th data segment.

[0159] In step S503, if the absolute value of the jump of the i-th sampling point is greater than or equal to the first jump threshold of the k-th data segment, then it is determined whether the absolute value of the jump of the i-th sampling point is the maximum value in the neighborhood of the i-th sampling point.

[0160] In step S504, if the absolute value of the jolt of the i-th sampling point is the maximum value in the neighborhood of the i-th sampling point, then the i-th sampling point is selected as a candidate sampling point.

[0161] It should be noted that the absolute value of the jump at the i-th sampling point is the maximum value in the neighborhood of the i-th sampling point. In other words, the absolute value of the jump at the i-th sampling point is a local maximum, and the i-th sampling point is a local maximum of the absolute value of the jump.

[0162] For example, if the absolute value of the jerk at a sampling point is the maximum value in its neighborhood, then that sampling point is considered a local maximum. The set of all local maxima in the k-th data segment is... .

[0163] For example, suppose the first phase window is The first velocity threshold of the k-th data segment is The first hop threshold for the k-th data segment is The set of all local maxima in the k-th data segment is Then the set of candidate sampling points for the k-th data segment As shown in formula (16).

[0164] (16)

[0165] in, The phase corresponding to the i-th sampling point in the k-th data segment. Let be the index of the i-th sampling point in the k-th data segment, where , The velocity of the i-th sampling point in the k-th data segment. Let be the judder of the i-th sampling point in the k-th data segment.

[0166] It should be noted that formula (16) means searching near the optimal phase (e.g., the optimal phase is 0.5), requiring the velocity to enter the first 20% high value area of ​​this segment; requiring the jump to enter the first 5% high value area of ​​this segment; taking the local peak of the absolute value of the jump to avoid continuous redundant points within the threshold.

[0167] The method described in the above embodiments, which utilizes a phase window, a first velocity threshold, and a first jolt threshold to search for candidate sampling points near the optimal phase of the k-th data segment, can improve the robustness of the bite vertical distance detection method. Furthermore, using the jolt as a local maximum as a screening condition can avoid the occurrence of consecutive redundant sampling points among the candidate sampling points.

[0168] In some embodiments, candidate sampling points for the k-th data segment are determined based on a first phase window, a first velocity threshold for the k-th data segment, and a first sag threshold for the k-th data segment, wherein... K is the total number of data segments. If the number of candidate sampling points for the k-th data segment is 0, then adjust at least one of the phase window, velocity threshold, and jog threshold until the number of candidate sampling points for the k-th data segment is greater than 0.

[0169] In some embodiments, adjusting at least one of the phase window, velocity threshold, and jump threshold includes: increasing the first phase window to obtain a second phase window; decreasing the first velocity threshold to obtain a second velocity threshold; and decreasing the first jump threshold to obtain a second jump threshold.

[0170] It should be noted here that if the set of candidate sampling points for the k-th data segment... If the number of sampling points is 0, the screening criteria can be gradually relaxed. For example, the first jolt threshold determined based on the 95th percentile of the jolt can be used. Adjusted to the second quiescent threshold determined based on the 90th percentile of the quiescent. Redundant values Adjust it to 0.15 to increase the phase window and obtain the second phase window.

[0171] In the method described in the above embodiments, by adjusting at least one of the phase window, velocity threshold, and jog threshold, the screening conditions are gradually relaxed until the number of candidate sampling points of the k-th data segment is greater than 0, thereby achieving adaptive threshold screening. This can maximize the accuracy of the detection results and reduce the risk of missed detection while ensuring that each data segment has enough candidate sampling points.

[0172] In step S404, representative sampling points for each data segment are determined from among the candidate sampling points for each data segment.

[0173] In some embodiments, a score for each candidate sampling point is determined based on the velocity, jerk, and phase of each candidate sampling point in the k-th data segment, wherein... K is the total number of data segments. The candidate sampling point with the highest score among the candidate sampling points of the k-th data segment is taken as the representative sampling point of the k-th data segment.

[0174] In some embodiments, the ratio of the absolute value of the velocity of each candidate sampling point to a first velocity threshold of the k-th data segment is calculated to obtain the relative velocity of each candidate sampling point. The ratio of the absolute value of the judder of each candidate sampling point to a first judder threshold of the k-th data segment is calculated to obtain the relative judder of each candidate sampling point. The difference between the phase of each candidate sampling point and the optimal phase is calculated to obtain the phase deviation of each candidate sampling point. Based on the relative velocity, relative judder, and phase deviation of each candidate sampling point, a score is obtained for each candidate sampling point.

[0175] In some embodiments, a weighted sum of the relative velocity, relative jerk, and phase deviation of each candidate sampling point is calculated to obtain a score for each candidate sampling point.

[0176] For example, suppose the i-th sampling point of the k-th data segment belongs to the candidate sampling point set of the k-th data segment. The first velocity threshold of the k-th data segment is The relative velocity of the i-th sampling point As shown in formula (17).

[0177] (17)

[0178] For example, suppose the first hop threshold for the k-th data segment is... The relative jerk of the i-th sampling point As shown in formula (18).

[0179] (18)

[0180] For example, suppose the relative velocity of the i-th sampling point of the k-th data segment is... The relative jump of the i-th sampling point is The phase corresponding to the i-th sampling point in the k-th data segment is The optimal phase is Then the score of the i-th sampling point of the k-th data segment As shown in formula (19).

[0181] (19)

[0182] in, As the weight of relative velocity, The weights for relative jumps, This represents the weight of the phase deviation.

[0183] In step S405, the representative value of each data segment is determined based on the representative sampling points of each data segment.

[0184] In some embodiments, for each data segment, the median distance between the nasal floor and chin floor of multiple sampling points within the neighborhood of the representative sampling point is taken as the representative value for each data segment. The size of the neighborhood is determined by the neighborhood radius.

[0185] For example, suppose the first tooth tapping data Each sampling point is a representative sampling point of the k-th data segment, where Then the representative value of the k-th data segment As shown in formula (20).

[0186] (20)

[0187] in, Let be the neighborhood radius, the th The sampling point is located at the th sampling point Within the neighborhood of each sampling point.

[0188] For example, neighborhood radius The value range is 2-4.

[0189] In step 25, the resting vertical distance of the target subject is determined based on the representative values ​​of all data segments.

[0190] In some embodiments, the median of the representative values ​​of all data segments is calculated to obtain the resting vertical distance of the target subject.

[0191] For example, suppose the representative value of the k-th data segment is Then the vertical distance of the target's resting point As shown in formula (21).

[0192] (twenty one)

[0193] By using the methods described in the above embodiments, numerical analysis is employed to explore the correlation between the distance between the nasal floor landmark and the chin floor landmark and the resting vertical distance during continuous teeth tapping, thereby obtaining the accurate resting vertical distance and achieving accurate detection of the occlusal vertical distance, thus establishing an objective and quantifiable method for detecting the occlusal vertical distance.

[0194] In some embodiments, a machine learning model is used to process the tapping data to determine the resting vertical distance of the target subject. The machine learning model utilizes, for example... Figure 3 The model training method shown is used to obtain the model.

[0195] Figure 3 This is a flowchart illustrating a model training method according to an embodiment of the present disclosure, including steps 31-35.

[0196] In step 31, during each subject's continuous teeth tapping, teeth tapping data samples are obtained for each subject.

[0197] For example, each subject sat upright, and markers were placed on the base of their nose and chin. Each subject then performed continuous teeth tapping at a comfortable frequency and force. The continuous teeth tapping process was recorded using a binocular high-speed camera, with a sampling frame rate of at least 500 frames per second. Accompanying software was used to identify the nasal and chin markers in each subject's teeth tapping video, and distance parameter calibration and marker tracking analysis were performed to obtain teeth tapping data samples for each subject.

[0198] It should be noted here that the naso-mental distance sequence can be extracted from the calibrated teeth-tapping video. It employs "trend removal + zero-phase low-pass" to suppress low-frequency drift and high-frequency noise.

[0199] In step 32, during each subject’s intermittent teeth tapping under the guidance of a doctor, intermittent teeth tapping data is acquired for each subject.

[0200] For example, each subject performed intermittent teeth tapping with a strong biting force as instructed by the doctor, and the mandible was required to return to the resting jaw position spontaneously after each tapping. The intermittent teeth tapping process of each subject was recorded using a binocular high-speed camera, resulting in intermittent teeth tapping videos for each subject. The acquisition parameters (e.g., video duration and sampling frame rate) of each subject's intermittent teeth tapping videos were kept consistent. Accompanying software was used to identify nasal and chin floor landmarks in each subject's intermittent teeth tapping videos, and distance parameter calibration and landmark tracking analysis were performed to obtain intermittent teeth tapping data for each subject.

[0201] In step 33, based on the intermittent tapping data of each subject, a resting vertical distance label corresponding to the tapping data sample of each subject is determined.

[0202] It should be noted that the resting vertical distance for each subject was determined by a clinically experienced physician based on their intermittent teeth-tapping data. This determined resting vertical distance was then used as a label corresponding to each subject's teeth-tapping data sample.

[0203] In step 34, a dataset is constructed using the teeth tapping data samples of all subjects and the corresponding resting vertical distance labels.

[0204] In step 35, the machine learning model is trained using the dataset.

[0205] In some embodiments, the tapping data samples in the dataset are cleaned.

[0206] For example, removing null values ​​and abrupt frames from the tapping teeth data samples.

[0207] In some embodiments, the nasal-mental distance sequence in the tapping data samples of the dataset is processed to obtain the corresponding velocity sequence, acceleration sequence, and jerk sequence. A machine learning model is then used to process the nasal-mental distance sequence and the corresponding velocity, acceleration, and jerk sequences of the tapping data samples in order to train the machine learning model.

[0208] In some embodiments, the tapping data samples in the dataset are divided into multiple data segments according to the method described in the above embodiments, and the resting vertical distance labels corresponding to the tapping data samples are used as labels for the multiple data segments. The machine learning model is trained using the multiple data segments and the corresponding resting vertical distance labels.

[0209] In some embodiments, each of the multiple data segments is normalized.

[0210] For example, amplitude normalization for each data segment includes: subtracting the median of all sampling points in the data segment from the nasomental distance of each sampling point in the data segment to obtain the normalized data segment; or, dividing the nasomental distance of each sampling point in the data segment by the peak-to-peak value of the nasomental distance in the data segment to obtain the normalized data segment.

[0211] For example, time normalization is performed on each data segment, including aligning the duration of the data segment to L.

[0212] The methods described in the above embodiments can maintain the consistency of training data for machine learning models and enhance the robustness of machine learning models across people and devices.

[0213] In some embodiments, the dataset is divided into a training set and a validation set based on the data source, such that the data samples in the training set and the validation set do not have overlapping data sources.

[0214] For example, multiple data segments belonging to the same subject may be distributed only in the training set or only in the validation set.

[0215] In some embodiments, the machine learning model is a Bidirectional Long Short-Term Memory (Bi-LSTM) network model.

[0216] It's important to note that the distance between the base of the nose and the base of the chin varies over time, exhibiting temporal characteristics. Considering these temporal features, a Bi-LSTM model suitable for temporal feature extraction is trained. This model combines forward and backward LSTMs to comprehensively consider historical and future trends from the complete sequence, learning the long-term dependencies and data distribution patterns in the time series of the distance between the base of the nose and the base of the chin. A Bi-LSTM temporal encoder encodes the aligned multi-channel sequence and performs global pooling, then connects to a regression head to output the predicted RVD. A parallel confidence head is also set up to output the confidence prediction.

[0217] It should also be noted that during the training of the machine learning model, the vertical distance between the nose and chin is used as the target value, and the model is trained by inputting feature data. During the training process, the model automatically learns the distance between the base of the nose and the base of the chin during continuous teeth tapping. The nonlinear mapping relationship between the variation pattern and the resting vertical distance. Key hyperparameters (e.g., learning rate, network depth and width, window length, regularized weights, etc.) are determined through Bayesian optimization or equivalent automated search to reduce the bias of manual parameter tuning. To further improve robustness, weak enhancement strategies such as amplitude micro-scaling, time stretching, and light noise injection can be introduced during the sample generation stage, and random dropout or weight decay can be used on the model side to control overfitting.

[0218] In some embodiments, the machine learning model may also be an LSTM, random forest, support vector machine, or other deep neural network.

[0219] By using the methods described in the above embodiments, a supervised learning task is constructed. Using occlusal data samples as input and manually determined resting vertical distance as labels, an end-to-end machine learning model is trained, which can accurately predict the resting vertical distance, thereby improving the accuracy and efficiency of occlusal vertical distance detection.

[0220] The numerical analysis method will now be described through a specific embodiment of this disclosure.

[0221] In some embodiments, the numerical analysis method includes the following steps S601-S606.

[0222] In step S601, during the process of the target subject performing intermittent teeth tapping under the guidance of a doctor, the intermittent teeth tapping data of the target subject is collected.

[0223] Figure 4 This is a schematic diagram illustrating the placement of marker points according to an embodiment of this disclosure. Figure 4 As shown, the nasal base marker 41 is pasted on the nasal base of the target subject, and the chin base marker 42 is pasted on the chin base of the target subject.

[0224] For example, a binocular high-speed camera is used to record video of the intermittent teeth-tapping process of the target subject, resulting in a video of the intermittent teeth-tapping of the target subject. Next, the accompanying software is used to identify the nasal base landmark 41 and the chin base landmark 42.

[0225] For example, the nasal base landmark 41 identified using the accompanying software... Figure 5 As shown.

[0226] Next, the distance between the nasal base and chin base is obtained by using the positions of the identified nasal base landmarks and chin base landmarks.

[0227] For example, using the accompanying software, the nasal floor landmarks and chin floor landmarks in the intermittent teeth tapping video of the target subject can be identified, and distance parameters can be calibrated and landmark tracking analysis can be performed to obtain the intermittent teeth tapping data of the target subject.

[0228] For example, intermittent tapping data includes frame number, timestamp (unit: s), X pixel coordinate (unit: px), Y pixel coordinate (unit: px), displacement (unit: mm), X component of displacement (unit: mm), Y component of displacement (unit: mm), velocity (unit: mm / s), X component of velocity (unit: mm / s), Y component of velocity (unit: mm / s), average velocity (unit: mm / s), and acceleration (unit: mm / s). ), acceleration X component (unit: ), acceleration Y component (unit: Table 1 shows some of the data on intermittent clenching.

[0229] Table 1

[0230] In step S602, the standard value of the resting vertical distance of the target body is determined based on the intermittent tapping data of the target body.

[0231] Figure 6 This is a schematic diagram of intermittent tapping data according to an embodiment of this disclosure. Figure 6 As shown, the horizontal axis represents the frame number, and the vertical axis represents the distance (unit: pixels). Figure 6 Curve 61 in the figure represents the change in the vertical distance between the nasal base landmark and the chin base landmark. Figure 6 The dashed line 62 in the figure is a constant curve, used to represent the standard value of the resting vertical distance determined based on the intermittent tapping data of the target subject.

[0232] It should be noted that the target subject was subjected to intermittent teeth tapping under the guidance of an experienced physician. Therefore, the resting vertical distance of the target subject, determined based on the intermittent teeth tapping data, will be used as the standard value for subsequent verification.

[0233] In step S603, during the continuous teeth tapping process of the target subject, the teeth tapping data of the target subject is collected.

[0234] For example, a binocular high-speed camera is used to record video of the continuous teeth-tapping process of a target subject, resulting in a continuous teeth-tapping video of the target subject. Accompanying software is used to identify nasal and chin floor landmarks in the continuous teeth-tapping video of the target subject, and distance parameter calibration and landmark tracking analysis are performed to obtain the target subject's teeth-tapping data.

[0235] In step S604, the nasal-mental distance of N sampling points in the teeth tapping data is processed to obtain the velocity, acceleration, and jerk of the N sampling points. Here, N is the total number of sampling points in the teeth tapping data.

[0236] In step S605, the correlation between the standard value of the resting vertical distance of the target subject and the nasal-mental distance, velocity, acceleration and jerk of N sampling points in the percussion data is analyzed to determine the correlation characteristics between the percussion data and the resting jaw position.

[0237] Figure 7 This is a schematic diagram of teeth tapping data according to an embodiment of this disclosure. Figure 7 As shown, the horizontal axis represents the frame number, the left vertical axis represents the distance (unit: px), and the right vertical axis represents the speed (unit: mm / s). Figure 7 Bar chart 71 in the figure is a bar chart of the distance between the nasal base and the chin base, corresponding to the left vertical axis. Figure 7 Curve 72 in the figure represents the velocity Y component curve, corresponding to the right ordinate. Figure 7 The straight line 73 in the figure is a constant curve, used to represent the standard value of the resting vertical distance determined based on the intermittent tapping data of the target subject, corresponding to the left vertical axis. Figure 7 The straight line 74 in the figure is a constant curve, which differs from the standard value of the vertical distance from the resting point by 2px. It is used to represent the upper limit of the observation range and corresponds to the left vertical axis. Figure 7 The straight line 75 in the figure is a constant curve, which differs from the standard value of the vertical distance from the resting point by 2px. It is used to represent the lower limit of the observation range and corresponds to the left vertical axis.

[0238] Figure 8 This is a schematic diagram of tapping data for a single cycle according to an embodiment of this disclosure. Figure 8 As shown, the horizontal axis represents time, the left vertical axis represents distance (unit: px), and the right vertical axis represents speed (unit: mm / s). Figure 8 Bar chart 81 in the figure is a bar chart of the distance between the nasal base and the chin base, corresponding to the left vertical axis. Figure 8 Curve 82 in the figure represents the velocity Y component curve, corresponding to the right ordinate. Figure 8 The MPP (Mandibular Postural Position) region is based on... Figure 7 The timestamp corresponding to the point where the distance between the base of the nose and the base of the chin in the bar chart 81 is within the MPP region, as determined by lines 74 and 75, is taken as the target timestamp. For example, bar 83 is one of several bars in the bar chart 81 corresponding to the target timestamp, and point 84 is the point on curve 82 corresponding to bar 83.

[0239] like Figure 8 As shown, based on the distance between the naso-mental base and the base of the nose, the data for a single cycle of tooth tapping is divided into four regions: Region 1 (85), Region 2 (86), Region 3 (87), and Region 4 (88). Region 1 (85) represents the mouth-opening process, where the mandible moves downward and the distance between the naso-mental base and the base of the nose increases. Region 2 (86) represents the mouth-closing process, where the mandible moves upward and the distance between the naso-mental base and the base of the nose decreases. Region 3 (87) represents the displacement caused by the biomechanical response of the periodontal ligament during tooth tapping. Region 4 (88) represents the stage where the mandible is in the intercuspal position (ICP), which is the position of the mandible when the upper and lower teeth reach the widest and closest contact. Figure 8 The dotted line 89 in the figure represents the distance between the nasal base and chin base when the mandible is in the position of intercuspal position.

[0240] like Figure 8 As shown, during a single jaw-tapping cycle, analyzing the points on curve 82 corresponding to the target timestamp reveals that these points are roughly distributed near the minimum point of curve 82 in the first region 85 or near the maximum point of curve 82 in the second region 86. In other words, when the absolute value of the mandibular movement speed is at its maximum during the opening and closing of the mouth, the distance between the two marker points exactly corresponds to the resting vertical distance.

[0241] Figure 9 This is a schematic diagram of tapping data according to another embodiment of this disclosure. Figure 9 As shown, the horizontal axis represents the frame number, the left vertical axis represents the distance (unit: px), and the right vertical axis represents the speed (unit: mm / s). Figure 9 Bar chart 91 in the figure is a bar chart of the distance between the nasal base and the chin base, corresponding to the left vertical axis. Figure 9 Curve 92 in the graph represents the velocity Y-component curve, corresponding to the right-hand ordinate. As an example, bar 93 is one of several bars in bar chart 91 corresponding to the target timestamp; point 94 is the point on curve 92 corresponding to the timestamp of bar 93.

[0242] like Figure 9As shown, analyzing the points corresponding to the target timestamps on curve 92 during multiple consecutive teeth-tapping cycles reveals that these points are precisely near the minimum or maximum values ​​of each cycle. In other words, when the absolute value of the mandibular movement speed is at its maximum during the opening and closing of the mouth, the distance between the two marker points exactly corresponds to the resting vertical distance. Based on these findings, it can be concluded that there are correlation features in the teeth-tapping data related to the resting jaw position.

[0243] In step S606, the estimated value of the resting vertical distance is determined based on the tapping data and the aforementioned associated features.

[0244] Figure 10 This is a schematic diagram of tapping data according to yet another embodiment of this disclosure. Figure 10 As shown, the horizontal axis represents the frame number, and the vertical axis represents the velocity (unit: mm / s). Figure 10 Curve 101 in the figure is the velocity Y component curve. As an example, point 102 is the maximum value point of the first cycle of curve 101, and point 103 is the minimum value point of the first cycle of curve 101.

[0245] like Figure 10 As shown, the frame number corresponding to the maximum value point (e.g., point 102) and minimum value point (e.g., point 101) of each cycle of curve 101 is taken as the target frame number. The average value of the distance between the base of the nose and the base of the chin corresponding to the target frame number is calculated as the resting vertical distance estimate, and the resting vertical distance estimate is 668.98px. According to Figure 6 The standard value of the resting vertical distance determined by the intermittent tapping data is 675px. Therefore, the estimated value of the resting vertical distance differs from the standard value by 6px, which is about 1mm, and is within the clinically acceptable range.

[0246] The above preliminary experiments show that there are correlation features in the occlusal data related to the resting jaw position, which can be used to determine the resting vertical distance, providing strong support for the effectiveness of the occlusal vertical distance detection method proposed in this disclosure.

[0247] Figure 11 This is a schematic diagram of the structure of a bite vertical distance detection device according to an embodiment of this disclosure. Figure 11 As shown, the bite vertical distance detection device 110 includes a memory 111, a processor 112, and a bus 113 connecting different system components.

[0248] The memory 111 may include, for example, system memory, non-volatile storage media, etc. The system memory may store, for example, an operating system, application programs, a boot loader, and other programs. The system memory may include volatile storage media, such as random access memory (RAM) and / or cache memory. The non-volatile storage media may store, for example, instructions for a corresponding embodiment of at least one bite perpendicular distance detection method being executed. Non-volatile storage media include, but are not limited to, disk storage, optical storage, flash memory, etc.

[0249] The processor 112 can be implemented using a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete hardware components such as discrete gates or transistors. Accordingly, the method in any of the above embodiments can be implemented by instructions in the central processing unit (CPU) running memory to execute the corresponding steps, or by dedicated circuitry to execute the corresponding steps.

[0250] For example, processor 112 is configured to implement memory-based instruction execution as follows: Figures 1 to 3 The method involved in any of the embodiments.

[0251] Bus 113 can use any of the various bus architectures. For example, bus architectures include, but are not limited to, Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MCA) bus, and Peripheral Component Interconnect (PCI) bus.

[0252] The interfaces 114, 115, and 116 of the bite vertical distance detection device 110, as well as the memory 111 and processor 112, can be connected via bus 113. Input / output interface 114 provides a connection interface for input / output devices such as monitors, mice, and keyboards. Network interface 115 provides a connection interface for various networked devices. Storage interface 116 provides a connection interface for external storage devices such as floppy disks, USB flash drives, and SD cards.

[0253] Various aspects of this disclosure are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus, and computer program products according to embodiments of this disclosure. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations thereof, can be implemented by computer-readable program instructions.

[0254] These computer-readable program instructions are provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable device to produce a machine, such that execution of the instructions by the processor produces means for implementing the functions specified in one or more boxes of the flowchart and / or block diagram.

[0255] These computer-readable program instructions may also be stored in a computer-readable storage medium. These instructions cause a computer to work in a particular manner to produce an article of manufacture, including instructions that implement the functions specified in one or more boxes in a flowchart and / or block diagram.

[0256] This disclosure may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects.

[0257] This disclosure also provides a computer-readable storage medium storing computer instructions that, when executed by a processor, implement... Figures 1 to 3 The method involved in any of the embodiments.

[0258] This disclosure also provides a computer program product, including computer instructions, wherein the computer instructions, when executed by a processor, implement as follows: Figures 1 to 3 The method involved in any of the embodiments.

[0259] In some embodiments, the functional units described above may be implemented as general-purpose processors, programmable logic controllers (PLCs), digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or any suitable combination thereof for performing the functions described herein.

[0260] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0261] The description in this disclosure is provided for illustrative and descriptive purposes only and is not intended to be exhaustive or to limit the disclosure to its forms. Many modifications and variations will be apparent to those skilled in the art. The embodiments were chosen and described in order to better illustrate the principles and practical application of this disclosure and to enable those skilled in the art to understand this disclosure and to design various embodiments with various modifications suitable for a particular purpose.

Claims

1. A method for detecting the vertical distance of occlusion, comprising: During the continuous teeth tapping process of the target subject, the teeth tapping data of the target subject is acquired; Based on the tapping data, determine the resting vertical distance of the target body; The engagement vertical distance of the target body is determined based on the resting vertical distance and the preset resting gap.

2. The method for detecting the vertical distance of occlusion according to claim 1, wherein, The step of determining the resting vertical distance of the target subject based on the tapping data includes: The tapping data is processed using numerical analysis methods to determine the resting vertical distance of the target subject; or The tapping data is processed using a machine learning model to determine the resting vertical distance of the target subject.

3. The method for detecting the vertical distance of occlusion according to claim 2, wherein, The process of using numerical analysis methods to process the tapping data and determine the resting vertical distance of the target subject includes: The distance between the nasal base and chin base of N sampling points in the tooth tapping data is processed to obtain the velocity, acceleration, and jerk of N sampling points in the tooth tapping data, where N is the total number of sampling points in the tooth tapping data; Based on the velocity of the N sampling points, the tapping data is divided into multiple data segments; Based on the multiple data segments, determine the optimal phase corresponding to the resting jaw position; Based on the optimal phase, determine the representative value of each of the plurality of data segments; Based on the representative values ​​of all data segments, the resting vertical distance of the target subject is determined.

4. The method for detecting the vertical distance of occlusion according to claim 3, wherein, Determining the optimal phase corresponding to the resting jaw position includes: Extract the nasomental distance from the sampling point corresponding to the preset initial phase in each of the multiple data segments to obtain the nasomental distance set; Determine the deviation of the set of nasal base-mental base distances; Based on the deviation of the set of nasal base-chin base distances, the preset initial phase is iteratively optimized to obtain the optimal phase, wherein the optimal phase minimizes the deviation of the set of nasal base-chin base distances.

5. The method for detecting the vertical distance of occlusion according to claim 3, wherein, The step of determining the representative value of each of the plurality of data segments based on the optimal phase includes: The first phase window is determined based on the optimal phase and redundancy value. Calculate a first velocity threshold and a first slap threshold for each data segment; Candidate sampling points for each data segment are determined based on the first phase window, the first velocity threshold for each data segment, and the first sag threshold for each data segment. Among the candidate sampling points of each data segment, a representative sampling point for each data segment is determined; The representative value of each data segment is determined based on the representative sampling points of each data segment.

6. The method for detecting the vertical distance of occlusion according to claim 5, wherein, The calculation of the first velocity threshold and the first spurt threshold for each data segment includes: Calculate the first velocity quantile of the absolute value of the velocity of all sampling points in each data segment to obtain the first velocity threshold of each data segment; The first hop quantile is calculated by taking the absolute value of the hops of all sampling points in each data segment, and the first hop threshold of each data segment is obtained.

7. The method for detecting the vertical distance of occlusion according to claim 5, wherein, The process of determining candidate sampling points for each data segment includes: If the i-th sampling point in the k-th data segment is located within the first phase window, then it is determined whether the absolute value of the velocity of the i-th sampling point is greater than or equal to the first velocity threshold of the k-th data segment, where... K is the total number of data segments. I is the total number of sampling points in the k-th data segment; If the absolute value of the velocity of the i-th sampling point is greater than or equal to the first velocity threshold of the k-th data segment, then determine whether the absolute value of the jog of the i-th sampling point is greater than or equal to the first jog threshold of the k-th data segment. If the absolute value of the jump of the i-th sampling point is greater than or equal to the first jump threshold of the k-th data segment, then it is determined whether the absolute value of the jump of the i-th sampling point is the maximum value in the neighborhood of the i-th sampling point. If the absolute value of the jerk of the i-th sampling point is the maximum value in the neighborhood of the i-th sampling point, then the i-th sampling point is selected as a candidate sampling point.

8. The method for detecting the vertical distance of occlusion according to claim 5, wherein, The process of determining candidate sampling points for each data segment further includes: Based on the first phase window, the first velocity threshold of the kth data segment, and the first sag threshold of the kth data segment, candidate sampling points for the kth data segment are determined, wherein... K is the total number of data segments; If the number of candidate sampling points for the k-th data segment is 0, then adjust at least one of the phase window, velocity threshold, and jog threshold until the number of candidate sampling points for the k-th data segment is greater than 0.

9. The method for detecting the vertical distance of occlusion according to claim 8, wherein, At least one of the adjusted phase window, velocity threshold, and jolt threshold includes: Increase the first phase window to obtain the second phase window; Decrease the first speed threshold to obtain the second speed threshold; Decrease the first hop threshold to obtain the second hop threshold.

10. The method for detecting the vertical distance of occlusion according to claim 5, wherein, Determining the representative sampling point for each data segment from the candidate sampling points for each data segment includes: The score of each candidate sampling point is determined based on the velocity, jerk, and phase of each candidate sampling point in the k-th data segment, where... K is the total number of data segments; The candidate sampling point with the highest score among the candidate sampling points of the k-th data segment is taken as the representative sampling point of the k-th data segment.

11. The method for detecting the vertical distance of occlusion according to claim 10, wherein, Determining the score for each candidate sampling point includes: The relative velocity of each candidate sampling point is obtained by calculating the ratio of the absolute value of the velocity of each candidate sampling point to the first velocity threshold of the k-th data segment. The relative judder of each candidate sampling point is obtained by calculating the ratio of the absolute value of the judder of each candidate sampling point to the first judder threshold of the k-th data segment. Calculate the difference between the phase of each candidate sampling point and the optimal phase to obtain the phase deviation of each candidate sampling point; The score for each candidate sampling point is obtained based on its relative velocity, relative jerk, and phase deviation.

12. The method for detecting the vertical distance of occlusion according to claim 3, wherein, The step of dividing the tapping data into multiple data segments based on the velocity of the N sampling points includes: Determine whether the velocity direction of the nth sampling point is the same as the velocity direction of the (n-1)th sampling point, where... ; If the velocity direction of the nth sampling point is not the same as the velocity direction of the (n-1)th sampling point, then it is determined that the velocity direction from the nth sampling point to the (n-1)th sampling point... Are the velocity directions of the sampling points the same? The hysteresis length; If from the nth sampling point to the nth sampling point If the velocity directions of all sampling points are the same, then the nth sampling point is taken as the segment boundary. The tapping data is divided according to the segmentation boundaries to obtain the multiple data segments.

13. The method for detecting the vertical distance of occlusion according to claim 12, wherein, The step of dividing the tapping data according to the segment boundaries to obtain the multiple data segments includes: Based on the segmentation boundaries, the tapping data is divided into multiple pre-selected segments; Calculate the amplitude ratio of each preselected segment relative to the reference amplitude among the plurality of preselected segments; Calculate the percentage of target sampling points for each pre-selected segment; The plurality of data segments are determined from the plurality of preselected segments based on at least one of the amplitude ratio of each preselected segment to the reference amplitude and the proportion of target sampling points of each preselected segment.

14. The method for detecting the vertical distance of occlusion according to claim 13, wherein, The calculation of the amplitude ratio of each preselected segment relative to the reference amplitude in the plurality of preselected segments includes: Calculate the amplitude of the nasal-mental distance at all sampling points in each pre-selected segment to obtain the amplitude of each pre-selected segment; The median amplitude of all pre-selected segments is used as the reference amplitude; Calculate the ratio of the amplitude of each preselected segment to the reference amplitude to obtain the amplitude ratio of each preselected segment relative to the reference amplitude.

15. The method for detecting the vertical distance of occlusion according to claim 13, wherein, The calculation of the target sampling point ratio for each preselected segment includes: Based on the absolute values ​​of the jumps at N sampling points in the tapping data, a jump reference value is determined; If the absolute value of the jerk at the nth sampling point is greater than the jerk reference value, then the nth sampling point is taken as the target sampling point; Calculate the ratio of the number of target sampling points in each pre-selected segment to the total number of sampling points in each pre-selected segment to obtain the target sampling point ratio of each pre-selected segment.

16. The method for detecting the vertical distance of occlusion according to claim 13, wherein, The step of determining multiple data segments from the multiple pre-selected segments includes: If the amplitude ratio of the preselected segment to the reference amplitude is within the preset amplitude ratio range, then the preselected segment is determined as a data segment; If the proportion of target sampling points in the pre-selected segment is less than or equal to a preset proportion threshold, then the pre-selected segment is determined as a data segment.

17. The method for detecting the vertical distance of occlusion according to claim 2, wherein, The methods for training the machine learning model include: During the continuous teeth tapping process of each subject, teeth tapping data samples were acquired for each subject; During the intermittent teeth tapping process performed by each subject under the guidance of a doctor, intermittent teeth tapping data of each subject was acquired. Based on the intermittent teeth tapping data of each subject, determine the resting vertical distance label corresponding to the teeth tapping data sample of each subject; A dataset was constructed using the teeth tapping data samples of all subjects and the corresponding resting vertical distance labels; The machine learning model is trained using the dataset.

18. A device for detecting the vertical distance of occlusion, comprising: Memory; A processor, coupled to a memory, is configured to execute instructions stored in the memory to implement the bite vertical distance detection method as described in any one of claims 1-17.

19. A computer-readable storage medium, wherein, The computer-readable storage medium stores computer instructions that, when executed by a processor, implement the bite vertical distance detection method as described in any one of claims 1-17.

20. A computer program product comprising computer instructions, wherein the computer instructions, when executed by a processor, implement the bite vertical distance detection method as described in any one of claims 1-17.

Images