A bite pad design method and system based on bite force distribution characteristics

Abstract

The application discloses a kind of based on occlusal force distribution characteristics and system of occlusal pad design method, belong to stomatology and biomechanics cross technical field.This method includes: the occlusal force distribution time series of subject from muscle fresh state to fatigue state and masticatory muscle group electromyogram are simultaneously acquired;Establish the statics equation set between muscle force, dentition contact force and joint counterforce;Through the muscle fatigue attenuation function of each muscle group is fitted by electromyogram time-frequency analysis, the joint counterforce of each time in entire fatigue process is calculated in combination with mechanical contact model;With the minimization of maximum value of joint load in fatigue process time domain as objective function, min-max robust optimization is carried out;Occlusal pad geometry generation, the thickness and material elastic modulus distribution of optimal stiffness distribution is reversely converted.The present application solves the technical problem that existing occlusal pad design cannot continuously protect temporomandibular joint in masticatory muscle fatigue process.

Description

Technical Field

[0001] This invention relates to the interdisciplinary field of oral medicine and biomechanics, specifically to a method and system for designing an occlusal pad based on the characteristics of occlusal force distribution. Background Technology

[0002] Occlusal splints are essential instruments in oral clinical practice for treating temporomandibular joint disorder, bruxism, and occlusal trauma. Their basic principle involves placing an elastic body with a specific geometry between the upper and lower teeth to alter the contact relationship and force distribution pattern between the teeth, thereby reducing joint load, protecting tooth structure, and alleviating masticatory muscle symptoms. With the development of digital dental technology, the design of occlusal splints has evolved from the traditional manual wax-imprinting and casting process to computer-aided design and manufacturing, significantly improving design accuracy and manufacturing efficiency.

[0003] Currently, the digital design method for occlusal pads typically includes the following steps: First, a three-dimensional digital model of the upper and lower dentition is obtained using an intraoral scanner; then, the thickness, occlusal surface morphology, and guide ramp parameters of the occlusal pad are set in CAD software based on clinical experience; some advanced methods also incorporate occlusal force detection devices to obtain information on the distribution of contact forces on the tooth surfaces, using this as a reference to guide adjustments to the design parameters, making the occlusal forces at each tooth position more uniform. These methods have achieved positive results in improving the matching accuracy between the occlusal pad and the individual dental arch morphology.

[0004] However, the aforementioned existing technologies have a fundamental technical flaw: for patients with temporomandibular joint disorder complicated by nocturnal bruxism, the existing design methods fail to address the problem of time-varying uncontrolled joint load during the masticatory muscle fatigue process. Specifically, this problem involves three levels of technical reasons. First, a misalignment at the optimization target level: existing methods aim to homogenize tooth surface contact forces, but uniform tooth surface force does not equate to minimizing joint load—the contact forces at different tooth positions contribute different amounts of torque to the joint through their respective torque arms. When the torque arms differ significantly, a uniform distribution of tooth surface force may actually lead to a larger resultant torque on the joint. Second, the lack of a temporal dimension: existing methods are designed based on the distribution of occlusal force at a specific moment, representing static optimization at a single instant. However, during prolonged and repetitive teeth grinding in patients with nocturnal bruxism, the masticatory muscles undergo a gradual change from a fresh state to a state of deep fatigue. Different muscle groups exhibit varying fatigue rates due to differences in the ratio of fast to slow muscle fibers, leading to continuous changes in the distribution of muscle force among different muscle groups. Consequently, the distribution of dental contact force and joint load exhibits unpredictable time-varying fluctuations. Third, the coupling effect of the first two levels: even if a statically designed occlusal pad initially keeps the joint load within a safe range, the joint load may exceed the safe threshold at a certain fatigue stage due to the dynamic reorganization of muscle force distribution. Current technology cannot predict or prevent this deterioration process. For patients with temporomandibular joint disorder whose joint structure is already in a fragile balance of damage and repair, this uncontrollable time-varying fluctuation of joint load may exacerbate disc displacement and retrodiscal tissue damage, severely affecting treatment outcomes. Summary of the Invention

[0005] Technical objective: To address the technical problem that existing bite pad designs cannot continuously protect vulnerable joint structures during masticatory muscle fatigue, this invention discloses a bite pad design method and system based on bite force distribution characteristics.

[0006] Technical solution: To achieve the above technical objectives, the present invention adopts the following technical solution:

[0007] A method for designing an occlusal pad based on occlusal force distribution characteristics includes the following steps:

[0008] S1, Multi-state occlusal force and electromyography combined acquisition: Subjects are subjected to a continuous occlusal task. The time series F(x,y,t) of the occlusal force distribution of the entire dental arch is simultaneously acquired through an intraoral occlusal force sensor array. The electromyography signals EMG(t) of each masticatory muscle group are simultaneously acquired through a bilateral masticatory muscle surface electromyography acquisition device. The duration of the continuous occlusal task covers the complete fatigue process from the fresh state of the muscles to the appearance of changes in electromyographic fatigue indicators.

[0009] S2, Construction of an individualized mandibular musculoskeletal model: Based on the craniofacial imaging data of the subjects, a three-dimensional model of the mandible is reconstructed, the positions of the hinge axes of the bilateral temporomandibular joints and the coordinates of the origin and insertion points and the direction of force lines of each masticatory muscle group are calibrated, and the results of the measurement of the maximum voluntary contraction force of each muscle group are combined to establish a set of static equilibrium equations for the mandible. The set of static equilibrium equations establishes the torque balance relationship between muscle force, dental contact force and joint reaction force.

[0010] S3, Time-varying prediction of joint load in the fatigue time domain: Time-frequency analysis of electromyography signals is performed to extract electromyography fatigue characteristic parameters of each muscle group, and the muscle force-fatigue decay function of each muscle group is fitted to obtain the time-varying muscle force vector M(t); the time-varying muscle force vector M(t) is substituted into the static equilibrium equation system, and the expected tooth force distribution at each moment is calculated by combining the occlusal pad stiffness distribution to be optimized through the mechanical contact model, and then the joint reaction force at each moment in the entire fatigue process is calculated to generate the joint load time curve R(t).

[0011] S4, Time-domain robust optimization of joint load: Taking the local stiffness value at each tooth position of the occlusal pad as the decision variable and minimizing the maximum value of the joint load in the entire fatigue process time domain as the objective function, robust optimization is performed under constraints to obtain the optimal stiffness distribution.

[0012] S5, Occlusal pad geometry generation: The optimal stiffness distribution is inversely converted into the thickness value and material elastic modulus value of the occlusal pad at each tooth position through the contact mechanics model, and a three-dimensional geometric model of the occlusal pad is generated by combining the subject's three-dimensional dental arch scan data.

[0013] On the other hand, the present invention provides an occlusal pad design system based on occlusal force distribution characteristics, including: a multi-state occlusal force and electromyography combined acquisition module, an individualized mandibular muscle and bone biomechanics model construction module, a fatigue time-domain joint load time-varying prediction module, a joint load time-domain robust optimization module, and an occlusal pad geometry generation module.

[0014] Beneficial Effects: The occlusal pad design method and system based on occlusal force distribution characteristics provided by this invention have the following beneficial effects:

[0015] 1) This invention elevates the design optimization target of the occlusal pad from the traditional uniformity of tooth surface contact force to minimizing the temporomandibular joint load. Through mandibular static inversion calculation, the force distribution that should be achieved on the dentition is deduced from the joint load target, which solves the problem of target misalignment in the prior art where tooth surface force is uniform but the joint is not necessarily safe.

[0016] 2) This invention introduces fatigue process modeling of masticatory muscle groups, establishes muscle force-fatigue decay function of each muscle group through time-frequency analysis of surface electromyography signals, incorporates time-varying muscle force into mandibular biomechanics model, realizes quantitative prediction of time-varying trajectory of joint load in fatigue process, and fills the gap in the time dimension of existing technology.

[0017] 3) This invention constructs a min-max robust optimization framework, with the minimization of the maximum joint load in the entire time domain of the fatigue process as the objective function, ensuring that the designed occlusal pad can control the joint load within the safe threshold at any time from the fresh state to the deep fatigue state, thus realizing joint protection in the entire time domain.

[0018] 4) This invention identifies the most sensitive muscle groups that are most affected by joint load fluctuations through fatigue-joint load cross-sensitivity analysis, and further improves the robustness of the design by setting additional stiffness buffers in the corresponding tooth positions. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below.

[0020] Figure 1 This is a schematic diagram of the overall process of the bite pad design method based on bite force distribution characteristics provided in an embodiment of the present invention.

[0021] Figure 2 This is a schematic diagram of a multi-state bite force and electromyography combined acquisition scheme provided in an embodiment of the present invention.

[0022] Figure 3 This is a schematic diagram illustrating the construction of an individualized mandibular muscle and bone biomechanics model provided in an embodiment of the present invention.

[0023] Figure 4 This is a schematic diagram of the muscle strength-fatigue decay function fitting provided in an embodiment of the present invention.

[0024] Figure 5 This is a schematic diagram comparing joint load-time curves provided in an embodiment of the present invention.

[0025] Figure 6 This is a schematic diagram of the joint load time-domain robust optimization process provided in an embodiment of the present invention.

[0026] Figure 7 This is a schematic diagram of the mapping between the thickness distribution and stiffness distribution of the bite pad provided in an embodiment of the present invention.

[0027] Figure 8 This is a schematic diagram of the module structure of the bite pad design system based on bite force distribution characteristics provided in an embodiment of the present invention. Detailed Implementation

[0028] The present invention will now be described more clearly and completely by way of a preferred embodiment in conjunction with the accompanying drawings, but this does not limit the invention to the scope of the described embodiment.

[0029] Before describing each step in detail, it is necessary to explain the inventiveness of the present invention compared to the prior art, and why those skilled in the art cannot directly obtain the technical inspiration of the present invention from the prior art.

[0030] In existing technologies, the technical approaches in the field of occlusal pad design can be summarized into two main directions: one is the "force distribution detection-uniformity adjustment" approach, represented by commercial occlusal force testing systems such as T-Scan, whose optimization goal is fixed on the uniformity of tooth surface contact forces; the other is the "stress simulation-structural optimization" approach, which uses finite element analysis as a tool, and typically performs stress analysis and shape optimization on the occlusal pad structure under given load conditions. The former's technical bias lies in equating "uniform tooth surface forces" with "optimal design," without questioning the mechanical correctness of this equation; the latter, although introducing finite element analysis, still takes load conditions from static measurements at a single moment, and the optimization goal is usually the stress distribution of the occlusal pad itself rather than joint loads.

[0031] The inventive contribution of this invention is reflected in the following three levels of systematic breakthroughs. These three levels are interdependent and indispensable, forming an organic whole:

[0032] The first level involves a leap in the dimensionality of the optimization objective—from the tooth surface force space to the joint force space. This invention recognizes that the distribution of tooth surface contact forces and joint loads are not simply linearly positively correlated, but rather form a nonlinear mapping through a torque arm relationship. This invention shifts the optimization objective from uniformizing tooth surface forces to minimizing joint reaction forces, using static inversion calculations to inversely deduce the dentition force distribution from the joint target load. This transformation is not merely a simple replacement of the objective function, but requires the establishment of a complete, individualized mandibular musculoskeletal biomechanics model, including a series of interrelated technical steps such as joint hinge axis positioning, multi-muscle group force line calibration, and the construction of a torque balance equation set. The long-held technical paradigm in the art is that "uniform occlusal force equals good occlusion." Elevating the optimization objective from the tooth surface to the joint level requires breaking free from this paradigm, which is neither common knowledge nor a conventional technical choice in the field.

[0033] The second aspect involves the introduction of the time dimension—from static snapshots to the fatigue time domain. This invention introduces the concept of masticatory muscle fatigue progression into occlusal pad design. This technological inspiration originates from the fields of exercise physiology and fatigue mechanics, rather than prosthodontics. Specifically, this invention utilizes time-frequency analysis of surface electromyography signals (short-time Fourier transform to extract the median frequency decay rate) as a quantitative indicator of muscle fatigue progression, establishing muscle force-fatigue decay functions for each muscle group, thus extending the problem of oral occlusal mechanics from a time-independent static problem to a time-dependent dynamic problem. This technological transfer transcends the disciplines of prosthodontics and exercise physiology and is not a commonly used technique in the field of prosthodontics.

[0034] The third level involves the construction of a robust optimization framework—from single-condition optimization to full-time-domain robustness. This invention employs a min-max robust optimization framework to handle muscle force uncertainties during the fatigue process. The framework's technical origin lies in robust optimization theory within structural reliability engineering, and its application to occlusal pad design is unprecedented in this field. Traditional occlusal pad design, even with the introduction of finite element optimization, only achieves deterministic optimization under a single load condition, lacking the ability to handle load uncertainties. The robust optimization framework ensures that the design result is not optimal at a "best moment" but fails at a "worst moment," instead satisfying joint safety constraints even under the worst-case conditions across the entire time domain.

[0035] There is a close technical dependency among the breakthroughs at these three levels: without the joint load inversion at the first level, the robust optimization at the third level would be performed in an incorrect target space; without the fatigue time-domain modeling at the second level, the multi-condition framework at the third level would lack time-varying load input; and without the robust optimization at the third level, the technical contributions at the first and second levels could only achieve joint optimization at a single fatigue moment, failing to guarantee full-process protection. The organic combination of these three aspects produces a synergistic effect that surpasses the sum of the individual technologies, and this synergistic effect is precisely the core inventiveness of this invention.

[0036] The steps of the present invention will be described in detail below with reference to the accompanying drawings.

[0037] This invention discloses a method for designing an occlusal pad based on the characteristics of occlusal force distribution, specifically including the following steps:

[0038] Step S1: Combined acquisition of multi-state occlusal force and electromyography

[0039] like Figure 2 As shown, the purpose of step S1 is to obtain the complete mechanical information evolution sequence of the subject from the fresh state to the fatigue state of the muscle, as the data basis for all subsequent analysis and optimization.

[0040] The technical principle behind this step is that the electromyographic (EMG) signals on the surface of the masticatory muscles can reflect the muscle fatigue state in real time. When muscle fibers are fatigued, the discharge frequency of the motor units decreases and the degree of synchronization increases, which is manifested as a shift in the median frequency of the EMG signal power spectrum towards lower frequencies and an initial increase followed by a decrease in the root mean square amplitude. Therefore, EMG signals can serve as a quantitative indicator of the muscle fatigue process. Combined with synchronously acquired bite force distribution data, a temporal correlation between "muscle fatigue state - force distribution pattern" can be established.

[0041] In practical implementation, the data acquisition device includes the following components:

[0042] Intraoral occlusal force sensor array: A piezoresistive thin-film sensor array with a thickness of no more than 0.3 mm is used to avoid significant interference from the sensor's own thickness on the occlusal vertical distance and force distribution. The sensor array covers the entire dental arch, with a spatial resolution of no less than 200 sensing points, ensuring the ability to distinguish the independent force contribution of adjacent cusps in the presence of multiple canines in the molar region.

[0043] Bilateral masticatory muscle surface electromyography (EMG) acquisition device: An eight-channel configuration is used, with acquisition channels located on the anterior bundle of the masseter, anterior bundle of the temporalis, posterior bundle of the temporalis, and medial pterygoid muscle bilaterally. The selection criteria for these four muscle groups are as follows: the masseter is the main muscle group generating closing force, primarily composed of fast-twitch fibers (type II fibers), and has a rapid fatigue rate; the anterior and posterior bundles of the temporalis participate in vertical and retractive closing movements respectively, primarily composed of slow-twitch fibers (type I fibers), and have a slow fatigue rate; the medial pterygoid muscle participates in lateral and protraction movements, is small in size, and has a long lever arm, making it one of the earliest muscle groups to show fatigue. By covering representative muscle groups that exhibit both rapid and slow fatigue, the dynamic reorganization process of muscle force distribution during the fatigue process can be comprehensively characterized.

[0044] Synchronization timestamp controller: Used to accurately align force signals and electromyographic (EMG) signals in time, with a sampling rate of no less than 1000Hz. The importance of time alignment lies in the fact that subsequent step S3 needs to correlate and model the EMG fatigue index at a specific moment with the force distribution pattern at the same moment. If there is a time deviation between the two signals, it will lead to errors in the correlation model.

[0045] Data preprocessing unit: Performs bandpass filtering (force signal: 0.1-50Hz; electromyography signal: 20-500Hz), baseline drift correction, and motion artifact removal on the acquired raw signals to generate a clean signal sequence that can be used for subsequent analysis.

[0046] During the data collection process, subjects performed a standardized continuous clenching and releasing task. In one specific embodiment, the task was designed as follows: subjects performed repeated clench-and-relax cycles at a frequency of 1.5 Hz (guided by a metronome) at 50% of their maximum clenching force until at least one of the following fatigue indicators was observed: the median frequency of any muscle group decreased by more than 15% of its initial value, or the root mean square amplitude of any muscle group reached an inflection point (from increasing to decreasing). The typical data collection duration was 15-20 minutes.

[0047] The reason for choosing 50% of the maximum closing force level is that this level can induce detectable muscle fatigue within a reasonable time (fatigue occurs too quickly and the force distribution changes insufficiently at the 100% level; significant fatigue may take several hours at the 20% level) and simulate the typical force level of nighttime teeth grinding—studies have shown that the average force value of nighttime teeth grinding is about 40-60% of the maximum closing force.

[0048] Step S2: Construction of an individualized mandibular muscle and bone biomechanics model

[0049] like Figure 3 As shown, the purpose of step S2 is to establish a quantitative mechanical model that can link muscle force input, dental contact force distribution and joint reaction force, serving as the core engine for subsequent joint load calculation and optimization design.

[0050] The technical principle behind constructing an individualized mandibular musculoskeletal biomechanics model is classical rigid body statics. The mandible is approximately in static equilibrium during occlusion (ignoring inertial forces due to the extremely slow closing motion). External forces acting on the mandible include: the muscle force vectors of each masticatory muscle group, the contact force distribution on the dental contact surfaces, and the reaction force vectors of the bilateral temporomandibular joints. Based on the rigid body statics equilibrium condition, where the net force is zero and the net torque about any point is zero, an analytical relationship between muscle forces, dental forces, and joint forces can be established.

[0051] The specific implementation process of model construction includes the following sub-steps:

[0052] S2.1, Mandibular 3D Reconstruction: Based on the subject's cone-beam computed tomography (CBCT) image data, a threshold segmentation algorithm is used to extract bone tissue from the surrounding soft tissue. After manual correction, a 3D surface mesh model of the mandible is generated. The reconstruction accuracy depends on the voxel resolution of the CBCT images, typically using CBCT images with a resolution of at least 0.2 mm.

[0053] S2.2, Hinge Axis Positioning: The geometric center points of the bilateral condyles are marked on the reconstructed mandibular model. The direction of the line connecting these two points is the hinge axis direction, and the midpoint between them is the reference origin of the hinge axis. The accuracy of hinge axis positioning directly affects the accuracy of moment arm calculation, and consequently, the accuracy of joint reaction force prediction. In a preferred embodiment, kinematic methods can be used for auxiliary correction: the mandibular movement trajectory during small opening and closing movements of the subject is recorded, and the instantaneous rotation center is fitted to improve the hinge axis positioning accuracy.

[0054] S2.3, Masturbatory Muscle Force Line Calibration: Based on identifiable anatomical landmarks and muscle attachment areas in CBCT images, and combined with prior anatomical knowledge, the origin coordinates (fixed end of the skull) and insertion coordinates (movable end of the mandible) of each masticatory muscle group are calibrated. The direction of the line connecting the two points is the force line direction vector of that muscle group. Individual differences exist in the origin and insertion positions of each muscle group; individualized calibration based on CBCT images significantly improves model accuracy compared to using population averages. The calibrated muscle groups include: bilateral anterior masseter muscles (origin: lower border of the zygomatic arch; insertion: lateral surface of the angle of the mandible), bilateral anterior temporalis muscles (origin: anterior part of the temporal fossa; insertion: anterior border of the coronoid process), bilateral posterior temporalis muscles (origin: posterior part of the temporal fossa; insertion: posterior border of the coronoid process), and bilateral medial pterygoid muscles (origin: medial surface of the lateral pterygoid plate; insertion: pterygoid tuberosity on the medial surface of the angle of the mandible).

[0055] S2.4, Calibration of Maximum Voluntary Contractile Force for Each Muscle Group: Subjects perform maximum voluntary clenching, and the surface electromyography (RMS) amplitude of each muscle group is recorded simultaneously. The maximum voluntary contractile force F of each muscle group is estimated by multiplying its RMS amplitude by the cross-sectional area and muscle specific strength coefficient (skeletal muscle specific strength is approximately 0.3-0.4 MPa). j0 The cross-sectional area can be measured from a section of the CBCT image perpendicular to the force line.

[0056] S2.5, Construction of Static Equilibrium Equations: The mandible is modeled as a rigid body, and the temporomandibular joint on one side is selected as the torque reference point to establish a three-dimensional torque equilibrium equation system. Let F be the force vector of the j-th muscle. j The position vector from the point of application to the reference point is r. j The force at the contact point of the i-th tooth row is f. i The position vector is p i The reaction force of the contralateral joint is R. contra The position vector is d contra The torque balance equation is then:

[0057]

[0058] Where × denotes the cross product operation. Combined force equilibrium equations ( (For ipsilateral joint reaction force), the reaction force vector of both joints can be uniquely determined under the given muscle force vector set and dental force distribution.

[0059] The reason for choosing to model the mandible as a rigid body rather than an elastic body is that the mandible is a dense bone tissue with an elastic modulus of approximately 15 GPa. Its deformation under occlusal force is extremely small (on the order of micrometers), and its impact on the torque balance is negligible. Simplifying the mandible as a rigid body reduces the mechanical model from partial differential equations to a system of algebraic equations, significantly reducing computational load and allowing subsequent multi-condition optimization to be completed within a reasonable timeframe.

[0060] Step S3: Time-varying prediction of fatigue time-domain joint load

[0061] like Figure 4 , Figure 5 As shown, step S3 is one of the core innovative steps that distinguishes this invention from the prior art. Its purpose is to establish the ability to predict joint loads at each moment during the fatigue process of the masticatory muscle.

[0062] Step S3 includes the following four sub-steps:

[0063] S3.1, Electromyographic fatigue feature extraction: Extracting electromyographic (EMG) signals from each muscle group acquired in step S1. j(t) Perform a short-time Fourier transform with a window length of 1 second and an overlap rate of 50% between adjacent windows. Extract two key metrics within each time window: Median Frequency (MDF). j (t) and root mean square magnitude RMS j (t).

[0064] The median frequency is the frequency value that bisects the power spectrum of an electromyographic (EMG) signal in the frequency domain. When muscle fibers become fatigued, the firing frequency of motor units decreases and the conduction velocity of action potentials slows down, causing the power spectrum to shift to lower frequencies overall, manifested as a monotonically decreasing median frequency. The rate of decrease in median frequency is positively correlated with the rate of muscle fatigue development and is unaffected by fluctuations in the amplitude of the EMG signal, making it the most robust EMG fatigue indicator.

[0065] The root mean square (RMS) amplitude reflects the overall energy level of the electromyographic (EMG) signal and is related to the number of active motor units and the synchronicity of their discharge. In the early stages of fatigue, the body compensates for the decline in force output by increasing the recruitment of motor units, resulting in an increase in the RMS amplitude. When fatigue reaches a certain level, the recruitable motor units approach saturation, and the RMS begins to decline. The inflection point of the RMS marks the critical state of compensation failure.

[0066] Median frequency MDF j (t) is fitted with a linear regression on time t, with a slope β j This represents the median rate of decrease in frequency for that muscle group. β j The larger the absolute value, the faster the muscle group fatigues.

[0067] S3.2, Fitting the Muscle Strength-Fatigue Decline Function: Based on the above electromyographic fatigue indices, a decay function of the maximum output force of each muscle group over time is established. For the j-th muscle, a modified exponential decay model is used:

[0068]

[0069] in, The initial maximum force value of the j-th muscle (defined by step S2.4). Let be the fatigue saturation coefficient of the j-th muscle. Let be the fatigue rate constant of the j-th muscle, and t be time.

[0070] The physical meaning of the above model is: muscle strength from the initial value It decays exponentially and eventually approaches zero. The steady-state value is maintained rather than decaying to zero. This model form is chosen based on the following physiological basis: skeletal muscle is composed of a mixture of fast-twitch and slow-twitch fibers. Fast-twitch fibers have high output but are easily fatigued, while slow-twitch fibers have low output but are more resistant to fatigue. During fatigue, fast-twitch fibers fail first, while slow-twitch fibers can maintain output for a longer period. Therefore, muscle strength eventually stabilizes at the baseline level contributed by slow-twitch fibers. Parameters This reflects the minimum proportion of sustainable output force determined by the percentage of slow-twitch muscle fibers in this muscle group.

[0071] parameter The values ​​are determined based on the fiber composition characteristics of each muscle group: the masseter muscle is predominantly composed of fast-twitch fibers, with an α value of approximately 0.3-0.5; the temporalis muscle is predominantly composed of slow-twitch fibers, with an α value of approximately 0.6-0.8; and the medial pterygoid muscle has a mixed fiber type, with an α value of approximately 0.4-0.6. These parameter ranges are based on existing muscle histology research literature and can be further corrected for individual patients using sEMG data.

[0072] parameter The method for determining it is: using the median frequency decreasing slope β j With fatigue rate constant The empirical relationship between them is converted. Existing research has shown that there is an approximately linear relationship between the rate of decline of median frequency and the rate of muscle strength decline, that is... , where γ is the conversion factor, which can be obtained by jointly calibrating the measured force-electromyography data of the subjects.

[0073] S3.3, Time-varying muscle force-time-varying dental force joint derivation: The key innovation of this sub-step is to introduce the occlusal pad stiffness distribution as a variable into the mechanical contact model.

[0074] At any time t, the time-varying muscle force vector Determined by the attenuation function in step S3.2. Given the occlusal pad stiffness distribution. Under these conditions, the contact force at each tooth position can be calculated using the following mechanical contact model:

[0075] The contact force at each tooth position is related to the total contact stiffness at that location and the occlusal interdental space. The total contact stiffness is contributed in series by three parts: periodontal ligament stiffness, periodontal ligament stiffness, and occlusal interdental space. (Depending on the periodontal support area and the elastic modulus of the periodontal ligament), stiffness of the hard tissues of the tooth. (The elastic modulus of tooth enamel is approximately 84 GPa, and its stiffness is extremely high, so it can be considered rigid) and the local stiffness of the occlusal pad. (Determined by the elastic modulus and local thickness of the occlusal pad material). Since the stiffness of the hard dental tissue is much greater than that of the periodontal ligament and the occlusal pad, it can be simplified to a series reaction of the periodontal ligament stiffness and the occlusal pad stiffness:

[0076]

[0077] Contact force at each tooth position This is determined as follows: Under a given resultant muscle force, the mandible moves along the closing direction until the resultant force of the dental contact forces balances the resultant muscle force. The contact force at each tooth position is proportional to its total contact stiffness, i.e.:

[0078]

[0079] in The contact deformation at the i-th tooth position depends on the closing displacement of the mandible and the initial gap at that location. The contact force distribution at each tooth position can be determined by iteratively solving for the equilibrium displacement of the mandible. This distribution represents the time-varying muscle force M(t) and occlusal pad stiffness distribution. The common function.

[0080] The reason for using a series stiffness model instead of an independent stiffness model is that the transmission chain of dental contact forces is "occlusal pad → tooth surface → tooth body → periodontal ligament → alveolar bone." The deformation of each link is sequential, and the total deformation is the sum of the deformations of each link. Therefore, the total stiffness is the series value of the stiffness of each link. The physical meaning of series stiffness is that the total stiffness is dominated by the softest link—when the stiffness of the occlusal pad is much smaller than the stiffness of the periodontal ligament, the total stiffness is approximately equal to the stiffness of the occlusal pad, and the occlusal pad has the greatest regulatory effect on force distribution; when the stiffness of the occlusal pad is much larger than the stiffness of the periodontal ligament, the total stiffness is approximately equal to the stiffness of the periodontal ligament, and the regulatory effect of the occlusal pad on force distribution degenerates to zero.

[0081] S3.4, Joint load time series calculation: Calculate the tooth force distribution at each time step obtained in step S3.3. Substituting the time-varying muscle force M(t) from step S3.2 into the static equilibrium equations established in step S2.5, we can solve for the bilateral joint reaction force vector R at each time step. left (t) and R right (t). The joint reaction force is decomposed into a compression component (the component perpendicular to the joint surface) and a shear component (the component parallel to the joint surface), and their time series are recorded respectively.

[0082] The final output is the joint load-time curve R(t), which contains the complete variation trajectory of compressive and shear loads on both joints throughout the fatigue process [0,T]. This curve represents the stiffness distribution K of the occlusal pad. splint The function is that different stiffness distributions will produce different joint load-time curves.

[0083] Step S4: Time-domain robust optimization of joint load

[0084] like Figure 6 As shown, step S4 is another core innovative step of the present invention, the purpose of which is to find a occlusal pad stiffness distribution that minimizes the maximum value of joint load throughout the fatigue process.

[0085] This step employs a min-max robust optimization framework. The reason for choosing the min-max framework instead of the traditional weighted summation objective function is that the weighted summation method optimizes the weighted average of multiple load conditions, which may result in a low average value but an extremely high value for a particular load condition. Furthermore, the damage mechanism of joint tissues is dominated by peak loads, not average loads. The min-max framework directly optimizes the worst-case scenario, ensuring that the joint load at any fatigue moment does not exceed the safe range, which perfectly matches the clinical needs for joint protection.

[0086] The specific mathematical form of the optimization problem is as follows:

[0087] Decision variable: Local stiffness of the occlusal pad at each tooth position control point , where N is the number of discretized tooth position control points. In a specific embodiment, N is between 28 and 32, meaning that each tooth corresponds to one control point.

[0088] Objective function: By introducing an auxiliary variable η, the original min-max problem is transformed into an equivalent constrained optimization problem.

[0089]

[0090] in This represents the comprehensive index of joint load at the k-th time slice.

[0091] In the above formula, and , , are the L2 norms of the reaction force vectors of the left and right temporomandibular joints at the k-th time slice, respectively, and represent the total compressive load borne by the joint; This represents the maximum value of the component of the joint reaction force parallel to the joint surface, and signifies the shear load.

[0092] Weighting coefficient , , The setting rules are as follows: For patients with temporomandibular joint disorder, the weight of the affected joint (disc displacement side) is set to 1.5 to 2.5 times the weight of the healthy joint to prioritize the protection of the damaged joint; the weight of the shear component is set as follows. The weighting is set to 1.2 to 2.0 times that of the compressive component. The physiological basis for this setting is that the shear strength of articular disc fibrocartilage is significantly lower than its compressive strength, making shear loads more harmful to patients with TMD. In one specific embodiment, it is set... =1.0 (unaffected side), =2.0 (affected side), =1.5.

[0093] Time slice selection: The fatigue process [0, T] is uniformly divided into M time slices. The value of M needs to balance calculation accuracy and calculation efficiency. In a specific embodiment, M is taken as 10 to 20. Too few time slices may miss the load peak, while too many time slices will increase the amount of calculation.

[0094] The constraints include the following five items:

[0095] Constraint C1, Non-negativity of Force: The contact force at all tooth positions at all time slices is non-negative, i.e. For all i and k, the physical meaning of this constraint is that the contact force between the teeth can only be compressive, not tensile.

[0096] Constraint C2, anterior tooth guiding function: The guiding force of the incisal guide ramp in the anterior tooth area during maximum lateral movement is not lower than a set threshold. This constraint ensures that the anterior tooth guiding function of the occlusal pad during lateral movement does not fail due to stiffness optimization. In one specific embodiment, Set it to 10N.

[0097] Constraint C3, Vertical Distance Limit: The thickness of the occlusal pad at any tooth position shall not exceed the set upper limit value h. max This constraint ensures wearing comfort and lip closure. In one specific embodiment, h max Set to 5mm.

[0098] Constraint C4, Stiffness Physical Feasibility: The local stiffness value at each tooth position is within the stiffness range corresponding to the elastic modulus of the available material. In one specific embodiment, the elastic modulus of the available material ranges from 20 MPa to 3 GPa, corresponding to soft EVA and hard PMMA, respectively.

[0099] Constraint C5, static consistency: The force system at each time slice satisfies the mandibular static equilibrium equation, that is, the force system composed of muscle force, dental force and joint force simultaneously satisfies the conditions of force balance and torque balance.

[0100] Solution Algorithm: A sequential quadratic programming algorithm is used to solve the problem. In each iteration, the sequential quadratic programming algorithm approximates the nonlinear constrained optimization problem at the current point, solves a quadratic programming subproblem to obtain the search direction, and then performs a line search along that direction to determine the step size. This algorithm has a superlinear convergence speed and is suitable for medium-sized (approximately 30-50 decision variables) nonlinear constrained optimization problems.

[0101] Based on the optimized solution, this step also includes two additional analyses:

[0102] Additional Analysis 1: Fatigue-Joint Load Cross-Sensitivity Analysis: Calculation of Joint Reaction Modulus The fatigue rate constant λ of each muscle group jpartial derivatives Muscle groups with large absolute values ​​of partial derivatives are considered fatigue-sensitive muscle groups under joint loads, meaning that even small changes in the fatigue rate of these muscle groups can lead to significant fluctuations in joint loads. For teeth controlled by these sensitive muscle groups, an additional stiffness buffer is set in the stiffness distribution: the stiffness constraint range for this tooth position is extended by a preset percentage margin from the optimal value towards safety; in a specific embodiment, this margin is set to 15% to 25%. The significance of this analysis is that even if the actual fatigue rate of an individual patient deviates from the model prediction, the additional buffer ensures that the robustness of the design is not compromised.

[0103] Additional Analysis 2, Joint Load Time-Domain Variance Penalty: Add a joint load time-domain variance penalty term to the objective function. ,in Let J be the variance of the joint load composite index across all time slices. This is a penalty factor. The purpose of this penalty term is to minimize peak load while making the joint load-time curve as flat as possible, reducing load fluctuations. The physiological basis for this is that joint tissues tolerate continuous moderate loads better than intermittent high peak loads, because the latter are more likely to cause cumulative micro-damage to the cartilage matrix.

[0104] Step S5: Geometric Generation of the Occlusal Pad

[0105] like Figure 7 As shown, step S5 will optimize the output of step S4—the optimal stiffness distribution. — Converted into a manufacturable three-dimensional geometric model of the bite pad.

[0106] Stiffness-thickness inverse conversion: Local stiffness of the occlusal pad at each tooth position It is determined by two geometric / material parameters: the elastic modulus E of the material at this location. i and the thickness h at that location i Based on the simplified Hertz contact theory, the local contact stiffness can be approximated as:

[0107]

[0108] in The equivalent elastic modulus is derived from the elastic modulus E of the occlusal pad material. i And the elastic modulus E of the opposing tooth material tooth Jointly determined, that is ν is Poisson's ratio; The equivalent contact radius is determined by the radius of curvature of the tooth cusp; δ i This represents the amount of contact deformation.

[0109] For a given target stiffness There are infinitely many sets (E) i ,h i This can be achieved through a combination of methods. Therefore, a material-thickness distribution strategy needs to be developed. In one specific embodiment, the following strategy is adopted:

[0110] when When the value is relatively large (belonging to the upper quartile of the stiffness value distribution), a material with a large elastic modulus (such as rigid PMMA, E≈3GPa) is selected to achieve the required stiffness with a smaller thickness and control the overall thickness; when When the stiffness is relatively small (belonging to the lower quartile of the stiffness value distribution), a material with a small elastic modulus (such as soft EVA, E≈20MPa) is selected to achieve the required low stiffness value at a reasonable thickness; when When the temperature is in the middle range, choose a material with a medium elastic modulus (such as medium hardness silicone rubber, E≈200MPa-1GPa).

[0111] Determine the material elastic modulus E for each tooth position i After that, the thickness h i This can be determined through numerical inverse calculation—adjusting h. i Until the contact stiffness at that point matches the target value The deviation is within the set tolerance.

[0112] 3D geometric model generation: Based on the subject's 3D dental arch scan data as the geometric reference for the occlusal pad tissue surface (lower surface), the calculated thickness h is used at each tooth position. i Offset towards the occlusal surface (upper surface) to generate the upper surface geometry of the occlusal pad. The connecting surface (marginal transition zone) between the upper and lower surfaces is rounded to avoid mucosal irritation during wear. Functional geometric features of the incisal guide ramp in the anterior tooth region and the guide groove in the molar region are also superimposed on the upper surface. The parameters of these features (bevel angle, groove width, groove depth) remain unchanged from the standard clinical design values ​​and are not involved in stiffness optimization.

[0113] In a preferred embodiment, when there are significant differences in the elastic modulus of the materials at different tooth positions, a multi-material 3D printing process is used to achieve a continuous material gradient transition within the same occlusal pad, thereby avoiding the risk of interface debonding between different material regions.

[0114] Step S6: Finite element multi-condition verification

[0115] After the three-dimensional geometric model is output in step S5, finite element multi-condition verification can be performed to check the accuracy and safety of the optimization results.

[0116] The verification method is as follows: the muscle force conditions of the M time slices in step S4 are respectively loaded as boundary conditions into the complete finite element model of occlusal pad-dental arch-mandible (an elastic body model is used here instead of a rigid body model to improve accuracy), and the reaction force and stress at both joints are extracted for each working condition.

[0117] If the joint reaction forces for all operating conditions are within the safety threshold, and the material stress of the occlusal pad itself does not exceed the fatigue strength limit (ensuring no material failure during long-term use), the verification is successful, and the final design model is output. If there are out-of-limit operating conditions, the constraints of the out-of-limit operating conditions are fed back to step S4, the corresponding load constraints are tightened, and robust optimization is performed again until all operating conditions pass the verification.

[0118] The finite element method (FEM) verification offers higher accuracy compared to the rigid body static model used in steps S2-S4 because it considers factors such as the nonlinear elastic deformation of the periodontal ligament, the large deformation effect of the occlusal pad material, and friction at the contact surface. The simplified rigid body model used in steps S2-S4 was chosen to meet computational efficiency requirements during the optimization iteration process (which requires evaluating hundreds to thousands of candidate stiffness distributions), while the FEM verification in step S6 only requires M analyses of the final optimal solution, making the computational load manageable.

[0119] Furthermore, this invention also provides a derivative function: personalized safety wearing time prediction. This is based on the joint load-time curve R(t) calculated in step S3 and a preset joint tissue damage threshold R. threshold Determine the time from the start of wearing until the joint load first reaches R. threshold Time .Will This "safe wearing duration" is recommended to clinicians to guide patients in managing their nighttime wearing time. For patients with shorter durations (e.g., less than 4 hours), clinicians should consider adjusting the treatment plan or increasing the frequency of follow-up visits.

[0120] To verify the effectiveness of the method of the present invention, the following comparative experiments were conducted:

[0121] The study included 10 adult patients diagnosed with temporomandibular joint disorder (unilateral reducible anterior displacement of the articular disc) complicated with nocturnal bruxism. Two occlusal splints were fabricated for each patient: one designed using the method of this invention (experimental group), and one designed using a traditional method based on tooth surface force homogenization (control group). Both occlusal splints used the same materials and were manufactured using the same process.

[0122] The evaluation indicators are: the compression load and shear load of the temporomandibular joint on the affected side calculated by finite element analysis under various muscle strength conditions.

[0123] Experimental results showed that under initial muscle strength (no fatigue) conditions, the compressive load on the affected joint of the experimental group was 72%-85% of that of the control group, and the shear load was 58%-76% of that of the control group, indicating that the joint load inversion design was superior to the traditional design from the initial state. Under simulated deep fatigue conditions (masseter muscle strength decreased to 40% of its initial value, temporalis muscle strength decreased to 70% of its initial value), the compressive load on the affected joint of the control group increased by 18%-35% compared to the initial condition, while the increase in the experimental group was only 3%-8%, indicating that the robust optimization framework effectively suppressed fatigue-induced joint load fluctuations. In all M = 15 time slice conditions, the maximum joint load of the experimental group was reduced by 28%-42% compared to the control group, and the load time-domain variance was reduced by 55%-70%.

[0124] like Figure 8 As shown, the present invention also provides a bite pad design system based on bite force distribution characteristics, comprising the following modules:

[0125] The multi-state occlusal force and electromyography (EMG) acquisition module includes an intraoral occlusal force sensor array, a bilateral masticatory muscle surface EMG acquisition device, a synchronization time stamp controller, and a data preprocessing unit. The intraoral occlusal force sensor array is a piezoresistive thin-film sensor array covering the entire dental arch. The bilateral masticatory muscle surface EMG acquisition device includes eight acquisition channels. The synchronization time stamp controller is used for precise time alignment of force signals and EMG signals. The data preprocessing unit is used for signal filtering, baseline drift correction, and motion artifact removal.

[0126] The individualized mandibular muscle skeletal biomechanics model construction module includes a CBCT 3D image data interface, a mandibular-condylar-glenoid 3D reconstruction engine, muscle geometric parameter calibration units, joint hinge axis positioning units, and a static equilibrium equation construction engine. The module takes CBCT image data and calibration data of the maximum voluntary contractile force of each muscle group as input, and outputs a complete static model containing muscle force lines, joint axis positions, and torque arm parameters.

[0127] The fatigue time-domain joint load time-varying prediction module includes an electromyography (EMG) fatigue feature extraction unit, a muscle force-fatigue decay model fitting engine, a time-varying muscle force-time-varying dental force joint derivation engine, and a joint load time series calculation engine. The EMG fatigue feature extraction unit performs short-time Fourier transform on EMG signals to extract the median frequency and root mean square amplitude. The muscle force-fatigue decay model fitting engine fits and corrects the parameters of the exponential decay model based on the time-domain variation of the median frequency. The time-varying muscle force-time-varying dental force joint derivation engine calculates the expected dental force distribution based on time-varying muscle force and occlusal pad stiffness distribution. The joint load time series calculation engine calculates the joint reaction force at each time step based on the time-varying dental force distribution and static equations.

[0128] The joint load time-domain robust optimization module includes a robust optimization solution engine and a fatigue-joint load cross-sensitivity analysis unit. The robust optimization solution engine uses a sequential quadratic programming algorithm to solve the min-max optimization problem. The fatigue-joint load cross-sensitivity analysis unit calculates the partial derivatives of the joint reaction force modulus with respect to the fatigue rate constants of each muscle group to identify sensitive muscle groups.

[0129] Occlusal pad geometry generation module: Includes a stiffness-to-thickness inverse conversion unit and a 3D geometry model generation engine. The stiffness-to-thickness inverse conversion unit is used to convert the optimal stiffness distribution into thickness and material elastic modulus distribution. The 3D geometry model generation engine is used to generate a manufacturable 3D geometry model by combining dental arch 3D scan data.

[0130] In a preferred embodiment, the system further includes a finite element verification module, which is used to perform finite element analysis verification on the generated three-dimensional geometric model under multiple fatigue time slice conditions, and to feed the constraints back to the joint load time-domain robust optimization module for iterative optimization when there are over-limit conditions.

[0131] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for designing an occlusal pad based on the characteristics of occlusal force distribution, characterized in that, Includes the following steps: S1. The subject was subjected to a continuous occlusal occlusion task. The time series F(x,y,t) of the occlusal force distribution of the entire dental arch was synchronously collected by an intraoral occlusal force sensor array, and the electromyographic signals EMG(t) of each masticatory muscle group were synchronously collected by a bilateral masticatory muscle surface electromyography acquisition device. The duration of the continuous occlusal occlusion task covered the complete fatigue process from the fresh state of the muscles to the appearance of changes in electromyographic fatigue indicators. S2. Based on the craniofacial imaging data of the subjects, a three-dimensional model of the mandible is reconstructed, the position of the hinge axis of the bilateral temporomandibular joint and the coordinates of the origin and insertion points and the direction of the force line of each masticatory muscle group are marked, and the results of the measurement of the maximum voluntary contraction force of each masticatory muscle group are combined to establish a set of static equilibrium equations of the mandible. The set of static equilibrium equations establishes the torque balance relationship between muscle force, dental contact force and joint reaction force. S3. Perform time-frequency analysis on the electromyographic signals, extract the electromyographic fatigue characteristic parameters of each masticatory muscle group, and fit the muscle force-fatigue decay function of each masticatory muscle group to obtain the time-varying muscle force vector M(t). Substitute the time-varying muscle force vector M(t) into the static equilibrium equations, and calculate the expected tooth force distribution at each moment by combining the occlusal pad stiffness distribution to be optimized through the mechanical contact model. Then calculate the joint reaction force at each moment in the entire fatigue process and generate the joint load time curve R(t). S4. Using the local stiffness values ​​at each tooth position of the occlusal pad as decision variables and minimizing the maximum joint load in the entire fatigue process time domain as the objective function, robust optimization is performed under constraints to obtain the optimal stiffness distribution. S5. The optimal stiffness distribution is converted into the thickness value and material elastic modulus value of the occlusal pad at each tooth position through the contact mechanics model, and a three-dimensional geometric model of the occlusal pad is generated by combining the three-dimensional scanning data of the subject's dental arch.

2. The occlusal pad design method based on occlusal force distribution characteristics according to claim 1, characterized in that, In step S1, the thickness of the intraoral occlusal force sensor array is no greater than 0.3 mm, and the spatial resolution is no less than 200 sensing points; the bilateral masticatory muscle surface electromyography acquisition device includes eight acquisition channels respectively set on the anterior bundle of the bilateral masseter muscle, the anterior bundle of the temporalis muscle, the posterior bundle of the temporalis muscle, and the medial pterygoid muscle; the intraoral occlusal force sensor array and the bilateral masticatory muscle surface electromyography acquisition device are time-aligned through a synchronous timestamp controller, and the sampling rate is no less than 1000 Hz.

3. The occlusal pad design method based on occlusal force distribution characteristics according to claim 1, characterized in that, In step S3, the time-frequency analysis of the electromyographic signals includes: performing a short-time Fourier transform on the electromyographic signals of each masticatory muscle group to extract the median frequency and root mean square amplitude within each time window; determining the fatigue decay rate parameter of each masticatory muscle group based on the slope of the median frequency over time; the muscle strength-fatigue decay function adopts a modified exponential decay model, the expression of which is: ,in, Let be the initial maximum force value of the j-th muscle. Let be the fatigue saturation coefficient of the j-th muscle. Let be the fatigue rate constant of the j-th muscle, and t be time.

4. The occlusal pad design method based on occlusal force distribution characteristics according to claim 1, characterized in that, In step S4, the objective function is: ； in, and These are the magnitudes of the reaction force vectors of the left and right temporomandibular joints at the k-th time slice, respectively. This represents the maximum value of the shear component in the joint reaction force. , , These are the weighting coefficients. M is the total number of discrete-time sampling points in the fatigue process, serving as an auxiliary decision variable.

5. The occlusal pad design method based on occlusal force distribution characteristics according to claim 4, characterized in that, The weighting coefficients are set according to the following rules: the weight of the affected joint in temporomandibular joint disorder is set to 1.5 to 2.5 times the weight of the healthy joint; the weight of the shear component is set as follows: Set the weight to 1.2 to 2.0 times that of the compressed component.

6. The occlusal pad design method based on occlusal force distribution characteristics according to claim 1, characterized in that, In step S4, the constraints include: the contact force of each tooth position at all time slices is non-negative; the guiding force of the incisal guide ramp in the anterior tooth region during lateral movement is not lower than a set threshold; the maximum thickness of the occlusal pad does not exceed a set upper limit value; the local stiffness value at each tooth position is within the stiffness range corresponding to the elastic modulus of the available material; and the force system at each time slice satisfies the mandibular static equilibrium equation.

7. The occlusal pad design method based on occlusal force distribution characteristics according to claim 1, characterized in that, Step S4 also includes: fatigue-joint load cross-sensitivity analysis, calculating the partial derivative of the joint reaction force modulus with respect to the fatigue rate constant of each muscle group, identifying joint load fatigue-sensitive muscle groups, setting additional stiffness buffers in the stiffness distribution for the tooth positions controlled by the joint load fatigue-sensitive muscle groups; and adding a joint load time-domain variance penalty term to the objective function to minimize the fluctuation amplitude of the optimized joint load time curve.

8. The occlusal pad design method based on occlusal force distribution characteristics according to claim 1, characterized in that, Step S5 is followed by a finite element multi-condition verification step: the muscle force conditions of each time slice in the fatigue process are loaded as load conditions onto the three-dimensional geometric model and finite element analysis is performed to verify whether the joint reaction force under each condition is within the set safety threshold. If there is an over-limit condition, it is fed back to step S4 to tighten the corresponding constraint conditions and re-optimize; and the safe wearing time is calculated based on the joint load time curve and the preset joint tissue damage threshold.

9. A bite pad design system based on bite force distribution characteristics, characterized in that, include: The multi-state occlusal force and electromyography (EMG) acquisition module is used to simultaneously acquire the time series of occlusal force distribution across the entire dental arch and the EMG signals of each masticatory muscle group when subjects are performing a continuous closed occlusal task. The module for constructing an individualized mandibular muscle and bone biomechanics model is used to establish a set of static equilibrium equations for the mandible based on the craniofacial imaging data of the subjects. The fatigue time-domain joint load time-varying prediction module is used to fit the muscle force-fatigue decay function of each muscle group and calculate the joint reaction force at each moment in the entire fatigue process. The joint load time-domain robust optimization module is used to perform robust optimization solutions with the objective function of minimizing the maximum joint load in the time domain during the fatigue process. The bite pad geometry generation module is used to inversely convert the optimal stiffness distribution into the thickness and material elastic modulus distribution and generate a three-dimensional geometric model.

10. The occlusal pad design system based on occlusal force distribution characteristics according to claim 9, characterized in that, The multi-state occlusal force and electromyography (EMG) acquisition module includes an intraoral occlusal force sensor array, a bilateral masticatory muscle surface EMG acquisition device, a synchronous timestamp controller, and a data preprocessing unit; the fatigue time-domain joint load time-varying prediction module includes an EMG fatigue feature extraction unit, a muscle force-fatigue decay model fitting engine, a time-varying muscle force-time-varying dental force joint deduction engine, and a joint load time series calculation engine. The joint load time-domain robust optimization module is solved using a sequential quadratic programming algorithm and includes a fatigue-joint load cross-sensitivity analysis unit.

Images