A design method and layered structure of a synergistic mouthguard based on musculoskeletal stability positioning optimization

Abstract

This application belongs to the field of medical devices and protective equipment, specifically involving a design method and layered structure for an enhanced mouthguard based on musculoskeletal stability optimization. The method includes: collecting multimodal data such as the full dentition morphology, jawbone structure, mandibular movement trajectory, surface electromyography, and plantar pressure of the subject; constructing a correlation system between mandibular position and overall stability using a radial basis function neural network; setting anatomical boundaries, joint spaces, and muscle symmetry constraints; and using a genetic algorithm to iteratively optimize and lock in a personalized musculoskeletal stability position; based on this, performing layered digital design of the mouthguard's flexible base layer and posterior tooth retention components, and then 3D printing manufacturing. This application, through precise optimization of jaw position, achieves a leap from passive protection to actively enhancing body stability in sports mouthguards, significantly improving exercise efficiency and wearing comfort while ensuring physical protection.

Description

Technical Field

[0001] This application belongs to the field of medical devices, specifically relating to a design method and layered structure for an enhanced mouthguard based on musculoskeletal stability optimization. Background Technology

[0002] With the increasing popularity of contact sports, sports mouthguards play an irreplaceable role in protecting athletes' teeth and maxillofacial tissues and preventing concussions. As a flexible protective device worn on the dental arch, mouthguards effectively absorb and disperse external impacts, thereby reducing the risk of jaw fractures and soft tissue injuries. In recent years, the cross-integration of digital oral technology and sports medicine has provided new technological pathways for the personalized design and manufacturing of mouthguards, resulting in significant progress in precise fitting and wearing comfort.

[0003] Among them, customized motion mouthguards based on 3D scanning and additive manufacturing achieve a leap from general-purpose devices to individualized fittings by accurately acquiring the anatomical morphology of the subject's dentition. This type of technology typically establishes the alignment relationship between the upper and lower jaws in virtual space and designs the occlusal surface morphology of the mouthguard according to the intercuspal position or clinical centric relationship, aiming to achieve passive protection of the oral system through precise physical coverage, while also taking into account the wearer's breathing and speech functions.

[0004] However, the core design of existing customized mouthguards remains limited to passive adaptation to anatomical morphology, lacking an active mechanism to enhance the user's overall biomechanical performance. Traditional design methods typically rely on static clinical jaw position standards, failing to establish a quantitative correlation model between jaw position and overall postural stability. This results in mouthguards being unable to functionally optimize for the motor performance of different individuals. Furthermore, existing occlusal surface design processes depend on empirical judgment, making it difficult to accurately optimize musculoskeletal stability within physiological movement boundaries. The lack of in-depth analysis and feature mapping of multi-source heterogeneous biomechanical data also leads to significant bottlenecks in mouthguards' ability to improve neuromuscular control.

[0005] Therefore, a design method and layered structure for an enhanced mouthguard based on musculoskeletal stability optimization is desired. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of the prior art and provide an enhanced mouthguard design method and layered structure based on musculoskeletal stability optimization, which can effectively solve the problems in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: A method for designing an enhanced mouthguard based on musculoskeletal stability optimization includes the following steps: S1: Acquire multimodal data of the subject, including: three-dimensional morphological data of the dentition obtained by an intraoral scanner, structural data of the jawbone and temporomandibular joint obtained by cone-beam CT, mandibular edge movement trajectory data obtained by a mandibular movement trajectory recorder, and surface electromyography signals and plantar pressure center trajectory data collected simultaneously under multiple preset functional jaw positions. S2: Construct a biomechanical functional correlation system between jaw position and overall stability: Spatial registration and standardization of the multimodal data are performed, muscle asymmetry index and plantar pressure dynamics index are extracted, and a radial basis function neural network model is trained with jaw position spatial coordinate vector as input and the biomechanical index as output to establish a nonlinear mapping relationship between jaw position and body stability performance. S3: Define the musculoskeletal stability optimization problem and construct a comprehensive fitness function: Based on S1 and S2, define the constraints of the optimization problem and construct a comprehensive fitness function to quantify the jaw position stability effectiveness; the constraints include: using the physiological motion envelope as the anatomical boundary constraint, and using the preset joint space range and muscle asymmetry index threshold as physiological safety constraints; the comprehensive fitness function is constructed based on the predicted plantar pressure dynamics index of the radial basis function neural network model trained in S2 as the prediction agent; S4: Optimizing musculoskeletal stable positions based on intelligent optimization algorithms: A genetic algorithm is used to iteratively optimize the jaw position solution space that satisfies all constraints. The neural network model of S2 is used to evaluate the fitness of candidate jaw positions. Through selection, crossover, and mutation operations, the personalized musculoskeletal stable position that maximizes the comprehensive fitness function value is locked. S5: Perform digital layered design of the sports mouthguard: Based on the personalized musculoskeletal stability obtained in S4, adjust the position of the mandibular model relative to the maxillary model in the digital design environment, and align the oral dental arch model with the jawbone model. Then, design the flexible base layer for impact protection and the posterior tooth retention components for maintaining the musculoskeletal stability. S6: The toothbrush is manufactured using 3D printing technology in a single piece.

[0008] Preferably, in step S1, the plurality of preset functional jaw positions include: the intercuspal position, which is defined as the initial zero point of the local anatomical coordinate system, and the sampling points determined after single-axis or multi-axis displacement in the vertical, anterior-posterior, and lateral directions based on the initial zero point; the sampling points specifically include the vertical elevation position, the protrusion functional position, the lateral balance position, the protrusion-vertical combined position, and the lateral-vertical combined position.

[0009] Preferably, in step S2, the construction of the biomechanical function correlation system specifically includes: S21: Data Alignment and Feature Extraction: The dental arch model and the skeletal structure model are registered using the iterative nearest-point algorithm; the root mean square value of the surface electromyography signal is calculated after bandpass filtering and full-wave rectification, and the muscle asymmetry index is calculated based on this value. The calculation formula is:

[0010] in, and These represent the root mean square values ​​of surface electromyography (EMG) signals for the corresponding muscle groups on the right and left sides, respectively. S22: Model Training: The model is trained using a jaw space vector containing the coordinates of the mandibular incisor points and the uppermost points of the bilateral condyles as input, and the muscle asymmetry index as input. Total length of trajectory Average speed and load distribution ratio To achieve the output target, a radial basis function neural network is trained. The training process includes normalizing the input vector, using a clustering algorithm to determine the centers of the radial basis functions, determining the width parameter based on the distance between the centers, and using the least squares method to solve for the weights of the output layer.

[0011] Preferably, in step S3, the definition of the optimization problem specifically includes: S31: Set constraints, including: anatomical boundary constraints, requiring the candidate jaw position to be located inside the jaw position edge motion envelope defined in step S1; joint space constraints, requiring the anterior, superior, and posterior spaces of the bilateral temporomandibular joints under the candidate jaw position to be within the physiologically safe range; and muscle symmetry constraints, requiring the muscle asymmetry index of the candidate jaw position predicted by the neural network to be lower than a preset threshold. S32: Construct the comprehensive fitness function F, whose expression is:

[0012] in, , and These represent the total lengths of the plantar pressure center trajectory predicted by the neural network. Average movement speed and bipedal load distribution ratio The scores after standardization These are weighting coefficients that are dynamically allocated based on the needs of the sport.

[0013] Preferably, in step S4, the optimization using a genetic algorithm specifically includes: S41: Initialization: Within the anatomical boundary constraints, a preset number of initial jaw position vectors are randomly generated using a real number encoding method to form an initial population; S42: Evaluation and Selection: Input the jaw position vector of each individual in the population into the radial basis function neural network trained in step S2. First, determine whether it satisfies the joint space constraint and muscle symmetry constraint, and eliminate individuals that do not meet the constraints; for individuals that meet the constraints, select the appropriate individuals based on their predictions. Value calculation of the comprehensive fitness function The value is used as its fitness; selection operations are performed based on fitness to retain superior individuals; S43: Iterative evolution: Perform crossover and mutation operations on the selected population to generate a new generation of population, and ensure that the new individuals still satisfy the anatomical boundary constraints; repeat steps S42 and S43 until the value of the comprehensive fitness function F tends to stabilize and no longer increases significantly. At this time, the individual with the highest fitness is the personalized musculoskeletal stable position obtained by optimization.

[0014] Preferably, in step S5, the digital hierarchical design specifically includes: S51: Realignment and Base Design: Based on the spatial transformation matrix corresponding to the personalized musculoskeletal stability position, adjust the position of the mandibular model relative to the maxillary model; and align the oral scan dental arch data with the realigned maxillary and mandibular models; then, based on the wearing lateral dental arch model, perform undercutting, edge line definition and smoothing to generate the flexible base layer that adapts to the dentition and gingival morphology. S52: Retention component design: In the posterior tooth region of the flexible base layer, an arc-shaped solid with a thickness equal to the vertical occlusal gap under the musculoskeletal stable position is generated along the dental arch morphology; using the cusp and pit morphology of the opposing dentition as the shear body, Boolean subtraction is performed on the occlusal surface of the arc-shaped solid to form a precise interlocking occlusal guiding surface, which constitutes the posterior tooth retention component.

[0015] Preferably, the weighting coefficient The dynamic allocation strategy is as follows: for sports that emphasize static balance, increase the load allocation ratio. weight For sports that emphasize dynamic agility and reaction speed, increase the average speed of movement. weight .

[0016] Preferably, in step S6, the integrated molding manufacturing specifically involves: using the digital model obtained in step S5, which includes a flexible base layer and a posterior tooth retention component, to print the model in one step using a biocompatible polymer resin material through photopolymerization 3D printing technology, and then performing post-curing treatment.

[0017] This invention also discloses an enhanced mouthguard designed and manufactured using the method described above, characterized in that it comprises an integrally molded component: The flexible base layer has an inner surface that perfectly conforms to the shape of the user's lateral dentition and gums, and is used to absorb and disperse impact forces. The posterior tooth retention component, connected to the posterior tooth region of the flexible base layer, has an occlusal surface with a three-dimensional morphology that precisely complements the cusp and pit fissure of the opposing tooth in a personalized musculoskeletal stable position, used to guide and lock the user's mandible into this optimal functional position during wear.

[0018] The present invention also discloses a non-volatile computer-readable storage medium storing a computer program, characterized in that the computer program, when executed by a processor, implements the above-mentioned enhanced mouthguard design method based on musculoskeletal stability position optimization.

[0019] Compared with the prior art, the present invention has the following beneficial effects: This invention actively enhances the user's neuromuscular control and body stability by optimizing jaw position, expanding the design goal of mouthguards from simple physical impact protection to biomechanical performance enhancement, which can effectively improve athletes' athletic performance and reduce the risk of injury.

[0020] Through multimodal data acquisition and intelligent algorithm optimization, this invention can accurately locate the unique musculoskeletal stability position for each user, solving the problem of the disconnect between anatomical morphology adaptation and biomechanical function in traditional mouthguard design, and realizing a deep evolution from morphological adaptation to functional adaptation.

[0021] By utilizing radial basis function neural networks and genetic algorithms, this invention transforms the jaw position determination process, which originally relied on clinical experience, into a data-driven scientific optimization problem. Through multidimensional constraints and a comprehensive fitness function, the objectivity, accuracy, and repeatability of the design results are ensured.

[0022] The layered structure design decouples the functional areas. The retention layer is responsible for maintaining the optimized jaw position, while the flexible base layer is responsible for absorbing impact. The two are combined in a digital environment for integrated modeling and 3D printing manufacturing, ensuring that the final product has extremely high manufacturing precision and wearing comfort.

[0023] By setting joint space constraints and muscle symmetry constraints, this invention, while pursuing maximum physical efficiency, fully considers the balance between the physiological health of the temporomandibular joint and muscle load, effectively preventing joint damage or muscle fatigue that may be caused by changes in jaw position. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of the overall technical solution architecture of the enhanced mouthguard design method based on musculoskeletal stability position optimization proposed in this invention. Figure 2 This is a graph obtained after multimodal data fusion in this invention; Figure 3 This is a schematic diagram of the overall structure of the tooth protector in this invention; Figure 4 This is a schematic diagram of the flexible substrate layer designed based on EXOCAD dental digital software in this invention; Figure 5 This is a schematic diagram of the posterior tooth retention component designed based on solidwork and geomgaic in this invention. Figure 6 This is a schematic diagram of the toothbrush after it has been inserted in this invention. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0026] The enhanced mouthguard design method based on musculoskeletal stability optimization includes the following steps: S1: Acquire multimodal physiological and biomechanical data of the subjects; Further, during the static data acquisition phase, an intraoral scanner with preset scanning accuracy was used to perform non-contact optical scanning of the subject's maxillary and mandibular dentition, as well as the occlusal relationship in the intercuspal position. The intraoral scanner acquired an STL-format triangular mesh model including the three-dimensional morphology of the tooth crowns, the anatomical contour of the gingival margin, and the mucosal transition. Simultaneously, a cone-beam computed tomography (CBCT) scan was performed on the subject in the intercuspal position (ICP) using panoramic and articular region scanning. The CBCT scanning parameters were set to preset tube voltage, preset tube current, and preset scan time, with the reconstructed slice thickness set to preset values. Medical image processing software was used to perform three-dimensional reconstruction of the skeletal structure from the DICOM-format image data, focusing on extracting the geometric boundaries of the condyles, glenoid fossa, and articular tuberosities of the bilateral temporomandibular joints. Based on this, the initial anatomical width values ​​of the anterior, superior, and posterior spaces of the condyles within the glenoid fossa were accurately measured and recorded; these values ​​were stored as the anatomical geometric constraint reference vector in the subsequent optimization algorithm.

[0027] During the dynamic boundary definition phase, subjects wore a mandibular motion trajectory recorder. This recorder, through sensors, captured the real-time motion trajectory of the mandible relative to the maxilla, guiding the subject to perform multiple consecutive movements, including maximum opening and closing of the mouth, maximum horizontal protrusion, and maximum left and right lateral movements. The system recorded the real-time coordinate sequence of the mandibular incisor points and the highest points of the bilateral condyles in three-dimensional space during the movements. By performing nonlinear fitting on the extreme points of the above motion paths, a closed three-dimensional geometric entity was generated, namely the jaw position edge motion envelope. This envelope represents the set of all physiological limit positions that the mandible can reach under anatomical constraints, and is used to establish the search space boundary for subsequent jaw position function optimization.

[0028] For the functional gradient sampling phase, multiple representative sampling target points are preset in three-dimensional space to cover typical functional areas of mandibular movement. These sampling points are divided into basic reference positions, uniaxial boundary positions, and multiaxial coupling positions. During the sampling process, occlusal recording materials with different thickness gradients (such as silicone rubber occlusal recording materials or pads of preset thickness) are used in conjunction with clinical guidance to stabilize the subject's mandible in the target spatial posture.

[0029] For the basic reference position, the intercuspal position is set as the initial zero position of the relative movement trajectory of the mandible, and data is collected under the resting jaw position as a physiological benchmark reference under the minimum muscle tension state.

[0030] For uniaxial boundary positions, these specifically include the vertical elevation position, the protrusion functional position, and the lateral balance position. The vertical elevation position, achieved by placing occlusal recording material with a preset thickness gradient, induces a linear vertical displacement of the mandible from the intercuspal position, thus establishing a functional mapping relationship along the vertical Z-axis. The protrusion functional position guides the mandible to protrude horizontally by a preset amount from the intercuspal position, establishing a functional mapping along the anteroposterior Y-axis. The lateral balance position guides the mandible to move to the left and right by preset amounts, respectively, establishing a functional mapping along the horizontal X-axis.

[0031] For multi-axial coupling positions, there are protrusion-vertical combination positions and lateral-vertical combination positions. The protrusion-vertical combination position guides the mandible to open vertically to a preset height while accompanied by a preset protrusion displacement; the lateral-vertical combination position guides the mandible to open vertically to a preset height while moving to the left and right by preset displacements, respectively.

[0032] In all the above sampling tests, subjects were required to maintain focus on a pre-set fixed point in front of them to eliminate the interference of visual feedback on body balance regulation. Subjects were required to maintain steady-state occlusion for a pre-set time period at each sampling point. During this period, the real-time three-dimensional coordinates of the mandibular incisor points and the superior points of the bilateral condyles were simultaneously captured using a mandibular motion trajectory recorder. Simultaneously, raw electromyographic signals of the bilateral masseter muscles and bilateral sternocleidomastoid muscles were recorded using a surface electromyography (EMG) acquisition device, with the sampling rate set to a pre-set high-frequency sampling value. Furthermore, the raw center of pressure (COP) trajectory data and bilateral foot pressure distribution data of the subjects under steady-state occlusion were simultaneously acquired using a plantar pressure analysis system.

[0033] S2: Construct a biomechanical functional correlation system between jaw position and overall stability; Specifically, the process involves: First, spatial alignment and standardization of multi-source data are performed. Using a medical image registration algorithm, the STL dental arch model obtained from intraoral scanning is registered at multiple points with the skeletal structure model reconstructed from cone-beam computed tomography (CBCT). During registration, anatomical landmarks on the crown surface are selected as feature points, and the Iterative Closest Point (ICP) algorithm is used to achieve high-precision fusion of different modalities in a unified three-dimensional coordinate system. Subsequently, a local anatomical coordinate system based on the intercuspal position is established for relative kinematic analysis. Specifically, the three-dimensional structure of the maxilla (or skull base) is used as a fixed reference frame, and the spatial pose of the mandible when the subject is in the intercuspal position is defined as the initial zero point. Under this standard, the spatial positions of the mandibular incisors and the superior points of the bilateral condyles in the intercuspal position are reset to their respective initial zero points for trajectory calculation. All dynamically acquired mandibular motion paths and sampling point coordinates are converted into three-dimensional displacement vectors relative to their respective initial zero points, thus representing the true spatial displacement of the mandible relative to the maxilla.

[0034] Next, feature extraction of physiological electrical signals and kinetic indices was performed. The raw surface electromyography (EMG) signals were bandpass filtered within a preset frequency range to remove power frequency interference and low-frequency artifacts, followed by full-wave rectification. The root mean square (RMS) value of the signal within each sampling period was calculated to characterize the muscle contraction intensity. Based on the RMS value, the asymmetry index at each sampling point was calculated. Its mathematical expression is as follows:

[0035] in, and These represent the root mean square (RMS) values ​​of surface electromyography (EMG) signals for the corresponding muscle groups (such as the jaw-raising muscles or neck posture muscles) on the right and left sides, respectively. The smaller the index, the better the symmetry of bilateral muscle contraction.

[0036] Simultaneously, the raw plantar pressure data is processed to calculate the total length of the plantar pressure center trajectory (PathLength). Average Velocity ) and weight distribution ratio (weight distribution, Among them, the total trajectory length and average speed reflect the intensity of the body's swaying at the microscopic level, while the load distribution ratio reflects the balance of the center of gravity between the left and right feet.

[0037] Subsequently, a training dataset of "jaw position-electromyography-plantar pressure center" was constructed. The relative spatial pose vector of jaw position after coordinate transformation was then used. The position vector serves as the input variable for a radial basis function (RBF) neural network. Composed of three-dimensional displacement components of the mandibular incisor points and the uppermost points of the bilateral condyles relative to the initial zero position, it contains a total of nine-dimensional feature data, which can completely define the translational and rotational posture of the mandible in three-dimensional space. The corresponding asymmetry index, total length of the plantar pressure center trajectory, average velocity, and load distribution ratio are used as target output variables.

[0038] An RBF neural network was used to train the aforementioned dataset using nonlinear mapping. During the training refinement process, the input vectors were normalized, and a clustering algorithm was used to determine the radial basis function centers of the neural network. The width parameter was determined based on the average distance between the centers. The output layer weights were solved using the least squares method. The training set and validation set were divided into a preset ratio, and the convergence of the validation set error was used as the training termination condition. After training, the mapping network can predict the corresponding muscle symmetry state and whole-body stability characteristics in real time for any continuous coordinate point within the jaw position edge motion envelope, thereby establishing a continuous biomechanical functional correlation system from "jaw position" to "body stability efficiency" in the digital space.

[0039] S3: Define the musculoskeletal stability optimization problem and construct a comprehensive fitness function; Specifically: First, the optimization variables are defined. Within a unified three-dimensional anatomical coordinate system, the variables to be optimized are defined as the spatial displacement vectors of the mandibular incisor points and the uppermost points of the bilateral condyles. This vector fully represents the six-degree-of-freedom positional relationship between the mandible and the maxilla.

[0040] To ensure that the optimization results improve efficiency while meeting the physiological safety requirements of the subjects, a multi-dimensional set of constraints is constructed, specifically including: S31, Anatomical Boundary Constraint. This requires that any candidate jaw position searched by the algorithm must be located within the jaw position edge motion envelope constructed in S1. This means that the optimization process is completely limited to the subject's actual physiological movement limits, and anatomically inaccessible jaw positions are strictly prohibited.

[0041] S32, Joint space constraint. Based on the initial joint space reference obtained by cone-beam computed tomography (CBCT), the displacement of the mandibular condyle relative to the glenoid fossa in the candidate mandibular position is calculated in real time. The anterior, superior, and posterior spaces of both condyles are required to be within a preset physiological safety range, which is set between 1.5-3.5 mm in this embodiment. This constraint aims to limit the amount of condylar displacement in the functional position to prevent overload stress on the articular disc and mechanical compression of the retrodiscal tissues, thereby protecting the health of the temporomandibular joint.

[0042] S33, Muscle Symmetry Constraint. Using the neural network surrogate model trained in S2, the muscle asymmetry index under candidate jaw positions is predicted. The activation difference between the two muscle groups must be lower than a preset asymmetry threshold, set to 15% in this embodiment, to ensure that jaw position optimization does not induce muscle fatigue or functional disorder.

[0043] Subsequently, a comprehensive fitness function is constructed. This function is used to quantify the synergistic effect of jaw position on overall stability. It standardizes and weights multiple core dynamic indicators from the neural network prediction output, and the specific algorithm is as follows:

[0044] In the above formula, , and These represent the standardized scores for the total length of the plantar pressure center trajectory, average movement speed, and load distribution ratio. The standardization process, through taking the reciprocal or linear mapping, transforms the original negative indicators into scores where "higher values ​​indicate better stability." Weighting coefficients. Dynamic allocation is performed based on the specific requirements of the sport. For example, for sports emphasizing static balance, such as shooting, the weight of the load distribution ratio is increased; for sports emphasizing agility and reaction time, such as football, the weight of the center of gravity shift speed index is increased. (Comprehensive fitness function) The larger the value, the closer the overall biomechanical efficiency of the jaw position is to the ideal "musculoskeletal stability".

[0045] S4: Optimize musculoskeletal stability based on intelligent optimization algorithm; Specifically, the candidate mandibular position population is first initialized. Based on the anatomical boundary constraints set by S3, the genetic algorithm population is initialized using real-number encoding within the physiological envelope of the subjects. A preset number of initial individuals are randomly generated, with each individual representing a potential mandibular position vector.

[0046] Next, a neural network-based performance evaluation and survival filtering are performed. Each candidate vector in the population is input into the RBF neural network model constructed in S2. The algorithm first performs hard constraint judgment, that is, using the muscle asymmetry index predicted by the model and the calculated joint space value, individuals that do not meet the physiological admission conditions are automatically eliminated. For compliant individuals that pass the hard constraint verification, the neural network is further used to predict their corresponding plantar pressure center dynamic index, and then substituted into the comprehensive fitness function defined in S3. In this process, the fitness score of each individual is calculated.

[0047] During the iterative evolution process, the algorithm performs selection, crossover, and mutation operations. Selection is based on fitness scores, ensuring that high-scoring jaw features have a higher probability of being inherited by the next generation. Crossover generates new jaw positions through a linear combination of two parent vectors. Mutation introduces random perturbations with preset probabilities to maintain population diversity and prevent getting trapped in local optima. During each generation, the algorithm continuously monitors whether newly generated jaw positions exceed anatomical boundary constraints to prevent the search from entering joint dislocation zones.

[0048] Through multiple generations of iterative evolution, when the comprehensive fitness function... When the value of the vector stabilizes and no longer increases significantly, the evolutionary calculation is terminated. At this point, the globally optimal jaw position vector is output, which is then locked as the subject's personalized "musculoskeletal stability position." This position represents the optimal mandibular spatial pose that maximizes the subject's body stability while satisfying all anatomical and physiological safety requirements.

[0049] S5: Implement digital layered design for sports mouthguards; Specifically, firstly, all multimodal data is imported into the digital design environment. Based on the globally optimal jaw position vector output by S4, rigid body transformations are performed on the already registered maxillary and mandibular dentition models of the subject. By applying a transformation matrix, the mandibular model is precisely translated and rotated relative to the maxillary model in virtual space to a preset musculoskeletal stable position, thereby establishing a new alignment relationship oriented towards enhanced efficiency. Then, the maxillary and mandibular dentition data from the intraoral scan are aligned with the re-aligned maxilla and mandible using a feature point registration method.

[0050] Subsequently, the flexible base layer component of the mouthguard was constructed using digital oral software. A dental arch model of the wearing side (e.g., maxilla) was selected as the design base. First, the expected path of insertion was set, and undercut analysis tools were used to identify undercut areas on the dental arch model. For deep undercuts that might hinder the smooth insertion and removal of the mouthguard, a virtual filling algorithm was used for smoothing. Next, the anatomical boundaries of the flexible base layer were defined based on the subject's gingival margin morphology and mucosal transition position. Chamfering and smoothing tools were applied to ensure continuity at all edges and transitions of the base layer to eliminate potential pressure concentration points. The design thickness of the flexible base layer was personalized according to the impact intensity of the sport. After completing the initial design of the base layer, the posterior occlusal area was virtually smoothed, and a pre-defined imprint depth was reserved to guide the mandibular insertion.

[0051] Next, a posterior tooth retention component was designed based on a flexible base layer. The anatomical edge trajectory of the posterior tooth functional area within the flexible base layer was extracted as a path, and the extrusion function of 3D modeling software was used to generate an arc-shaped solid with a constant support thickness. The thickness of this solid precisely filled the vertical space between the upper and lower posterior teeth under musculoskeletal stability. Finally, using the cusp, fissure, and incisal edge morphology of the contralateral dentition (mandibular) as a "digital shear body," Boolean operations were performed to cut the occlusal surface of the posterior tooth retention component. In this way, the generated retention layer occlusal surface can form a precise locking relationship with the contralateral teeth under optimal occlusal position, thereby forcibly guiding and maintaining the mandible in the optimized musculoskeletal stability position during wear.

[0052] S6: Digital integrated manufacturing using 3D printing technology.

[0053] Specifically, the model is imported into the pre-processing software of the 3D printing system to set the printing support structure and layout. Biocompatible polymer resin material is used, and the device is manufactured as a single piece through photopolymerization 3D printing technology. During the printing process, the dimensional accuracy of the mouthguard is ensured by precisely controlling the exposure time and layer thickness. After molding, the mouthguard undergoes post-curing treatment to achieve the preset mechanical strength and biocompatibility requirements. The resulting personalized, enhanced mouthguard not only possesses excellent impact absorption capabilities but also actively improves the wearer's motor function by precisely maintaining optimal jaw alignment.

[0054] In the aforementioned design process, data accuracy and algorithm convergence are crucial to ensuring the final oral brace's effectiveness. The dental arch model acquired by the intraoral scanner is not only used for the morphological adaptation of the oral brace but also serves as the spatial benchmark for all biomechanical calculations. The skeletal data provided by cone-beam computed tomography (CBCT) sets insurmountable physiological red lines for the optimization algorithm. By integrating these multimodal data into a surrogate model based on an RBF neural network, this embodiment successfully transforms the originally complex and nonlinear "jaw position-stability" correlation into a quantifiable mathematical problem for optimization.

[0055] During the optimization process, the genetic algorithm continuously evaluates thousands of virtual jaw position combinations. The final musculoskeletal stable position found is not a simple geometric center point, but the optimal solution tailored to the athlete's neuromuscular control characteristics. This deep customization considers not only tooth alignment but also the force distribution of the temporomandibular joint and the coordination of whole-body postural reflexes.

[0056] During the digital design phase, undercut filling technology ensures the mouthguard fits smoothly into dentitions with complex anatomical structures, while smoothed edges greatly enhance wearing comfort during high-intensity activities. The core logic of the layered design lies in the spatial decoupling and integration of the two functional goals of "protection" and "efficiency enhancement." The flexible base layer dissipates impact energy, protecting the soft and hard tissues of the oral cavity; while the retention layer uses precise occlusal guiding surfaces to lock the mandible in the spatial coordinates of optimal dynamic performance. This combination of structure and function transforms the mouthguard from a traditional passive shielding device into an active biomechanical regulator.

[0057] During the manufacturing stage, this invention employs 3D printing technology to ensure that the spatial pose and anatomical features in the digital design are accurately reproduced at the physical level. In particular, the high-precision printing of the posterior tooth occlusal guide component provides precise occlusal guidance to stabilize the target jaw position. Furthermore, by selecting medical-grade biocompatible polymer materials, not only is the chemical stability of the mouthguard ensured in the complex oral microenvironment, but it also endows it with excellent resistance to alternating stress fatigue and mechanical durability.

[0058] The technical solution provided in this embodiment, through deep fusion of multimodal data, construction of nonlinear mapping networks, and iteration of intelligent optimization algorithms, has achieved a leap from simple "anatomical adaptation" to "functional enhancement" for sports mouthguards. It not only provides athletes with top-notch physical protection, but also taps into and releases the human body's potential stability and control by optimizing jaw position, a core biomechanical lever, and has extremely high clinical application value and engineering practice significance.

[0059] In further implementation, the weight allocation strategy of this embodiment can be adjusted specifically for different types of sports. For example, for rugby, a highly competitive sport, the weight allocation strategy can be adjusted when constructing the comprehensive fitness function. Furthermore, an impact force distribution uniformity index can be introduced. By simulating the force distribution when the mandible is impacted from different directions, the geometric thickness distribution of the mouthguard can be optimized. This maximizes the absorption and dissipation of mechanical impact energy while maintaining musculoskeletal stability, effectively reducing the risk of craniocerebral and joint injuries caused by impact force transmission from the temporomandibular joint to the skull base. For sports involving a large number of rapid head rotations, such as boxing or karate, this embodiment can specifically adjust the judgment logic of muscle constraint conditions. Based on the original "muscle activation symmetry," the "minimum co-contraction threshold" of the neck muscle group is introduced as a composite constraint condition. With the help of RBF neural network prediction, the system will prioritize the selection of candidate jaw positions that take into account both fatigue resistance symmetry and high impact resistance stiffness. Only when the jaw position simultaneously meets this dual constraint can it proceed to the subsequent objective function for fitness calculation.

[0060] In terms of material selection for digital manufacturing, in addition to biocompatible polymer resin materials with a single hardness, multi-material gradient printing technology can also be used. The digital model generated in step S5, including a flexible base layer and posterior tooth retention components, is imported into a multi-material photopolymerization 3D printing device. At least two biocompatible polymer resin materials with different Shore hardnesses are used for one-time printing. Specifically, the flexible base layer uses a resin material with a lower Shore hardness, and the posterior tooth retention components use a resin material with a higher Shore hardness. This gradient distribution of material properties can further optimize the energy absorption characteristics and retention accuracy of the dental prosthesis.

[0061] Furthermore, the design method of this embodiment can also be combined with a real-time biofeedback system. During training while athletes wear mouthguards, embedded micro-sensors monitor their bite force distribution and body balance data in real time. This real-time data can be fed back to the design system for periodic fine-tuning and optimization of musculoskeletal stability to adapt to changes in the athlete's physiological state at different training stages.

[0062] Through the detailed technical specifications and engineering implementation outlined above, this embodiment fully demonstrates a data-driven, intelligent optimization-based end-to-end design system for enhancing mouthguard performance. This system not only solves the problem of reliance on experience in traditional mouthguard design, but also establishes a new standard for mouthguards as a tool for improving athletic performance through rigorous biomechanical modeling and multi-dimensional constraint optimization.

[0063] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A method for designing an enhanced mouthguard based on musculoskeletal stability optimization, characterized in that, Includes the following steps: S1: Acquire multimodal data of the subject, including: three-dimensional morphological data of the dentition obtained by an intraoral scanner, structural data of the jawbone and temporomandibular joint obtained by cone-beam CT, mandibular edge movement trajectory data obtained by a mandibular movement trajectory recorder, and surface electromyography signals and plantar pressure center trajectory data collected simultaneously under multiple preset functional jaw positions. S2: Construct a biomechanical functional correlation system between jaw position and overall stability: Spatial registration and standardization of the multimodal data are performed, muscle asymmetry index and plantar pressure dynamics index are extracted, and a radial basis function neural network model is trained with jaw position spatial coordinate vector as input and the biomechanical index as output to establish a nonlinear mapping relationship between jaw position and body stability performance. S3: Define the musculoskeletal stability optimization problem and construct a comprehensive fitness function: Based on S1 and S2, define the constraints of the optimization problem and construct a comprehensive fitness function to quantify the jaw position stability effectiveness; the constraints include: using the physiological motion envelope as the anatomical boundary constraint, and using the preset joint space range and muscle asymmetry index threshold as physiological safety constraints; the comprehensive fitness function is constructed based on the predicted plantar pressure dynamics index of the radial basis function neural network model trained in S2 as the prediction agent; S4: Optimizing musculoskeletal stable positions based on intelligent optimization algorithms: A genetic algorithm is used to iteratively optimize the jaw position solution space that satisfies all constraints. The neural network model of S2 is used to evaluate the fitness of candidate jaw positions. Through selection, crossover, and mutation operations, the jaw position that maximizes the comprehensive fitness function value is locked, which is the personalized musculoskeletal stable position. S5: Perform digital layered design of the sports mouthguard: Based on the personalized musculoskeletal stability obtained in S4, adjust the position of the mandibular model relative to the maxillary model in the digital design environment, and align the oral dental arch model with the jawbone model. Then, design a flexible base layer for impact protection and posterior tooth retention components for maintaining the musculoskeletal stability. S6: The toothbrush is manufactured using 3D printing technology in a single piece.

2. The method according to claim 1, characterized in that, In step S1, the multiple preset functional jaw positions include: the intercuspal position, which is defined as the initial zero point of the local anatomical coordinate system, and the sampling points determined after single-axis or multi-axis displacement in the vertical, anterior-posterior, and lateral directions based on the initial zero point; the sampling points specifically include the vertical elevation position, the protrusion functional position, the lateral balance position, the protrusion-vertical combination position, and the lateral-vertical combination position.

3. The method according to claim 1 or 2, characterized in that, In step S2, the construction of the biomechanical function correlation system specifically includes: S21: Data Alignment and Feature Extraction: The dental arch model and the skeletal structure model are registered using the iterative nearest-point algorithm; the root mean square value of the surface electromyography signal is calculated after bandpass filtering and full-wave rectification, and the muscle asymmetry index is calculated based on this value. The calculation formula is: ； in, and These represent the root mean square values ​​of surface electromyography (EMG) signals for the corresponding muscle groups on the right and left sides, respectively. S22: Model Training: The model is trained using a jaw space vector containing the coordinates of the mandibular incisor points and the uppermost points of the bilateral condyles as input, along with the muscle asymmetry index. Total length of trajectory Average speed and load distribution ratio To achieve the output target, a radial basis function neural network is trained. The training process includes normalizing the input vector, using a clustering algorithm to determine the centers of the radial basis functions, determining the width parameter based on the distance between the centers, and using the least squares method to solve for the weights of the output layer.

4. The method according to claim 3, characterized in that, In step S3, the definition of the optimization problem specifically includes: S31: Set constraints, including: anatomical boundary constraints, requiring the candidate jaw position to be located inside the jaw position edge motion envelope defined in step S1; joint space constraints, requiring the anterior, superior, and posterior spaces of the bilateral temporomandibular joints under the candidate jaw position to be within the physiologically safe range; and muscle symmetry constraints, requiring the muscle asymmetry index of the candidate jaw position predicted by the neural network to be lower than a preset threshold. S32: Construct the comprehensive fitness function F, whose expression is: ； in, , and These represent the total lengths of the plantar pressure center trajectory predicted by the neural network. Average movement speed and bipedal load distribution ratio The scores after standardization These are weighting coefficients that are dynamically allocated based on the needs of the sport.

5. The method according to claim 4, characterized in that, In step S4, the optimization using a genetic algorithm specifically includes: S41: Initialization: Within the anatomical boundary constraints, a preset number of initial jaw position vectors are randomly generated using a real number encoding method to form an initial population; S42: Evaluation and Selection: Input the jaw position vector of each individual in the population into the radial basis function neural network trained in step S2. First, determine whether it satisfies the joint space constraint and muscle symmetry constraint, and eliminate individuals that do not meet the constraints; for individuals that meet the constraints, select the appropriate individuals based on their predictions. , , The value of the comprehensive fitness function F is calculated as its fitness; a selection operation is performed based on the fitness to retain superior individuals; S43: Iterative Evolution: Perform crossover and mutation operations on the selected population to generate a new generation of individuals, ensuring that the new individuals still satisfy the anatomical boundary constraints; repeat steps S42 and S43 until the fitness function is integrated. The value tends to stabilize and no longer increases significantly. At this point, the individual with the highest fitness is the personalized musculoskeletal stable position obtained through optimization.

6. The method according to claim 1, characterized in that, In step S5, the digital hierarchical design specifically includes: S51: Realignment and Base Design: Based on the spatial transformation matrix corresponding to the personalized musculoskeletal stability position, adjust the position of the mandibular model relative to the maxillary model; and align the oral scan dental arch data with the realigned maxillary and mandibular models; then, based on the wearing lateral dental arch model, perform undercutting, edge line definition and smoothing to generate the flexible base layer that adapts to the dentition and gingival morphology. S52: Retention component design: In the posterior tooth region of the flexible base layer, an arc-shaped solid with a thickness equal to the vertical occlusal gap under the musculoskeletal stable position is generated along the dental arch morphology; using the cusp and pit morphology of the opposing dentition as the shear body, Boolean subtraction is performed on the occlusal surface of the arc-shaped solid to form a precise interlocking occlusal guiding surface, which constitutes the posterior tooth retention component.

7. The method according to claim 4, characterized in that, The weighting coefficient The dynamic allocation strategy is as follows: for sports that emphasize static balance, increase the weight of the load allocation ratio W. For sports that emphasize dynamic agility and reaction speed, increase the weight of average speed V. .

8. The method according to claim 1, characterized in that, In step S6, the integrated molding manufacturing specifically involves: using 3D printing technology to print the digital model obtained in step S5, which includes a flexible base layer and a posterior tooth retention component, using a biocompatible resin material, and then performing post-curing treatment.

9. A power-enhancing mouthguard designed and manufactured using the method described in any one of claims 1-8, characterized in that, Including one-piece molded: The flexible base layer has an inner surface that perfectly conforms to the shape of the user's lateral dentition and gums, and is used to absorb and disperse impact forces. The posterior tooth retention component, connected to the posterior tooth region of the flexible base layer, has an occlusal surface with a three-dimensional morphology that precisely complements the cusp and pit fissure of the opposing tooth in a personalized musculoskeletal stable position, used to guide and lock the user's mandible into this optimal functional position during wear.

10. A non-volatile computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the enhanced mouthguard design method based on musculoskeletal stability position optimization as described in any one of claims 1-8.

Images