A risk scoring and recommendation system for malocclusion during the mixed dentition period in children
The multi-source data fusion-based malocclusion risk scoring system for children during the mixed dentition period enables dynamic tracking and individualized assessment of maxillofacial development. It solves the problems of single assessment mode and lack of future trend prediction in existing technologies, and provides accurate risk assessment and intervention recommendations.
Patent Information
- Application Number
- CN202611213878.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-11
- Publication Date
- 2026-11-10
AI Technical Summary
Current technologies for assessing the risk of malocclusion during the mixed dentition period in children are based on a single snapshot, which cannot capture the dynamic changes in maxillofacial development over several years. They lack cross-validation of multi-source data, rely on subjective experience, lack a closed-loop management mechanism, and each assessment method is independent, making it impossible to identify hidden risk factors and lacking future trend prediction and individualized assessment.
The study employs a perioral biomechanics monitoring module, a maxillofacial development imaging tracking module, a jawbone growth potential assessment module, and a maxillofacial development dynamics analysis module. By fusing multi-source data, a dynamic model is constructed to achieve dynamic tracking and individualized assessment of maxillofacial development, and to output a comprehensive risk score and intervention recommendations.
It enables longitudinal tracking and dynamic trend analysis of maxillofacial development during the mixed dentition period, quantifies oral bad habits, assesses individualized growth potential, predicts the severity of future malocclusion, identifies hidden risk factors, and provides precise intervention recommendations, thereby improving the comprehensiveness and reliability of the assessment and forming a closed-loop management system.
Smart Images

Figure CN123072707A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of orthodontics and medical monitoring technology, and in particular relates to a risk scoring and recommendation system for malocclusion during the mixed dentition period in children. Background Technology
[0002] The mixed dentition period (6-12 years old) is a critical period for children's maxillofacial growth and development, and also the optimal window for early intervention of malocclusion. Currently, clinical risk assessment for malocclusion in children during the mixed dentition period mainly relies on clinical examination and imaging examinations during a single visit. Clinical examination assesses facial symmetry, dentition alignment, and occlusion through visual inspection and palpation. Oral habits (such as thumb sucking, lip biting, and tongue thrusting) are obtained by questioning parents and children, and their presence and approximate degree are recorded qualitatively. Imaging examinations involve taking lateral cephalometric X-rays, panoramic radiographs, or intraoral scans, followed by cephalometric analysis by the physician to obtain static maxillofacial morphological parameters such as the SNA angle, SNB angle, ANB angle, and mandibular plane angle. Some medical institutions also use wrist bone X-rays or cervical spine maturity assessments to determine the child's developmental stage, aiding in determining the timing of intervention. These methods constitute the main technical means of malocclusion risk assessment in current clinical practice.
[0003] However, the existing technologies mentioned above have the following shortcomings: First, the assessment model is a single snapshot, with intervals between visits ranging from 6 to 12 months. This fails to capture the dynamic changes in maxillofacial development over several years during the mixed dentition period, making it difficult to detect missed optimal intervention windows. Second, the assessment of oral habits is crude, only qualitatively recording their presence without considering quantitative dimensions such as frequency, duration, and intensity, resulting in the inability to objectively quantify the cumulative effect of bad habits. Third, bone age assessment is only used to qualitatively determine the growth and development stage (e.g., "before growth spurt" or "after growth spurt"), without quantitatively calculating remaining growth, and lacking core indicators to determine the urgency of intervention. Fourth, it can only assess the current severity of malocclusion, lacking the ability to predict future malocclusion trends, and cannot answer the key clinical question of "how severe will it become if intervention is not implemented now?" Fifth, relying solely on perioral muscle strength monitoring or imaging morphology assessment lacks a multi-source data cross-validation mechanism, failing to identify hidden risk factors not directly monitored. Sixth, risk assessment relies on the physician's subjective experience, lacking a multi-dimensional, objective, and quantitative scoring system, leading to poor consistency in judgments among different physicians. Seventh, risk assessment and intervention recommendations are separated, lacking a closed-loop management mechanism based on dynamic monitoring data. Eighth, existing technologies employ independent assessment methods, lacking a collaborative architecture that integrates and analyzes multi-source data such as muscle strength, imaging, and growth potential, thus limiting the overall assessment effectiveness. Summary of the Invention
[0004] To overcome the aforementioned deficiencies of the prior art, this invention provides a risk scoring and recommendation system for malocclusion during the mixed dentition period in children. This system addresses the problems in the prior art, which features a single, static assessment model, qualitative description of bad habits, no quantitative calculation of remaining growth, no prediction of future trends, no cross-validation of multi-source data, reliance on subjective experience for risk assessment, no closed-loop management mechanism, and the fact that various assessment methods are independent and lack integration.
[0005] To achieve the above objectives, the present invention provides the following technical solution:
[0006] A risk scoring and recommendation system for malocclusion in children during the mixed dentition period includes:
[0007] The perioral biomechanics monitoring module is used to collect perioral muscle strength data of children and calculate the cumulative exposure dose of abnormal pressure based on the perioral muscle strength data.
[0008] The maxillofacial development image tracking module is used to guide the acquisition of children's maxillofacial images via a mobile terminal, and to perform image processing on the maxillofacial images to extract time-series data of maxillofacial morphological parameters.
[0009] The jawbone growth potential assessment module is used to calculate the remaining growth equivalent of the jawbone based on the child's bone age assessment results.
[0010] The maxillofacial development dynamics analysis module is used to construct a dynamic model of maxillofacial development based on the cumulative exposure dose, the time series data, and the remaining growth equivalent of the jawbone, and to perform forward prediction and inverse dynamics inversion analysis.
[0011] The risk scoring and recommendation output module is used to calculate a comprehensive risk score based on the analysis results of the maxillofacial development dynamics analysis module and output intervention recommendations.
[0012] The perioral biomechanics monitoring module is communicatively connected to the maxillofacial development image tracking module. When the cumulative exposure dose exceeds a preset threshold, the perioral biomechanics monitoring module sends a trigger signal to the maxillofacial development image tracking module to control the maxillofacial development image tracking module to acquire maxillofacial images. The preset threshold is 1.5 times the average cumulative exposure dose of people of the same age and sex.
[0013] Preferably, the perioral biomechanics monitoring module includes at least one perioral biomechanics sensing unit, which comprises an electromyographic sensing element and a pressure sensing element.
[0014] Preferably, the sensing unit is a sensing patch, which includes: a lip patch for attaching to the inner side of the upper lip and the inner side of the lower lip to collect lip muscle pressure signals; a tongue patch for attaching to the back of the tongue to collect tongue pressure signals; and a cheek patch for attaching to the buccal mucosa on both sides to collect buccal muscle pressure signals.
[0015] The pressure sensing element has no fewer than 200 pressure-sensitive units, and its sensitivity in the pressure range of 0 to 2.5 kPa is no less than 42 kPa⁻. 1 The sampling frequency of the sensor patch is not less than 100 Hz, and it is used to continuously collect abnormal pressure of the lip muscles, abnormal pressure of the tongue, and abnormal pressure of the cheek muscles.
[0016] Preferably, the perioral biomechanical monitoring module calculates the cumulative exposure dose according to the following method:
[0017] The dose value of a single abnormal stress event is obtained by multiplying the difference between the peak pressure of a single abnormal stress event and the normal pressure threshold of a group of people of the same age and sex by the duration of the abnormal stress event, and then by the frequency weighting coefficient of the abnormal stress event.
[0018] The cumulative exposure dose is obtained by summing the dose values of all abnormal stress events.
[0019] Preferably, the maxillofacial development image tracking module performs the following functions through a mobile terminal application: guiding the shooting angle through the mobile terminal's gyroscope and facial recognition algorithm, guiding the shooting distance through the facial detection frame, guiding the child to complete the actions of naturally closing the lips, smiling and showing teeth, and opening the mouth to expose the dental arch through voice and animation, and performing quality checks on the clarity, exposure, and angle deviation of the photos; the facial recognition algorithm guides the shooting angle by recognizing the angle between the line connecting the two pupils and the horizontal direction, and determines the oral cavity exposure status by recognizing the integrity of the lip contour and the crown area;
[0020] The maxillofacial development image tracking module collects the following images according to a preset cycle: frontal smiling photo, frontal tooth-showing photo, left 45-degree side view photo, right 45-degree side view photo, maxillary dental arch view photo, and mandibular dental arch view photo; the preset cycle is dynamically adjusted according to the comprehensive risk score.
[0021] Preferably, the maxillofacial development image tracking module extracts the following maxillofacial morphological parameters from the maxillofacial image using an image processing algorithm: dental arch width, anterior overbite depth, anterior overjet distance, deviation between the maxillary dental midline and the facial midline, deviation between the mandibular dental midline and the facial midline, dental arch symmetry index, and facial symmetry index; the image processing algorithm identifies and marks the pixel coordinates of facial soft tissue landmarks and crown landmarks in the maxillofacial image, and calculates the maxillofacial morphological parameters based on the Euclidean distance between the pixel coordinates;
[0022] The extracted maxillofacial morphological parameters were arranged in a time series, and the rate of change and acceleration of change of each parameter were calculated.
[0023] Preferably, the jawbone growth potential assessment module obtains the child's bone age in the following ways: determining the cervical bone maturity stage based on the morphological changes of the second, third, and fourth cervical vertebrae in the child's lateral cephalometric X-ray; or determining the bone age based on the degree of ossification center development in the child's left wrist anteroposterior X-ray.
[0024] The jawbone growth potential assessment module calculates the remaining growth equivalent of the jawbone in the following way: obtain the reference value of the maximum jawbone growth of the child's population, multiply the reference value by the remaining growth ratio function, which is obtained by fitting population growth and development data with bone age, actual age and gender as independent variables.
[0025] Preferably, the dynamic model of the maxillofacial development dynamic analysis module is as follows: subtract the product of the tissue damping coefficient and the current displacement velocity from the net abnormal external force acting on the jawbone or dental arch, and then divide by the effective mass of the jawbone or dental arch to obtain the displacement acceleration of the jawbone or dental arch in three-dimensional space; the tissue damping coefficient is determined according to the remaining growth equivalent of the jawbone, and the larger the remaining growth equivalent of the jawbone, the smaller the tissue damping coefficient;
[0026] The maxillofacial development dynamics analysis module, based on the dynamic model, takes the current state as the initial condition, inputs the time series of the net abnormal external force and the remaining growth equivalent of the jawbone, and positively extrapolates the displacement of the jawbone or dental arch at each future time point to predict the future development trend of malocclusion severity.
[0027] Preferably, the maxillofacial development dynamics analysis module substitutes the jawbone displacement trajectory measured by the maxillofacial development image tracking module into the dynamic equation, and reversely derives the equivalent abnormal resultant force actually borne by the child in the past period of time, that is: multiply the effective mass of the jawbone or dental arch by the measured displacement acceleration, and add the product of the tissue damping coefficient and the measured displacement velocity to obtain the equivalent abnormal resultant force.
[0028] The equivalent abnormal resultant force is compared with the cumulative exposure dose calculated by the perioral biomechanics monitoring module; when the equivalent abnormal resultant force is greater than the cumulative exposure dose and the difference exceeds 20% of the cumulative exposure dose, a deep screening command is triggered; when the cumulative exposure dose is greater than the equivalent abnormal resultant force and the difference exceeds 30% of the cumulative exposure dose, the comprehensive risk score is reduced.
[0029] Preferably, the risk scoring and suggestion output module calculates the developmental deviation index as follows: subtract the mean displacement acceleration of the individual jawbone or dental arch from the mean displacement acceleration of children of the same bone age and sex, and then divide by the standard deviation of displacement acceleration of children of the same bone age and sex to obtain the developmental deviation index; when the developmental deviation index is greater than 1.5, it is judged as a significant deviation;
[0030] The risk scoring and suggestion output module calculates the comprehensive risk score as follows: the developmental deviation risk sub-score is multiplied by a first dynamic weighting coefficient, the cumulative exposure risk sub-score is multiplied by a second dynamic weighting coefficient, the growth potential depletion risk sub-score is multiplied by a third dynamic weighting coefficient, and the trend risk sub-score is multiplied by a fourth dynamic weighting coefficient to obtain the comprehensive risk score; the first, second, third, and fourth dynamic weighting coefficients are automatically adjusted according to the child's bone age stage; the cumulative exposure risk sub-score is corrected based on the remaining growth equivalent of the jawbone, and the smaller the remaining growth equivalent of the jawbone, the higher the cumulative exposure risk sub-score;
[0031] The risk scoring and recommendation output module calculates the critical intervention cutoff point based on the forward prediction results of the maxillofacial development dynamics analysis module, predicts the time point at which the displacement of the jawbone or dental arch reaches the preset skeletal malocclusion threshold, and outputs the following intervention recommendations: current bone age, estimated remaining equivalent growth of the jawbone, current displacement acceleration value, estimated remaining time to reach the preset skeletal malocclusion threshold, and recommended time range for initiating intervention.
[0032] The technical effects and advantages of the present invention, a risk scoring and recommendation system for malocclusion during the mixed dentition period in children:
[0033] 1. This invention uses a maxillofacial development image tracking module to collect children's maxillofacial images at preset intervals. It then arranges the extracted maxillofacial morphological parameters in a time series, calculating the rate and acceleration of change for each parameter, thus achieving longitudinal tracking and dynamic trend analysis of maxillofacial development during the mixed dentition period. This overcomes the shortcomings of traditional methods that only provide snapshot-style assessments during a single visit and cannot capture dynamic changes over several years during the mixed dentition period. Furthermore, the preset interval is dynamically adjusted based on a comprehensive risk score, with more frequent data collection during high-risk periods and longer intervals during low-risk periods, optimizing the operational burden on home users while ensuring monitoring effectiveness.
[0034] 2. This invention collects perioral muscle strength data in children through a perioral biomechanics monitoring module and calculates the cumulative exposure dose based on the peak pressure, duration, and frequency weighting coefficients of abnormal pressure events. For the first time, it elevates oral hygiene habits from a qualitative description (presence / absence) to a calculable and traceable quantitative indicator. Simultaneously, by identifying and marking the pixel coordinates of facial soft tissue landmarks and crown landmarks in the maxillofacial images, and calculating maxillofacial morphological parameters based on the Euclidean distance between the pixel coordinates, it achieves precise quantification of maxillofacial morphological changes, providing an objective data basis for risk scoring.
[0035] 3. This invention obtains a child's individualized bone age (based on cervical vertebral bone maturity stage or wrist bone ossification degree) through a jawbone growth potential assessment module, and calculates the remaining growth equivalent of the jawbone based on bone age, actual age, and gender, thus incorporating individualized growth potential into the risk assessment system. Simultaneously, the cumulative exposure risk sub-score in this invention is adjusted based on the remaining growth equivalent of the jawbone; the smaller the remaining growth equivalent, the higher the cumulative exposure risk sub-score. This ensures that the same undesirable habit carries different risk weights for children with different growth potentials, truly achieving individualized and accurate assessment.
[0036] 4. This invention constructs a dynamic model through a maxillofacial development dynamics analysis module. Based on net abnormal external force, tissue damping coefficient, current displacement velocity, and effective jawbone mass, it calculates displacement acceleration and uses the current state as initial conditions to forward extrapolate jawbone displacement at future time points. For the first time, it upgrades the assessment of malocclusion risk from "how severe it is now" to "how severe it will be in the future if no intervention is provided now." Simultaneously, by calculating the critical intervention cutoff point and predicting the time point when jawbone displacement reaches the preset skeletal malocclusion threshold, it outputs intervention recommendations including current bone age, estimated remaining growth equivalent, current displacement acceleration value, estimated remaining time to reach the preset skeletal malocclusion threshold, and suggested intervention time range, providing precise time window guidance for clinical decision-making.
[0037] 5. This invention, by substituting the jawbone displacement trajectory measured by the maxillofacial development imaging tracking module into the dynamic equation, reversely derives the equivalent abnormal resultant force actually borne by the child, and compares this equivalent abnormal resultant force with the cumulative exposure dose directly monitored by the perioral biomechanics monitoring module. When the equivalent abnormal resultant force is significantly greater than the cumulative exposure dose, a deep screening command is triggered, effectively identifying hidden risk factors (such as nocturnal bruxism, sleep apnea, etc.) that cannot be detected by a single monitoring method, making up for the shortcomings of relying solely on surface muscle strength monitoring, and improving the comprehensiveness and reliability of risk identification.
[0038] 6. This invention obtains a comprehensive risk score by multiplying the developmental deviation risk sub-score, cumulative exposure risk sub-score, growth potential depletion risk sub-score, and trend risk sub-score by their respective dynamic weighting coefficients and then summing them. Each dynamic weighting coefficient automatically adjusts according to the child's bone age stage, constructing a multi-dimensional objective quantitative scoring system. This overcomes the shortcomings of traditional methods, such as reliance on physicians' personal experience for risk assessment, lack of unified objective standards, high subjectivity, and poor repeatability.
[0039] 7. This invention establishes an instantaneous causal relationship between abnormal external force and jawbone morphological response by connecting the perioral biomechanics monitoring module and the maxillofacial development imaging tracking module. When the cumulative exposure dose exceeds a preset threshold, image acquisition is automatically triggered. Simultaneously, by simulating developmental deviation trends under different abnormal pressure reduction ratios and correcting prediction maps, this prediction map is sent to a mobile terminal for display, allowing parents to intuitively see the specific benefits of behavioral correction and significantly improving treatment adherence at home. Furthermore, the system continues to operate after intervention, recalculating the comprehensive risk score based on new data and automatically adjusting the acquisition cycle and intervention recommendations, forming a closed-loop management system of "monitoring → assessment → intervention → remonitoring," providing a systematic solution for the long-term management of malocclusion during the mixed dentition period.
[0040] 8. The invention's perioral biomechanics monitoring module, maxillofacial development imaging tracking module, jawbone growth potential assessment module, maxillofacial development dynamics analysis module, and risk scoring and recommendation output module work together synergistically: the perioral biomechanics monitoring module provides quantitative input of abnormal external forces; the maxillofacial development imaging tracking module provides measured responses of jawbone displacement; the jawbone growth potential assessment module provides individualized parameters of tissue compliance; the maxillofacial development dynamics analysis module integrates the above three sources of data to construct dynamic equations to achieve forward prediction and inverse dynamics inversion; and the risk scoring and recommendation output module outputs a comprehensive risk score and intervention recommendations based on the analysis results. These five modules form a complete causal chain closed loop, and their overall technical effect far exceeds the simple sum of the effects of each module, realizing intelligent decision support throughout the entire process from data collection to precise intervention. Attached Figure Description
[0041] Figure 1 This is an overall system block diagram of a risk scoring and recommendation system for malocclusion during the mixed dentition period proposed in this invention;
[0042] Figure 2 This is an event-driven image acquisition triggering flowchart for a risk scoring and recommendation system for malocclusion during the mixed dentition period proposed in this invention. Detailed Implementation
[0043] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.
[0044] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include," "contain," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "includes..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.
[0045] refer to Figure 1-2 This invention provides a risk assessment and recommendation system for malocclusion during the mixed dentition period in children. The invention is described in detail below with reference to specific embodiments, but the implementation of the invention is not limited thereto. Those skilled in the art can make similar modifications without departing from the spirit of the invention; therefore, the invention is not limited to the specific embodiments disclosed below.
[0046] The following embodiments are described with specific numerical values and operating steps, and are only intended to help those skilled in the art fully understand the present invention, and should not be regarded as limiting the scope of protection of the present invention. Where specific technical conditions are not specified in the embodiments, they should be performed according to conventional technical conditions in the art or conditions recommended by the manufacturer.
[0047] Example 1
[0048] Purpose of implementation:
[0049] The purpose of this embodiment is to provide a specific implementation of the perioral biomechanical monitoring module in a risk assessment and recommendation system for malocclusion during the mixed dentition period in children. It utilizes a flexible wearable sensor patch to continuously collect and quantify the perioral muscle strength of children, elevating the qualitative description of oral habits to a calculable cumulative exposure dose, thus providing accurate quantitative input data for subsequent risk assessment. Simultaneously, it verifies the wearing comfort of the sensor patch, the stability of signal acquisition, and the feasibility of the event-driven acquisition mechanism.
[0050] Implementation System:
[0051] The perioral biomechanics monitoring module in this embodiment includes three sensing patches: a lip patch, a tongue patch, and a buccal patch. Each sensing patch uses medical-grade silicone as a flexible substrate with a thickness of 0.3 mm. Each sensing patch comprises a metal electrode layer and a hydrogel layer, with the metal electrode layer positioned between the flexible substrate and the hydrogel layer. The pressure sensor is a flexible piezoresistive sensor with 256 pressure-sensitive units and a sensitivity of 45 kPa⁻ within a pressure range of 0 to 2.5 kPa.1 The pressure-sensitive unit employs a composite microporous structure of multi-walled carbon nanotubes and polydimethylsiloxane as the piezoresistive sensitive layer, which exhibits linear response characteristics in the low-pressure range. The surface electromyography electrodes are a silver / silver chloride wet electrode array with an electrode diameter of 5 mm and an electrode spacing of 10 mm.
[0052] The lip patches are attached to the inner mucosa of the upper and lower lips to collect lip muscle pressure signals. Specifically, the upper lip patch is located 5 mm on each side of the upper lip frenulum, and the lower lip patch is located 5 mm on each side of the lower lip frenulum. The tongue patch is attached to the back of the tongue approximately 15 mm from the tip, with a suction cup-like fixation structure on the side facing away from the electrode layer to adhere to the back of the tongue and prevent displacement during swallowing and speech. The buccal patches are attached to the buccal mucosa corresponding to the first molar region on both sides, one on each side.
[0053] Each sensor patch continuously collects lip muscle pressure, tongue pressure, and cheek muscle pressure signals at a sampling frequency of 100Hz. The sensor patches are connected to a signal acquisition and processing unit via flexible circuitry. This unit is integrated into the external wearable main unit, which transmits data with the home gateway device, i.e., a smartphone, via Bluetooth Low Energy protocol.
[0054] The signal acquisition and processing unit performs the following preprocessing on the raw pressure signal: First, a 50Hz notch filter is used to eliminate power frequency interference; second, a 0.5Hz high-pass filter is used to eliminate baseline drift; third, a finite impulse response low-pass filter (cutoff frequency 10Hz) is used to filter out high-frequency noise.
[0055] The rules for identifying abnormal stress events are as follows:
[0056] Abnormal lip muscle pressure: When the resting lip pressure consistently exceeds the normal pressure threshold for the same age and sex for more than 2 seconds, it is defined as an abnormal lip closure pressure event; when the peak pressure generated by sucking action exceeds the normal sucking pressure threshold, it is defined as a sucking habit event. The normal pressure threshold is based on the 95th percentile of the epidemiological survey data on resting lip pressure in children aged 3 years and above.
[0057] Abnormal tongue pressure: When the resting tongue pressure exceeds the normal tongue pressure threshold for the same age and sex for more than 3 seconds, it is considered an abnormal tongue position event; when the center of tongue pressure distribution is located in the anterior teeth area rather than the palate during swallowing, it is considered an abnormal swallowing event. The normal tongue pressure threshold is based on the 95th percentile of the epidemiological survey data of resting tongue pressure in children over 3 years old.
[0058] Abnormal cheek muscle pressure: When cheek pressure exhibits a periodic increase pattern and the peak value exceeds the normal threshold, it is identified as a cheek-sucking habit event.
[0059] The data processing unit of the perioral biomechanics monitoring module calculates the cumulative exposure dose as follows: the difference between the peak pressure of a single abnormal pressure event and the normal pressure threshold of the same age and sex population is multiplied by the duration of the abnormal pressure event, and then multiplied by the frequency weighting coefficient of the abnormal pressure event to obtain the dose value of the abnormal pressure event; the dose values of all abnormal pressure events are summed to obtain the cumulative exposure dose.
[0060] Taking an 8-year-old male child with a thumb-sucking habit as an example: the child sucked his thumb 15 times in 24 hours. For the third thumb-sucking event, the peak pressure was 2.1 kPa, the normal pressure threshold for the same age and sex was 0.3 kPa (the upper limit of normal pressure for thumb-sucking), the duration was 8 seconds, and the frequency weighting coefficient was 1.0 (thumb-sucking is a high-frequency bad habit, with the highest weighting coefficient). Therefore, the dose value for this event was: (2.1-0.3)×8×1.0=14.4 Pa·s.
[0061] The dose values from all 15 events were summed to obtain the child's cumulative exposure dose for the day as 156.8 Pa·s. This value was compared with the average cumulative exposure dose for the same age and sex population, which was 85.3 Pa·s based on epidemiological survey data from over 1000 individuals of the same age and sex. Since 156.8 Pa·s > 85.3 × 1.5 = 127.95 Pa·s, the cumulative exposure dose exceeded the preset threshold. The perioral biomechanics monitoring module sent a trigger signal to the maxillofacial development image tracking module, controlling the module to immediately acquire maxillofacial images, thus achieving event-driven acquisition.
[0062] Implementation results:
[0063] This embodiment achieves highly sensitive, multi-channel continuous monitoring of children's perioral muscle strength using a flexible piezoresistive sensor array. The sensor sensitivity is 45 kPa⁻ 1 This ensures the detectability of minute pressure changes; 256 pressure-sensitive units guarantee the spatial resolution of pressure distribution; and the distributed deployment of three patches (lip, tongue, and cheek) enables comprehensive capture of multi-source abnormal pressure. More importantly, through the quantitative calculation of cumulative exposure dose, oral hygiene habits are elevated from qualitative descriptions (whether thumb-sucking exists) to calculable and traceable quantitative indicators (daily cumulative exposure dose of 156.8 Pa·s), providing accurate input data for subsequent kinetic analysis and risk scoring. The event-driven acquisition mechanism automatically triggers image acquisition when the cumulative exposure dose exceeds the threshold, establishing an instantaneous causal relationship between abnormal external force and jawbone morphological response, providing a time-aligned data foundation for subsequent inverse kinetic inversion and cross-validation.
[0064] Clinical trial results show that the sensor patch is comfortable to wear in children, with an adverse skin reaction rate of 2.3% after 24 hours of continuous wear, and an effective signal acquisition rate of 97.5%, meeting the practical requirements for long-term home monitoring.
[0065] Example 2
[0066] Purpose of implementation:
[0067] The purpose of this embodiment is to provide specific implementation methods for the maxillofacial development image tracking module and the maxillofacial development dynamics analysis module, to verify the feasibility and accuracy of standardized home-based image acquisition, to verify the effectiveness of the method for extracting the trend of maxillofacial morphological parameters based on time-series image analysis, to verify the clinical accuracy of the forward prediction and inverse dynamics analysis methods based on dynamic models, and to verify the effectiveness of the cross-validation mechanism between inverse dynamics inversion and measured data in identifying hidden risk factors.
[0068] Implementation System:
[0069] In this embodiment, the maxillofacial development image tracking module is implemented in the form of a mobile terminal application. The application includes the following guidance functions.
[0070] Regarding shooting angle guidance, the application uses the mobile terminal's gyroscope sensor to obtain the phone's posture data, and simultaneously uses a facial recognition algorithm to identify the position of the child's pupils in the frame. Specifically, the facial recognition algorithm calculates the angle between the line connecting the pupils and the horizontal direction. When the angle exceeds 2°, the application prompts the user with arrows and text on the interface to adjust the phone angle until the angle is less than 2°. At the same time, the algorithm identifies the tip of the nose and the chin point to ensure that the deviation between the facial midline and the vertical midline of the image does not exceed 3mm.
[0071] Regarding shooting distance guidance, the application uses a face detection bounding rectangle to provide real-time prompts on the shooting distance. Specifically, the algorithm identifies the bounding rectangle of the face, and when the height of the rectangle is not within 60% to 80% of the screen height, it prompts the user to move closer to or further away from the child until the requirement is met.
[0072] Regarding oral exposure guidance, the app uses voice and animation prompts to guide children in performing standard actions. For the smiling action, the algorithm identifies the curvature of the lip contour and the exposed area of the crown region. When less than 2 / 3 of the crown of the maxillary anterior teeth is exposed, the app prompts the child to open their mouth wider and smile. For the mouth-opening action to expose the dental arch, the algorithm identifies the distance between the upper and lower lips and the integrity of the crown region. When the distance between the upper and lower lips is less than 15mm or the crown region is obscured, the app prompts the child to open their mouth wider.
[0073] Regarding quality verification, the application automatically detects the image's sharpness (based on the image's Laplacian variance, with a threshold of 50), exposure (based on the image's brightness histogram, ensuring that the midtone area accounts for no less than 60%), and angle deviation. If the image fails to meet the requirements, a retake is prompted, with a maximum of 3 retakes.
[0074] Six standard images are collected according to a preset cycle: a frontal smiling photo, a frontal photo showing teeth, a 45° side view from the left, a 45° side view from the right, a maxillary dental arch view, and a mandibular dental arch view. The preset cycle is initially set to once a month. When the comprehensive risk score rises to 60 points or above, the system automatically shortens the cycle to once every two weeks.
[0075] For capturing maxillary dental arch facial images, the application instructs the child to tilt their head back until their face is at approximately a 60° angle to the ground, and then use an oral retractor to expose the dental arch before taking the image. For mandibular dental arch facial images, the application instructs the child to tilt their head down until their face is at approximately a 60° angle to the ground before taking the image.
[0076] The application encrypts and uploads the captured images to a cloud server. The cloud server deploys a deep learning-based facial landmark detection model (using a stacked hourglass network architecture, trained on a dataset of 5000 annotated facial landmark images), which automatically identifies and labels the pixel coordinates of the following landmarks: facial soft tissue landmarks including the center of the left pupil, the center of the right pupil, the tip of the nose, the left corner of the mouth, the right corner of the mouth, the midpoint of the upper lip, the midpoint of the lower lip, the anterior point of the chin, the left angle of the mandible, and the right angle of the mandible; and dental crown landmarks including the cusp of the left maxillary canine, the cusp of the right maxillary canine, the mesobuccal cusp of the left maxillary first molar, the mesobuccal cusp of the right maxillary first molar, the cusp of the left mandibular canine, the cusp of the right mandibular canine, the mesobuccal cusp of the left mandibular first molar, the mesobuccal cusp of the right mandibular first molar, the midpoint of the incisal edge of the maxillary central incisor, and the midpoint of the incisal edge of the mandibular central incisor.
[0077] The maxillofacial morphological parameters are calculated based on the Euclidean distance between the pixel coordinates of the aforementioned markers. Taking the dental arch width as an example, the width between the maxillary canines is equal to the pixel distance between the cusps of the maxillary left canine and the maxillary right canine multiplied by the calibration scale (in mm / pixel); the width between the maxillary first molars is equal to the pixel distance between the mesiobuccal cusps of the maxillary left first molar and the mesiobuccal cusps of the maxillary right first molar multiplied by the calibration scale.
[0078] The extracted maxillofacial morphological parameters were arranged in chronological order. Taking the follow-up data of an 8-year-old male child in the mixed dentition stage over 6 consecutive months as an example, at the initial month, the intercanine width was 28.4 mm, the overbite depth was 3.2 mm, and the maxillary midline deviation was 0.8 mm; at the first month, the intercanine width was 28.6 mm, the overbite depth was 3.5 mm, and the maxillary midline deviation was 0.9 mm; at the second month, the intercanine width was 28.3 mm, the overbite depth was 3.8 mm, and the maxillary midline deviation was... The midline deviation was 1.1 mm; at the 3rd month, the intercanine width was 28.1 mm, the overbite depth was 4.2 mm, and the midline deviation was 1.4 mm; at the 4th month, the intercanine width was 27.8 mm, the overbite depth was 4.7 mm, and the midline deviation was 1.8 mm; at the 5th month, the intercanine width was 27.5 mm, the overbite depth was 5.1 mm, and the midline deviation was 2.1 mm.
[0079] Calculate the rate of change (change per unit time) and acceleration (change in rate of change per unit time) of each parameter. The rate of change of the anterior overbite depth is (5.1-3.2) / 5 = 0.38 mm / month; the acceleration of the anterior overbite depth is approximately 0.06 mm / month. 2 (The trend is accelerating and deepening). The rate of change in the width between the maxillary canines is (27.5-28.4) / 5 = -0.18 mm / month (showing a narrowing trend). The rate of change in the maxillary midline deviation is (2.1-0.8) / 5 = 0.26 mm / month (the deviation is accelerating and increasing).
[0080] In this embodiment, the child's actual age is 8 years and 2 months. The jawbone growth potential assessment module determines the cervical vertebral maturity stage based on the morphological changes of the 2nd, 3rd, and 4th cervical vertebrae in the child's lateral cephalometric X-ray. Specifically, the system's built-in deep learning model (based on a convolutional neural network, trained on a dataset of 3000 lateral X-rays labeled with cervical vertebral maturity stages) automatically identifies the cervical vertebral body morphology and outputs the cervical vertebral maturity stage. The assessment determined that the child's cervical vertebral maturity stage is Stage 3 (accelerated growth phase).
[0081] The jawbone growth potential assessment module obtains the reference value for the maximum jawbone growth in the child's population. Based on epidemiological data of mandibular growth and development in 8-year-old male children in China (sample size: 500), the reference value for maximum growth is 12.5 mm. The residual growth ratio function is calculated using bone age, chronological age, and sex as independent variables, based on a Logistic growth curve model fitted to the population growth and development data. The child's bone age is 9.0 years (the bone age corresponding to stage 3 of cervical spine maturity), chronological age is 8.17 years, and sex is male. Substituting these values, the residual growth ratio function value is calculated to be 0.214. Therefore, the residual growth equivalent = 12.5 × 0.214 ≈ 2.68 mm.
[0082] The tissue damping coefficient is determined based on the residual growth equivalent of the jawbone; the larger the residual growth equivalent, the smaller the tissue damping coefficient. Specifically, the tissue damping coefficient = 100 × exp(-0.15 × RGE), where RGE is the residual growth equivalent. When the residual growth equivalent is 2.68 mm, the tissue damping coefficient = 100 × exp(-0.15 × 2.68) = 100 × exp(-0.402) ≈ 66.9 N·mm / mm.
[0083] The dynamic model of the maxillofacial development dynamics analysis module is as follows: subtract the product of the tissue damping coefficient and the current displacement velocity from the net abnormal external force acting on the jawbone or dental arch, and then divide by the effective mass of the jawbone or dental arch to obtain the displacement acceleration of the jawbone or dental arch in three-dimensional space.
[0084] The net abnormal external force was calculated from the cumulative exposure dose (156.8 Pa·s) provided by the perioral biomechanical monitoring module, with a conversion factor of 0.02 N / (Pa·s), i.e., net abnormal external force = 156.8 × 0.02 = 3.14 N. The current displacement velocity is synthesized from the rate of change of the intercanine width of the maxillary teeth (-0.18 mm / month) and the rate of change of the anterior overbite depth (0.38 mm / month), v(t) = sqrt(0.18... 2 +0.38 2 =0.42mm / month. The effective mass of the jawbone is estimated based on the child's weight and jawbone ratio, with M=0.15kg.
[0085] Therefore, the current displacement acceleration a(t) = (3.14 - 66.9 × 0.42) / 0.15 = (3.14 - 28.10) / 0.15 = -166.4 mm / month 2 A negative value indicates that jawbone development is being inhibited by abnormal external forces, and the width of the dental arch is narrowing at an accelerated rate.
[0086] Using the current state as the initial condition, the time series of net abnormal external forces (extrapolated from the past 6 months of perioral biomechanical monitoring data) and the remaining growth equivalent are input to forward extrapolate the jawbone displacement at each future time point. The prediction results show that without intervention, the width between the maxillary canines of this child will narrow to 25.2 mm after 12 months (2.3 mm lower than the normal value of 27.5 mm for the same age), and the depth of the anterior overbite will deepen to 8.7 mm after 12 months (4-6 mm beyond the normal range), reaching the threshold of skeletal malocclusion.
[0087] The maxillofacial development dynamics analysis module substitutes the jawbone displacement trajectory measured by the maxillofacial development imaging tracking module into the dynamic equation, and reversely derives the equivalent abnormal resultant force actually borne by the child in the past 6 months. That is, the effective mass of the jawbone or dental arch is multiplied by the measured displacement acceleration, and the product of the tissue damping coefficient and the measured displacement velocity is added to obtain the equivalent abnormal resultant force.
[0088] Taking the data from the 5th month as an example, the measured displacement acceleration was 0.06 mm / month for the change in the overbite depth of the anterior teeth. 2 The acceleration of the width change between the maxillary canine and the maxillary canine is -0.04 mm / month. 2 The synthesis, a_measured = sqrt(0.06) 2 +(-0.04) 2 )=0.072mm / month 2 The measured displacement velocity v_measured = 0.42 mm / month. The equivalent abnormal resultant force F̂_net = 0.15 × 0.072 + 66.9 × 0.42 = 0.0108 + 28.10 ≈ 28.11 N.
[0089] The equivalent abnormal resultant force (28.11 N) was compared with the converted value of the cumulative abnormal pressure dose directly monitored by the perioral biomechanics monitoring module (3.14 N). The equivalent abnormal resultant force was significantly greater than the cumulative exposure dose, with a difference of approximately 24.97 N, exceeding 20% of the cumulative exposure dose (i.e., 0.63 N), indicating the presence of undetected abnormal functional factors. The system triggered a deep screening command, recommending that the clinician perform nocturnal sleep apnea monitoring and temporomandibular joint examination on the child.
[0090] Further clinical examination revealed that the child had a nighttime teeth grinding habit (the biting force generated by the masseter muscle during teeth grinding can reach 200-500N, far exceeding the thumb-sucking pressure monitored during the day), a factor that was not captured by the daytime perioral biomechanics monitoring module. The system successfully alerted to the potential, previously undetected risk factor.
[0091] Implementation results:
[0092] This embodiment achieves longitudinal tracking and dynamic trend analysis of maxillofacial morphological parameters during the mixed dentition period through standardized home-based image acquisition and cloud-based automatic image processing. Specific quantitative indicators set in the facial recognition algorithm, such as an angle between the two pupils less than 2°, a facial midline deviation of less than 3mm, and a crown exposure area greater than 2 / 3, ensure good comparability and repeatability of images acquired at different time points and by different operators. In the clinical trial (sample size 120 cases), the success rate for parents' first operation was 82.5%, which increased to 96.7% after one guidance session, and the proportion of images meeting clinical analysis requirements was 94.2%.
[0093] Through forward modeling using a kinetic model, a quantitative prediction of future malocclusion risk was achieved—specifically, the intercanine width of the maxillary teeth was predicted to narrow to 25.2 mm after 12 months, and the overbite depth of the anterior teeth was predicted to deepen to 8.7 mm after 12 months. Cross-validation with inverse kinetic modeling and measured data (a significant difference between the equivalent abnormal synergistic force of 28.11 N and the cumulative exposure dose conversion value of 3.14 N) successfully identified a potential risk factor (nocturnal bruxism) that could not be detected by a single monitoring method, demonstrating the synergistic advantages of multi-source data fusion analysis. In a clinical trial (sample size 120 cases, monitoring period 6 months), this cross-validation mechanism identified 11 occult risk factors (including 6 cases of nocturnal bruxism, 3 cases of sleep-disordered breathing, and 2 cases of abnormal swallowing not captured by daytime monitoring), with a positive predictive value of 78.6%, indicating that the mechanism has good risk warning capabilities.
[0094] Example 3
[0095] Purpose of implementation:
[0096] The purpose of this embodiment is to provide a complete application process for a risk scoring and recommendation system for malocclusion during the mixed dentition period in children, to verify the feasibility of the system's collaborative operation throughout the entire process from initialization, data collection, risk scoring, intervention recommendations to dynamic monitoring, to verify the clinical applicability and accuracy of functions such as developmental deviation index, comprehensive risk score, intervention time window prediction and behavioral intervention biofeedback, to verify the rationality of the adaptive mechanism that dynamically adjusts the preset period according to the comprehensive risk score, and to verify the superiority of this system over traditional assessment methods in real clinical scenarios.
[0097] Implementation System:
[0098] Patient Li, male, 8 years and 2 months old, presented to the clinic because his parents noticed that his upper anterior teeth were gradually protruding. During the initial oral examination, it was found that the child had a thumb-sucking habit (approximately 10-20 times per day), deep overbite (overbite depth 4.5 mm), and an intercanine width of 28.4 mm. The system initialization process is as follows.
[0099] First, the perioral biomechanics monitoring module: The child wears lip, tongue, and cheek patches for continuous monitoring for 72 hours to establish baseline data. Second, the jawbone growth potential assessment module: A lateral cephalometric X-ray is taken, and the system automatically assesses the cervical vertebrae maturity stage as Stage 3 (bone age 9.0 years), calculating the remaining growth equivalent to 2.68 mm. Third, the maxillofacial development image tracking module: Parents take six standard images using a mobile application, and the system extracts baseline maxillofacial morphological parameters.
[0100] Based on the standard curve of displacement acceleration in children of the same bone age and sex (established based on longitudinal follow-up data of more than 1,000 male children aged 8-10 years), the mean displacement acceleration μ_a (BA=9.0, Sex=Male) in this population is -50.0 mm / month. 2 (During normal development, there is a slight natural tendency for the width of the dental arch to narrow), standard deviation σ_a (BA=9.0, Sex=Male) = 20.0 mm / month 2 The child's current displacement acceleration a(t) = -166.4 mm / month. 2 (Calculated from Example 2).
[0101] The developmental deviation index (DDI) is calculated by subtracting the mean displacement acceleration of the individual's jawbone or dental arch from the mean displacement acceleration of children of the same bone age and sex, and then dividing by the standard deviation of displacement acceleration of children of the same bone age and sex. This child's DDI = (-166.4 - (-50.0)) / 20.0 = -5.82. Since the absolute value of DDI = -5.82 is much greater than 1.5, it is considered a significant deviation (a negative value indicates that the dental arch narrowing is significantly faster than the normal level for children of the same age).
[0102] The risk scoring and recommendation output module calculates the comprehensive risk score as follows: The developmental deviation risk sub-score is multiplied by the first dynamic weighting coefficient, plus the cumulative exposure risk sub-score multiplied by the second dynamic weighting coefficient, plus the growth potential depletion risk sub-score multiplied by the third dynamic weighting coefficient, and plus the trend risk sub-score multiplied by the fourth dynamic weighting coefficient. The first, second, third, and fourth dynamic weighting coefficients are automatically adjusted according to the child's bone age stage. The cumulative exposure risk sub-score is corrected based on the remaining growth equivalent of the jawbone; the smaller the remaining growth equivalent of the jawbone, the higher the cumulative exposure risk sub-score.
[0103] The specific values for each sub-score and weighting coefficient are as follows:
[0104] The developmental deviation risk sub-score is calculated based on the developmental deviation index. The developmental deviation index is -5.82, which corresponds to a percentile far below the 2.5 percentile (corresponding to a developmental deviation index of approximately -1.96), indicating an extreme deviation. The developmental deviation risk sub-score is 95 (out of 100, with higher scores indicating greater risk).
[0105] The cumulative exposure risk sub-score was calculated based on a cumulative exposure dose of 156.8 Pa·s. The mean cumulative exposure dose for the same age and sex population was 85.3 Pa·s, meaning this child's cumulative exposure dose was approximately 1.84 times the mean. Considering the remaining growth equivalent of 2.68 mm (in a period of accelerated growth with high tissue plasticity), the cumulative exposure risk sub-score was adjusted: the reference value for the remaining growth equivalent (mean for the same age population) is 3.5 mm, and the correction factor = 3.5 / 2.68 ≈ 1.31. The cumulative exposure risk sub-score = 70 × 1.84 × 1.31 ≈ 168.7, and the upper limit of 100 points was taken.
[0106] The risk sub-score for growth potential exhaustion was calculated based on a remaining growth equivalent of 2.68 mm. The normal range for remaining growth equivalent in children of the same age and sex is 2.0–5.0 mm. This child's score was within the normal range but slightly low, resulting in a risk sub-score of 65 for growth potential exhaustion.
[0107] The risk sub-score based on the acceleration of change in anterior overbite depth (0.06 mm / month) 2 (showing an accelerating and deepening trend) and the acceleration of the change in the width between the maxillary canines (-0.04 mm / month). 2 The calculation shows an accelerating narrowing trend. Both parameters exhibit an unfavorable development trend, with a risk sub-score of 80 points.
[0108] The dynamic weighting coefficients are automatically adjusted based on the child's bone age stage (Cervical spine bone maturity stage 3, growth acceleration period). During the growth acceleration period, the weights for the risks of growth potential depletion and trend changes are higher, while the weight for the risk of cumulative exposure is relatively lower (because growth plasticity can compensate for the influence of bad habits to some extent). The dynamic weighting coefficients are set as follows: Dynamic weighting coefficient 1 w1 = 0.25 (developmental deviation risk), Dynamic weighting coefficient 2 w2 = 0.15 (cumulative exposure risk), Dynamic weighting coefficient 3 w3 = 0.30 (growth potential depletion risk), and Dynamic weighting coefficient 4 w4 = 0.30 (trend changes risk).
[0109] The overall risk score R = 0.25×95 + 0.15×100 + 0.30×65 + 0.30×80 = 23.75 + 15 + 19.5 + 24 = 82.25 points.
[0110] The risk scoring and recommendation output module calculates the critical intervention cutoff point based on the forward prediction results from the maxillofacial development dynamics analysis module. The preset threshold for skeletal malocclusion is a maxillary canine interdental width of less than 26.0 mm or anterior overbite depth of greater than 6.5 mm.
[0111] Orthogonal modeling predicts that without intervention, the child's maxillary intercanine width will reach 26.0 mm (narrowing threshold) after 8 months, and the anterior overbite depth will reach 6.5 mm (deepening threshold) after 7 months. Taking the earlier of the two (7 months), it is predicted that irreversible skeletal malocclusion threshold will be reached after 7 months.
[0112] The system outputs the following intervention recommendations: The child's current bone age is 9.0 years, the estimated remaining growth equivalent is approximately 2.68 mm, and the current displacement acceleration is -166.4 mm / month. 2 (Accelerated narrowing of the dental arch and accelerated deepening of the overbite) It is expected that the irreversible skeletal malocclusion threshold will be reached after 7 months. It is recommended to start functional appliance intervention within 3 months, with myofunctional therapy devices combined with perioral muscle function training as the first choice. At the same time, it is recommended that parents help the child to quit the thumb-sucking habit. If thumb-sucking can be successfully stopped within 3 months and the cumulative exposure dose can be reduced by more than 50%, the intervention window is expected to be extended to 10-12 months.
[0113] The system simulates and visualizes dynamic predictions of how developmental deviations will be corrected if abnormal stress is immediately reduced by different proportions: if abnormal stress is reduced by 0% (no intervention), the skeletal malocclusion threshold is reached after 7 months; if abnormal stress is reduced by 30%, the skeletal malocclusion threshold is reached after 10 months; if abnormal stress is reduced by 50%, the skeletal malocclusion threshold is reached after 13 months; and if abnormal stress is reduced by 70%, the skeletal malocclusion threshold is reached after 16 months. This visualization is presented to parents via a mobile application.
[0114] The system continued to operate after the intervention was initiated. Images were collected monthly, and perioral muscle strength was continuously monitored daily. At the follow-up examination the following month, new imaging data showed that the width between the maxillary canines was 28.0 mm (narrowed by 0.4 mm from the previous month, with the narrowing rate decreasing from 0.18 mm / month to 0.10 mm / month), and the depth of the anterior overbite was 4.7 mm (deepening rate decreasing from 0.38 mm / month to 0.20 mm / month). Perioral biomechanical monitoring showed that the cumulative exposure dose decreased from 156.8 Pa·s to 89.2 Pa·s (a reduction of 43%).
[0115] The system recalculated the overall risk score based on the new data, resulting in a score of 68.5 (a significant decrease from the initial score of 82.25). The preset cycle was automatically adjusted from once a month to once every two months, and the intervention recommendation was updated to "continue the current treatment plan, with the expected intervention window extended to 11 months."
[0116] Implementation results:
[0117] This embodiment demonstrates the complete workflow of the system, from initialization, data collection, risk scoring, intervention recommendations, to dynamic monitoring. Through multi-dimensional quantification of the comprehensive risk score (covering four dimensions: developmental deviation, cumulative exposure, depletion of growth potential, and trend, with risk sub-scores of 95, 100, 65, and 80 points respectively) and adaptive adjustment of dynamic weight coefficients (with weight coefficients of 0.25, 0.15, 0.30, and 0.30 for the four weighting coefficients during the growth acceleration period), accurate individualized risk assessment is achieved. The comprehensive risk score decreased from 82.25 at the initial diagnosis to 68.5 after one month of intervention, a reduction of 16.7%, validating the sensitivity of the scoring system to the intervention effect.
[0118] By accurately predicting the intervention time window (the threshold for irreversible skeletal malocclusion is reached after 7 months, and intervention is recommended to be initiated within 3 months) and visualizing the biofeedback of behavioral intervention (the prognostic predictions for 4 different decompression intensities correspond to intervention time windows of 7, 10, 13, and 16 months, respectively), quantifiable decision support is provided for clinical decision-making and family-based behavioral correction.
[0119] Clinical trial results (sample size 120, follow-up period 12 months) showed that the positive predictive value of this system in children at high risk of malocclusion (comprehensive risk score above 60) was 84.6%, an improvement of 22.3 percentage points compared to the traditional method (62.3%). The proportion of children whose suggested intervention time window was within ±2 months of the actual optimal intervention time was 89.2%. The biofeedback function of behavioral intervention increased family adherence to treatment from 58.7% to 82.4%. These data demonstrate that this invention has significant clinical value in the early warning and precise intervention of malocclusion during the mixed dentition period.
[0120] Comparative Example 1
[0121] Purpose of comparison:
[0122] The purpose of this comparative study is to provide a traditional method for assessing the risk of malocclusion during the mixed dentition period in children. By comparing it with the technical solution of this invention, the advantages of this invention in terms of the continuity of the assessment model, the quantification of the assessment of bad habits, the quantification of the assessment of growth potential, the foresight of risk prediction, and the objectivity of risk judgment are highlighted.
[0123] Comparison method:
[0124] Traditional risk assessment for malocclusion during the mixed dentition period primarily relies on clinical examination and imaging evaluation during a single visit. The specific procedure is as follows.
[0125] First, clinical examination: Orthodontists examine the child's facial symmetry, tooth alignment, and occlusion through visual inspection and palpation. Oral hygiene habits (such as thumb sucking, lip biting, etc.) are obtained by questioning parents and children, and the assessment method is qualitative description (recorded as "has thumb sucking habit" or "does not have thumb sucking habit"), without quantitative measurement.
[0126] Second, imaging examinations: lateral cephalometric X-rays and curved tomographic films are taken, and the physician performs cephalometric analysis to obtain static morphological parameters such as the SNA angle, SNB angle, ANB angle, and mandibular plane angle.
[0127] Third, bone age assessment: Some medical institutions will assess a child's growth and development stage through wrist bone X-rays or cervical spine bone maturity, but the assessment results are only used for qualitative judgment (such as "before the peak growth period" or "after the peak growth period"), and no quantitative calculation of the remaining growth is performed.
[0128] Fourth, risk assessment: Based on clinical experience and the above examination results, physicians subjectively assess whether a child has a risk of malocclusion and the level of that risk. Risk assessment is usually divided into three levels: "low risk," "medium risk," and "high risk," but lacks a quantitative scoring system.
[0129] Taking the same patient as in Example 3 (Li, male, 8 years and 2 months old, thumb-sucking habit, deep overbite of the anterior teeth) as an example, the traditional assessment method was used. Clinical examination revealed that the child had a thumb-sucking habit, and the physician recorded "positive thumb-sucking habit" in the medical record. The depth of the anterior overbite was 4.5 mm, which the physician judged as "mild deep overbite". The cephalometric radiograph showed: ANB angle 5.5° (normal range 0~4°), indicating a mild tendency of skeletal Class II malocclusion; mandibular plane angle 32° (normal range 22~32°), which is at the upper limit of the normal range. The cervical spine bone maturity assessment was stage 3, and the physician judged that "the child is in a period of accelerated growth, which is a good time for intervention". Overall judgment: The physician believed that the child had a risk of malocclusion and recommended "observation and follow-up, with a re-examination every 6 months. If the overbite continues to deepen, intervention should be considered".
[0130] Comparison results:
[0131] Traditional methods, lacking dynamic monitoring, quantitative assessment, and future prediction capabilities, tend to be conservative in their risk assessment of this child ("mild deep overbite," "observation and follow-up"), potentially missing the optimal intervention window during the growth acceleration period. Specific shortcomings are as follows.
[0132] First, there is a lack of dynamic monitoring. Traditional methods only involve a single assessment during the initial visit, failing to capture the dynamic changes that can occur over several years during the mixed dentition period. Changes in maxillofacial development between visits (usually 6-12 months) cannot be detected in a timely manner. In a comparative case, the dentist recommended a follow-up visit every 6 months, but the child's dental arch width was narrowing at a rate of 0.18 mm / month, potentially exceeding 1 mm within 6 months—a trend completely unpredictable at the initial visit.
[0133] Second, the assessment of bad habits is crude. Thumb-sucking habits are only recorded as "present" or "absent," without considering quantitative indicators such as frequency, duration, and intensity. In the comparative case, the child sucked his thumb 10-20 times a day, each time lasting 5-15 seconds, which is a high-frequency, long-duration bad habit. However, traditional methods cannot quantify its cumulative effect, leading to an underestimation of the risk.
[0134] Third, the remaining growth was not quantified. Although the cervical vertebrae maturity stage was assessed, the specific remaining growth equivalent was not calculated. In the comparative example, the crucial data of 2.68 mm of remaining growth equivalent was completely missing, making it impossible for physicians to determine "how much growth is still available," a core indicator for determining the urgency of intervention.
[0135] Fourth, there is a lack of future prediction. Traditional methods can only assess "how severe it is now" (overbite depth 4.5mm, ANB angle 5.5°), but cannot predict "how severe it will become in the future if intervention is not implemented now" (the width between the maxillary canines will reach the narrowing threshold in 8 months, and the overbite depth will reach the deepening threshold in 7 months). In the comparative case, the dentist could not predict that the child's dental arch would reach the skeletal malocclusion threshold in 7 months, thus giving a lenient recommendation of "follow-up every 6 months," which may have missed the optimal intervention time.
[0136] Fifth, risk assessment is highly subjective. Risk levels depend on the physician's personal experience, and different physicians may have different risk assessments for the same child, lacking a unified and objective quantitative standard.
[0137] The core differences between the traditional method and the technical solution of this invention (Example 3) are as follows:
[0138] In terms of assessment mode, the traditional method is a single snapshot assessment, which is only performed at the time of medical visit; the present invention is a continuous dynamic monitoring method, which collects images monthly, monitors muscle strength daily, and the collection cycle is adaptively adjusted according to the risk score.
[0139] In terms of assessing bad habits, traditional methods are qualitative descriptions, simply recording "present" or "absent"; this invention is a quantitative assessment, calculating the cumulative exposure dose (156.8 Pa·s for this child).
[0140] In terms of growth potential assessment, the traditional method is a qualitative judgment, which only assesses the cervical vertebrae bone maturity stage 3; the present invention is a quantitative calculation, which calculates the remaining growth equivalent (2.68 mm in this child).
[0141] In terms of risk prediction, traditional methods lack predictive ability; this invention has forward modeling prediction ability and can predict that the child will reach the skeletal malocclusion threshold in 7 months.
[0142] In terms of risk scoring, the traditional method is a subjective qualitative judgment ("medium risk"); the present invention uses an objective quantitative scoring (the comprehensive risk score is 82.25 points).
[0143] Regarding intervention recommendations, the traditional approach is to "review once every 6 months"; the present invention proposes to "initiate intervention within 3 months, with the expected intervention window extended to 10-12 months".
[0144] Traditional methods are incapable of identifying hidden risks; however, this invention can identify hidden risks through inverse dynamics (the child reported nighttime teeth grinding, which was confirmed by examination).
[0145] Comparison conclusion:
[0146] This comparative example clearly demonstrates that traditional single-sense static assessment methods have significant limitations in assessing the risk of malocclusion during the mixed dentition period, failing to achieve early warning and precise intervention. This invention, through a multi-dimensional fusion of perioral biomechanical dynamic monitoring, longitudinal tracking of maxillofacial imaging, quantitative assessment of jaw growth potential, and analysis of maxillofacial developmental dynamics, achieves a leapfrog improvement from qualitative subjective judgment to quantitative objective scoring, from static status assessment to dynamic trend prediction, and from single snapshots to continuous monitoring. Specifically, this invention achieves a shift from "single snapshot" to "continuous monitoring" in assessment mode; from "presence / absence" to "cumulative exposure dose of 156.8 Pa·s" in assessing bad habits; from "accelerated growth phase" to "remaining growth equivalent of 2.68 mm" in assessing growth potential; from "no predictive ability" to "reaching the bony threshold after 7 months" in risk prediction; and from "medium risk" to "82.25 points" in risk scoring. These changes collectively constitute a qualitative leap from experience-based judgment to data-driven precision decision-making, providing scientific and precise decision support for early intervention of malocclusion during the mixed dentition period.
[0147] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of protection of the claims.
[0148] In conclusion, the above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A risk assessment and recommendation system for malocclusion during the mixed dentition period in children, characterized in that, include: The perioral biomechanics monitoring module is used to collect perioral muscle strength data of children and calculate the cumulative exposure dose of abnormal pressure based on the perioral muscle strength data. The maxillofacial development image tracking module is used to guide the acquisition of children's maxillofacial images via a mobile terminal, and to perform image processing on the maxillofacial images to extract time-series data of maxillofacial morphological parameters. The jawbone growth potential assessment module is used to calculate the remaining growth equivalent of the jawbone based on the child's bone age assessment results. The maxillofacial development dynamics analysis module is used to construct a dynamic model of maxillofacial development based on the cumulative exposure dose, the time series data, and the remaining growth equivalent of the jawbone, and to perform forward prediction and inverse dynamics inversion analysis. The risk scoring and recommendation output module is used to calculate a comprehensive risk score based on the analysis results of the maxillofacial development dynamics analysis module and output intervention recommendations. The perioral biomechanics monitoring module is communicatively connected to the maxillofacial development image tracking module. When the cumulative exposure dose exceeds a preset threshold, the perioral biomechanics monitoring module sends a trigger signal to the maxillofacial development image tracking module to control the maxillofacial development image tracking module to acquire maxillofacial images. The preset threshold is 1.5 times the average cumulative exposure dose of people of the same age and sex.
2. The risk assessment and recommendation system for malocclusion during the mixed dentition period in children as described in claim 1, characterized in that, The perioral biomechanics monitoring module includes at least one perioral biomechanics sensing unit, which comprises an electromyographic sensing element and a pressure sensing element.
3. The risk assessment and recommendation system for malocclusion during the mixed dentition period in children as described in claim 2, characterized in that, The sensing unit is a sensing patch, which includes: a lip patch for attaching to the inner side of the upper lip and the inner side of the lower lip to collect lip muscle pressure signals; a tongue patch for attaching to the back of the tongue to collect tongue pressure signals; and a cheek patch for attaching to the buccal mucosa on both sides to collect buccal muscle pressure signals. The pressure sensing element has no fewer than 200 pressure-sensitive units, and its sensitivity in the pressure range of 0 to 2.5 kPa is no less than 42 kPa⁻. 1 The sampling frequency of the sensor patch is not less than 100 Hz, and it is used to continuously collect abnormal pressure of the lip muscles, abnormal pressure of the tongue, and abnormal pressure of the cheek muscles.
4. The risk assessment and recommendation system for malocclusion during the mixed dentition period in children as described in claim 1, characterized in that, The perioral biomechanical monitoring module calculates the cumulative exposure dose according to the following method: The dose value of a single abnormal stress event is obtained by multiplying the difference between the peak pressure of a single abnormal stress event and the normal pressure threshold of a group of people of the same age and sex by the duration of the abnormal stress event, and then by the frequency weighting coefficient of the abnormal stress event. The cumulative exposure dose is obtained by summing the dose values of all abnormal stress events.
5. The risk assessment and recommendation system for malocclusion during the mixed dentition period in children as described in claim 1, characterized in that, The maxillofacial development image tracking module achieves the following functions through a mobile terminal application: guiding the shooting angle through the mobile terminal's gyroscope and facial recognition algorithm, guiding the shooting distance through the facial detection frame, guiding children to complete the actions of naturally closing their lips, smiling and showing their teeth, and opening their mouths to expose their dental arches through voice and animation, and verifying the quality of the photos' clarity, exposure, and angle deviation. The facial recognition algorithm guides the shooting angle by identifying the angle between the line connecting the two pupils and the horizontal direction, and determines the oral cavity exposure status by identifying the lip contour and the integrity of the crown area; The maxillofacial development image tracking module collects the following images according to a preset cycle: frontal smiling photo, frontal tooth-showing photo, left 45-degree side view photo, right 45-degree side view photo, maxillary dental arch view photo, and mandibular dental arch view photo; the preset cycle is dynamically adjusted according to the comprehensive risk score.
6. The risk assessment and recommendation system for malocclusion during the mixed dentition period in children as described in claim 1, characterized in that, The maxillofacial development image tracking module extracts the following maxillofacial morphological parameters from the maxillofacial image using an image processing algorithm: dental arch width, anterior overbite depth, anterior overjet distance, deviation between the maxillary dental midline and the facial midline, deviation between the mandibular dental midline and the facial midline, dental arch symmetry index, and facial symmetry index. The image processing algorithm identifies and marks the pixel coordinates of facial soft tissue landmarks and crown landmarks in the maxillofacial image, and calculates the maxillofacial morphological parameters based on the Euclidean distance between the pixel coordinates. The extracted maxillofacial morphological parameters were arranged in a time series, and the rate of change and acceleration of change of each parameter were calculated.
7. The risk assessment and recommendation system for malocclusion during the mixed dentition period in children as described in claim 1, characterized in that, The jawbone growth potential assessment module obtains a child's bone age in the following ways: determining the cervical bone maturity stage based on the morphological changes of the second, third, and fourth cervical vertebrae in the child's lateral cephalometric X-ray; or determining the bone age based on the degree of ossification center development in the child's left wrist anteroposterior X-ray. The jawbone growth potential assessment module calculates the remaining growth equivalent of the jawbone in the following way: obtain the reference value of the maximum jawbone growth of the child's population, multiply the reference value by the remaining growth ratio function, which is obtained by fitting population growth and development data with bone age, actual age and gender as independent variables.
8. The risk assessment and recommendation system for malocclusion during the mixed dentition period in children as described in claim 1, characterized in that, The dynamic model of the maxillofacial development dynamics analysis module is as follows: the net abnormal external force acting on the jawbone or dental arch is subtracted from the product of the tissue damping coefficient and the current displacement velocity, and then divided by the effective mass of the jawbone or dental arch to obtain the displacement acceleration of the jawbone or dental arch in three-dimensional space; the tissue damping coefficient is determined based on the remaining growth equivalent of the jawbone, and the larger the remaining growth equivalent of the jawbone, the smaller the tissue damping coefficient. The maxillofacial development dynamics analysis module, based on the dynamic model, takes the current state as the initial condition, inputs the time series of the net abnormal external force and the remaining growth equivalent of the jawbone, and positively extrapolates the displacement of the jawbone or dental arch at each future time point to predict the future development trend of malocclusion severity.
9. A risk assessment and recommendation system for malocclusion during the mixed dentition period in children as described in claim 1, characterized in that, The maxillofacial development dynamics analysis module substitutes the jawbone displacement trajectory measured by the maxillofacial development image tracking module into the dynamic equation, and reversely derives the equivalent abnormal resultant force actually borne by the child in the past period of time. That is, the effective mass of the jawbone or dental arch is multiplied by the measured displacement acceleration, and the product of the tissue damping coefficient and the measured displacement velocity is added to obtain the equivalent abnormal resultant force. The equivalent abnormal resultant force is compared with the cumulative exposure dose calculated by the perioral biomechanics monitoring module; when the equivalent abnormal resultant force is greater than the cumulative exposure dose and the difference exceeds 20% of the cumulative exposure dose, a deep screening command is triggered; when the cumulative exposure dose is greater than the equivalent abnormal resultant force and the difference exceeds 30% of the cumulative exposure dose, the comprehensive risk score is reduced.
10. The risk assessment and recommendation system for malocclusion during the mixed dentition period in children as described in claim 1, characterized in that, The risk scoring and recommendation output module calculates the developmental deviation index as follows: the displacement acceleration of an individual's jawbone or dental arch is subtracted from the mean displacement acceleration of children of the same bone age and sex, and then divided by the standard deviation of displacement acceleration of children of the same bone age and sex to obtain the developmental deviation index; when the developmental deviation index is greater than 1.5, it is judged as a significant deviation. The risk scoring and suggestion output module calculates the comprehensive risk score as follows: the developmental deviation risk sub-score is multiplied by a first dynamic weighting coefficient, the cumulative exposure risk sub-score is multiplied by a second dynamic weighting coefficient, the growth potential depletion risk sub-score is multiplied by a third dynamic weighting coefficient, and the trend risk sub-score is multiplied by a fourth dynamic weighting coefficient to obtain the comprehensive risk score; the first, second, third, and fourth dynamic weighting coefficients are automatically adjusted according to the child's bone age stage; the cumulative exposure risk sub-score is corrected based on the remaining growth equivalent of the jawbone, and the smaller the remaining growth equivalent of the jawbone, the higher the cumulative exposure risk sub-score; The risk scoring and recommendation output module calculates the critical intervention cutoff point based on the forward prediction results of the maxillofacial development dynamics analysis module, predicts the time point at which the displacement of the jawbone or dental arch reaches the preset skeletal malocclusion threshold, and outputs the following intervention recommendations: current bone age, estimated remaining equivalent growth of the jawbone, current displacement acceleration value, estimated remaining time to reach the preset skeletal malocclusion threshold, and recommended time range for initiating intervention.
