Oral cleaning data collection method and oral health management system

Abstract

This invention provides a method for collecting oral hygiene data and an oral health management system, comprising the following steps: S1, collecting toothbrush movement data and obtaining toothbrush movement trajectory data based on the movement data; S2, collecting pressure data from the toothbrush head; S3, dividing the movement trajectory data into regional trajectory data based on oral cavity regions; S4, aligning the regional trajectory data obtained in S3 with the pressure data collected in S2 to form oral hygiene data. The beneficial effects of this invention are: by dividing the data into regional trajectories, the oral hygiene data includes cleaning time and pressure data for each region, making the oral hygiene data more detailed and easier for users or medical personnel to view and analyze.

Description

Technical Field

[0001] This invention belongs to the field of data processing technology, and in particular relates to a method for collecting oral hygiene data and an oral health management system. Background Technology

[0002] Currently, periodontal disease, mainly caused by dental plaque and involving multiple factors, is a major killer threatening human health. Periodontal disease is characterized by a long course, continuous aggravation if left untreated, and the need for lifelong treatment. People find it difficult to accurately describe their oral hygiene data during daily brushing, and medical personnel are unable to optimize oral hygiene methods based on current oral hygiene data and oral health conditions, leading to recurrence or aggravation of periodontal disease.

[0003] The invention patent with patent application number CN202110669700.X obtains oral images and uses artificial intelligence technology to identify and analyze the images to adjust the cleaning parameters of the electric toothbrush, providing users with targeted oral cleaning services. This technical solution is mainly used to improve the cleaning effect of toothbrushes. However, it has the following problems: it cannot collect the user's real oral cleaning data, it cannot modify and optimize the user's current oral cleaning data, and it lacks online guidance from professionals. It does not improve patient compliance and self-management, and it does not form a systematic management of patients through a standardized and clear task process.

[0004] Smart brushing products have become an important application in oral health care. Most products on the market rely on sensor data, allowing users to unconsciously follow their habitual routines without a clear mental pattern or sequence. They provide feedback on brushing effectiveness in the form of scores, but these scores are only related to brushing time and cannot provide a detailed description of the cleaning process for each area of ​​the teeth. The efficiency of brushing depends not only on brushing time but also on the duration, pressure, and number of brushing cycles in the oral cavity, resulting in insufficient comprehensiveness and rationality in their assessment and guidance. Summary of the Invention

[0005] In view of this, the present invention aims to provide an oral hygiene data acquisition method and an oral health management system, in order to solve at least one of the above-mentioned technical problems.

[0006] To achieve the above objectives, the technical solution of the present invention is implemented as follows: The first aspect of this invention provides a method for collecting oral hygiene data, comprising the following steps: S1. Collect the movement data of the toothbrush and obtain the movement trajectory data of the toothbrush based on the movement data; S2. Collect pressure data from the toothbrush head; S3. Based on the oral cavity region division, the movement trajectory data is used to form regional trajectory data; S4. Align the regional trajectory data obtained in S3 with the pressure data collected in S2 to form oral cleaning data.

[0007] Furthermore, S1 includes the following steps: The toothbrush handle is equipped with a multi-axis sensor module; S11. Acceleration data of the toothbrush is acquired through the accelerometer of the multi-axis sensor module; S12. Perform noise reduction processing on the data collected in S11; S13. Integrate the acceleration data to obtain the movement trajectory data of the toothbrush head.

[0008] Furthermore, S3 includes the following steps; S31. Divide the oral cavity into multiple main regions, including the upper left region, lower left region, upper right region, lower right region, upper anterior region, and lower anterior region; S32. Divide each main region into multiple sub-regions, specifically as follows: The upper left region is divided into the occlusal surface of the upper left posterior teeth, the buccal surface of the upper left, and the palatal surface of the upper left. The lower left region is divided into the occlusal surface of the lower left posterior teeth, the buccal surface of the lower left, and the palatal surface of the lower left. The upper right region is divided into the upper right posterior occlusal surface region, the upper right buccal surface region, and the upper right palatal surface region. The lower right region is divided into the occlusal surface of the lower right posterior teeth, the buccal surface of the lower right teeth, and the palatal surface of the lower right teeth. The anterior region is divided into the lateral region of the anterior teeth and the medial region of the anterior teeth. The lower anterior region is divided into the lateral region of the lower anterior teeth and the medial region of the lower anterior teeth. S33. The toothbrush's posture data is acquired by the gyroscope of the multi-axis sensor module, and the movement trajectory data is divided into corresponding sub-regions by combining the posture data.

[0009] Furthermore, S4 includes the following steps: S41. Align the pressure data collected in S2 with the trajectory data in the sub-region according to the timestamp, and calculate the average pressure in the sub-region. S42. The total duration of the trajectory data is used as the total cleaning duration, and the trajectory duration in each sub-region is used as the cleaning duration of the sub-region. S43. Calculate the trajectory interval time within the sub-region. If the interval time exceeds the threshold, it is determined to increase the number of cleaning area changes by one. S44. The cleanliness consistency rate is calculated using the following formula: ; in, Indicates the cleanliness consistency rate. Indicates the number of times the cleaning area changes. Indicates the total cleaning time. Represents the total number of reachable sub-regions, using It is 18. This represents the number of sub-regions reached during the cleaning process; S45. The cleaning isochronity is calculated using the following formula: ; in, Indicates the cleaning time rate. Indicates the total cleaning time. Represents the total number of reachable sub-regions, using It is 18. Indicate the cleaning duration of the i-th sub-region; S45. Calculate the brushing quality index for each sub-region using the following formula: ; in, This indicates the brushing quality index. Indicates the theoretical maximum mass index. Indicates time weighting, Indicates pressure weight. Indicates the cleaning duration of the sub-region. Indicates the standard cleaning time; This represents the average pressure. This indicates the standard cleaning pressure.

[0010] Furthermore, step S33 includes the following steps: Attitude data and acceleration data are concatenated into a high-dimensional vector, and then low-pass filtering and gravity separation are performed on the high-dimensional vector. By using the sliding window technique, the movement trajectory data is divided into time windows to form training samples, and each sub-region is used as a class in the classification task. Subjects were instructed to swipe through each region as directed to label each time window, and the labeled time windows were used to train the neural network model. The trained neural network model is used to divide the currently collected movement trajectory data into corresponding sub-regions.

[0011] Furthermore, step S33 includes the following steps: Construct a binary classification tree using sub-regions as leaf nodes; Real-time acquisition of motion trajectory data, forming time series input samples according to time windows, and inputting the time series samples into a neural network model to obtain the corresponding motion features; The motion features are input into the top-level node of the binary tree, and the current motion trajectory is divided into the corresponding sub-region by the layer-by-layer classification path to reach the leaf node.

[0012] A second aspect of the present invention provides an oral health management system, comprising: The toothbrush includes a toothbrush body, a data acquisition module, and a wireless communication module. The data acquisition module includes a multi-axis sensor module and a pressure sensor. The multi-axis sensor module is mounted on the toothbrush handle and is used to collect the movement data of the toothbrush. The pressure sensor is mounted at the connection between the toothbrush head and the handle and is used to collect the pressure data of the toothbrush. The wireless communication module is configured to transmit the movement data and pressure data of the toothbrush. The data processing device includes a data processing module configured to obtain toothbrush movement trajectory data based on movement data, divide the movement trajectory data into oral cavity regions to form regional trajectory data, and align the regional trajectory data with pressure data to form oral cleaning data.

[0013] Furthermore, the data processing device includes: a user management module, a health management module, and an online management module; The user management module includes an inspection and evaluation unit, a medical record unit, and a contract grouping unit. The assessment unit is used to collect standardized scales of patients' periodontal disease. The medical record file unit is used to collect the medical records of patients with periodontal disease; The contract grouping unit is used to group periodontal disease patients according to their condition. The specific steps are as follows: Collect patients’ health record information based on standardized scales or medical records; Patients with a clear disease type or characteristics are grouped according to their disease type; For patients without a clear disease type or characteristic, the dentist communicates with the patient through the doctor's terminal to clarify the patient's periodontal disease type and manually groups them. Each group has a corresponding oral health management plan.

[0014] Furthermore, the health management module also includes an analysis and evaluation unit and a reminder service unit; The analysis and evaluation unit is configured to obtain the patient's oral cleaning data from the analysis and evaluation unit and the data processing module. It analyzes the oral cleaning data based on the standardized scale or medical records collected by the user management module, determines whether the current oral cleaning data is correct, and if it is incorrect, plans an oral cleaning plan. The reminder service unit will send the planned oral hygiene schedule and current oral hygiene data to the user's device.

[0015] A third aspect of the present invention provides an electronic device including a processor and a memory communicatively connected to the processor and used to store processor-executable instructions, the processor being used to perform the method described in the first aspect above.

[0016] Compared with existing technologies, the oral hygiene data acquisition method and oral health management system described in this invention have the following beneficial effects: (1) The oral cleaning data collection method of the present invention includes cleaning time and pressure data of each area through the regional trajectory obtained by dividing the oral cleaning data. The oral cleaning data is more detailed and convenient for users or medical staff to view and analyze.

[0017] (2) The oral health management system described in this invention can assess the quality of daily oral hygiene based on objective brushing data without relying on the patient's subjective description, thereby improving the scientific nature of follow-up, group management and intervention decisions. Attached Figure Description

[0018] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings: Figure 1 This is a schematic diagram of the method described in an embodiment of the present invention. Detailed Implementation

[0019] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0020] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined with "first," "second," etc., may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0021] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0022] The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0023] Example 1: like Figure 1 As shown, a method for collecting oral hygiene data includes the following steps: S1. Collect the movement data of the toothbrush and obtain the movement trajectory data of the toothbrush based on the movement data; S1 includes the following steps: The toothbrush handle is equipped with a multi-axis sensor module; the multi-axis sensor module adopts, but is not limited to, the existing nine-axis sensor module, specifically the MPU-9250.

[0024] S11. Acceleration data of the toothbrush is acquired through the accelerometer of the multi-axis sensor module; S12. Perform noise reduction processing on the data collected in S11; Includes the following steps: High-pass filter models can suppress low-frequency drift, or slow drift processing can be performed using the sliding window averaging method; Low-pass filtering preserves the main frequency band of toothbrush movement and removes high-frequency noise.

[0025] S13. Integrate the acceleration data to obtain the movement trajectory data of the toothbrush head; Acceleration data is integrated to obtain velocity data using, but not limited to, a short-window integration method. Position data is then obtained by integrating the velocity data. Continuous position data is the trajectory data within a short window, with the short window duration being 0.5-2 seconds.

[0026] By using a multi-axis sensor module to collect data and introducing high-pass filtering, low-pass filtering, and short-window integration methods, low-frequency drift and high-frequency noise are effectively suppressed, avoiding the position divergence problem caused by traditional direct integration. This makes the movement trajectory of the toothbrush head more stable and usable within a short time window.

[0027] S2. Collect pressure data from the toothbrush head; S3. Based on the oral cavity region division, the movement trajectory data is used to form regional trajectory data; S3 includes the following steps; S31. Divide the oral cavity into multiple main regions, including the upper left region, lower left region, upper right region, lower right region, upper anterior region, and lower anterior region; S32. Divide each main region into multiple sub-regions, specifically as follows: The upper left region is divided into the occlusal surface of the upper left posterior teeth, the buccal surface of the upper left, and the palatal surface of the upper left. The lower left region is divided into the occlusal surface of the lower left posterior teeth, the buccal surface of the lower left, and the palatal surface of the lower left. The upper right region is divided into the upper right posterior occlusal surface region, the upper right buccal surface region, and the upper right palatal surface region. The lower right region is divided into the occlusal surface of the lower right posterior teeth, the buccal surface of the lower right teeth, and the palatal surface of the lower right teeth. The anterior region is divided into the lateral region of the anterior teeth and the medial region of the anterior teeth. The lower anterior region is divided into the lateral region of the lower anterior teeth and the medial region of the lower anterior teeth. S33. The toothbrush's posture data is acquired by the gyroscope of the multi-axis sensor module, and the movement trajectory data is divided into corresponding sub-regions by combining the posture data.

[0028] By dividing the oral cavity into main and sub-regions and combining posture data and time-series motion patterns, rather than judging a single instantaneous angle, dynamic brushing behavior recognition improves the accuracy of region identification.

[0029] S33 includes the following steps: Attitude data and acceleration data are concatenated into a high-dimensional vector, and then low-pass filtering and gravity separation are performed on the high-dimensional vector. By using the sliding window technique, the movement trajectory data is divided into time windows to form training samples, and each sub-region is used as a class in the classification task. Subjects were instructed to swipe through each region as directed to label each time window, and the labeled time windows were used to train the neural network model. The trained neural network model is used to divide the currently collected movement trajectory data into corresponding sub-regions.

[0030] In some embodiments, the specific execution flow of step S33 is as follows: At each moment, the multi-axis sensor module acquires acceleration and attitude data, including angular velocity and Euler angles. The acceleration, angular velocity, and Euler angles are then concatenated into a high-dimensional vector.

[0031] The high-dimensional vector is low-pass filtered (e.g., 10–15 Hz) to preserve the main frequency of the brushing motion and to separate gravity. The direction of gravity is projected onto the sensor coordinate system using attitude, and the linear acceleration is calculated.

[0032] By concatenating acceleration, angular velocity, and Euler angles into a high-dimensional vector, rich information about brushing motion can be provided. By simultaneously considering data from multiple sensors, the model can acquire more comprehensive motion features, improving the accuracy of region recognition. Separating the gravity component from the posture data yields true linear acceleration, reducing interference from gravity changes and making the motion data more accurate, thus helping the model accurately determine brushing motion.

[0033] By using low-pass filtering, the dominant frequency of the brushing motion is preserved while high-frequency noise is removed. This helps reduce the impact of noise caused by hand movements, allowing focus to be placed on the main frequencies and patterns of the brushing motion. This step helps ensure the stability and consistency of the characteristics.

[0034] Neural network models require a sequence of events; therefore, the motion trajectory is divided into time windows, with the window length... A point (e.g., a 1-second window: 50Hz →) Or 1.5 seconds: ), step size (For example, 0.25 seconds:) or Therefore, each sample is a matrix representing the brushing motion pattern within one second. The differences between sub-regions are not determined by the posture at a single instant, but by how the action is performed; hence, time windows are used. Each time window is assigned a label, and the subject naturally obtains window-level labels by brushing in sections as instructed (brushing only one section at a time). By dividing the motion trajectory into continuous small windows using time window segmentation technology, the temporal sequence features of the brushing action can be captured. Each time window contains a fixed length of motion data, helping to maintain temporal consistency and avoiding the loss of long-term dependent features in the data. Labeling each time window ensures that each time window corresponds to a brushing area, and the label of each time window matches the actual brushing action area, improving the model's learning accuracy.

[0035] The architecture of the neural network model is 1DTemporalCNN+BiLSTM; Local motion textures (such as short vibrations, reciprocating rhythms, and frequency band patterns) are extracted using CNN, and LSTM is used to understand how the preceding and following actions are connected (for example, there is usually a wrist flip transition when entering the palate). The loss function enables the model to accurately identify brushing areas; The loss function is ,in, It's a one-hot tag. It represents the probability of each partition.

[0036] In some areas (such as the inner side of the front teeth), users brush less frequently, resulting in insufficient data. This can be addressed by using weighted cross-entropy to give more weight to classes with fewer samples or by using FocalLoss to make the model focus more on difficult samples.

[0037] The process of region partitioning is prone to instantaneous region jumps, therefore post-processing operations are introduced, including: Based on certain region switching patterns, penalties are applied to transitions to make the output more consistent with actual brushing paths. For example, transitions between the occlusal, buccal, and palatal surfaces of the posterior teeth on the same side are more common, while a sudden jump from the upper left to the lower right is rare. A rule-based state machine is used to penalize these transitions. This avoids unreasonable region jumps and improves the rationality and naturalness of region switching. Common region transitions (such as from posterior to anterior teeth) are reinforced, while uncommon transitions (such as jumping directly from the upper left to the lower right) are penalized, thus ensuring that the brushing path output by the model more closely matches actual brushing behavior.

[0038] The neural network model shares features and outputs results through multiple classification heads, including HeadA (main region), HeadB (tooth surface type), and HeadC (final 16 categories, main task). By using multiple classification heads, the model can predict the brushing area from different angles, which can refine the model's prediction task, thereby improving the accuracy of each task and avoiding excessive model complexity.

[0039] It outputs the predicted area and confidence level at each moment (each window) in real time, and can accumulate the coverage time of each area (for brushing coverage feedback).

[0040] In other embodiments, the classification method that uses multiple classification heads to output results and shares features in the neural network model is replaced by a 16-partition recognition scheme based on binary classification: The binary tree is organized into 16 partitions, with each internal node being a binary classifier. The partitions with the most similar shared motion features are grouped together first, and then the partitions are split layer by layer. During inference, only one path is taken. The complexity is reduced from performing a 16-class softmax operation to performing 3 to 5 binary classification operations. Furthermore, the average classification cost can be minimized using the idea of ​​an "optimal binary tree".

[0041] The specific working process is as follows: Each sub-region is treated as a leaf node. Each binary classifier receives the same time series (a high-dimensional vector obtained by concatenating acceleration, angular velocity, and Euler angles). The motion pattern is extracted using a time encoder, and then the binary classification probability is output to obtain the probability of the left and right branches of the node. Each tree node shares a lightweight Enc (computed in one operation). Then, each tree node only needs to make a very small MLP binary classification head.

[0042] Based on this approach, a four-level binary classification tree is constructed as follows: L1 includes: N1: Anterior teeth (4) / Posterior teeth (12); The front teeth move more vertically, with smaller amplitude and closer to the front, while the back teeth move more laterally, with wrist flips and greater angle variations.

[0043] L2 includes: For posterior branch N2: left posterior teeth (6) / right posterior teeth (6); For anterior branch N3: upper anterior teeth (2) / lower anterior teeth (2); The long-term bias direction and reciprocating direction of the Yaw / Roll brush are mirror images of each other on the left and right sides (the brushing methods on the right and left sides often show a systematic mirror image in the direction of wrist rotation). The main characteristics are: differences in pitch distribution and entry method (the upper front teeth are more likely to show a raised / upward tilting posture distribution).

[0044] L3 includes: Left posterior teeth (6) N4: left upper posterior teeth (3) / left lower posterior teeth (3); right posterior teeth (6) N5: right upper posterior teeth (3) / right lower posterior teeth (3); upper anterior teeth (2) N6: upper anterior lateral / upper anterior medial; lower anterior teeth (2) N7: lower anterior lateral / lower anterior medial.

[0045] L4 includes: N8: Occlusal surface / (buccal + palatal), the occlusal surface usually shows a short stroke, a near-vertical reciprocating motion, and less Roll fluctuation; N9: Buccal / palatal, mainly based on the system difference of "brush head orientation / wrist turning direction" (Roll sign bias, Yaw bias, angular velocity mode when entering / exiting), similarly the above classification is applied to the lower left, upper right, and lower right to obtain N10~N15.

[0046] After obtaining the confusion matrix by training a temporary 16-class model, the classification and merging of binary trees are performed by constructing a classification binary tree. The process includes: Train a baseline 16-class time series model to obtain the confusion matrix. (Frequency of i being misclassified as j), define inter-class distance: Hierarchical clustering places the most easily confused classes on the same side to form a binary clustering tree. The resulting tree is naturally the optimal approximation of the layer-by-layer splitting based on similarity.

[0047] For any node n in the tree, it is responsible for distinguishing between two sets of categories. and The training samples are constructed as follows: If sample label Then node label ;like ,but Other samples that do not belong to the subtree of this node do not participate in the training of this node (or are used as "rejection" strategy data).

[0048] Calculate once for each time window Starting from the root node, calculate the binary classification probability sequentially. Select the left or right subtree to proceed to the next node, until the leaf (one of the 16 areas); At each node, if If the value is very close to 0 or 1 (e.g., >0.95), proceed directly to the next range; if it is close to 0.5 (uncertain), retain the previous stable region (to prevent jumps). The leaf output is voted on using the nearest N windows (N=3~7).

[0049] By organizing 16 regions into a binary classification tree, the complexity of each inference is reduced, requiring only 3 to 5 binary classification inferences per instance. Compared to softmax classification with 16 classes, this significantly reduces computational overhead and improves the efficiency of real-time response.

[0050] In other embodiments, there is a certain error in inferring the vertical position of the toothbrush head by changes in pitch angle and acceleration data, especially when brushing the front of the teeth, where the pitch angle may indeed be small or close to zero. Therefore, the vertical position is determined by combining pitch angle, acceleration and movement trajectory. 1. Extract pitch angle and acceleration changes: Calculate the pitch angle and acceleration changes at each moment, including the vertical component of acceleration. By comparing the pitch angle changes at each moment and combining them with the acceleration changes, determine the approximate trend of the vertical position.

[0051] 2. Analyze the motion trajectory: Calculate straightness based on the toothbrush's trajectory ,if A larger value suggests that the current action is relatively vertical, and the toothbrush may be positioned at the top. A smaller value indicates a curved trajectory, meaning the toothbrush may be positioned at the bottom.

[0052] 3. Comparison and judgment of the time periods before and after: Compare the pitch angle and acceleration data at the current moment with those at previous and subsequent moments. If the pitch angle is small and the acceleration changes little at the current moment, but the pitch angle increases and the acceleration increases significantly at subsequent moments, it indicates that the toothbrush is likely at the bottom and moving upward. If the pitch angle is large and the acceleration increases at the current moment, but the pitch angle decreases and the acceleration decreases at subsequent moments, it means that the toothbrush has likely moved from above to below.

[0053] The following example is provided to illustrate the above judgment process: Assuming you are brushing the outer side of your front teeth, the motion data you input is as follows: The current pitch angle is The Z-direction component of the acceleration is At the previous moment, the pitch angle was The acceleration changes little in the Z direction, and the path straightness is high (e.g., ).

[0054] At this point, it can be inferred that the current pitch angle is large and the acceleration change is small, possibly because the toothbrush is turning from below to above, and the path is relatively straight, indicating that the brushing action is relatively vertical and the acceleration change in the Z direction is relatively stable. Therefore, it can be determined that the toothbrush is likely above and its path is relatively straight.

[0055] By combining pitch angle, acceleration data, and motion trajectory to determine the up-down position of the toothbrush, the error in orientation determination when brushing the front can be effectively addressed. By analyzing the straightness of the toothbrush's trajectory, it can be determined whether the toothbrush is positioned above or below, which helps improve the accuracy of area division. In particular, it can make more accurate predictions when judging changes in brushing direction and motion. By comparing data from different moments, the changes in the up-down position of the toothbrush can be determined more accurately.

[0056] S4. Align the regional trajectory data obtained in S3 with the pressure data collected in S2 to form oral cleaning data.

[0057] S4 includes the following steps: S41. Align the pressure data collected in S2 with the trajectory data in the sub-region according to the timestamp, and calculate the average pressure in the sub-region. S42. The total duration of the trajectory data is used as the total cleaning duration, and the trajectory duration in each sub-region is used as the cleaning duration of the sub-region. S43. Calculate the trajectory interval time within the sub-region. If the interval time exceeds the threshold, it is determined to increase the number of cleaning area changes by one. S44. The cleanliness consistency rate is calculated using the following formula: ; in, Indicates the cleanliness consistency rate. Indicates the number of times the cleaning area changes. Indicates the total cleaning time. Represents the total number of reachable sub-regions, using It is 18. This represents the number of sub-regions reached during the cleaning process; S45. The cleaning isochronity is calculated using the following formula: ; in, Indicates the cleaning time rate. Indicates the total cleaning time. Represents the total number of reachable sub-regions, using It is 18. Indicate the cleaning duration of the i-th sub-region.

[0058] S45. Calculate the brushing quality index for each sub-region using the following formula: ; in, This indicates the brushing quality index. Indicates the theoretical maximum mass index. Indicates time weighting, Indicates pressure weight. Indicates the cleaning duration of the sub-region. Indicates the standard cleaning time; This represents the average pressure. This indicates the standard cleaning pressure.

[0059] The standard cleaning time can be a fixed value or it can be adjusted according to the user's oral health condition and age.

[0060] Standard cleaning pressure can be a fixed value or it can be adjusted individually based on the user's oral health and age.

[0061] By comprehensively calculating cleaning time, average pressure, and area coverage, the effectiveness of cleaning can be evaluated without relying solely on total brushing time or a single indicator.

[0062] Cleaning consistency rate, cleaning time rate, and brushing quality index can quantitatively reflect the regional balance, integrity, and standardization of the brushing process. The results are intuitive, interpretable, and easy for users to understand and doctors to evaluate.

[0063] The oral hygiene data obtained by dividing the area trajectory includes data such as cleaning time and pressure for each area, making the oral hygiene data more detailed and easier for users or medical staff to view and analyze.

[0064] An oral health management system, comprising: The toothbrush includes a toothbrush body, a data acquisition module, and a wireless communication module. The data acquisition module includes a multi-axis sensor module and a pressure sensor. The multi-axis sensor module is mounted on the toothbrush handle and is used to collect the movement data of the toothbrush. The pressure sensor is mounted at the connection between the toothbrush head and the handle and is used to collect the pressure data of the toothbrush. The wireless communication module is configured to transmit the movement data and pressure data of the toothbrush. The data processing device includes a data processing module configured to obtain toothbrush movement trajectory data based on movement data, divide the movement trajectory data into oral cavity regions to form regional trajectory data, and align the regional trajectory data with pressure data to form oral cleaning data.

[0065] The data processing device includes: a user management module, a health management module, and an online management module; The user management module includes an inspection and evaluation unit, a medical record unit, and a contract grouping unit. The assessment unit is used to collect standardized scales of patients' periodontal disease. The medical record file unit is used to collect the medical records of patients with periodontal disease; The contract grouping unit is used to group periodontal disease patients according to their condition. The specific steps are as follows: Collect patients’ health record information based on standardized scales or medical records; Patients with a clear disease type or characteristics are grouped according to their disease type; For patients without a clear disease type or characteristic, the dentist communicates with the patient through the doctor's terminal to clarify the patient's periodontal disease type and manually groups them. Each group has a corresponding oral health management plan.

[0066] The health management module also includes an analysis and evaluation unit and a reminder service unit; The analysis and evaluation unit is configured to obtain the patient's oral cleaning data from the analysis and evaluation unit and the data processing module. It analyzes the oral cleaning data based on the standardized scale or medical records collected by the user management module, determines whether the current oral cleaning data is correct, and if it is incorrect, plans an oral cleaning plan. The reminder service unit will send the planned oral hygiene schedule and current oral hygiene data to the user's device.

[0067] By combining brushing process data with patients' periodontal disease scales and medical records, a complete closed loop is achieved from "brushing behavior collection" to "health risk assessment" and then to "cleaning plan development".

[0068] Doctors can assess the quality of daily oral hygiene based on objective brushing data without relying on patients' subjective descriptions, thus improving the scientific basis of follow-up, group management, and intervention decisions.

[0069] By providing real-time or post-brushing feedback on brushing coverage, weak areas, and improvement suggestions, users' perception of brushing quality and their willingness to proactively improve can be enhanced, which is beneficial for long-term oral health maintenance.

[0070] Example 2: An electronic device includes a processor and a memory communicatively connected to the processor and used to store processor-executable instructions, the processor being used to execute the method described in Embodiment 1 above.

[0071] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and specification of the present invention.

[0072] The above description is only a preferred embodiment of the present invention and is 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 method for collecting oral hygiene data, characterized in that, Includes the following steps: S1. Collect the movement data of the toothbrush and obtain the movement trajectory data of the toothbrush based on the movement data; S2. Collect pressure data from the toothbrush head; S3. Based on the oral cavity region division, the movement trajectory data is used to form regional trajectory data; S4. Align the regional trajectory data obtained in S3 with the pressure data collected in S2 to form oral cleaning data.

2. The oral hygiene data acquisition method according to claim 1, characterized in that: S1 includes the following steps: The toothbrush handle is equipped with a multi-axis sensor module; S11. Acceleration data of the toothbrush is acquired through the accelerometer of the multi-axis sensor module; S12. Perform noise reduction processing on the data collected in S11; S13. Integrate the acceleration data to obtain the movement trajectory data of the toothbrush head.

3. The oral hygiene data acquisition method according to claim 2, characterized in that: S3 includes the following steps; S31. Divide the oral cavity into multiple main regions, including the upper left region, lower left region, upper right region, lower right region, upper anterior region, and lower anterior region; S32. Divide each main region into multiple sub-regions, specifically as follows: The upper left region is divided into the occlusal surface of the upper left posterior teeth, the buccal surface of the upper left, and the palatal surface of the upper left. The lower left region is divided into the occlusal surface of the lower left posterior teeth, the buccal surface of the lower left, and the palatal surface of the lower left. The upper right region is divided into the upper right posterior occlusal surface region, the upper right buccal surface region, and the upper right palatal surface region. The lower right region is divided into the occlusal surface of the lower right posterior teeth, the buccal surface of the lower right teeth, and the palatal surface of the lower right teeth. The anterior region is divided into the lateral region of the anterior teeth and the medial region of the anterior teeth. The lower anterior region is divided into the lateral region of the lower anterior teeth and the medial region of the lower anterior teeth. S33. The toothbrush's posture data is acquired by the gyroscope of the multi-axis sensor module, and the movement trajectory data is divided into corresponding sub-regions by combining the posture data.

4. The oral hygiene data acquisition method according to claim 3, characterized in that: S4 includes the following steps: S41. Align the pressure data collected in S2 with the trajectory data in the sub-region according to the timestamp, and calculate the average pressure in the sub-region. S42. The total duration of the trajectory data is used as the total cleaning duration, and the trajectory duration in each sub-region is used as the cleaning duration of the sub-region. S43. Calculate the trajectory interval time within the sub-region. If the interval time exceeds the threshold, it is determined to increase the number of cleaning area changes by one. S44. The cleanliness consistency rate is calculated using the following formula: ； in, Indicates the cleanliness consistency rate. Indicates the number of times the cleaning area changes. Indicates the total cleaning time. Represents the total number of reachable sub-regions, using It is 18. This represents the number of sub-regions reached during the cleaning process; S45. The cleaning isochronity is calculated using the following formula: ； in, Indicates the cleaning time rate. Indicates the total cleaning time. Represents the total number of reachable sub-regions, using It is 18. Indicate the cleaning duration of the i-th sub-region; S45. Calculate the brushing quality index for each sub-region using the following formula: ； in, This indicates the brushing quality index. Indicates the theoretical maximum mass index. Indicates time weighting, Indicates pressure weight. Indicates the cleaning duration of the sub-region. Indicates the standard cleaning time; This represents the average pressure. This indicates the standard cleaning pressure.

5. The oral hygiene data acquisition method according to claim 3, characterized in that, S33 includes the following steps: Attitude data and acceleration data are concatenated into a high-dimensional vector, and then low-pass filtering and gravity separation are performed on the high-dimensional vector. By using the sliding window technique, the movement trajectory data is divided into time windows to form training samples, and each sub-region is used as a class in the classification task. Subjects were instructed to swipe through each region as directed to label each time window, and the labeled time windows were used to train the neural network model. The trained neural network model is used to divide the currently collected movement trajectory data into corresponding sub-regions.

6. The oral hygiene data acquisition method according to claim 3, characterized in that, S33 includes the following steps: Construct a binary classification tree using sub-regions as leaf nodes; Real-time acquisition of motion trajectory data, forming time series input samples according to time windows, and inputting the time series samples into a neural network model to obtain the corresponding motion features; The motion features are input into the top-level node of the binary tree, and the current motion trajectory is divided into the corresponding sub-region by the layer-by-layer classification path to reach the leaf node.

7. An oral health management system, characterized in that, include: The toothbrush includes a toothbrush body, a data acquisition module, and a wireless communication module. The data acquisition module includes a multi-axis sensor module and a pressure sensor. The multi-axis sensor module is mounted on the toothbrush handle and is used to collect the movement data of the toothbrush. The pressure sensor is mounted at the connection between the toothbrush head and the handle and is used to collect the pressure data of the toothbrush. The wireless communication module is configured to transmit the movement data and pressure data of the toothbrush. The data processing device includes a data processing module configured to obtain toothbrush movement trajectory data based on movement data, divide the movement trajectory data into oral cavity regions to form regional trajectory data, and align the regional trajectory data with pressure data to form oral cleaning data.

8. An oral health management system according to claim 7, characterized in that, The data processing device includes: a user management module, a health management module, and an online management module; The user management module includes an inspection and evaluation unit, a medical record unit, and a contract grouping unit. The assessment unit is used to collect standardized scales of patients' periodontal disease. The medical record file unit is used to collect the medical records of patients with periodontal disease; The contract grouping unit is used to group periodontal disease patients according to their condition. The specific steps are as follows: Collect patients’ health record information based on standardized scales or medical records; Patients with a clear disease type or characteristics are grouped according to their disease type; For patients without a clear disease type or characteristic, the dentist communicates with the patient through the doctor's terminal to clarify the patient's periodontal disease type and manually groups them. Each group has a corresponding oral health management plan.

9. An oral health management system according to claim 8, characterized in that, The health management module also includes an analysis and evaluation unit and a reminder service unit; The analysis and evaluation unit is configured to obtain the patient's oral cleaning data from the analysis and evaluation unit and the data processing module. It analyzes the oral cleaning data based on the standardized scale or medical records collected by the user management module, determines whether the current oral cleaning data is correct, and if it is incorrect, plans an oral cleaning plan. The reminder service unit will send the planned oral hygiene schedule and current oral hygiene data to the user's device.

10. An electronic device comprising a processor and a memory communicatively connected to the processor and used for storing processor-executable instructions, characterized in that: The processor is used to execute the method described in any one of claims 1-6.

Images