Pressure detection and control method for toothbrush head

Abstract

The invention provides a pressure detection and control method for a toothbrush head, and belongs to the technical field of oral cavity cleaning appliances, and the method comprises the steps: obtaining pressure data during tooth brushing through a flexible pressure sensor arranged on the toothbrush head; obtaining posture data of the toothbrush through a motion sensor arranged on the electric toothbrush; on the basis of the pressure data and the posture data, tooth position information corresponding to the current toothbrush head is obtained through calculation; comparing the pressure data with a pressure threshold value preset for the tooth position information; when the pressure data exceeds the corresponding preset pressure threshold value, a control instruction is generated to adjust the working state of the electric toothbrush. According to the invention, the specific tooth position which is currently cleaned by the brush head can be judged in real time and with high precision, the extensive management mode that a traditional electric toothbrush uses a single pressure threshold for the whole oral cavity is thoroughly changed, and refined and personalized nursing is realized.

Description

Technical Field

[0001] This invention relates to the field of oral hygiene appliance technology, and in particular to a method for detecting and controlling the pressure of a toothbrush head. Background Technology

[0002] When the brush head of an electric toothbrush comes into contact with the tooth surface, it vibrates at a high frequency. When the contact pressure is too high, the friction generated by the high-frequency vibration of the brush head can easily damage the tooth surface and even the gums. Moreover, this damage is often irreversible. Therefore, during the operation of an electric toothbrush, it is necessary to detect and control the pressure on the brush head to prevent users from being injured while using the toothbrush.

[0003] In addition, the gums and enamel of different ages or special patients have different tolerance levels, so different brushing pressures need to be selected according to different people and different toothbrush heads, which is more conducive to the protection of oral health.

[0004] Existing methods for testing electric toothbrush pressure primarily rely on detecting mechanical changes (including displacement or vibration) in the toothbrush's transmission components to indirectly measure the pressure applied between the brush head and teeth. However, this indirect measurement method is susceptible to interference with the pressure measurement results when the brush head contacts other parts of the mouth, when the electric toothbrush's transmission components lack sufficient power, or when different toothbrush heads exhibit varying or inconsistent force transmission performance.

[0005] Furthermore, and more importantly, different teeth in the mouth (such as incisors and molars) have different shapes and gum conditions, and therefore their ideal brushing force should also differ. Current pressure control solutions mostly use a "one-size-fits-all" approach, applying a fixed pressure threshold, which fails to provide personalized pressure control for different teeth. This results in a poor user experience and prevents truly precise oral care. Summary of the Invention

[0006] This invention provides a method for pressure detection and control of a toothbrush head, which solves the problem that the pressure detection of electric toothbrushes in the prior art is too general and cannot distinguish the different pressure tolerance of different teeth (such as sensitive incisors and strong molars), resulting in inaccurate pressure control, poor user experience, and even potential tooth damage. The invention enables personalized and precise pressure control based on real-time tooth position recognition, thereby improving brushing safety and comfort.

[0007] This invention provides a method for pressure detection and control of a toothbrush head. The toothbrush head is detachably mounted on the handle of an electric toothbrush. Pressure data during brushing is acquired by a flexible pressure sensor mounted on the toothbrush head; posture data during brushing is acquired by a motion sensor mounted on the electric toothbrush; based on the pressure data and the posture data, the tooth position information corresponding to the current toothbrush head is calculated; the pressure data is compared with a preset pressure threshold for the tooth position information; when the pressure data exceeds the corresponding preset pressure threshold, a control command is generated to adjust the working state of the electric toothbrush.

[0008] According to the toothbrush head pressure detection and control method provided by the present invention, tooth position information is calculated based on the pressure data and the posture data, specifically including: converting the pressure data into a pressure distribution matrix, converting the posture data into a vector representing the spatial posture of the toothbrush; inputting the pressure distribution matrix and the vector into a trained neural network model; and obtaining the tooth position recognition result output by the neural network model.

[0009] According to the toothbrush head pressure detection and control method provided by the present invention, the neural network model includes a first feature extraction network for processing the pressure distribution matrix, a second feature extraction network for processing the vector, and a classifier for fusing the two types of features and outputting a classification result.

[0010] According to the pressure detection and control method for the toothbrush head provided by the present invention, the preset pressure threshold for different tooth position information is a fixed value, or a personalized parameter set by the user for the tooth position information through an associated application.

[0011] According to the toothbrush head pressure detection and control method provided by the present invention, the control command is used to perform at least one of the following operations: adjusting the vibration intensity of the motor, changing the vibration mode, generating an alarm signal, or recording and displaying pressure data for the current teeth through an application.

[0012] According to the toothbrush head pressure detection and control method provided by the present invention, the training and / or updating process of the neural network model includes the following steps: acquiring multiple pressure distribution matrix sequences, posture data sequences, and corresponding real tooth position label data during user brushing processes as training samples; inputting the pressure distribution matrix sequences into the first feature extraction network, inputting the posture data sequences into the second feature extraction network, and adjusting the parameters of the entire network based on the difference between the output of the classifier and the real tooth position labels using a backpropagation algorithm; distributing the trained model parameters to the electric toothbrush or a smart terminal communicating with it; the real tooth position label data is obtained by: guiding the user to clean designated teeth in a preset order through the application of the smart electric toothbrush system; recording the pressure distribution matrix and posture data generated by the user during cleaning the designated teeth; and using the preset position information for the designated teeth as the real tooth position label corresponding to the recorded data.

[0013] The toothbrush head pressure detection and control method provided by the present invention further includes: simultaneously or subsequently providing interface elements to the user through the application program to receive user confirmation or correction feedback on the accuracy of the currently identified tooth position; using the user-confirmed or corrected tooth position information as a new real label, and together with the current pressure data and posture data, forming an incremental training sample; and using the incremental training sample to perform online or periodic incremental learning on the neural network model.

[0014] According to the toothbrush head pressure detection and control method provided by the present invention, the motion sensor is a nine-axis sensor.

[0015] The toothbrush head pressure detection and control method provided by this invention has the following advantages compared with the prior art: (1) By integrating the data from the flexible pressure sensor on the toothbrush head (pressure distribution matrix) and the motion sensor data on the brush handle (spatial attitude vector), and using a neural network model for intelligent calculation, this invention can determine the specific tooth position (such as incisors, canines, molars, etc.) that the brush head is currently cleaning in real time with high precision. This allows the invention to call up a personalized pressure threshold preset for that position for comparison, changing the traditional electric toothbrush's extensive management mode of using a single pressure threshold for the entire oral cavity, and realizing refined and personalized care.

[0016] (2) This invention can use user confirmation or correction feedback as new training samples to continuously optimize the neural network model for specific users' oral structure and brushing habits. This makes the toothbrush "understand" the user better the more it is used, and the accuracy of tooth position recognition and pressure control experience will continue to improve with the increase of usage time, forming a virtuous cycle of intelligent evolution. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating the pressure detection and control method for the toothbrush head provided by the present invention. Figure 2 This is a schematic diagram of the framework of the intelligent toothbrush system provided by the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0020] It should be noted that in the description of the embodiments of the present invention, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises 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 limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element. Those skilled in the art can understand the specific meaning of the above terms in the present invention according to the specific circumstances. Furthermore, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following related objects are in an "or" relationship.

[0021] The following is combined with Figures 1-2 This invention describes a method for detecting and controlling the pressure of a toothbrush head provided in an embodiment of the invention.

[0022] Figure 1 This is a flowchart illustrating the pressure detection and control method for the toothbrush head provided by the present invention, as shown below. Figure 1 As shown, including but not limited to the following steps: Step 101: Obtain pressure data during brushing by using a flexible pressure sensor mounted on the toothbrush head.

[0023] Flexible pressure sensors are directly mounted on the brush head (e.g., on the back of the bristles or at the connection to the handle) to measure the positive pressure applied to the teeth and gums in real time during brushing. Depending on the required level of precision, different numbers or various specifications of flexible pressure sensors can be designed, ranging from one to N.

[0024] Step 102: Acquire the toothbrush's posture data using a motion sensor mounted on the electric toothbrush.

[0025] Motion sensors (such as inertial measurement units, including accelerometers and gyroscopes) are built into the toothbrush body (toothbrush head or handle) to capture the toothbrush's posture data in real time.

[0026] Step 103: Based on the pressure data and the posture data, calculate the tooth position information corresponding to the current toothbrush head.

[0027] This invention fuses and analyzes pressure and posture data. By identifying specific motion patterns (e.g., horizontal reciprocating motion in the incisor region, and circular or vertical brushing in the molar region) and combining this with the angle of the brush head in space, the position of the tooth being brushed can be inferred.

[0028] Step 104: Compare the pressure data with the pressure threshold preset for the tooth position information.

[0029] This invention pre-sets pressure thresholds for different tooth positions. These thresholds can be set based on dental knowledge (e.g., the tolerance of the gingival region and the crown differs), or they can be customized by the user through an application associated with the electric toothbrush. This invention compares real-time pressure data with the threshold corresponding to the current tooth position.

[0030] Step 105: When the pressure data exceeds the corresponding preset pressure threshold, a control command is generated to adjust the working state of the electric toothbrush.

[0031] Based on the generated control commands, the electric toothbrush automatically adjusts its operating status to avoid injury caused by excessive force. Common adjustment methods include: adjusting the motor's vibration intensity, changing the vibration mode, generating alarm signals, or recording and displaying pressure data for the current teeth via an application.

[0032] For example, when the pressure exceeds the limit, the motor drive circuit immediately reduces the output power, significantly weakening the vibration amplitude and frequency; it automatically switches from a powerful mode such as "cleaning" or "whitening" to a gentle mode such as "sensitive" or "massage"; the LED light on the toothbrush handle changes from green to flashing red, or the ring light strip lights up in a red area; the toothbrush synchronizes the pressure data (including the time of exceeding the limit, location, and maximum pressure value) during each brushing process to the mobile app via Bluetooth.

[0033] Based on the above embodiments, as an optional embodiment, the toothbrush head pressure detection and control method provided by the present invention calculates tooth position information based on the pressure data and the posture data, specifically including: (1) Convert the pressure data into a pressure distribution matrix and convert the posture data into a vector representing the spatial posture of the toothbrush.

[0034] This invention can use multiple flexible pressure sensors to form a flexible pressure sensor array, directly generating a two-dimensional pressure distribution map (an M x N matrix).

[0035] The matrix form can reflect not only the magnitude of pressure but also its central location and distribution pattern. For example, the pressure distribution pattern differs when the bristle edge presses against the gum line versus when the bristle center presses against the crown. This provides neural networks with richer feature information.

[0036] (2) Input the pressure distribution matrix and the vector into the trained neural network model.

[0037] Optionally, the neural network model includes a first feature extraction network for processing the pressure distribution matrix, a second feature extraction network for processing the vector, and a classifier for fusing the two types of features and outputting a classification result.

[0038] Specifically, the neural network model consists of three parts: an L-layer convolutional neural network feature extractor, a K-layer fully connected feature extractor, and a J-layer fully connected classifier. The L-layer convolutional neural network feature extractor receives an M×N stress matrix, the K-layer fully connected feature extractor receives multi-dimensional pose data, and the J-layer fully connected classifier receives the outputs of the two features mentioned above and outputs the final tooth position.

[0039] (3) Obtain the tooth position recognition result output by the neural network model.

[0040] The model directly outputs the most probable tooth location label, or outputs the confidence score for each location. The system selects the one with the highest confidence score as the recognition result. This result will be used for subsequent pressure threshold comparison.

[0041] Based on the above embodiments, as an optional embodiment, the pressure detection and control method for the toothbrush head provided by the present invention has a fixed preset pressure threshold for different tooth position information, or a personalized parameter set by the user for the tooth position information through an associated application.

[0042] For example, a set of scientifically researched pressure threshold data can be pre-stored in the toothbrush firmware or system. These thresholds are typically established by dental experts based on the tolerance of tooth enamel and gums, setting different safe upper limits for different tooth areas (such as the sensitive gum line and the wear-resistant chewing surface of molars).

[0043] For example, a mobile application (APP) can provide users with an interactive interface that allows them to adjust the sensitivity or pressure threshold of specific tooth areas according to their own situation.

[0044] Based on the above embodiments, as an optional embodiment, the toothbrush head pressure detection and control method provided by the present invention includes the following steps in the training and / or updating process of the neural network model: (1) Obtain the pressure distribution matrix sequence, posture data sequence and corresponding real tooth position label data of multiple users during the brushing process as training samples; (2) Input the pressure distribution matrix sequence into the first feature extraction network, input the posture data sequence into the second feature extraction network, and adjust the parameters of the entire network through the backpropagation algorithm based on the difference between the output of the classifier and the real tooth position label; (3) Send the trained model parameters to the electric toothbrush or the smart terminal that communicates with it; The actual tooth location label data is obtained through the following method: (1) The user is guided to clean designated teeth in a preset order through the application of the intelligent electric toothbrush system; (2) Record the pressure distribution matrix and posture data generated by the user during the cleaning of the specified teeth; (3) The preset position information of the specified tooth is used as the real tooth position label corresponding to the recorded data.

[0045] This invention provides a low-cost, high-accuracy data annotation scheme. Through continuous data collection and model updates, the intelligence level of the product will continuously accumulate and evolve, improving the user experience.

[0046] Based on the above embodiments, as an optional embodiment, the present invention provides a method for detecting and controlling pressure on a toothbrush head, further comprising: (1) While generating control commands to adjust the working state of the electric toothbrush, or afterward, the application provides the user with interface elements to receive the user's confirmation or correction feedback on the accuracy of the currently identified tooth position.

[0047] At the same time as or after the system intervenes due to excessive pressure (such as reducing speed and pressure).

[0048] Interface Elements: The app interface can be very simple. For example, it could display a message in the corresponding position on the virtual tooth model: "The system has detected that you are brushing your lower left molar. Is this correct?" and provide two buttons: "Correct" and "Fix". If the user clicks "Fix", a list should pop up allowing the user to select the correct tooth position.

[0049] (2) The tooth position information confirmed or corrected by the user is used as the new real label, and together with the pressure data and posture data at the current moment, they form an incremental training sample.

[0050] User confirmations or corrections are extremely valuable supervisory signals. They provide real training data labels, annotated by users themselves, in real-world usage scenarios, to form authentic incremental training samples for training neural network models with superior performance.

[0051] (3) Use the incremental training samples to perform online or periodic incremental learning on the neural network model.

[0052] Optionally, these feedback samples are temporarily stored locally on the toothbrush and periodically (e.g., after each brushing session) synchronized to the cloud. The cloud then aggregates all users' anonymous feedback data, performs centralized training to generate a more powerful new version model, and pushes it to all users via OTA updates.

[0053] The pressure detection and control method described above for toothbrush heads can be applied to specific electric toothbrushes. Figure 2 This is a schematic diagram of the framework of the intelligent toothbrush system provided by the present invention. The following is in conjunction with... Figure 2 and one A complete embodiment will be provided to illustrate the application of the technical solution of the present invention.

[0054] I. Hardware System Composition This system consists of two main parts: a replaceable smart toothbrush head and a smart toothbrush handle.

[0055] 1. Toothbrush head Pressure sensing module: Multiple flexible pressure sensors are arranged in an array beneath the toothbrush bristle base. These sensors directly convert the physical pressure of the bristles in contact with the teeth / gingiva into analog electrical signals. The number of sensors (M×N) can be configured according to the required detection accuracy.

[0056] Signal processing module: A low-power MCU acquires analog signals from multiple flexible pressure sensors through its built-in ADC (analog-to-digital converter) and converts them into a digital pressure distribution matrix.

[0057] Wireless communication and power supply module: This module consists of an RFID tag chip and its connected magnetic coil. It has two core functions: Wireless power supply: It receives electromagnetic energy from the brush handle, and after rectification and voltage regulation by the RFID chip, it powers the entire brush head module (low-power MCU, pressure sensor), realizing the "passive" nature of the brush head.

[0058] Data communication: The low-power MCU sends the processed pressure data to the RFID tag chip. The chip encodes the data using load modulation technology and then wirelessly transmits it back to the brush handle via a magnetic coil.

[0059] 2. Toothbrush handle Energy and data interaction module: RFID reader and its magnetic coil. This module generates a radio frequency field to wirelessly power the brush head; simultaneously, it demodulates the data returned by the brush head and transmits the pressure data to the main control MCU via serial port.

[0060] Attitude perception module: The main control board integrates a nine-axis sensor (not shown separately in the figure, usually connected to the main control MCU) to acquire the toothbrush's attitude data (9-dimensional vector) such as acceleration and angular velocity in space in real time.

[0061] The nine-axis sensor consists of a three-axis accelerometer (X / Y / Z axis gravitational acceleration), a three-axis gyroscope (X / Y / Z axis angular velocity), and a three-axis magnetometer (X / Y / Z axis magnetic field strength), and achieves spatial attitude perception through data fusion.

[0062] Core Computing and Control Module: The main control MCU is the brain of the system, responsible for running the core algorithms. It internally stores a pre-trained multimodal neural network model. It receives pressure and posture data, calculates tooth position and pressure, and generates control commands.

[0063] Execution module: Based on the judgment result, the main control MCU controls the motor drive circuit, adjusts the vibration intensity or mode of the motor, and sends an alarm to the user via LED lights or Bluetooth connection.

[0064] II. Working Methods and Algorithm Flow The method flow of the present invention specifically includes the following steps: S1: System Initialization The user attaches the toothbrush head to the handle. The RFID reader on the handle then activates, powering the brush head via electromagnetic induction, which in turn activates the various electronic components within the brush head.

[0065] S2: Real-time acquisition of multimodal data During brushing, the flexible pressure sensor array on the brush head generates a pressure distribution matrix (M×N) in real time.

[0066] The nine-axis motion sensor inside the brush handle generates attitude data vectors (9-dimensional) in real time.

[0067] S3: Intelligent tooth position calculation The main control MCU inputs the collected pressure matrix and attitude vector into a pre-trained multimodal neural network model for calculation.

[0068] The first feature extraction network (L-layer CNN) is specifically designed to process the pressure distribution matrix and extract its spatial distribution features.

[0069] The second feature extraction network (K-layer fully connected network) is specifically designed to process pose vectors and extract their motion trajectory features.

[0070] Classifier (J-layer fully connected network): It fuses the high-level features output by the two feature extraction networks and finally outputs a recognition result, namely the current tooth position information (such as "left lower first molar buccal surface").

[0071] S4: Personalized Stress Threshold Comparison and Decision Making The main control MCU queries the preset pressure threshold for the identified tooth position. This threshold can be: Fixed value: Default value based on dental medical consensus.

[0072] Personalized parameters: Values ​​that users can set individually for different areas of the teeth via a mobile app.

[0073] Compare real-time pressure data with the corresponding threshold.

[0074] S5: Intelligent Feedback and Control When the real-time pressure exceeds the preset threshold, the main control MCU immediately generates control commands to adjust the electric toothbrush's operating status, including but not limited to: Adjust the motor: reduce vibration intensity or switch to a gentle mode.

[0075] Generate an alarm: make the brush handle LED light flash red, or send a notification via the APP.

[0076] Meanwhile, the app interface provides feedback options, allowing users to confirm or correct tooth positioning results. This feedback data can be used for incremental learning of the model, making the system more and more accurate with use.

[0077] III. Advantages of the Embodiments of the Invention (1) Direct and accurate measurement: The pressure is measured directly under the bristles, avoiding the error caused by indirect measurement through the transmission components.

[0078] (2) Truly passive brush head: wireless power supply and contactless communication solve the problems of waterproofing, corrosion and battery replacement, with long life and high reliability.

[0079] (3) Artificial intelligence positioning: Using a multimodal neural network, it can accurately calculate the position of teeth, laying the foundation for personalized care of "each tooth".

[0080] (4) Complete intelligent closed loop: From perception, analysis, decision-making to control, a complete intelligent closed loop is formed, which effectively protects the user's oral health.

[0081] This invention provides a complete solution from hardware to software, achieving precise and personalized control of brushing pressure through innovative passive brush head design and advanced artificial intelligence algorithms.

[0082] 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 of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method of pressure detection and control of a toothbrush head, characterized by, The method comprises: acquiring pressure data during brushing through a flexible pressure sensor arranged on the toothbrush head; acquiring posture data during brushing through a motion sensor arranged on the electric toothbrush; based on the pressure data and the posture data, obtaining the tooth position information corresponding to the current toothbrush head through calculation; comparing the pressure data with the preset pressure threshold for the tooth position information; when the pressure data exceeds the corresponding preset pressure threshold, generating a control instruction to adjust the working state of the electric toothbrush.

2. The pressure detection and control method of a toothbrush head according to claim 1, wherein, Based on the pressure data and the posture data, the tooth position information is calculated, specifically including: convert the pressure data into a pressure distribution matrix, and convert the posture data into a vector representing the spatial posture of the toothbrush; input the pressure distribution matrix and the vector into a trained neural network model; obtain the recognition result of the tooth position output by the neural network model.

3. The method of claim 2, wherein the pressure of the toothbrush head is detected by a pressure sensor. The neural network model includes a first feature extraction network for processing the pressure distribution matrix, a second feature extraction network for processing the vector, and a classifier for fusing the two types of features and outputting a classification result.

4. The method of claim 1, wherein the method further comprises: The pressure threshold preset for different tooth position information is a fixed value, or a personalized parameter set by the user through the associated application program for the tooth position information.

5. The method of claim 1, wherein the method further comprises: The control instruction is used to perform at least one of the following operations: adjust the vibration intensity of the motor, change the vibration mode, generate an alarm signal, or record and display the pressure data for the current tooth through the application program.

6. The method of claim 2, wherein the pressure of the toothbrush head is detected by a pressure sensor. The training and / or updating process of the neural network model includes the following steps: obtain a sequence of pressure distribution matrices, a sequence of posture data, and corresponding real tooth position label data in the tooth brushing process of multiple users as training samples; input the sequence of pressure distribution matrices into the first feature extraction network, input the sequence of posture data into the second feature extraction network, and based on the difference between the output of the classifier and the real tooth position label, adjust the parameters of the entire network through the back propagation algorithm; distribute the trained model parameters to the electric toothbrush or the intelligent terminal in communication with the electric toothbrush; The real tooth position label data is obtained by the following method: guide the user to clean the specified teeth in a preset order through the application program of the intelligent electric toothbrush system; record the pressure distribution matrix and the posture data generated by the user during cleaning the specified teeth; take the position information preset for the specified teeth as the real tooth position label corresponding to the recorded data.

7. The method of claim 6, wherein the pressure of the toothbrush head is detected by a pressure sensor. Further comprising: at the same time or after generating the control instruction to adjust the working state of the electric toothbrush, provide interface elements to the user through the application program for receiving the user's confirmation or correction feedback on the accuracy of the currently identified tooth position; take the tooth position information confirmed or corrected by the user as a new real label, and together with the pressure data and the posture data at the current time, form an incremental training sample; use the incremental training sample to perform online or periodic incremental learning on the neural network model.

8. The method of claim 1, wherein the method further comprises: The motion sensor is a nine-axis sensor.

Images