Bruxism event detection system and bruxism data processing method

Abstract

The application provides a tooth grinding event detection system and a tooth grinding data processing method. The method comprises the following steps: acquiring facial electromyography data and auxiliary data collected by a wearable recorder worn on the face of a detection object; determining a suspected tooth grinding event according to the facial electromyography data and a tooth grinding threshold; and verifying the suspected tooth grinding event by using the auxiliary data to determine a tooth grinding event. The wearable recorder collects auxiliary data while collecting facial electromyography data. When analyzing a tooth grinding event, the facial electromyography data and a tooth grinding threshold are used to preliminarily screen out a suspected tooth grinding event, and then the auxiliary data is combined with the suspected tooth grinding event to analyze and verify an actual tooth grinding event one by one. As a result, artifacts in the electromyography data can be removed, and the accuracy of tooth grinding event recognition is improved.

Description

[0001] This application is a divisional application of the patent application No. 2023101352524, titled "Bruxism event detection system and bruxism data processing method", filed on February 20, 2023. TECHNICAL FIELD

[0002] The present application relates to the field of medical devices, in particular to a bruxism event detection system and a bruxism data processing method. BACKGROUND

[0003] Bruxism is a common and frequently-occurring disease in stomatology, and the most common type is that patients grind their teeth after falling asleep at night. Long-term tooth grinding can cause abnormal tooth wear, and cause various diseases such as toothache, loose tooth, broken tooth, and inflammation. In order to diagnose and treat bruxism, doctors need to know the tooth grinding actions of patients during the whole sleep period. Facial electromyography data can be used to detect tooth grinding actions, and the detection object can wear an electromyography detection device all night to collect facial electromyography data during sleep.

[0004] When analyzing electromyography data, it is generally to compare facial electromyography data with a set threshold to screen out data segments that may be tooth grinding actions. However, in actual application, the detection object may have various conditions during sleep, such as turning over, facial movements, etc., and these non-tooth grinding actions may also cause obvious fluctuations in facial electromyography data, so it is easy to misjudge. SUMMARY

[0005] Therefore, the present application provides a bruxism data processing method, which comprises: acquiring facial electromyography data and auxiliary data collected by a wearable recorder worn on the face of a detection object; determining a suspected tooth grinding event according to the facial electromyography data and a tooth grinding threshold; and verifying the suspected tooth grinding event by using the auxiliary data to determine a tooth grinding event.

[0006] Optionally, the auxiliary data comprises acceleration data.

[0007] Optionally, verifying the suspected tooth grinding event by using the auxiliary data to determine a tooth grinding event further comprises: determining time information of the suspected tooth grinding event; extracting acceleration data corresponding to the time information; and determining whether the suspected tooth grinding event is a tooth grinding event according to the variation of the extracted acceleration data.

[0008] Optionally, the time information comprises a start time and an end time of the suspected tooth grinding event; and in the step of verifying the suspected tooth grinding event by using the auxiliary data to determine a tooth grinding event, the auxiliary data between the start time and the end time is extracted.

[0009] Optionally, the determining the suspected bruxism event according to the facial myoelectric data and the bruxism threshold further comprises: screening all data segments exceeding the bruxism threshold from the facial myoelectric data; and identifying at least two types of suspected bruxism events according to time lengths of each of the screened data segments, wherein the types include a persistent attack and a phase attack, and the length of the data segment of the persistent attack is greater than the length of the data segment of the phase attack.

[0010] Optionally, the identifying at least two types of suspected bruxism events according to the time lengths of each of the screened data segments further comprises: for each of the all data segments, respectively judging whether the duration t2-t1 of the data segment satisfies t tonic >t2-t1≥t phasic , or t2-t1≥t tonic ; recording the data segment satisfying t tonic >t2-t1≥t phasic as the suspected bruxism event of the phase attack, and recording the data segment satisfying t2-t1≥t tonic as the suspected bruxism event of the persistent attack.

[0011] Optionally, after the identifying at least two types of suspected bruxism events according to the time lengths of each of the screened data segments, the method further comprises: for the suspected bruxism event of the phase attack, judging whether a time interval between every two adjacent suspected bruxism events is lower than an interval threshold; and when the time interval between two adjacent suspected bruxism events is lower than the interval threshold, merging the two adjacent suspected bruxism events.

[0012] Optionally, the merging is iteratively performed until the time interval between every two adjacent suspected bruxism events is higher than the interval threshold; and after the merging, the method further comprises: for each suspected bruxism event obtained by the merging, acquiring a number of the suspected bruxism events merged by the suspected bruxism event; and deleting a merging result whose number is lower than a number threshold.

[0013] Correspondingly, the present application provides a bruxism data processing device, comprising: a processor and a memory connected with the processor; wherein the memory stores instructions executable by the processor, and the instructions are executed by the processor to make the processor execute the above bruxism data processing method.

[0014] Correspondingly, the present application provides a bruxism event detection system, comprising: a host and a wearable recorder, wherein the wearable recorder is used to collect facial myoelectric data and auxiliary data; and the host is used to execute the above bruxism data processing method.

[0015] According to the bruxism event detection system and bruxism data processing method provided by the present invention, an auxiliary data is collected simultaneously with facial electromyography (EMG) data by a wearable recorder. When analyzing bruxism events, suspected bruxism events are first preliminarily screened using facial EMG data and bruxism thresholds. Then, the suspected bruxism events are analyzed one by one in combination with the auxiliary data to verify the actual bruxism events. This can remove artifacts in the EMG data and improve the accuracy of bruxism event identification. Attached Figure Description

[0016] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0017] Figure 1 This is a schematic diagram of a bruxism event detection system;

[0018] Figure 2 This is a circuit architecture diagram of the wearable recorder in an embodiment of the present invention;

[0019] Figure 3 This diagram shows the connection status between the wearable recorder and the flexible electromyography recording electrodes.

[0020] Figure 4 This is a flowchart of the molar data processing method in an embodiment of the invention.

[0021] Figure 5 This is a segment of electromyography data containing two seizure types. Detailed Implementation

[0022] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] Furthermore, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0024] Figure 1 A bruxism event detection system is illustrated, comprising a host device 1 and a wearable recorder 2, wherein the wearable recorder 2 is adapted to be worn on the face for collecting electromyographic (EMG) data and auxiliary data. The host device 1 can acquire EMG data and auxiliary data through the wearable recorder 2. The host device 1 may be equipped with one or more wearable recorders 2. Figure 1The image shows two wearable recorders 2. The wearable recorders 2 can be removed from the charging slot of the main unit 1, connected to flexible electromyography (EMG) electrodes, and worn on the masseter muscle area on both sides of the face.

[0025] In operation, the main unit 1 and the wearable recorder 2 communicate wirelessly, transmitting control commands and data collected by the recorder. When the main unit 1 is off, the wearable recorder 2 is in sleep mode. When the main unit 1 is powered on, the main unit 1 and the equipped wearable recorder 2 automatically establish a wireless connection. The user can use the software system of the main unit 1 to disconnect one or more wearable recorders 2 and put them into sleep mode, using only the wearable recorders 2 that are not in sleep mode. When the main unit 1 is off, all wearable recorders 2 disconnect and enter sleep mode.

[0026] Data collected by the wearable recorder 2 can also be transmitted to general electronic devices, such as smartphones. Electronic devices pre-installed with the accompanying software can achieve all the same software functions as the main unit 1.

[0027] Figure 2 This is the circuit architecture diagram of the wearable recorder 2, which contains three sensing modules: an electromyography (EMG) acquisition module, which directly collects raw EMG signals from flexible EMG recording electrodes plugged into the recorder and inputs them into the recorder, where they undergo further signal processing such as amplification, filtering, and analog-to-digital conversion sampling; a body motion acquisition module, which uses an accelerometer to collect the user's head movement information to determine whether EMG signal fluctuations are artifacts introduced by head movements; and a sound acquisition module, which uses a microphone to collect ambient sound to determine whether EMG signal fluctuations are accompanied by teeth grinding, thus helping to determine whether the EMG signal is generated by teeth grinding. All data collected by the three sensing modules is input to a microcontroller, which performs data compression, packet segmentation, and other processing before inputting it to a wireless communication module for transmission.

[0028] The recorder has an internal memory. Data output from the microcontroller can be written to this local memory as a backup while being wirelessly transmitted. If data loss is detected during wireless transmission, the microcontroller can retrieve the data from the memory. There are two data export modes: the microcontroller reads the data from the memory and outputs it to the wireless communication module for wireless transmission; or the recorder is placed in the charging slot of the main unit, and the data in the memory is transmitted to the main unit 1 through contacts on the bottom of the recorder and contacts in the charging slot.

[0029] The wireless communication module has transmit and receive functions. In addition to wirelessly transmitting the data input by the microcontroller, it can also receive control commands from external sources (such as host 1). After the commands are input into the microcontroller, they can control the wireless connection and disconnection of the recorder, the activation and deactivation of each sensor module, and the reading and writing of the memory.

[0030] In addition, the power module inside the recorder includes a battery, power management circuitry, and charging circuitry, which supplies power to the other modules of the recorder.

[0031] In other embodiments of the recorder circuit architecture, a sound acquisition module may not be included. The sound acquisition module may be located inside the instrument host 1 to realize the function of acquiring ambient sound.

[0032] like Figure 3 As shown, the wearable recorder 2 needs to be used in conjunction with a matching flexible electromyography (EMG) recording electrode 3. The EMG recording electrode 3 consists of a recorder adhesive backing 31, a connector 32, a skin adhesive backing 33, and a flexible circuit 34. The recorder adhesive backing 31 is used to attach and fix the wearable recorder 2. When not in use, its surface is covered with a release film; when in use, the release film is removed and the electrode is attached to the recorder. The connector 32 can be plugged into the connector 22 on the recorder to input the EMG signals collected by the electrode into the recorder's internal circuitry. The skin adhesive backing 33 is made of medical tape; when not in use, its surface is covered with a release film; when in use, the release film is removed and the electrode is attached to the skin of the masseter muscle area on the face. The flexible circuit 34 is the core structure of the flexible EMG recording electrode 3, and it has electrode points and leads on it, responsible for collecting and transmitting EMG signals.

[0033] like Figure 4 As shown, this embodiment of the invention provides a method for processing molar data, which can be executed by the host 1 or a general electronic device, and includes the following operations:

[0034] S1. Acquire facial electromyography (EMG) data and auxiliary data from a wearable recorder worn on the subject's face. Typically, this method is used to detect involuntary teeth grinding during sleep, and the acquired EMG and auxiliary data are data collected by the subject during sleep.

[0035] The auxiliary data described in this application may be one or both of acceleration data and sound data. After monitoring, the software system processes the received raw data, including digital filtering through a preset filter and time alignment of the facial electromyography data and auxiliary data based on the time information in the data packet.

[0036] S2. Suspected bruxism events are determined based on facial electromyography (EMG) data and a bruxism threshold. The bruxism threshold is a preset value. Facial EMG data during the monitoring period is compared with the bruxism threshold. Data segments exceeding the bruxism threshold are identified as evidence of suspected bruxism, thus generating a suspected bruxism event. Multiple suspected bruxism events can also be merged based on time; for example, two suspected bruxism events that are very close in time can be merged into a single suspected bruxism event.

[0037] S3. Utilize auxiliary data to verify and confirm suspected bruxism events. Analyze each suspected bruxism event individually. Each suspected bruxism event corresponds to a segment of electromyography (EMG) data, and each suspected bruxism event has start and end time points. Auxiliary data can be extracted within the start and end time range, or the scope can be expanded to include longer auxiliary data beyond the start and end time points. Alternatively, only auxiliary data near the start time point or only auxiliary data near the end time point of the EMG data segment can be extracted.

[0038] In one embodiment, only acceleration data is used as auxiliary data, and step S3 specifically includes:

[0039] S31A, determine the time information of the suspected teeth grinding event, which may be one or both of the start time and end time;

[0040] S32A, extract the acceleration data corresponding to the time information, which may be acceleration data within a period of time before and after the start time point, acceleration data within a period of time before and after the end time point, or acceleration data between the start time and the end time.

[0041] S33A, determine whether the suspected teeth grinding event is indeed a teeth grinding event based on the extracted acceleration data values. Specifically, the extracted acceleration data is also a segment of acceleration data that changes over time. The change in acceleration data segment is compared with a preset threshold. The change refers to the difference between the acceleration value at a certain moment and the initial acceleration value of the data segment. If the change in acceleration at any moment within this time period exceeds the preset threshold, it indicates that the wearer has performed a significant action during this time period, such as turning over, turning the head, or getting up at night. In this case, the suspected teeth grinding event is determined to be a spoofing. If the change in acceleration at any moment within this time period does not reach the preset threshold, it indicates that the wearer has remained still during this time period. In this case, it is determined that the electromyographic fluctuations during this time period are not caused by abnormal facial movements of the tested subject, and therefore, a teeth grinding event can be confirmed.

[0042] In another embodiment, only sound data is used as auxiliary data, and step S3 specifically includes:

[0043] S31B, Determine the time information of the suspected teeth grinding event, for details please refer to step S31A;

[0044] S32B, extract the sound data corresponding to the time information, please refer to step S32A for details;

[0045] S33B, determine whether a suspected teeth grinding event is indeed a teeth grinding event based on the characteristics of the extracted sound data. Specifically, the extracted sound data is also a time-varying segment of sound data. If the subject grinds their teeth during this time segment, a teeth grinding sound may be produced. This sound is characterized by accompanying the teeth grinding action, starting with the beginning of the grinding action, increasing in intensity with the increase in intensity of the grinding action, and finally disappearing with the disappearance of the grinding action. This is significantly different from the sounds produced by other actions such as rubbing against a pillow. (See reference...) Figure 5 In the electromyography (EMG) data segments (EMG data within a time period) corresponding to suspected teeth grinding events, the EMG data exhibits obvious fluctuation characteristics, that is, it gradually increases from a low level and then gradually decreases. Therefore, the sound intensity change characteristics should also follow this pattern. If the extracted sound intensity change matches the set sound intensity change characteristics, that is, it matches the EMG intensity change characteristics of teeth grinding, it indicates that the subject has teeth grinding, and the suspected teeth grinding event is indeed a teeth grinding event. If the extracted sound intensity change characteristics do not match the EMG intensity change characteristics of teeth grinding, it indicates that the subject has not grinded teeth, and other body movements may have caused the facial EMG data to exceed the teeth grinding threshold. Therefore, the suspected teeth grinding event is determined to be an artifact and not a real teeth grinding event.

[0046] In the third embodiment, acceleration data and sound data are used as auxiliary data. Step S3 specifically includes:

[0047] S31C, Determine the time information of the suspected teeth grinding event; for details, please refer to step S31A.

[0048] S32C, extract the acceleration data corresponding to the time information, please refer to step S32A for details;

[0049] S33C: Determine whether the change in the extracted acceleration data is greater than a threshold. See step S33A for details. If the change in the extracted acceleration data is greater than the threshold, proceed to step S4C; otherwise, classify the suspected teeth grinding event as a teeth grinding event (confirming it as a teeth grinding action).

[0050] S34C, extract the sound data corresponding to the time information, please refer to step S32A for details;

[0051] S35C, determine whether the characteristics of the sound data match the characteristics of teeth grinding sounds. Specifically, refer to step S33B. If the characteristics of the sound data do not match the characteristics of teeth grinding sounds, proceed to step S6C; otherwise, determine the suspected teeth grinding event as a teeth grinding event (confirm that it belongs to the teeth grinding action).

[0052] S36C was determined to be a non-molar event (artifact).

[0053] This embodiment can be regarded as a combination of the first two embodiments. This embodiment uses two kinds of auxiliary information to comprehensively analyze suspected teeth grinding events, which can further improve the accuracy of the judgment results.

[0054] According to the teeth grinding data processing method provided in the embodiments of the present invention, an auxiliary data is collected simultaneously with facial electromyography data by a wearable recorder. When analyzing teeth grinding events, suspected teeth grinding events are first preliminarily screened using facial electromyography data and teeth grinding thresholds. Then, the suspected teeth grinding events are analyzed one by one in combination with the auxiliary data to verify the actual teeth grinding events. This can remove artifacts in the electromyography data and improve the accuracy of teeth grinding event identification.

[0055] In an optional embodiment, this solution distinguishes between two types of suspected bruxism events: tonic events and phasic events. A tonic event is defined as a single masseter muscle activity with electromyographic data exceeding the molar threshold and a relatively long duration (e.g., 1–3 seconds); a phasic event is defined as at least N consecutive electromyographic activities with short durations (e.g., 0.1–0.5 seconds), short intervals, and electromyographic data exceeding the molar threshold.

[0056] The above step S2 specifically includes:

[0057] All data segments exceeding the molar threshold were filtered out from the facial electromyography data.

[0058] Based on the time length of each selected data segment, at least two types of suspected bruxism events were identified, including continuous attacks and phased attacks, wherein the length of the data segment for continuous attacks is greater than the length of the data segment for phased attacks.

[0059] Two time thresholds t can be set. tonic and t phasic To distinguish between two types of suspected teeth grinding events, t tonic >t phasic For each data segment in the entire data segment, determine whether the duration t2-t1 of the data segment satisfies t. tonic >t2-t1≥t phasic , or t2-t1≥t tonic ; where t2 represents the time point corresponding to the end of the data segment, and t1 represents the time point corresponding to the beginning of the data segment.

[0060] For satisfying t tonic >t2-t1≥t phasic The data segment records suspected bruxism events with phase-dependent onset, for those satisfying t2-t1≥t tonic The data segment records suspected bruxism events that occur repeatedly.

[0061] Figure 5 Typical electromyographic data segments for two types of events are shown. Based on the above process, a single segment of continuous bruxism events and multiple segments of phased bruxism events can be identified. For the identified continuous bruxism type data segment, it can be directly determined as a suspected continuous bruxism; for the phased bruxism type data segments, the number and interval can be counted as results, or further processing can be performed.

[0062] As an optional embodiment, the following processing may also be performed:

[0063] For suspected bruxism events identified as having phase-related attack patterns, it is determined whether the time interval between two adjacent suspected bruxism events is lower than an interval threshold. When the time interval between two adjacent suspected bruxism events is lower than the interval threshold, the two adjacent suspected bruxism events are merged. It should be noted that merging two events in this scheme does not mean fusing two data segments into a single data segment, but rather treating the two data segments as a single event.

[0064] Specifically, the first step is to extract the data segments of the two initial suspected phase attacks (referred to as phasic data segments), and then determine whether the time interval Δt between the two phasic data segments is less than the interval threshold t. interval If the conditions are met, the two phasic data segments belong to the same phasic seizure, and the two phasic data segments are merged into one phasic action record, which is then used to continue the Δt judgment with the next phasic data segment. If the conditions are not met, the two phasic data segments do not belong to the same phasic seizure, and the second phasic data segment is used to continue the Δt judgment with the next phasic data segment. This process is repeated until the Δt judgment of the last two phasic data segments is completed, resulting in the merged phasic action record. The above merging process is executed iteratively, and the time interval between adjacent phasic data segments after merging is higher than the interval threshold.

[0065] Based on this, the following further processing can be performed:

[0066] For each suspected bruxism event obtained through merging, the number of merging events is determined. During the phasic data segment merging process, the merging count for each new phasic data segment is recorded. Merging results with a count exceeding a threshold are classified as suspected phase-related events. Figure 5 Taking the electromyography (EMG) data shown as an example, initially five phasic data segments could be selected. After merging, they were merged into one phasic event, resulting in a merging count of 5. Assuming a threshold of 3, then... Figure 5The example shown contains one suspected phased attack event; if there is a merge result with a merge count of less than 3, then the merge result does not belong to the suspected phased attack event.

[0067] The preferred embodiments provided by the present invention can distinguish between two different bruxism events, especially accurately identify phase-related events. The frequency and timing of the two events play a good auxiliary role in the diagnosis and treatment of bruxism.

[0068] After the above series of processes, step S2 identified two suspected bruxism events, and step S3 analyzed each of these events individually. It should be noted that in step S3, the time information for the suspected tonic event is... Figure 5 The t shown tonic For suspected phasic events obtained from the merger, the time information is from the start time of the first phasic data segment before the merger to the end time of the last phasic data segment before the merger.

[0069] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0070] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0071] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1The function specified in one or more boxes.

[0072] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0073] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A method for processing molar data, characterized in that, include: Acquire facial electromyography data and auxiliary data collected by a wearable recorder worn on the face of the test subject, wherein the auxiliary data includes acceleration data; Suspected bruxism events were determined based on the facial electromyography data and bruxism threshold. Determine the timing information of the suspected teeth grinding event; Extract acceleration data corresponding to the time information; this acceleration data is obtained by the accelerometer collecting the user's head movement information and is used to determine whether the electromyographic signal fluctuation is an artifact introduced by head movement; The suspected teeth grinding event is determined to be a teeth grinding event based on the change in the extracted acceleration data; If the change in acceleration exceeds a preset threshold, the corresponding suspected teeth grinding event is determined to be a fake. If the change in acceleration does not reach a preset threshold, the corresponding suspected teeth grinding event is determined to be a teeth grinding event.

2. The method according to claim 1, characterized in that, The time information includes the start and end times of the suspected teeth grinding event; in the step of verifying the suspected teeth grinding event using the auxiliary data to determine the teeth grinding event, the auxiliary data between the start and end times is extracted.

3. The method according to claim 1, characterized in that, Identifying suspected bruxism events based on the facial electromyography data and bruxism threshold further includes: Filter out all data segments exceeding the molar threshold from the facial electromyography data; Based on the time length of each of the selected data segments, at least two types of suspected bruxism events are identified, including continuous attacks and phased attacks, wherein the length of the data segment for continuous attacks is greater than the length of the data segment for phased attacks.

4. The method according to claim 3, characterized in that, Based on the time length of each of the selected data segments, at least two types of suspected bruxism events are identified, further including: For each of the data segments, determine whether the duration t2-t1 of the data segment satisfies t tonic >t2-t1≥t phasic , or t2-t1≥t tonic Among them, two time thresholds t tonic and t phasic Used to distinguish between two types of suspected bruxism events, t tonic >t phasic t2 represents the time point corresponding to the end of the data segment, and t1 represents the time point corresponding to the beginning of the data segment. For satisfying t tonic >t2-t1≥t phasic The data segment records suspected bruxism events with phase-dependent onset, for those satisfying t2-t1≥t tonic The data segment records suspected bruxism events that occur repeatedly.

5. The method according to claim 3, characterized in that, After identifying at least two types of suspected bruxism events based on the time length of each of the selected data segments, the process further includes: For suspected bruxism events of phase-related occurrence type identified, determine whether the time interval between two adjacent suspected bruxism events is lower than the interval threshold; When the time interval between two adjacent suspected molar events is less than the interval threshold, the two adjacent suspected molar events are merged.

6. The method according to claim 5, characterized in that, The merging process is performed iteratively until the time interval between any two adjacent suspected molar events is higher than the interval threshold. After the merge process, it also includes: For each suspected bruxism event obtained from the merging process, obtain the number of suspected bruxism events merged with it; Delete merge results where the number of merges is less than the threshold.

7. A molar data processing device, characterized in that, include: A processor and a memory connected to the processor; wherein the memory stores instructions executable by the processor, the instructions being executed by the processor to cause the processor to perform the molar data processing method as described in any one of claims 1-6.

8. A bruxism event detection system, characterized in that, include: The main unit and the wearable recorder, among which The wearable recorder is used to collect facial electromyography data and auxiliary data; The host computer is used to execute the molar data processing method as described in any one of claims 1-6.

Images