Non-contact vital sign monitoring method and system applied to dental chair

Abstract

The present application relates to the technical field of vital sign monitoring, and discloses a non-contact vital sign monitoring method and system applied to a dental chair, comprising: synchronously collecting a physiological acoustic signal of a headrest and a mechanical vibration signal of a chair back, determining a reference time point of each heart cycle by a first heart sound event of the acoustic signal, and then separating and subtracting a low-frequency respiratory baseline of the mechanical vibration signal for correction, and then determining a mechanical response time point according to the reference time point, and then calculating a time interval between the two, the present application constructs a new monitoring dimension of calibrating a mechanical signal by an internal physiological time mark, the time interval parameter obtained by the present application is free from the influence of the drift of a mechanical transfer function caused by the change of a patient's posture, thereby avoiding the measurement artifacts in amplitude and shape of a traditional vibration signal, and providing a stable and reliable vital sign state evaluation for a diagnosis and treatment process.

Description

TECHNICAL FIELD

[0001] The application relates to a non-contact vital sign monitoring method and system applied to a dental chair and belongs to the technical field of vital sign monitoring. BACKGROUND

[0002] At present, continuous monitoring of the vital signs of a patient during diagnosis and treatment is one of the technical means for guaranteeing the safety of the patient, and in the prior art, a common method is to use a vibration sensor arranged on a bearing platform such as a chair or a bed to collect weak vibration signals of the body caused by physiological activities such as heartbeats, and analyze parameters such as heart rate from the signals, and this method has clear application value in specific scenarios such as dental diagnosis and treatment due to its convenience in application.

[0003] However, the application effect of the above method is restricted by a basic physical condition, and the effectiveness of the technical solution depends on a basic assumption that the patient chair coupling system is regarded as a transmission medium with short-term constant physical characteristics, but in actual diagnosis and treatment, the fine adjustment of the patient's body posture or the change of muscle tension will cause dynamic changes of the mechanical transfer function, and the direct consequence is that the same mechanical force generated by the heartbeat will generate vibration signals with different shapes and amplitudes when transmitted to the sensor.

[0004] To solve this problem, the usual improvement ideas focus on the signal processing level, such as using adaptive filtering or pattern recognition algorithms to try to extract information from the distorted signal form, and the basis of such ideas is still to regard the change of the signal form as external noise interference, but it cannot distinguish whether the signal fluctuation is caused by the real change of the physiological state of the patient or the signal artifact generated by the change of the mechanical transmission path. Specifically, the prior art mainly has the following deficiencies: 1. The consistency of the monitoring results depends on the premise that the patient's posture remains stable for a period of time, which does not conform to the actual situation of the diagnosis and treatment process; 2. When the signal amplitude changes are monitored, it is impossible to distinguish whether they belong to physiological changes or measurement artifacts in mechanism, so that the monitoring results have diagnostic ambiguity; 3. All longitudinal comparisons that depend on the signal waveform form and amplitude are difficult to establish reliable correlations due to the lack of stable measurement criteria, which limits the development of this technology from basic parameter counting to dynamic state evaluation. Therefore, how to establish a monitoring method that can avoid the influence caused by the dynamic change of the mechanical transmission path to obtain stable and continuously comparable vital sign state evaluation parameters has become a technical problem to be solved by the application. SUMMARY

[0005] The application provides a non-contact vital sign monitoring method and system applied to a dental chair, which mainly aims to solve the problem that the evaluation of vital sign state is affected by signal distortion due to dynamic changes in the mechanical transmission path between the patient and the dental chair, thereby producing inconsistency and ambiguity.

[0006] To achieve the above-mentioned purpose, the application provides a non-contact vital sign monitoring method applied to a dental chair, comprising the following steps:

[0007] Step a: synchronously collecting a first physiological acoustic signal through a physiological acoustic sensor arranged in the headrest or neck support of the dental chair, and synchronously collecting a first mechanical vibration signal through a mechanical vibration sensor arranged on the back plate of the dental chair;

[0008] Step b: identifying a first heart sound event representing the start of ventricular contraction in each cardiac cycle of the first physiological acoustic signal, and determining the occurrence time point of the event as a reference time point;

[0009] Step c: before processing the first mechanical vibration signal, separating a low-frequency baseline signal representing respiratory motion from the first mechanical vibration signal through a low-pass filter, and then subtracting the low-frequency baseline signal from the first mechanical vibration signal point by point to obtain a second mechanical vibration signal after baseline correction;

[0010] Step d: identifying a heart impact main wave peak associated with the first heart sound event in the second mechanical vibration signal, and determining the peak time point as a mechanical response time point;

[0011] Step e: when the reference time point cannot be determined in a certain cardiac cycle, a candidate mechanical response time point is predicted in the second mechanical vibration signal according to the rhythm of historical cardiac cycles, and a verification time window is reversely determined in the first physiological acoustic signal based on the candidate mechanical response time point, and the reference time point of the cardiac cycle is searched and confirmed only in the verification time window;

[0012] Step f: calculating the time interval between the reference time point and the mechanical response time point in each cardiac cycle, and monitoring the vital sign state of the patient based on the change of the time interval.

[0013] Preferably, the first mechanical vibration signal is a ballistocardiogram signal, the first heart sound event is the first heart sound S1, the heart impact main wave peak is the J wave, and the time interval is the heart sound impact conduction time , and the calculation method is as follows: , wherein, is the mechanical response time point, is the reference time point.

[0014] Preferably, in step c, the cut-off frequency of the low-pass filter is set to a preset value lower than the main energy frequency range of the ballistocardiogram signal and higher than the respiration frequency range, the preset value being 0.5 Hz to 1 Hz.

[0015] Preferably, in step e, the determination of the verification time window further comprises: obtaining a time interval average calculated based on a preset number of preceding valid cardiac cycles, and combining the candidate mechanical response time point to calculate a time range in which the reference time point is expected to occur, taking the time range as the verification time window, the method further comprising a dynamic response baseline calibration step, the calibration step comprising: controlling the driving component of the dental chair to perform a preset body position changing program from a semi-recumbent position to a horizontal position before or during a diagnosis and treatment interval; continuously performing steps b to f during the whole process of the body position changing program and a recovery period thereafter to obtain a dynamic response trajectory of the time interval; extracting at least one dynamic characteristic parameter representing cardiovascular regulation capacity from the dynamic response trajectory, and establishing an individualized dynamic response baseline for the patient based on the dynamic characteristic parameter, which is used to calibrate the assessment of the vital sign state in subsequent monitoring.

[0016] Preferably, the dynamic characteristic parameter comprises at least one of: a maximum change amplitude of the time interval in the dynamic response trajectory, a response time required for the maximum change amplitude to be reached, and a recovery time required for the maximum change amplitude to recover to an initial stable level.

[0017] Preferably, the method further comprises a physiological model correction step based on heart rate, the correction step comprising: based on the first physiological acoustic signal, distinguishing a treatment stress period in which the dental instrument is working and a physiological resting period in which the instrument is stopped by analyzing the energy in the high frequency band; in the physiological resting period, recording data pairs of heart rate values and corresponding time interval values under natural heart rate fluctuations of the patient, and establishing an individualized physiological baseline model representing the correlation between the time interval and the heart rate through regression analysis; in the treatment stress period, substituting the real-time heart rate value into the individualized physiological baseline model to calculate a predicted time interval baseline value, subtracting the predicted time interval baseline value from the actually calculated time interval to obtain a time interval correction value in which the influence of heart rate change is eliminated, and monitoring the vital sign state of the patient based on the time interval correction value.

[0018] Preferably, it further comprises a step of determining the heart rate of the patient based on the time period between two consecutive reference time points in the first physiological acoustic signal.

[0019] Preferably, the processing frequency band of the first physiological acoustic signal is limited to 20 Hz to 150 Hz, and the processing frequency band of the first mechanical vibration signal is limited to 1 Hz to 20 Hz.

[0020] Preferably, the signals collected by the physiological acoustic sensor and the mechanical vibration sensor are set with a uniform timestamp to ensure that the reference time point and the mechanical response time point are determined under the same time reference.

[0021] A non-contact vital sign monitoring system applied to a dental chair, comprising:

[0022] A physiological acoustic sensor arranged in a headrest or a neck support of the dental chair;

[0023] A mechanical vibration sensor arranged on a backboard of the dental chair;

[0024] A processor;

[0025] A memory having computer executable instructions stored thereon, the instructions, when executed by the processor, causing the system to implement the functions of the following modules:

[0026] A data acquisition module configured to synchronously acquire a first physiological acoustic signal and a first mechanical vibration signal through the physiological acoustic sensor and the mechanical vibration sensor;

[0027] A reference time point determination module configured to identify a first heart sound event representing the start of ventricular contraction in each cardiac cycle of the first physiological acoustic signal, and determine the occurrence time point of the event as the reference time point;

[0028] A baseline correction module configured to separate a low-frequency baseline signal representing respiratory motion from the first mechanical vibration signal through a low-pass filter, and then subtract the low-frequency baseline signal from the first mechanical vibration signal point by point to obtain a second mechanical vibration signal;

[0029] A mechanical response time point determination module configured to identify a heart impact main wave peak associated with the first heart sound event in the second mechanical vibration signal, and determine the peak time point thereof as the mechanical response time point;

[0030] A reverse verification module configured to, when the reference time point cannot be determined in a certain cardiac cycle, predict a candidate mechanical response time point in the second mechanical vibration signal according to the rhythm of historical cardiac cycles, and based on the candidate mechanical response time point, determine a verification time window in the first physiological acoustic signal in reverse, and only search and confirm the reference time point of the cardiac cycle within the verification time window;

[0031] A time interval calculation and monitoring module configured to calculate the time interval between the reference time point and the mechanical response time point in each cardiac cycle, and based on the change of the time interval, monitor the vital sign state of the patient.

[0032] Compared with the prior art, the present application has the following beneficial effects:

[0033] 1. By synchronously collecting the physiological acoustic signal and the mechanical vibration signal of the patient, and taking a specific physiological event in the acoustic signal as a time reference point, the time position of the corresponding mechanical response feature point in the mechanical vibration signal is calibrated, establishing a monitoring dimension with time interval as the core. This method uses the time sequence locking relationship between the cardiac acoustic event and the mechanical event in the physiological source, so that the measurement result is only related to the time difference between the two event points, and is not affected by the changes in the mechanical transmission path caused by the posture adjustment muscle tension of the patient's body and the dental chair. In this way, the artifacts caused by unstable transmission function when relying on the vibration signal waveform or amplitude for judgment are avoided, providing a basis for obtaining continuous comparative diagnostic information flow under dynamic contact conditions.

[0034] 2. Before identifying the mechanical response feature point in the mechanical vibration signal, the signal is first subjected to low-pass filtering to separate the low-frequency baseline signal generated by the patient's respiratory movement, and then the baseline signal is subtracted from the original mechanical vibration signal. This process uses the same sensor signal source to convert the respiratory part with lower frequency from an interference component into a point-by-point correction basis. In this way, while preserving the original form information of the heart impact signal, baseline drift caused by respiratory movement is avoided, enabling subsequent identification of the mechanical response feature point on a stable baseline, thereby improving the consistency of the time interval parameter calculated finally.

[0035] 3. In the case where the reference time point cannot be identified in the acoustic signal, a reverse verification process will be enabled. This process first predicts the possible position of the mechanical response feature point in the current cardiac cycle based on the historical rhythm of the mechanical vibration signal, and then combines the recent stable time interval average recorded by the system to reversely calculate the verification time window in which the acoustic reference time point should appear. The system only searches the acoustic signal within this narrow verification window. This logic closed loop of forward calibration and reverse verification forms a complementary redundant relationship between the two signal channels. When the dominant acoustic signal source has temporary information loss, the system can logically compensate based on the information of the mechanical signal channel, thereby ensuring the continuity of the monitoring data stream. BRIEF DESCRIPTION OF DRAWINGS

[0036] Fig. 1 The present application combines reverse verification and heart rate correction for vital sign monitoring method flowchart;

[0037] Fig. 2 The performance verification diagram of the present application method state interference;

[0038] Fig. 3 The hardware configuration and function module structure diagram of the monitoring system of the present application. DETAILED DESCRIPTION

[0039] In order to make the technical solutions and advantages of the present application clearer, the technical solutions of the present application will be described clearly and completely below. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the protection scope of the present application.

[0040] The disclosed application is a non-contact vital sign monitoring method and system applied to a dental chair. The system architecture includes a data acquisition module, a reference time point determination module, a baseline correction module, a mechanical response time point determination module, a reverse verification module, and a time interval calculation and monitoring module. In a dental diagnosis and treatment application scenario, the patient lies on the dental chair. The slight changes in the patient's body posture and muscle tension will cause the mechanical transfer function of the patient's dental chair coupling system to change, which is a common technical problem faced by non-contact vital sign monitoring. To solve this problem, the system of the present application synchronously acquires two different modal physiological signals and uses one signal to provide a time reference for the other signal, so that the monitoring result is not affected by the change of the mechanical transfer function. The operation of the system starts from the data acquisition module. The module realizes synchronous acquisition of signals through two sensors configured on the dental chair. Specifically, a physiological acoustic sensor is integrated inside the headrest or neck support of the dental chair to acquire the first physiological acoustic signal conducted through the patient's skull and cervical spine. At the same time, a mechanical vibration sensor is configured on the back plate of the dental chair to acquire the first mechanical vibration signal, i.e. the ballistocardiogram signal, caused by body movement due to heart beating. The two sensors perform synchronous data acquisition at a unified sampling frequency, for example , and the system sets aligned time stamps for each group of sampling points from the two channels, thereby constructing a time-synchronized original data set for subsequent cross-modal time series analysis.

[0041] After obtaining the synchronous data, the reference time point determination module processes the first physiological acoustic signal. Since the first heart sound event is a physiological marker of the start of ventricular contraction and is a clear acoustic event in terms of time point, the system passes the first physiological acoustic signal through a band-pass filter with a passband set between to to filter out interference in other frequency bands. Subsequently, a peak detection algorithm searches for a pulse in the filtered signal that meets the first heart sound waveform characteristics and determines the peak time point of the pulse as the reference time point of the cardiac cycle. This reference time point The determination of the baseline time point, which characterizes a specific physiological event in each cardiac cycle, is completed in the method of this invention. This provides a fixed time reference derived from the physiological event for the analysis of each cardiac cycle. Simultaneously, the system processes the first mechanical vibration signal. The patient's respiratory movements generate a low-frequency signal, which is superimposed on the cardiac impactor signal, forming a dynamically changing baseline. This baseline affects the accurate identification of the subsequent main peak of the cardiac impact. To address this, the baseline correction module employs a feedforward compensation method. The system passes the first mechanical vibration signal through a low-pass filter. The cutoff frequency of this low-pass filter is set to preserve the frequency components of the respiratory signal while avoiding attenuation of the effective frequency band of the cardiac impactor signal. Given that the human respiratory rate is typically lower than... The main energy of the cardiac impact signal is concentrated in to Frequency band, therefore the cutoff frequency is set to a preset value between the two, for example The system can effectively separate a low-frequency baseline signal that characterizes respiratory motion. Then, the system subtracts this low-frequency baseline signal point by point from the original first mechanical vibration signal to obtain a second mechanical vibration signal that has been baseline corrected. This process eliminates baseline drift while preserving the complete information of the cardiac impaction signal, and provides an input signal for the stable identification of subsequent mechanical response points.

[0042] After obtaining a baseline-stable second mechanical vibration signal, the mechanical response time point determination module begins operation, with the system using the reference time point for each cardiac cycle. Starting from a pre-defined time window, for example... After Within the search center, the peak of the main shock wave, i.e., the peak point of the J wave, is determined as the mechanical response time point. Subsequently, the time interval calculation and monitoring module calculates the time interval between these two time points, which is defined as the heart sound impact conduction time. The calculation method is as follows: In the formula, The mechanical response time point, Using a numerical example as a reference time point, let's illustrate how the system identifies the reference time point within a cardiac cycle. The timestamp is The mechanical response time point was identified in the baseline-corrected second mechanical vibration signal. The timestamp is Then the time interval of this period Calculated as The system continuously calculates the cardiac cycle. value, and based on that The sequence changes of values ​​are used to monitor the patient's vital signs. It is a time-based measurement parameter, independent of the signal amplitude or shape, thus suitable for monitoring under dynamic contact conditions. To improve the system's operational stability in a diagnostic and therapeutic environment, a reverse verification module is also configured to handle the impact of a sudden drop in the signal-to-noise ratio of the first physiological acoustic signal on the reference time point. In cases where a reliable identification is not possible, the system predicts the candidate mechanical response time point of the current cycle's J wave from the second mechanical vibration signal based on the average rhythm of historical cardiac cycles. Then, the system acquires a set number of prior valid cardiac cycles. The average time interval calculated is combined with the predicted candidate mechanical response time points to inversely calculate the baseline time point. The system searches for the first physiological acoustic signal only within the expected time window, using an adjusted threshold, to confirm the baseline time point of the cardiac cycle. This combination of forward calibration and reverse verification makes the two signal channels complementary, ensuring the continuity of monitoring data.

[0043] In addition, the system can perform a dynamic response baseline calibration step and a heart rate-based physiological model correction step. The dynamic response baseline calibration step is performed before treatment; the system controls the dental chair drive to execute a preset postural change program, such as from a semi-recumbent position to a horizontal position, thereby applying a standardized cardiovascular load to the patient. During this process, the system continuously records... The system tracks the dynamic response trajectory of the values ​​and extracts dynamic characteristic parameters, such as the maximum amplitude of change and recovery time, to establish an individualized dynamic response baseline for the patient. This baseline is used as a reference in subsequent monitoring, enabling the system to better understand and respond to changes in the patient's condition. The assessment of changes can be combined with the patient's own cardiovascular regulatory capacity, and a physiological model correction step based on heart rate is used to address the effects of heart rate changes. The system utilizes physiological acoustic sensors to analyze high-frequency energy to distinguish between the therapeutic stress period of dental instrument operation and the physiological rest period when the instrument is stopped. During the physiological rest period, the system establishes a characterization of the patient's physiological fluctuations through online regression analysis. Value according to heart rate Individualized physiological baseline model of changing relationships Once the treatment stress period begins, for each real-time measured data point... The system uses the established model to calculate the predicted values. Baseline value And obtain a time interval correction value that eliminates the influence of heart rate variability. Finally, the system monitors the vital sign state of the patient based on the time interval correction value.

[0044] In a dental treatment application scenario, a patient involuntarily adjusts the tension of the back muscles due to nervousness during treatment, and the heart rate also rises due to emotional fluctuations; in this working condition, the monitoring method based on the shape and amplitude of the ballistocardiogram signal will face the problem of interpretation, that is, it cannot distinguish whether the monitored signal fluctuation is caused by the change of the physiological state of the patient or the measurement artifact caused by the adjustment of the muscle posture changing the mechanical transmission path; the operation process of the contactless vital sign monitoring system and method applied to the dental chair is as follows: the physiological acoustic sensor arranged in the headrest and the mechanical vibration sensor arranged on the back of the chair are used to synchronously collect the first physiological acoustic signal and the first mechanical vibration signal with aligned time stamps; at this time, even if the change of the tension of the back muscles of the patient causes the waveform of the first mechanical vibration signal to be distorted, the reference time point determination module can still identify the first heart sound event of each cardiac cycle from the first physiological acoustic signal which is basically not affected by the change and determine the reference time point thereof ; this stably identified time reference provides a reference anchor point that is not affected by the transmission function for subsequent processing of the mechanical vibration signal with changed shape.

[0045] Further, when processing the first mechanical vibration signal, the baseline correction module separates the low-frequency baseline signal generated by the respiratory motion of the patient from the original signal by low-pass filtering, and subtracts it from the original signal point by point to obtain a second mechanical vibration signal with the respiratory baseline drift eliminated; this step enables the subsequent identification of the ballistocardiogram main peak to be performed on a stable baseline and is not affected by the change of the respiratory rhythm; then, the mechanical response time point determination module searches in the second mechanical vibration signal after baseline correction, taking the reference time point as the starting point, identifies the J-peak value and determines the mechanical response time point thereof ; the system calculates the time interval therefrom; this processing process converts a measurement problem depending on the waveform amplitude and shape into a measurement problem related only to the time difference between two event points, and the result is not affected by the distortion of the signal shape; in the case of the patient's heart rate rising, the system performs the physiological model correction step based on the heart rate; using the individualized physiological baseline model established for the patient during the physiological rest period before treatment , the system substitutes the current real-time heart rate value into the model to calculate the expected time interval baseline value under the heart rate; then, the system subtracts the expected value from the actually measured time interval to obtain a time interval correction value eliminating the influence of the change of the heart rate The operator is presented with a trend graph on the monitoring interface, showing the time interval correction value The trend graph, and the real-time heart rate curve that changes synchronously; ultimately, the operator can distinguish between two different physiological phenomena, i.e., the patient's heart rate increases due to tension, but the time interval correction value reflecting the state of myocardial contraction remains stable, indicating that the patient's cardiovascular system has not appeared abnormal physiological state fluctuations; the system provides data support for clinical judgment by providing a continuously comparable time series parameter that has been calibrated multiple times.

[0046] To further verify the superiority of the method of the present application over the prior art, especially the specific effect of avoiding measurement artifacts caused by changes in the mechanical transmission path, the following comparative test was conducted.

[0047] Comparative Example 1: This comparative example aims to simulate the actual working conditions in dental diagnosis and treatment, and compare the stability and reliability of the output parameters of the method of the present application and the conventional single vibration monitoring method mentioned in the background art when the patient unconsciously adjusts the body state; the test conditions are basically the same as those of Example 1, the same dental treatment chair and data acquisition system with installed sensors are used, the difference is that in the control group of this comparative example, the monitoring system only enables the mechanical vibration sensor configured on the back plate of the dental chair, and uses the conventional signal processing method in the art, i.e., by extracting the peak amplitude of the J wave in the ballistocardiogram (BCG) signal, to evaluate the patient's vital signs, the J wave peak amplitude, in a physical sense, is related to the cardiac output per beat, and is the core evaluation index of conventional vibration monitoring technology, the sample group of the present application is completely executed according to the method described in Example 1, in order to provide an objective physiological benchmark, the subject's heart rate is monitored by standard three-lead electrocardiogram (ECG) throughout the test to confirm that the subject's cardiac physiological state remains stable during the test; the test procedure is as follows: the subject lies on the dental chair to simulate receiving a dental examination lasting 120 seconds, during which the subject is required to perform some unconscious body state fine adjustment actions that are common in real diagnosis and treatment at specific time points, and the data is recorded in Table 2 below.

[0048] Table 2: Comparison of parameters under body state fine adjustment interference by different monitoring methods.

[0049]

[0050] From the data in Table 2, during the entire 120-second test period, the reference electrocardiogram shows that the heart rate of the subject remains stable, indicating that the cardiovascular physiological state does not actually fluctuate dramatically; the core parameter J-wave peak amplitude of the conventional monitoring method used in the control group remains stable during the static relaxation period (0-30 seconds), however, once entering the dynamic interference period, the parameter appears a dramatic artifact that is seriously inconsistent with the true physiological state of the patient; at the 65th second, only because the patient's muscle tension increases the mechanical transmission efficiency, the J-wave amplitude increases by 64% pseudo, which is enough to trigger a false alarm of abnormal heartbeat, causing interference to the operator; and at the 108th second, due to the patient's arching back, the mechanical transmission efficiency decreases, and the amplitude decreases by 52% pseudo, which may lead to a false negative of the system as a weakened heartbeat; in contrast, the aortic valve impact conduction time (ABI) calculated by the sample group of the application remains stable within a very narrow range of 55.9ms to 56.2ms during the entire test process, including the dynamic interference period, and the fluctuation rule has no significant difference with the static reference period, which accurately reflects the objective fact that the subject's heart physiological state remains stable; the test results show that under the simulated real diagnosis and treatment working conditions, the conventional monitoring method relying on the amplitude of a single mechanical vibration signal is easily contaminated by the change of the mechanical transmission path between the patient's teeth and the chair, resulting in non-physiological measurement artifacts, and cannot provide reliable vital sign state evaluation for clinical use. The monitoring dimension of the application is calibrated by the internal physiological time marker (first heart sound S1) and the mechanical signal (J-wave), and the time interval parameter ABI obtained by the application can provide stable and reliable vital sign state evaluation under the dynamic contact condition of body posture fine adjustment interference.

[0051] Example 2: In order to verify the stability of the output parameter of the method of the application under the condition of dynamic change of the mechanical transmission path, the following comparative test is designed and performed, which aims to quantify the respective performances of the time interval parameter calculated by the method of the application and the signal amplitude parameter obtained by relying on a single mechanical vibration sensor when the patient fine-tunes the body posture in the simulated diagnosis and treatment scene; the test is carried out on a standard dental treatment chair, and the test platform includes a data acquisition system and a reference benchmark; the data acquisition system includes a physiological acoustic sensor configured in the headrest of the dental chair, a mechanical vibration sensor configured on the back plate of the dental chair, and a data acquisition card for synchronously acquiring two signals, and the sampling frequency is set to ; the reference benchmark is a standard three-lead electrocardiogram monitor, which synchronously records the electrocardiogram signal during the test to confirm the stability of the heart rate of the subject during the test.

[0052] The test procedure is divided into two stages, the first stage is a static reference period, the subject lies on the dental chair in a relaxed posture and keeps still; the second stage is a dynamic interference period, the subject slowly performs the alternating actions of arching and compacting the waist and back under the premise of keeping the upper body position unchanged, to simulate the slight adjustment of the patient to the contact pressure of the dental chair back plate in the clinic, thereby introducing changes in the mechanical transfer function; the processing of test data is divided into two groups, the first group is the control group, which only analyzes the first mechanical vibration signal collected by the mechanical vibration sensor, extracts the peak amplitude of the J wave in each cardiac cycle, as the monitoring parameter of this group; the second group is the sample group of the application, which uses the method of the application, synchronously uses the first physiological acoustic signal and the first mechanical vibration signal, determines the reference time point and the mechanical response time point , calculates the heart sound impact conduction time as the monitoring parameter of this group, part of the data is recorded in Table 1 below.

[0053] Table 1: Comparison of monitoring parameters of two groups in different test stages.

[0054]

[0055] Referring to Table 1, the test data shows that in the static reference period, the peak amplitude of J wave of the control group and the value of the sample group of the application both fluctuate within a small range; after entering the dynamic interference period, the peak amplitude of J wave of the control group appears fluctuations related to the action of the waist and back of the subject, its value decreases to about 0.0011g when arching the waist and back, and increases to about 0.0038g when compacting the waist and back, while the electrocardiogram monitoring during the same period shows that the heart rate of the subject does not change, indicating that this amplitude fluctuation is not caused by physiological state changes; in contrast, the value calculated by the sample group of the application remains in the interval of 55.8ms to 56.4ms throughout the dynamic interference period, and its fluctuation has no difference compared with the static reference period; the test results show that the monitoring parameter relying on signal amplitude will produce non-physiological fluctuations due to changes in the mechanical transfer path of the patient's dental chair coupling system; and the method of the application calculates a mechanical conduction time interval with a physiological acoustic event as the time reference, the measurement result is not affected by the stability of the mechanical transfer function, so it can still output a stable monitoring parameter under the working condition of body posture adjustment interference.

[0056] Example 3: This example combines Figs. 1 to 3 to explain the non-contact vital sign monitoring method and system applied to the dental chair, such as Fig. 1As shown, the process begins by simultaneously acquiring a first physiological acoustic signal and a first mechanical vibration signal with a uniform timestamp through a physiological acoustic sensor configured on the headrest or neck brace of the dental chair and a mechanical vibration sensor configured on the back of the dental chair. The first mechanical vibration signal first enters the baseline correction module, which generates a second mechanical vibration signal by separating and subtracting the low-frequency baseline caused by respiratory movements, and then transmits it to the mechanical response time point determination module to identify the J wave, the main peak of the cardiac impact, and obtain its mechanical response time point. Simultaneously, the first physiological acoustic signal is sent to the reference time point determination module to identify the first heart sound S1 and obtain the reference time point. System check Whether it was successfully identified; if so, the confirmed baseline time point. With mechanical response time point They were sent together to the time interval calculation module to calculate the heart sounds and impulse conduction time. ,like If identification fails, the reverse verification module is activated. This module starts from the candidate mechanical response time points predicted by the mechanical signal, searches backwards, and confirms the reference time point of the current cycle. To ensure the continuity of the data stream; the calculated time interval The time interval is further corrected using a heart rate-based physiological model to avoid the physiological influence of heart rate variations on the time interval, resulting in a corrected time interval. Ultimately, the system is based on the corrected time interval. Changes in vital signs are monitored and assessed.

[0057] like Fig. 2 As shown, the horizontal axis of the graph represents the cardiac cycle, with cycles 1 to 4 representing the static baseline period and cycles 5 to 8 representing the dynamic interference period. The vertical axis contains two monitoring parameters corresponding to the left and right axes: the J-wave amplitude (in grams), representing the control group, indicated by a solid black line with a filled dot, and the parameter representing the sample group of this invention. The time interval is measured in milliseconds (ms) and is represented by a black dashed line with hollow diamond-shaped dots. Experimental data shows that during the dynamic interference period, when subjects performed the lumbar arching cycle 5 and 7 and the compaction cycle 6 and 8, the J-wave amplitude parameter of the control group fluctuated drastically, while that of the sample group of this invention... The time interval parameter remains stable throughout, and its fluctuation range is not significantly different from that of the static baseline period. This indicates that the time interval parameter calculated by the present invention can effectively avoid the influence of mechanical transfer function drift caused by changes in the patient's body posture.

[0058] like Fig. 3As shown, the system architecture includes a dental chair as a physical carrier, on which a physiological acoustic sensor located at the headrest / neck support and a mechanical vibration sensor located at the back of the chair are integrated, which respectively collect physiological acoustic signals and mechanical vibration signals, and transmit the signals to the monitoring system host, which internally contains a processor and a memory, wherein the memory has an individualized physiological baseline model and dynamic response baseline data solidified therein, and the processor executes instructions of multiple functional modules including data acquisition module, baseline correction module, reference time point determination module, mechanical response time point determination module, reverse verification module, and time interval calculation and monitoring module, to complete the processing and analysis of the original signals, and finally output the vital sign evaluation results to the monitoring terminal.

[0059] Embodiment 4: This embodiment discloses a calibration procedure for determining the key processing parameters in the method of the present application, to solve the problem that due to the differences in mechanical structure characteristics, the general parameters may not achieve the expected monitoring performance in different models of dental chairs or specific installation environments; when the monitoring system is first installed on a dental chair or is subjected to periodic maintenance, an offline parameter calibration mode can be started; this mode aims to determine a set of signal processing parameters that match the current physical environment for subsequent online monitoring; to determine the cutoff frequency of the low-pass filter in the baseline correction module for separating the respiratory signal, the system collects a first mechanical vibration signal generated by a subject lying on the dental chair for a duration of 120 seconds; the system performs a fast Fourier transform on the signal to obtain its power spectrum density graph; on the power spectrum graph, the system identifies two frequency boundary points, one is the upper limit of the respiratory energy band , which is determined by finding the frequency point below whose power spectrum energy decays to 10% of the peak value, and the other is the lower limit of the heart impact signal energy band , which is determined by finding the frequency point above whose power spectrum energy first exceeds the preset noise floor; the cutoff frequency of the low-pass filter is set to the geometric mean of the two frequency boundary points, i.e. ; if is , is , then the cutoff frequency used by the system will be set to .

[0060] is the peak detection threshold for identifying the first heart sound event in the calibration reference time point determination module, the system collects a first physiological acoustic signal for a duration of 30 seconds; after band-pass filtering the signal to to , the system calculates the statistical characteristics of the signal amplitude in this period, including its mean value with standard deviation Main detection threshold It is set to be a fixed standard deviation above the background noise level, i.e. Meanwhile, the auxiliary search threshold used in the reverse verification module It is set to a lower value, that is When the system is within a predicted cardiac cycle, the master detection threshold is used. The reference time point could not be identified. At that time, the reverse verification process is activated; the system verifies the data based on the most recent 5 valid cardiac cycles. Interval average value, predict the current cycle The approximate time and location, combined with recent events. The average value is used to locate candidate mechanical response time points in the second mechanical vibration signal. Subsequently, the system based on the five most recent valid... average value with standard deviation A verification time window is calculated from the first physiological acoustic signal. The system only uses an auxiliary search threshold within this verification time window. A search is performed, and if a signal peak that meets the criteria is found, it is identified as the reference time point for that cycle. By executing the above calibration procedure, the system obtains and stores a set of signal processing parameters adapted to the specific dental chair installation environment. This enables data processing with an optimized configuration in subsequent routine monitoring work, thereby improving the stability of monitoring results.

[0061] Example 5: This example discloses an engineering procedure for establishing an individualized physiological baseline model; during the physiological resting period before the monitoring task begins, the system enters a 90-second model learning phase, during which the system continuously collects and calculates heart rate. Corresponding heart sound impulse conduction time After constructing the data pairs and removing outliers, the least squares method was used to perform linear regression analysis on the valid data points, thereby determining a physiological baseline model for the current patient. slope coefficient in With intercept coefficient These two coefficients were then stored by the system and used to evaluate the measured values ​​during subsequent treatment stress periods. The value is used to correct for heart rate effects.

[0062] The embodiment further discloses a pre-check procedure for performing the dynamic response baseline calibration step; before triggering the dental chair to perform the preset body position change program, the system enters a 30-second physiological state stabilization verification stage; in this stage, the system continuously monitors and calculates the standard deviation of the patient's heart rate and the standard deviation of the value ; only when the values of and at the end of the verification stage are both lower than the preset stability threshold, that is, the heart rate standard deviation is less than 3 beats per minute and the value standard deviation is less than 2 milliseconds, the system determines that the current patient is in a stable state suitable for baseline calibration, and further controls the dental chair driving part to lower the chair back from a 45-degree angle to a horizontal position within 15 seconds with a preset sinusoidal speed curve; the pre-check step makes the subsequent dynamic response trajectory measurement based on a verified stable physiological baseline.

[0063] Embodiment 6: The embodiment discloses an updating procedure for maintaining the effectiveness of the individualized physiological baseline model online; in the continuous monitoring task, the system uses its working condition perception ability to automatically identify the physiological resting period appearing between treatments, and when the length of any physiological resting period exceeds a preset value, which is 60 seconds in this embodiment, the system automatically collects a new set of heart rate and heart sound impact conduction time data pairs in this period.

[0064] The system substitutes the new data pairs into the currently stored individualized physiological baseline model , calculates the root mean square error, and if the error value exceeds a preset mismatch threshold, the system determines that the current model cannot accurately represent the patient's physiological state, and uses the latest collected data pairs to perform least squares linear regression analysis again to update the values of the slope coefficient and the intercept coefficient ; this periodic model verification and updating mechanism is used to adjust the correction of the heart rate effect according to the changes in the patient's physiological state in long-term monitoring.

[0065] It is obvious to those skilled in the art that the present application is not limited to the details of the above exemplary embodiments, and can be implemented in other specific forms without departing from the spirit or essential characteristics of the present application.

[0066] Finally, it should be noted that the above examples are merely intended to illustrate the technical solutions of the present application and not to limit the present application. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or equivalently replaced without departing from the spirit and scope of the present application.

Claims

1. A non-contact vital sign monitoring method applied to dental chairs, characterized in that, Includes the following steps: Step a: The first physiological acoustic signal is simultaneously acquired by a physiological acoustic sensor configured in the headrest or neck brace of the dental chair, and the first mechanical vibration signal is simultaneously acquired by a mechanical vibration sensor configured on the back of the dental chair. Step b: In each cardiac cycle of the first physiological acoustic signal, identify the first heart sound event that characterizes the start of ventricular contraction, and determine the time point of occurrence of this event as a reference time point; Step c: Before processing the first mechanical vibration signal, a low-frequency baseline signal representing respiratory motion is separated from it by a low-pass filter. Then, the low-frequency baseline signal is subtracted from the first mechanical vibration signal point by point to obtain the second mechanical vibration signal after baseline correction. Step d: In the second mechanical vibration signal, identify the main wave peak of cardiac impact associated with the first heart sound event, and determine its peak time point as a mechanical response time point; Step e: When the reference time point cannot be determined within a certain cardiac cycle, a candidate mechanical response time point is predicted in the second mechanical vibration signal based on the rhythm of the historical cardiac cycle, and a verification time window is determined in reverse in the first physiological acoustic signal based on the candidate mechanical response time point, and the reference time point of the cardiac cycle is searched and confirmed only within the verification time window. Step f: Calculate the time interval between the baseline time point and the mechanical response time point within each cardiac cycle, and monitor the patient's vital signs based on the changes in the time interval. Furthermore, in step e, the determination of the verification time window further includes: obtaining an average time interval calculated based on a preset number of prior effective cardiac cycles, and combining it with candidate mechanical response time points to calculate the expected time range of the baseline time point, using this time range as the verification time window. The method also includes a dynamic response baseline calibration step, which includes: before or between treatment sessions, controlling the drive component of the dental chair to execute a preset positional change procedure from a semi-recumbent position to a horizontal position; continuously executing steps b to f throughout the entire positional change procedure and during a subsequent recovery period to obtain a dynamic response trajectory for one time interval; extracting at least one dynamic characteristic parameter characterizing cardiovascular regulatory capacity from the dynamic response trajectory, and establishing an individualized dynamic response baseline for the patient based on the dynamic characteristic parameter, for calibrating the assessment of vital signs in subsequent monitoring.

2. The non-contact vital sign monitoring method applied to a dental chair according to claim 1, characterized in that, The first mechanical vibration signal is the cardiac impact signal, and the first heart sound event is the first heart sound. The main peak of the cardiac impact is the J wave, and the time interval is determined to be the conduction time of the cardiac sound impact. The calculation method is as follows: ,in, The mechanical response time point, This is the baseline time point.

3. The non-contact vital sign monitoring method applied to a dental chair according to claim 1, characterized in that, In step c, the cutoff frequency of the low-pass filter is set to a preset value that is lower than the main energy frequency range of the cardiac impaction signal and higher than the respiratory rate range, with the preset value being 0.5Hz to 1Hz.

4. The non-contact vital sign monitoring method applied to a dental chair according to claim 1, characterized in that, The dynamic characteristic parameters include at least one of the following: the response time required for the maximum change of the time interval in the dynamic response trajectory to reach the maximum change, and the recovery time required to recover from the maximum change to the initial stable level.

5. The non-contact vital sign monitoring method applied to a dental chair according to claim 1, characterized in that, The method also includes a heart rate-based physiological model correction step, which includes: based on the first physiological acoustic signal, by analyzing the energy of the high-frequency band, distinguishing between the therapeutic stress period of dental instrument operation and the physiological rest period when the instrument is stopped; during the physiological rest period, recording data pairs of heart rate values ​​and corresponding time interval values ​​under the patient's natural heart rate fluctuations, and establishing an individualized physiological baseline model characterizing the correlation between time interval and heart rate through regression analysis; during the therapeutic stress period, substituting the real-time heart rate value into the individualized physiological baseline model, calculating a predicted time interval baseline value, and subtracting the predicted time interval baseline value from the actual calculated time interval.

6. The non-contact vital sign monitoring method applied to a dental chair according to claim 5, characterized in that, It also includes a step of determining the patient’s heart rate based on the time period between two consecutive reference time points in the first physiological acoustic signal.

7. The non-contact vital sign monitoring method applied to a dental chair according to claim 1, characterized in that, The processing frequency band for the first physiological acoustic signal is limited to between 20 Hz and 150 Hz, and the processing frequency band for the first mechanical vibration signal is limited to between 1 Hz and 20 Hz.

8. The non-contact vital sign monitoring method applied to a dental chair according to claim 1, characterized in that, The signals collected by the physiological acoustic sensor and the mechanical vibration sensor are all set with uniformly aligned timestamps.

9. A non-contact vital signs monitoring system for use in dental chairs, wherein the system performs the method of claim 1, characterized in that, include: Physiological acoustic sensors are installed in the headrest or neck support of the dental chair; Mechanical vibration sensor mounted on the back of the dental chair; processor; The memory stores computer-executable instructions that, when executed by the processor, enable the system to perform the functions of the following modules: The data acquisition module is configured to simultaneously acquire a first physiological acoustic signal and a first mechanical vibration signal through a physiological acoustic sensor and a mechanical vibration sensor; The reference time point determination module is configured to: identify the first heart sound event characterizing the start of ventricular contraction within each cardiac cycle of the first physiological acoustic signal, and determine the occurrence time of this event as the reference time point; The baseline correction module is configured to: separate a low-frequency baseline signal characterizing respiratory motion from the first mechanical vibration signal through a low-pass filter, and then subtract the low-frequency baseline signal from the first mechanical vibration signal point by point to obtain the second mechanical vibration signal; The mechanical response time point determination module is configured to: identify the cardiac impact main wave peak associated with the first heart sound event in the second mechanical vibration signal, and determine its peak time point as the mechanical response time point; The reverse verification module is configured to: when the reference time point cannot be determined within a certain cardiac cycle, predict a candidate mechanical response time point in the second mechanical vibration signal based on the rhythm of the historical cardiac cycle, and determine a verification time window in the first physiological acoustic signal based on the candidate mechanical response time point, and search for and confirm the reference time point of the cardiac cycle only within the verification time window. The time interval calculation and monitoring module is configured to calculate the time interval between the baseline time point and the mechanical response time point within each cardiac cycle, and monitor the patient's vital signs based on the changes in the time interval.

Images