Biofeedback-based respiratory therapy system and method

Abstract

The invention discloses a respiratory therapy system and method based on biofeedback, and belongs to the field of medical instruments, and the therapy method comprises the following specific steps: S1, collecting and recording multiple paths of user physiological parameters; s2, respiratory rhythm extraction is carried out according to the characteristics of the multi-path data; s3, analyzing each piece of feature data to obtain the confidence coefficient of each piece of data, calculating the total confidence coefficient, and rejecting non-ideal data; according to the invention, by realizing coordination of electrical stimulation between diaphragm and abdominal muscles, the effect of synchronizing with breathing of a user is achieved, the expiration ability is improved, the recovery of respiratory muscle functions of a patient is promoted, the muscle strength and endurance are gradually improved, the dyspnea symptom is relieved, the exercise ability and exercise endurance are improved, and the life quality of the patient is improved.

Description

Technical Field

[0001] The present invention relates to the field of medical devices, and in particular to a respiratory therapy system and method based on biofeedback. Background Art

[0002] Breathing is an important physiological function of the human body, and its rhythm and pattern directly reflect and affect psychological and physical health. In recent years, breathing regulation technology has received widespread attention in the fields of anxiety relief, insomnia treatment, and chronic respiratory disease management. However, due to individual differences and external interference, traditional single treatment methods often fail to achieve personalized effects. In order to cope with the failure of ventilator withdrawal, the current clinical practice mainly uses external diaphragm pacemakers for respiratory muscle training. However, current products can only output treatment waveforms at a fixed frequency and cannot detect the patient's actual breathing movements. This will cause the signal given by the therapeutic device to be out of sync with the patient's actual physiological signal, resulting in a human-machine confrontation. Not only will there be no therapeutic effect, but it will also damage the patient's originally weak respiratory function.

[0003] After searching, Chinese patent number CN113907725A discloses a multi-parameter biofeedback therapeutic apparatus and treatment system. Although the invention guarantees the accuracy and consistency of the collected data, timely reminders, and the realization of training effects, has no requirements for the use occasions, and is low-cost and easy to carry, it cannot achieve the effect of synchronizing the user's breathing by coordinating the electrical stimulation between the diaphragm and the abdominal muscles, reduces the patient's exhalation ability, and cannot promote the recovery of the patient's respiratory muscle function. For this reason, we propose a respiratory treatment system and method based on biofeedback. Summary of the invention

[0004] The purpose of the present invention is to solve the defects in the prior art and to propose a respiratory therapy system and method based on biofeedback.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] A breathing therapy method based on biofeedback, the specific steps of the therapy method are as follows:

[0007] S1. Collect and record multiple user physiological parameters;

[0008] S2, extracting breathing rhythm according to the characteristics of multi-channel data;

[0009] S3, analyzing each feature data, obtaining the confidence of each data, calculating the overall confidence, and eliminating undesirable data;

[0010] S4, matching the respiratory treatment plan pre-stored in the database according to the data characteristics;

[0011] S5. Trigger stimulation output using a determined treatment plan according to the actually detected beat.

[0012] As a further solution of the present invention, the specific steps of collecting and recording multiple channels of user physiological parameters in S1 are as follows:

[0013] Q1.1: The patient lies on the treatment bed, and the multi-parameter breathing sensor is tied to the part of the chest or abdomen where the human body rises the most when breathing. The sensor's electromyographic interface is connected to two pairs of ECG electrodes and applied to the intercostal muscles on both sides of the chest. At the same time, the nasal breathing interface is fixed to the patient's mouth or nose for breathing. If the patient is undergoing tracheotomy, it is fixed to the three-way connection of the tracheotomy mask.

[0014] Q1.2: After the multi-parameter respiratory sensor is fixed, connect the host interface of the multi-parameter respiratory sensor to the corresponding host interface of the whole machine, connect the output wire to the host, apply the surface electrodes for physical therapy to the corresponding treatment part of the patient, and then synchronously collect various physiological signals and transmit them to the physiological parameter monitoring module.

[0015] As a further solution of the present invention, the specific steps of analyzing each feature data, obtaining the confidence of each data, and calculating the overall confidence, and eliminating undesirable data in S3 are as follows:

[0016] Q2.1: The data feature extraction module extracts the characteristic value of each physiological signal from the raw data collected by the physiological parameter monitoring module, evaluates the data quality of each channel according to the waveform consistency, noise level and physiological rationality of each group of physiological signals, quantifies the results of each indicator into a score, and generates a preliminary confidence value for each channel;

[0017] Q2.2: The data comprehensive analysis module compares the consistency of data from different acquisition channels for the same physiological parameter. If there is a significant deviation between the data of a certain channel and other channels, the confidence of the channel is reduced. Based on the confidence values ​​of all channels, combined with weighted average or Bayesian inference method, the confidence of the overall data is calculated to generate the final overall confidence value;

[0018] S2.3: According to the preset confidence threshold, channel data below the preset confidence threshold is eliminated, and abnormal information of the eliminated data is recorded and marked for subsequent quality review. If there is a key parameter signal that does not reach the threshold, the user is prompted to re-collect data or adjust the sensor, and the log information of the elimination process is stored to analyze the cause of the bad data.

[0019] As a further solution of the present invention, the specific steps of matching the respiratory therapy scheme pre-stored in the database according to the data characteristics in S4 are as follows:

[0020] Q3.1: Standardize the high-quality data retained after confidence analysis to eliminate the dimensional differences of different physiological signals, extract key features related to respiratory therapy from the standardized data, and form a feature vector containing all the core data for matching;

[0021] Q3.2: The treatment plan matching module loads the treatment plan matching models pre-stored in the database. Each treatment plan matching model contains feature templates of multiple treatment plans. The pattern recognition algorithm is used to match the user's feature vector with each treatment plan template in the database.

[0022] Q3.3: Score the matching results of each treatment plan and sort them by score. If the score of the plan with the highest matching degree is lower than the set threshold, the user will be prompted that the data is abnormal or the matching fails, and the data needs to be collected again. Among multiple plans with similar scores, the final plan is selected based on the user's treatment goals or preferences;

[0023] Q3.4: Output the matched treatment plan and its specific parameter settings, and record the matching results and plan selection process. If the user's real-time status is found to have changed significantly during the treatment process, re-match the plan or adjust the parameters, and combine the user's real-time data with the historical matching results to update the matching model.

[0024] As a further solution of the present invention, the specific steps of triggering stimulation output using a determined treatment plan according to the actually detected beat in S5 are as follows:

[0025] Q4.1: Turn on the host, select the treatment mode, and start the treatment. The physiological parameter monitoring module detects the patient's effective data in real time and outputs the data to the feature extraction module. The feature extraction module then extracts the feature information from each group of data and performs comprehensive analysis through the data comprehensive analysis module;

[0026] Q4.2: When the credibility of the data is greater than the set value, the corresponding treatment output signal starts at the corresponding respiratory beat. When the patient's respiratory beat changes dramatically, the credibility of the respiratory beat data drops below the set value, disrupting the original respiratory rhythm, and stopping the data output of the physiological parameter monitoring module;

[0027] Q4.3: When the patient's breathing is stable and the data credibility is greater than the set value again, the feedback treatment output module is triggered to give the treatment signal again. After the treatment time reaches the prescribed duration, the host sends a treatment completion signal, removes the detection and treatment related sensors and wires, and shuts down.

[0028] A respiratory therapy system based on biofeedback, comprising a physiological parameter monitoring module, a data feature extraction module, a data comprehensive analysis module, a treatment plan matching module and a trigger feedback treatment output module;

[0029] The physiological parameter monitoring module is used for collecting physiological data of respiratory electromyography, nasal air pressure, nasal air temperature, and chest and abdominal movement;

[0030] The data feature extraction module is used to filter each channel of physiological information, calculate the frequency and amplitude characteristics of the filtered data, and extract the frequency and amplitude stability and repeatability of each signal;

[0031] The data comprehensive analysis module is used to calibrate the peak and trough time of multiple data, and calculate the time difference of each data, and comprehensively obtain the key information of respiratory rate, respiratory ratio, exhalation start time, inhalation start time, exhalation period, breath holding period, and inhalation period, and at the same time summarize the frequency and amplitude stability and repeatability of each data to calculate the credibility;

[0032] The treatment plan matching module is used to match the prescription set by the doctor in the database according to the current user's respiratory rate and respiratory ratio parameters;

[0033] The trigger feedback treatment output module is used to provide a corresponding treatment output signal within a corresponding breathing time period according to the treatment prescription setting when the signal credibility is higher than a set value.

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] After the present invention turns on the host, the treatment mode is selected and the treatment is started. The physiological parameter monitoring module detects the effective data of the patient in real time and outputs the data to the feature extraction module. Then the feature extraction module extracts the feature information in each group of data and performs comprehensive analysis through the data comprehensive analysis module. When the credibility of the data is greater than the set value, the corresponding treatment output signal is started at the corresponding breathing beat. When the patient's breathing beat changes drastically, the credibility of the breathing beat data drops to below the set value, and the original breathing rhythm is disrupted, and the data output of the physiological parameter monitoring module is stopped. When the patient's breathing is stable and the data credibility is greater than the set value again, the feedback treatment output module is triggered to give the treatment signal again. After the treatment time reaches the duration set in the prescription, the host sends a treatment completion signal, removes the detection and treatment related sensors and wires, and shuts down. The electrical stimulation between the diaphragm and the abdominal muscles can be coordinated to achieve the effect of synchronization with the user's breathing, improve the exhalation ability, promote the recovery of the patient's respiratory muscle function, gradually improve muscle strength and endurance, relieve the symptoms of dyspnea, improve exercise ability and exercise endurance, and improve the patient's quality of life. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0037] Figure 1 A flowchart of a respiratory therapy method based on biofeedback proposed by the present invention;

[0038] Figure 2 This is a system block diagram of a respiratory therapy system based on biofeedback proposed by the present invention. DETAILED DESCRIPTION

[0039] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0040] Example 1

[0041] Reference Figure 1 , a breathing therapy method based on biofeedback, the specific steps of this therapy method are as follows:

[0042] Collect and record multiple channels of user physiological parameters.

[0043] Specifically, the patient lies on the treatment bed, and the multi-parameter respiratory sensor is tied to the part of the chest or abdomen where the human body rises the most when breathing. The sensor's electromyographic interface is connected to two pairs of ECG electrodes and applied to the intercostal muscles on both sides of the chest. At the same time, the nasal breath interface is fixed to the patient's mouth or nose where he breathes. If the patient is undergoing tracheotomy, it is fixed to the three-way connection of the tracheotomy mask. After the multi-parameter respiratory sensor is fixed, the host interface of the multi-parameter respiratory sensor is connected to the corresponding host interface of the whole machine, and the output wire is connected to the host. Physical therapy surface electrodes are applied to the corresponding treatment part of the patient, and then various physiological signals are synchronously collected and transmitted to the physiological parameter monitoring module.

[0044] Respiratory beat extraction is performed based on the characteristics of multi-channel data.

[0045] Analyze each feature data, obtain the confidence of each data, calculate the overall confidence, and eliminate undesirable data.

[0046] Specifically, the data feature extraction module extracts the characteristic value of each physiological signal from the original data collected by the physiological parameter monitoring module, evaluates the data quality of each channel according to the waveform consistency, noise level and physiological rationality of each group of physiological signals, quantifies the results of various indicators into scores, and generates a preliminary confidence value for each channel. The data comprehensive analysis module compares the consistency of data from different acquisition channels of the same physiological parameter. If there is a significant deviation between the data of a certain channel and other channels, the confidence of the channel is reduced. Based on the confidence values ​​of all channels, combined with weighted average or Bayesian inference methods, the confidence of the overall data is calculated to generate a final overall confidence value. According to the preset confidence threshold, the channel data below the preset confidence threshold is eliminated, and the abnormal information of the eliminated data is recorded and marked for subsequent quality review. If there is a key parameter signal that does not reach the threshold, the user is prompted to re-collect data or adjust the sensor, and the log information of the elimination process is stored to analyze the cause of the bad data.

[0047] Match the respiratory treatment plans pre-stored in the database according to the data characteristics.

[0048] Specifically, the high-quality data retained after the confidence analysis is standardized to eliminate the dimensional differences of different physiological signals, extract key features related to respiratory therapy from the standardized data, and form a feature vector containing all core data for matching. The treatment plan matching module loads the treatment plan matching models pre-stored in the database. Each treatment plan matching model contains feature templates of multiple treatment plans. The pattern recognition algorithm is used to match the user's feature vector with each treatment plan template in the database, and the matching results of each treatment plan are scored and sorted according to the score. If the score of the plan with the highest matching degree is lower than the set threshold, the user is prompted that the data is abnormal or the matching fails and the data needs to be re-collected. Among multiple plans with similar scores, the final plan is selected according to the user's treatment goals or preferences, and the matched treatment plan and its specific parameter settings are output. The matching results and the plan selection process are recorded at the same time. If the user's real-time status is found to have changed significantly during the treatment process, the plan is re-matched or the parameters are adjusted, and the user's real-time data is combined with the historical matching results to update the matching model.

[0049] The stimulation output is triggered using a determined treatment plan according to the actual detected beat.

[0050] Specifically, turn on the host, select the treatment mode, and start the treatment. The physiological parameter monitoring module detects the patient's effective data in real time and outputs the data to the feature extraction module. The feature extraction module then extracts the feature information in each group of data, and performs a comprehensive analysis through the data comprehensive analysis module. When the credibility of the data is greater than the set value, the corresponding treatment output signal is started at the corresponding respiratory beat. When the patient's respiratory beat changes dramatically, the credibility of the respiratory beat data drops below the set value, and disrupts the original respiratory rhythm, and stops the data output of the physiological parameter monitoring module. When the patient's breathing is stable and the data credibility is greater than the set value again, the feedback treatment output module is triggered to give the treatment signal again. After the treatment time reaches the duration set in the prescription, the host sends a treatment completion signal, removes the detection and treatment-related sensors and wires, and shuts down.

[0051] Example 2

[0052] Reference Figure 2 , a respiratory therapy system based on biofeedback, includes a physiological parameter monitoring module, a data feature extraction module, a data comprehensive analysis module, a treatment plan matching module and a trigger feedback treatment output module.

[0053] The physiological parameter monitoring module is used for collecting physiological data such as respiratory electromyography, nasal air pressure, nasal air temperature, and chest and abdominal movement; the data feature extraction module is used to filter each channel of physiological information, calculate the frequency and amplitude characteristics of the filtered data, and extract the frequency and amplitude stability and repeatability of each signal; the data comprehensive analysis module is used to calibrate the peak and trough time of multiple data, and calculate the time difference of each data, and comprehensively obtain the key information of respiratory rate, respiratory ratio, exhalation start time, inhalation start time, exhalation period, breath holding period, and inhalation period, and at the same time summarize the frequency and amplitude stability and repeatability of each data to calculate the credibility; the treatment plan matching module is used to match the prescription set by the doctor in the database according to the current user's respiratory rate and respiratory ratio parameters; the trigger feedback treatment output module is used to give the corresponding treatment output signal in the corresponding breathing time period according to the treatment prescription setting when the signal credibility is higher than the set value.

Claims

1. A breathing therapy method based on biofeedback, characterized in that: The specific steps of this treatment method are as follows: S1. Collect and record multiple user physiological parameters; S2, extracting breathing rhythm according to the characteristics of multi-channel data; S3, analyzing each feature data, obtaining the confidence of each data, calculating the overall confidence, and eliminating undesirable data; S4, matching the respiratory treatment plan pre-stored in the database according to the data characteristics; S5. Trigger stimulation output using a determined treatment plan according to the actually detected beat.

2. A method of breathing therapy based on biofeedback according to claim 1, characterized in that: The specific steps of collecting and recording multiple channels of user physiological parameters in S1 are as follows: Q1.1: The patient lies on the treatment bed, and the multi-parameter breathing sensor is tied to the part of the chest or abdomen where the human body rises the most when breathing. The sensor's electromyographic interface is connected to two pairs of ECG electrodes and applied to the intercostal muscles on both sides of the chest. At the same time, the nasal breathing interface is fixed to the patient's mouth or nose for breathing. If the patient is undergoing tracheotomy, it is fixed to the three-way connection of the tracheotomy mask. Q1.2: After the multi-parameter respiratory sensor is fixed, connect the host interface of the multi-parameter respiratory sensor to the corresponding host interface of the whole machine, connect the output wire to the host, apply the surface electrodes for physical therapy to the corresponding treatment part of the patient, and then synchronously collect various physiological signals and transmit them to the physiological parameter monitoring module.

3. A method of breathing therapy based on biofeedback according to claim 2, characterized in that: The specific steps of analyzing each feature data, obtaining the confidence of each data, and calculating the overall confidence, and eliminating undesirable data as described in S3 are as follows: Q2.1: The data feature extraction module extracts the characteristic value of each physiological signal from the raw data collected by the physiological parameter monitoring module, evaluates the data quality of each channel according to the waveform consistency, noise level and physiological rationality of each group of physiological signals, quantifies the results of each indicator into a score, and generates a preliminary confidence value for each channel; Q2.2: The data comprehensive analysis module compares the consistency of data from different acquisition channels for the same physiological parameter. If there is a significant deviation between the data of a certain channel and other channels, the confidence of the channel is reduced. Based on the confidence values ​​of all channels, combined with weighted average or Bayesian inference method, the confidence of the overall data is calculated to generate the final overall confidence value; S2.3: According to the preset confidence threshold, channel data below the preset confidence threshold is eliminated, and abnormal information of the eliminated data is recorded and marked for subsequent quality review. If there is a key parameter signal that does not reach the threshold, the user is prompted to re-collect data or adjust the sensor, and the log information of the elimination process is stored to analyze the cause of the bad data.

4. A method of breathing therapy based on biofeedback according to claim 3, characterized in that: The specific steps of matching the respiratory treatment plan pre-stored in the database according to the data characteristics described in S4 are as follows: Q3.1: Standardize the high-quality data retained after confidence analysis to eliminate the dimensional differences of different physiological signals, extract key features related to respiratory therapy from the standardized data, and form a feature vector containing all the core data for matching; Q3.2: The treatment plan matching module loads the treatment plan matching models pre-stored in the database. Each treatment plan matching model contains feature templates of multiple treatment plans. The pattern recognition algorithm is used to match the user's feature vector with each treatment plan template in the database. Q3.3: Score the matching results of each treatment plan and sort them by score. If the score of the plan with the highest matching degree is lower than the set threshold, the user will be prompted that the data is abnormal or the matching fails, and the data needs to be collected again. Among multiple plans with similar scores, the final plan is selected based on the user's treatment goals or preferences; Q3.4: Output the matched treatment plan and its specific parameter settings, and record the matching results and plan selection process. If the user's real-time status is found to have changed significantly during the treatment process, re-match the plan or adjust the parameters, and combine the user's real-time data with the historical matching results to update the matching model.

5. A method of breathing therapy based on biofeedback according to claim 4, characterized in that: The specific steps of triggering stimulation output using the determined treatment plan according to the actually detected beat in S5 are as follows: Q4.1: Turn on the host, select the treatment mode, and start the treatment. The physiological parameter monitoring module detects the patient's effective data in real time and outputs the data to the feature extraction module. The feature extraction module then extracts the feature information from each group of data and performs comprehensive analysis through the data comprehensive analysis module; Q4.2: When the credibility of the data is greater than the set value, the corresponding treatment output signal starts at the corresponding respiratory beat. When the patient's respiratory beat changes dramatically, the credibility of the respiratory beat data drops below the set value, disrupting the original respiratory rhythm, and stopping the data output of the physiological parameter monitoring module; Q4.3: When the patient's breathing is stable and the data credibility is greater than the set value again, the feedback treatment output module is triggered to give the treatment signal again. After the treatment time reaches the prescribed duration, the host sends a treatment completion signal, removes the detection and treatment related sensors and wires, and shuts down.

6. A respiratory therapy system based on biofeedback, used to implement the respiratory therapy method based on biofeedback according to any one of claims 1 to 5, characterized in that: It includes a physiological parameter monitoring module, a data feature extraction module, a data comprehensive analysis module, a treatment plan matching module, and a trigger feedback treatment output module; The physiological parameter monitoring module is used for collecting physiological data of respiratory electromyography, nasal air pressure, nasal air temperature, and chest and abdominal movement; The data feature extraction module is used to filter each channel of physiological information, calculate the frequency and amplitude characteristics of the filtered data, and extract the frequency and amplitude stability and repeatability of each signal; The data comprehensive analysis module is used to calibrate the peak and trough time of multiple data, and calculate the time difference of each data, and comprehensively obtain the key information of respiratory rate, respiratory ratio, exhalation start time, inhalation start time, exhalation period, breath holding period, and inhalation period, and at the same time summarize the frequency and amplitude stability and repeatability of each data to calculate the credibility; The treatment plan matching module is used to match the prescription set by the doctor in the database according to the current user's respiratory rate and respiratory ratio parameters; The trigger feedback treatment output module is used to provide a corresponding treatment output signal within a corresponding breathing time period according to the treatment prescription setting when the signal credibility is higher than a set value.

Images