Online analysis system based on hydrogen and methane expiration monitoring

Abstract

The invention discloses an online analysis system based on hydrogen and methane expiration monitoring, and belongs to the technical field of respiration monitoring. Comprising a zero calibration module used for obtaining a first collection environment background carbon dioxide concentration; the original signal acquisition module is used for acquiring an expiration sample; the data processing module is used for processing the original signal data stream; the signal processing module is used for outputting the filtered concentration signal; the signal compensation module is used for obtaining the compensated hydrogen and methane concentration; the secondary zero calibration module is used for acquiring a reference carbon dioxide zero point; the baseline correction module is used for carrying out baseline correction; and the model building module is used for outputting trend judgment. By means of the online hydrogen and methane expiration analysis system, the technical effect that high-concentration precision is still kept under the expiration volume fluctuation and sensor nonlinear conditions is achieved, and the problem of measurement deviation caused by volume change, nonlinear response and carbon dioxide zero drift in the prior art is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of respiratory monitoring, and in particular to an online analysis system based on hydrogen and methane breath monitoring. Background Art

[0002] With the growing demand for intestinal microbiome research and noninvasive diagnosis of functional gastrointestinal diseases, the hydrogen-methane breath test has become an important tool for noninvasively assessing conditions such as carbohydrate malabsorption and small intestinal bacterial overgrowth. Currently, mainstream breath analysis systems use a combination of flow sensors and electrochemical or infrared gas sensors to collect and calculate the concentrations of hydrogen, methane, and carbon dioxide in exhaled breath to determine the subject's metabolic status.

[0003] In the existing technology, linear dilution models and fixed zero-point calibration strategies are generally adopted. When the subject's exhaled volume fluctuates or the sensor exhibits a nonlinear response in the high concentration area, the measurement results are prone to significant deviations. At the same time, the carbon dioxide zero-point drift is not compensated in real time, resulting in baseline drift of hydrogen and methane concentrations, which in turn affects diagnostic accuracy. Summary of the Invention

[0004] The embodiments of the present application provide an online analysis system based on hydrogen and methane breath monitoring, which solves the problem of hydrogen and methane measurement deviation caused by exhaled volume fluctuations, sensor nonlinear response and carbon dioxide zero point drift in the prior art, and realizes high-precision real-time compensation and dynamic baseline correction of exhaled breath concentration.

[0005] The embodiment of the present application provides an online analysis system based on hydrogen and methane breath monitoring, comprising: a zero point calibration module for acquiring and recording a background carbon dioxide concentration in a first acquisition environment as a zero point A;

[0006] The original signal acquisition module is used to obtain the breath sample of the subject and obtain the original signal data stream containing the concentrations of hydrogen, methane and carbon dioxide;

[0007] A data processing module, configured to process the original signal data stream to generate a denoised time domain signal;

[0008] The signal processing module is used to perform state estimation and smoothing based on the denoised time domain signal and output a filtered concentration signal;

[0009] a signal compensation module, configured to obtain a flow-integrated expiratory volume value and perform compensation on the filtered concentration signal to obtain compensated hydrogen and methane concentrations;

[0010] A secondary zero point calibration module is used to collect and record the background carbon dioxide concentration of the second collection environment as zero point B to obtain a calibrated reference carbon dioxide zero point;

[0011] A baseline correction module is used to perform baseline correction on the compensated hydrogen and methane concentrations based on the corrected reference carbon dioxide zero point to obtain a final accurate concentration value;

[0012] The model building module is used to establish a time series prediction model based on multiple sets of historical final precise concentration values ​​and output trend judgments on future changes in hydrogen and methane concentrations.

[0013] Furthermore, the steps of obtaining a breath sample from a subject in real time and obtaining a raw signal data stream containing hydrogen, methane, and carbon dioxide concentrations include:

[0014] The hydrogen sensor, methane sensor and carbon dioxide sensor arranged in the sampling channel continuously output electrical signals, and the electrical signals are converted into digital sampling values ​​according to a fixed sampling frequency;

[0015] The original signal data stream is formed by writing the digital sample values ​​into the ring buffer in time sequence;

[0016] By synchronously recording the timestamp corresponding to each sampling point during the writing process of the ring buffer, the timestamp corresponds to the digital sampling value one by one;

[0017] The original signal data stream includes a hydrogen digital sampling value sequence, a methane digital sampling value sequence, and a carbon dioxide digital sampling value sequence. The three sequences are of the same length and aligned in time sequence.

[0018] When the ring buffer is full, the three sequences are packaged into a data frame with a frame number and a start timestamp.

[0019] Furthermore, the steps of performing state estimation and smoothing processing based on the denoised time domain signal and outputting a filtered concentration signal include:

[0020] Initialize the state vector and error covariance matrix through the received denoised time domain signal;

[0021] Based on each time point in the denoised time domain signal, a recursive calculation loop is performed;

[0022] In the prediction phase of the calculation cycle, the state vector and error covariance matrix of the previous moment are projected forward through the preset system process model to obtain the prior state estimate and prior error covariance of the current moment;

[0023] In the update phase of the calculation loop, the Kalman gain is obtained;

[0024] Based on the denoised time domain signal, the denoised signal value is obtained, the denoised signal value is combined with the prior state estimate, and a weighted correction is performed through the Kalman gain to obtain the posterior state estimate at the current moment;

[0025] The error covariance matrix is ​​updated, and the output posterior state estimation sequence is the filtered concentration signal.

[0026] Furthermore, the step of obtaining the flow-integrated expiratory volume value includes:

[0027] A flow sensor located in the expiratory airflow channel measures the instantaneous flow rate of the gas passing through the sensor at fixed time intervals and records the flow rate measurement result at each moment as a data point;

[0028] Continuously monitor the instantaneous flow rate. When the measured instantaneous flow rate is lower than the preset flow rate termination threshold for a period of time, the exhalation process is determined to be over.

[0029] After the exhalation process is judged to be completed, all instantaneous flow velocity data points recorded during this exhalation are retrieved, and all instantaneous flow velocity data points together constitute a complete flow velocity time series;

[0030] Perform numerical integration on the flow rate time series to obtain the total amount of gas passing through the sensor during the entire exhalation process;

[0031] The total amount of gas passing through the sensor is defined as the flow-integrated expiratory volume value, and the flow-integrated expiratory volume value is stored.

[0032] Furthermore, the step of performing compensation on the filtered concentration signal to obtain compensated hydrogen and methane concentrations includes:

[0033] extracting filtered hydrogen and methane concentration signals based on the filtered concentration signals;

[0034] Through the preset gas dilution physical model, the dilution ratio coefficient is obtained based on the acquired flow-integrated exhaled volume value and the calibration volume of the quantitative sampling loop known to the device itself.

[0035] Construct a likelihood function and set a prior probability distribution. By combining the likelihood function and the prior probability distribution, apply the Bayesian inference rule to obtain the posterior probability distribution.

[0036] The maximum probability value is extracted from the posterior probability distribution as the compensated hydrogen and methane concentrations.

[0037] Furthermore, the likelihood function It is a gas concentration observation model based on non-ideal mixing and sensor response characteristics. Its expected observation concentration is calculated by the following formula:

[0038] ;

[0039] in, is the expected observed concentration, is the actual gas concentration to be estimated, is the actual measured concentration value extracted from the filtered concentration signal, is the flow-integrated expiratory volume value obtained, is the dilution ratio coefficient, is the residual gas concentration in the quantitative sampling loop, is the nonlinear response coefficient of the sensor.

[0040] Furthermore, the steps of performing baseline correction on the compensated hydrogen and methane concentrations based on the corrected reference carbon dioxide zero point to obtain the final accurate concentration value include:

[0041] By presetting a fixed reference carbon dioxide zero point value in the initial calibration stage, based on the corrected reference carbon dioxide zero point, the carbon dioxide zero point drift is obtained by calculating the difference between the corrected reference carbon dioxide zero point and the fixed reference carbon dioxide zero point;

[0042] Through a preset multivariate transfer relationship model, the calculated carbon dioxide zero-point drift is used as an input feature to obtain the correction amounts required for hydrogen concentration and methane concentration respectively;

[0043] Based on subtracting the corresponding hydrogen correction amount from the compensated hydrogen concentration value, and subtracting the corresponding methane correction amount from the compensated methane concentration value, the final accurate hydrogen concentration value and the final accurate methane concentration value of this detection are obtained.

[0044] Furthermore, the multivariable transfer relationship model calculates the correction amount of a specific gas by the following formula:

[0045] ;

[0046] Where, is the calculated concentration correction for a specific gas, is the carbon dioxide zero drift, is the first-order drift transfer coefficient, is the second-order drift transfer coefficient, is the compensated target gas concentration before baseline correction, is the concentration-dependent cross-coupling coefficient.

[0047] Furthermore, based on multiple sets of historical final accurate concentration values, the steps of establishing a time series prediction model include:

[0048] Extract the final accurate concentration value sequence of hydrogen and methane corresponding to past exhalation through the historical result database;

[0049] Initialize the level component and trend component of the final precise concentration value sequence based on the extracted final precise concentration value sequence;

[0050] Through iterative calculation, starting from the first data point of the sequence, processing backward point by point;

[0051] For each data point in the sequence, an update calculation of the horizontal component and the trend component is performed;

[0052] When all historical data points are processed and the smoothing coefficient is optimized, the final horizontal component, trend component and optimized smoothing coefficient together constitute a complete time series forecasting model.

[0053] Furthermore, the steps of outputting a trend judgment of future hydrogen and methane concentration changes include:

[0054] Obtaining an input parameter representing a prediction time span through the time series prediction model;

[0055] For each future time point that needs to be predicted, a prediction calculation is performed;

[0056] The forecast calculation is repeated until the entire forecast time span is covered, thus generating a sequence of multiple future forecast concentration values.

[0057] Based on the generated future predicted concentration value sequence, obtaining the overall slope of the future predicted concentration value sequence;

[0058] Based on the obtained overall slope, if the overall slope is positive, the trend of future changes in hydrogen and methane concentrations is judged to be increasing;

[0059] If the overall slope is negative, the trend of future changes in hydrogen and methane concentrations is judged to be downward.

[0060] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0061] 1. By performing a two-stage calibration of the ambient carbon dioxide zero point and dynamically compensating for zero-point drift in combination with a baseline correction step, the accuracy of hydrogen and methane concentration measurements is significantly improved, thereby achieving high-precision detection of online breath concentration analysis under complex environmental changes, effectively solving the problem of increased measurement error caused by ambient zero-point drift in existing technologies.

[0062] 2. By applying Kalman filtering and recursive state estimation to the raw exhaled breath signal, the instantaneous concentration data is rapidly denoised and smoothed, thereby outputting a high signal-to-noise ratio hydrogen and methane concentration series in real time. This enables rapid response and stable tracking of sudden exhaled breath fluctuations, significantly improving the response speed and stability of the online analysis system.

[0063] 3. The sensor sampling values ​​are jointly corrected by calculating the dilution coefficient based on the sampling volume and combining it with Bayesian inference to obtain the maximum a posteriori estimated compensation concentration value, thereby achieving accurate correction for the gas dilution effect and the nonlinear response of the sensor, significantly improving the reliability and repeatability of the detection results. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 A schematic structural diagram of an online analysis system based on hydrogen and methane breath monitoring provided in an embodiment of the present application. DETAILED DESCRIPTION

[0065] The embodiments of the present application provide an online analysis system based on hydrogen and methane breath monitoring, which solves the problem of hydrogen and methane measurement deviation caused by exhaled volume fluctuations, sensor nonlinear response and carbon dioxide zero drift in the prior art. By introducing a nonlinear dilution compensation model, hyperbolic tangent saturation correction, Bayesian drift compensation and a multivariate zero drift transfer model, high-precision real-time compensation and dynamic baseline correction of exhaled breath concentration are achieved.

[0066] In order to better understand the above technical solution, the above technical solution will be described in detail below with reference to the accompanying drawings and specific implementation methods.

[0067] like Figure 1 FIG. 1 is a schematic diagram of the structure of an online analysis system based on hydrogen and methane breath monitoring provided by an embodiment of the present application, comprising: a zero point calibration module for obtaining and recording the background carbon dioxide concentration of the first collection environment as a zero point A when the sampling inlet is closed before the breath sample is collected;

[0068] The original signal acquisition module is used to obtain the subject's breath sample in real time and obtain the original signal data stream containing the concentrations of hydrogen, methane and carbon dioxide;

[0069] A data processing module, configured to process the original signal data stream using wavelet transform to generate a denoised time domain signal;

[0070] The signal processing module is used to apply the Kalman filter algorithm to perform state estimation and smoothing based on the denoised time domain signal, and output the filtered concentration signal;

[0071] a signal compensation module, configured to obtain a flow-integrated exhaled volume value and perform Bayesian dilution compensation on the filtered concentration signal to correct for concentration dilution caused by insufficient exhaled volume, thereby obtaining compensated hydrogen and methane concentrations;

[0072] The secondary zero point calibration module is used to collect and record the background carbon dioxide concentration of the second collection environment as zero point B again with the sampling inlet closed after the sample analysis is completed, and calculate the arithmetic mean of zero point A and zero point B to obtain the corrected baseline carbon dioxide zero point;

[0073] A baseline correction module, configured to perform baseline correction on the compensated hydrogen and methane concentrations based on the corrected reference carbon dioxide zero point to obtain a final accurate concentration value;

[0074] The model building module is used to establish a time series prediction model based on multiple sets of historical final precise concentration values ​​and output trend judgments on future changes in hydrogen and methane concentrations.

[0075] Furthermore, the steps of obtaining a breath sample from a subject in real time and obtaining a raw signal data stream containing hydrogen, methane, and carbon dioxide concentrations include:

[0076] The hydrogen sensor, methane sensor, and carbon dioxide sensor provided in the sampling channel continuously output electrical signals, which are converted into digital sampling values ​​by an analog-to-digital converter at a fixed sampling frequency;

[0077] The original signal data stream is formed by writing the digital sampling values ​​into the circular buffer in time sequence. The buffer capacity is sufficient to cover a complete exhalation cycle. When the buffer is full, the old data is overwritten by the new data.

[0078] By synchronously recording the timestamp corresponding to each sampling point during the writing process of the ring buffer, the timestamp corresponds to the digital sampling value one by one;

[0079] The original signal data stream includes a hydrogen digital sampling value sequence, a methane digital sampling value sequence, and a carbon dioxide digital sampling value sequence. The three sequences are of the same length and aligned in time sequence.

[0080] When the circular buffer is full, the three sequences are packaged into data frames with frame numbers and start timestamps. By sending the data frames to the data processing module, the original signal data stream maintains its timing unchanged and is not compressed during transmission.

[0081] Furthermore, a state vector is initialized using the received denoised time domain signal, which is used to describe the estimated gas concentration, and an error covariance matrix, which quantifies the uncertainty of the estimated value of the state vector.

[0082] Based on each time point in the denoised time domain signal, a recursive calculation loop is performed;

[0083] In the prediction phase of the calculation cycle, the state vector and error covariance matrix of the previous moment are projected forward through the preset system process model to obtain the prior state estimate and prior error covariance of the current moment;

[0084] In the update phase of the calculation loop, the Kalman gain is obtained. The value of the gain depends on the covariance of the prior error and the covariance of the measurement noise, which is used to balance the weight between the prior estimate and the new measurement value.

[0085] Based on the denoised time domain signal, the denoised signal value is obtained, the denoised signal value is combined with the prior state estimate, and a weighted correction is performed through the Kalman gain to obtain the posterior state estimate at the current moment;

[0086] The error covariance matrix is ​​updated to reflect the reduced uncertainty due to the adoption of the new measurement value. This prediction and update cycle will be carried out throughout the entire time domain signal sequence, and the final output posterior state estimate sequence is the filtered concentration signal.

[0087] Furthermore, the step of obtaining the flow-integrated expiratory volume value includes:

[0088] A flow sensor located in the expiratory airflow channel measures the instantaneous flow rate of the gas passing through the sensor at fixed time intervals and records the flow rate measurement result at each moment as a data point;

[0089] Continuously monitor the instantaneous flow rate. When the measured instantaneous flow rate is lower than the preset flow rate termination threshold for a period of time, the exhalation process is determined to be over.

[0090] After the exhalation process is judged to be completed, all instantaneous flow velocity data points recorded during this exhalation are retrieved, and all instantaneous flow velocity data points together constitute a complete flow velocity time series;

[0091] Perform a numerical integration operation on the flow rate time series by multiplying the value of each flow rate data point by the sampling interval and then accumulating all these product results. The sum of these accumulation operations is the total amount of gas passing through the sensor during the entire exhalation process;

[0092] The total amount of gas passing through the sensor is defined as a flow-integrated expiratory volume value, and the flow-integrated expiratory volume value is stored for use by the signal compensation module in subsequent steps.

[0093] Furthermore, the step of performing compensation on the filtered concentration signal to obtain compensated hydrogen and methane concentrations includes:

[0094] extracting filtered hydrogen and methane concentration signals based on the filtered concentration signals;

[0095] The device uses a pre-defined gas dilution model to describe the physical process of exhaled gas mixing with residual gas within the device's internal sampling loop. Based on the acquired flow-integrated exhaled volume and the device's known calibration volume for the sampling loop, a dilution factor is derived, representing the actual proportion of exhaled gas within the sampling loop.

[0096] A likelihood function is constructed, representing the probability of observing the current filtered concentration signal given the true concentration and the calculated dilution factor. A prior probability distribution is also set; this distribution represents a preliminary estimate of the true concentration before the measurement. By combining the likelihood function and the prior probability distribution, Bayesian inference rules are applied to obtain a posterior probability distribution; this posterior probability distribution comprehensively describes the updated understanding of the true gas concentration after taking into account both the measured value and the dilution effect.

[0097] The maximum probability value is extracted from the posterior probability distribution as the final result after compensating for the dilution effect caused by insufficient exhalation, that is, the compensated hydrogen and methane concentrations.

[0098] Furthermore, the likelihood function It is a gas concentration observation model based on non-ideal mixing and sensor response characteristics. Its expected observation concentration is calculated by the following formula:

[0099] ;

[0100] in, is the expected observed concentration, which represents the concentration value that is most likely to be measured by the sensor in theory under a given true concentration and exhaled volume. is the true gas concentration to be estimated, is the target variable of Bayesian inference, is the actual measured concentration value extracted from the filtered concentration signal, is the flow-integrated expiratory volume value obtained, is the dilution ratio coefficient, To quantify the residual gas concentration in the sampling loop, the calibrated reference carbon dioxide zero point is usually used for dynamic assignment or set to the ambient air concentration. is the nonlinear response coefficient of the sensor.

[0101] Furthermore, the steps of performing baseline correction on the compensated hydrogen and methane concentrations based on the corrected reference carbon dioxide zero point to obtain the final accurate concentration value include:

[0102] A fixed reference CO2 zero point value is preset during the initial calibration phase. This is based on a corrected baseline CO2 zero point, which is obtained by calculating the arithmetic mean of the background CO2 concentration of the first and second collected environments. The CO2 zero drift is calculated by calculating the difference between the corrected baseline CO2 zero point and the fixed reference CO2 zero point.

[0103] A pre-defined multivariate transfer relationship model describes the quantitative relationship between the zero drift of the CO2 sensor and the reading offsets of the hydrogen and methane sensors. The calculated CO2 zero drift is used as an input feature to determine the required corrections for the hydrogen and methane concentrations.

[0104] Based on subtracting the corresponding hydrogen correction amount from the compensated hydrogen concentration value, and subtracting the corresponding methane correction amount from the compensated methane concentration value, the final accurate hydrogen concentration value and the final accurate methane concentration value of this detection are obtained.

[0105] Furthermore, the multivariable transfer relationship model calculates the correction amount of a specific gas by the following formula:

[0106] ;

[0107] Where, is the calculated value for a specific gas ( or ) concentration correction amount, is the CO2 zero point drift, that is, the difference between the corrected baseline CO2 zero point and the fixed reference CO2 zero point. is the first-order drift transfer coefficient, which describes the linear effect of the carbon dioxide zero drift on the target gas concentration and is determined by multivariable calibration experiments. is the second-order drift transfer coefficient, which describes the nonlinear effect of the carbon dioxide zero drift on the target gas concentration. It is used to correct the response relationship under large drift and is also determined through calibration experiments. is the compensated target gas concentration before baseline correction, is a natural constant, The concentration-dependent cross-coupling coefficient reflects the relationship between the impact of zero drift and the target gas concentration level. For example, against a high methane background, carbon dioxide drift may cause a more significant reading error. This coefficient is calibrated by performing drift experiments at different background concentrations.

[0108] Furthermore, based on multiple sets of historical final accurate concentration values, the steps of establishing a time series prediction model include:

[0109] The final precise concentration value sequence of hydrogen and the final precise concentration value sequence of methane corresponding to past exhalations are extracted through the historical result database; the sequence is for a specific gas, such as hydrogen or methane.

[0110] Based on the extracted final precise concentration value sequence, the horizontal component and trend component of the final precise concentration value sequence are initialized; the horizontal component represents the baseline value of the sequence, and the trend component represents the growth or decrease rate of the sequence over time.

[0111] Through iterative calculation, starting from the first data point of the sequence, processing backward point by point;

[0112] For each data point in the sequence, an update calculation of the horizontal component and the trend component is performed;

[0113] The horizontal component is updated using a weighted averaging process that combines the value of the current data point with the sum of the horizontal and trend components at the previous moment. The trend component is also updated using a weighted averaging process that combines the difference between the current updated horizontal component and the previous horizontal component, as well as the trend component at the previous moment. The weighting factors used in these two weighted averaging processes, or smoothing coefficients, are determined using an optimization algorithm. The goal of this algorithm is to find a set of smoothing coefficients that minimizes the total error generated by the model's one-step forecast based on historical data.

[0114] When all historical data points are processed and the smoothing coefficient is optimized, the final horizontal component, trend component and optimized smoothing coefficient together constitute a complete time series forecasting model.

[0115] Furthermore, the steps of outputting a trend judgment of future hydrogen and methane concentration changes include:

[0116] The time series forecasting model includes the latest horizontal component, the latest trend component, and the optimized smoothing coefficient. An input parameter representing the forecast time span is obtained; the parameter defines the number of future time points to be forecasted.

[0117] For each future time point to be predicted, a forecast calculation is performed by adding the latest horizontal component to a delta that is the product of the latest trend component and the number of steps away from the future time point relative to the current time.

[0118] The forecast calculation is repeated until the entire forecast time span is covered, thus generating a sequence of multiple future forecast concentration values.

[0119] Based on the generated future predicted concentration value sequence, the system analyzes the sequence to determine its overall evolution direction. This analysis process includes comparing the predicted values ​​in the later stages of the sequence with the predicted values ​​in the earlier stages to obtain the overall slope of the future predicted concentration value sequence;

[0120] Based on the obtained overall slope, if the overall slope is positive, the trend of future changes in hydrogen and methane concentrations is judged to be increasing;

[0121] If the overall slope is negative, the trend of future changes in hydrogen and methane concentrations is judged to be downward.

[0122] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0123] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts 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, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0124] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0125] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0126] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0127] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. An online analysis system based on hydrogen and methane breath monitoring, characterized in that: include: A zero point calibration module, used to obtain and record the background carbon dioxide concentration of the first acquisition environment as zero point A; The original signal acquisition module is used to obtain the breath sample of the subject and obtain the original signal data stream containing the concentrations of hydrogen, methane and carbon dioxide; A data processing module, configured to process the original signal data stream to generate a denoised time domain signal; The signal processing module is used to perform state estimation and smoothing based on the denoised time domain signal and output a filtered concentration signal; a signal compensation module, configured to obtain a flow-integrated expiratory volume value and perform compensation on the filtered concentration signal to obtain compensated hydrogen and methane concentrations; A secondary zero point calibration module is used to collect and record the background carbon dioxide concentration of the second collection environment as zero point B to obtain a calibrated reference carbon dioxide zero point; A baseline correction module is used to perform baseline correction on the compensated hydrogen and methane concentrations based on the corrected reference carbon dioxide zero point to obtain a final accurate concentration value; The model building module is used to establish a time series prediction model based on multiple sets of historical final precise concentration values ​​and output trend judgments on future changes in hydrogen and methane concentrations.

2. An online analysis system based on hydrogen and methane breath monitoring as claimed in claim 1, characterized in that: The steps of obtaining a breath sample from a subject in real time and obtaining a raw signal data stream containing hydrogen, methane, and carbon dioxide concentrations include: The hydrogen sensor, methane sensor and carbon dioxide sensor arranged in the sampling channel continuously output electrical signals, and the electrical signals are converted into digital sampling values ​​according to a fixed sampling frequency; The original signal data stream is formed by writing the digital sample values ​​into the ring buffer in time sequence; By synchronously recording the timestamp corresponding to each sampling point during the writing process of the ring buffer, the timestamp corresponds to the digital sampling value one by one; The original signal data stream includes a hydrogen digital sampling value sequence, a methane digital sampling value sequence, and a carbon dioxide digital sampling value sequence. The three sequences are of the same length and aligned in time sequence. When the ring buffer is full, the three sequences are packaged into a data frame with a frame number and a start timestamp.

3. The online analysis system based on hydrogen and methane breath monitoring according to claim 1, characterized in that: Based on the denoised time domain signal, the steps of performing state estimation and smoothing processing and outputting the filtered concentration signal include: Initialize the state vector and error covariance matrix through the received denoised time domain signal; Based on each time point in the denoised time domain signal, a recursive calculation loop is performed; In the prediction phase of the calculation cycle, the state vector and error covariance matrix of the previous moment are projected forward through the preset system process model to obtain the prior state estimate and prior error covariance of the current moment; In the update phase of the calculation loop, the Kalman gain is obtained; Based on the denoised time domain signal, the denoised signal value is obtained, the denoised signal value is combined with the prior state estimate, and a weighted correction is performed through the Kalman gain to obtain the posterior state estimate at the current moment; The error covariance matrix is ​​updated, and the output posterior state estimation sequence is the filtered concentration signal.

4. The online analysis system based on hydrogen and methane breath monitoring according to claim 1, characterized in that: The steps for obtaining the flow-integrated expiratory volume value include: A flow sensor located in the expiratory airflow channel measures the instantaneous flow rate of the gas passing through the sensor at fixed time intervals and records the flow rate measurement result at each moment as a data point; Continuously monitor the instantaneous flow rate. When the measured instantaneous flow rate is lower than the preset flow rate termination threshold for a period of time, the exhalation process is determined to be over. After the exhalation process is judged to be completed, all instantaneous flow velocity data points recorded during this exhalation are retrieved, and all instantaneous flow velocity data points together constitute a complete flow velocity time series; Perform numerical integration on the flow rate time series to obtain the total amount of gas passing through the sensor during the entire exhalation process; The total amount of gas passing through the sensor is defined as the flow-integrated expiratory volume value, and the flow-integrated expiratory volume value is stored.

5. The online analysis system based on hydrogen and methane breath monitoring according to claim 1, characterized in that: The step of performing compensation on the filtered concentration signal to obtain compensated hydrogen and methane concentrations comprises: extracting filtered hydrogen and methane concentration signals based on the filtered concentration signals; The dilution ratio coefficient is obtained through a preset gas dilution physical model based on the acquired flow-integrated exhaled volume value and the calibration volume of the device's own known quantitative sampling loop; Construct a likelihood function and set a prior probability distribution. By combining the likelihood function and the prior probability distribution, apply the Bayesian inference rule to obtain the posterior probability distribution. The maximum probability value is extracted from the posterior probability distribution as the compensated hydrogen and methane concentrations.

6. The online analysis system based on hydrogen and methane breath monitoring according to claim 5, characterized in that: Likelihood function It is a gas concentration observation model based on non-ideal mixing and sensor response characteristics. Its expected observation concentration is calculated by the following formula: ； in, is the expected observed concentration, is the actual gas concentration to be estimated, is the actual measured concentration value extracted from the filtered concentration signal, is the flow-integrated expiratory volume value obtained, is the dilution ratio coefficient, is the residual gas concentration in the quantitative sampling loop, is the nonlinear response coefficient of the sensor.

7. The online analysis system based on hydrogen and methane breath monitoring according to claim 1, characterized in that: The steps of performing baseline correction on the compensated hydrogen and methane concentrations based on the corrected reference carbon dioxide zero point to obtain a final accurate concentration value include: By presetting a fixed reference carbon dioxide zero point value in the initial calibration stage, based on the corrected reference carbon dioxide zero point, the carbon dioxide zero point drift is obtained by calculating the difference between the corrected reference carbon dioxide zero point and the fixed reference carbon dioxide zero point; Through a preset multivariate transfer relationship model, the calculated carbon dioxide zero-point drift is used as an input feature to obtain the correction amounts required for hydrogen concentration and methane concentration respectively; Based on subtracting the corresponding hydrogen correction amount from the compensated hydrogen concentration value, and subtracting the corresponding methane correction amount from the compensated methane concentration value, the final accurate hydrogen concentration value and the final accurate methane concentration value of this detection are obtained.

8. An online analysis system based on hydrogen and methane breath monitoring as claimed in claim 7, characterized in that: The multivariable transfer relationship model calculates the correction amount for a specific gas using the following formula: ； Where, is the calculated concentration correction for a specific gas, is the carbon dioxide zero drift, is the first-order drift transfer coefficient, is the second-order drift transfer coefficient, is the compensated target gas concentration before baseline correction, is the concentration-dependent cross-coupling coefficient.

9. The online analysis system based on hydrogen and methane breath monitoring according to claim 1, characterized in that: The steps to establish a time series prediction model based on multiple sets of historical final accurate concentration values ​​include: Extract the final accurate concentration value sequence of hydrogen and methane corresponding to past exhalation through the historical result database; Initialize the level component and trend component of the final precise concentration value sequence based on the extracted final precise concentration value sequence; Through iterative calculation, starting from the first data point of the sequence, processing backward point by point; For each data point in the sequence, an update calculation of the horizontal component and the trend component is performed; When all historical data points are processed and the smoothing coefficient is optimized, the final horizontal component, trend component and optimized smoothing coefficient together constitute a complete time series forecasting model.

10. The online analysis system based on hydrogen and methane breath monitoring according to claim 1, characterized in that: The steps for outputting the trend judgment of future hydrogen and methane concentration changes include: Obtaining an input parameter representing a prediction time span through the time series prediction model; For each future time point that needs to be predicted, a prediction calculation is performed; Repeat the forecast calculation until the entire forecast time span is covered, thereby generating a sequence containing multiple future forecast concentration values; Based on the generated future predicted concentration value sequence, obtaining the overall slope of the future predicted concentration value sequence; Based on the obtained overall slope, if the overall slope is positive, the trend of future changes in hydrogen and methane concentrations is judged to be increasing; If the overall slope is negative, the trend of future changes in hydrogen and methane concentrations is judged to be downward.

Images