Oxygen-free threshold calculation method and system based on portable end-expiratory carbon dioxide concentration dynamic change

Abstract

The invention provides an anaerobic threshold value calculation method and system based on portable end-expiratory carbon dioxide concentration dynamic change, belongs to the technical field of anaerobic threshold value calculation, and solves the problems that an existing AT measurement mode is invasive, complex in operation, poor in portability and the like. The method comprises the following steps: enabling a subject to carry out load progressive increase movement on a power vehicle, and collecting end-expiratory carbon dioxide concentration and respiratory rate of the subject and movement power of the power vehicle in real time; the collected sequences of the end-expiratory carbon dioxide concentration and the respiratory rate of the subject are preprocessed, and interpolation sequences of the end-expiratory carbon dioxide concentration and the respiratory rate with the motion power of the power vehicle as the X axis are obtained; according to the interpolation sequence of the end-expiratory carbon dioxide concentration and the respiratory rate, the characteristic inflection point of the motion power is recognized, and the characteristic inflection point serves as the calculation result of the oxygen-free threshold value of the subject.

Description

Technical Field

[0001] This invention relates to the field of anaerobic threshold calculation technology, and in particular to a method and system for calculating anaerobic threshold based on the dynamic changes in end-tidal carbon dioxide concentration using a portable device. Background Technology

[0002] The gold standard for anaerobic threshold (AT) – the blood lactate assay – determines this core indicator of exercise physiology by measuring blood biochemical parameters. It marks a key turning point in exercise intensity, from a predominantly aerobic metabolism to a significant increase in the involvement of anaerobic metabolism.

[0003] Currently, the gold standard for determining the anaerobic threshold (AT) is through invasive arterial blood sampling to measure blood lactate concentration or to perform blood gas analysis (such as venous blood lactate / lactate threshold testing, or ventilation equivalent method). However, this method has the following inherent limitations: (1) Invasive: It requires multiple arterial punctures, which will cause pain to the subject and increase the risk of infection, and it is difficult to perform the procedure frequently during exercise; (2) Lag: Changes in blood lactate concentration are delayed by tens of minutes compared to metabolic changes, making real-time feedback impossible; (3) Discontinuity: It can only provide data at discrete time points and cannot continuously monitor the dynamic changes of AT; (4) Complex equipment: It requires the use of expensive blood lactate analyzers or blood gas analyzers, which requires a high level of professional expertise in operation.

[0004] Furthermore, some existing non-invasive methods are mainly based on ventilation parameters, such as the V-Slope method and the ventilation equivalent method. However, these methods rely on precise gas metabolism analysis (i.e., cardiopulmonary exercise testing, CPET), which not only involves large and expensive equipment and complex algorithms, but also requires extremely high environmental conditions and operator skills, making it difficult to widely apply in scenarios such as bedside, sports fields, or ordinary gyms. Summary of the Invention

[0005] Based on the above analysis, the present invention aims to provide a method and system for calculating the anaerobic threshold based on the dynamic changes in end-tidal carbon dioxide concentration using a portable device, in order to solve the problems of existing AT measurement methods being invasive, complex to operate, and poorly portable.

[0006] On one hand, the present invention provides a method for calculating the anaerobic threshold based on the dynamic changes in end-tidal carbon dioxide concentration using a portable device, the method comprising: Subjects were instructed to perform progressively increasing exercise on a power bike, and their end-tidal carbon dioxide concentration, respiratory rate, and power output of the power bike were collected in real time. The sequences of end-tidal carbon dioxide concentration and respiratory rate of the subjects were preprocessed to obtain interpolated sequences of end-tidal carbon dioxide concentration and respiratory rate with the power of the trolley as the X-axis. The characteristic inflection point of exercise power is identified based on the interpolation sequence of end-tidal carbon dioxide concentration and respiratory rate, and the characteristic inflection point is used as the calculation result of the subject's anaerobic threshold.

[0007] Based on the above solution, the present invention also makes the following improvements: Furthermore, the preprocessing includes: Signal cleaning was performed on the sequences of end-tidal carbon dioxide concentration and respiratory rate, respectively. The sequences of end-tidal carbon dioxide concentration and respiratory rate after washing were smoothed with the power of the dynamometer as the X-axis. The sequences of end-tidal carbon dioxide concentration and respiratory rate after cleaning were subjected to motion power mapping to form a sequence of end-tidal carbon dioxide concentration and respiratory rate with the motion power of the power vehicle as the X-axis.

[0008] Furthermore, the signal smoothing with the vehicle's motion power as the X-axis is performed as follows: The sequences of end-tidal carbon dioxide concentration and respiratory rate after cleaning were subjected to motion power mapping to form a sequence of end-tidal carbon dioxide concentration and respiratory rate with the motion power of the power vehicle as the X-axis. The SG filter was used to smooth the sequences of end-tidal carbon dioxide concentration and respiratory rate with the kinetic power of the vehicle as the X-axis.

[0009] Furthermore, the SG filter is used to smooth the signals of the end-tidal carbon dioxide concentration and respiratory rate sequences with the vehicle's kinetic power as the X-axis, respectively. Within each local window of the power axis, the relationship between the end-tidal carbon dioxide concentration sequence, the respiratory rate sequence and the exercise power is parametrically fitted using a k-order polynomial. The end-tidal carbon dioxide concentration sequence within a local window is smoothed using a fitted k-order polynomial corresponding to the end-tidal carbon dioxide concentration, and the respiratory rate sequence within a local window is smoothed using a fitted k-order polynomial corresponding to the respiratory rate.

[0010] Furthermore, the identification of the characteristic inflection point of motion power is performed as follows: Calculate the second derivative of the relative power of the vehicle's motion within each local window for the interpolation sequences of end-tidal carbon dioxide concentration and respiratory rate. The negative of the second derivative of the relative power of the vehicle within each local window was used to normalize the interpolated sequence of end-tidal carbon dioxide concentration, and the second derivative of the relative power of the vehicle within each local window was used to normalize the interpolated sequence of respiratory rate. Within each local window, the data normalization results of the interpolation sequence of end-tidal carbon dioxide concentration relative to the negative of the second derivative of the power of the vehicle and the data normalization results of the interpolation sequence of respiratory rate relative to the second derivative of the power of the vehicle are weighted and fused. The weighted fusion results of all local windows are then concatenated in order of increasing power to construct the fusion objective function. The motion power at which the fusion objective function reaches its maximum value is taken as the feature inflection point.

[0011] Furthermore, the method also includes: The rotational speed of the power vehicle was collected during the subjects' exercise with increasing load. Plot the power curve of the vehicle's motion power relative to time, and the speed curve of the vehicle's rotational speed relative to time; If the power curve increases at a constant rate within the preset power deviation range, and the speed curve remains constant within the preset speed deviation range, then the preprocessing is performed.

[0012] Furthermore, the method also includes: If the power curve does not maintain a constant increase within the preset power deviation range, or the speed curve does not maintain a constant speed within the preset speed deviation range, the collected data is invalid, and the subject must restart the progressively increasing load exercise on the power vehicle.

[0013] On the other hand, the present invention also provides an anaerobic threshold calculation system based on the dynamic change of end-tidal carbon dioxide concentration in a portable system, the system comprising: The data acquisition module is used to enable subjects to perform progressively increasing exercise on the exercise vehicle and to collect the subjects' end-tidal carbon dioxide concentration, respiratory rate, and exercise power of the exercise vehicle in real time. The preprocessing module is used to preprocess the sequences of end-tidal carbon dioxide concentration and respiratory rate of the subjects respectively to obtain the interpolated sequences of end-tidal carbon dioxide concentration and respiratory rate with the power of the dynamometer as the X-axis. The anaerobic threshold calculation module is used to identify the characteristic inflection point of exercise power based on the interpolation sequence of end-tidal carbon dioxide concentration and respiratory rate, and use the characteristic inflection point as the calculation result of the subject's anaerobic threshold.

[0014] Based on the above solution, the present invention also makes the following improvements: Furthermore, the data acquisition module includes: An end-tidal carbon dioxide monitor is used to collect data on the end-tidal carbon dioxide concentration and respiratory rate of subjects in real time. The power vehicle's feedback sensor is used to collect the vehicle's motion power in real time.

[0015] Furthermore, the system also includes a motion detection module; at this time, the preprocessing module also collects the rotational speed of the power vehicle during the subject's exercise with increasing load; The motion detection module performs the following: Plot the power curve of the vehicle's motion power relative to time, and the speed curve of the vehicle's rotational speed relative to time; If the power curve increases at a constant rate within the preset power deviation range, and the speed curve remains constant within the preset speed deviation range, then the preprocessing is performed; otherwise, the collected data is invalid, and the subject restarts the progressively increasing load exercise on the power vehicle.

[0016] Compared with the prior art, the present invention can achieve at least one of the following beneficial effects: This invention provides a method for calculating the anaerobic threshold based on the dynamic changes in end-tidal carbon dioxide concentration using a portable device. By real-time acquisition of the subject's end-tidal carbon dioxide concentration, respiratory rate, and power output during incremental exercise, it eliminates the need for invasive arterial blood sampling, avoiding pain and infection risks for the subject, thus achieving non-invasive measurement. Furthermore, this method performs real-time analysis of dynamically changing physiological parameters, overcoming the lag and discontinuity issues of blood lactate measurement, and can reflect metabolic state changes more promptly and continuously.

[0017] In terms of technical implementation, the accuracy and reliability of anaerobic threshold calculation are improved by preprocessing the collected data such as signal cleaning, smoothing and motion power mapping, and combining second derivative calculation and weighted fusion to construct a fusion objective function.

[0018] In addition, the core of the entire system relies on a portable end-tidal carbon dioxide monitor and a power bike feedback sensor. The device is small in size and easy to operate, which gets rid of the limitations of traditional CPET equipment, which is bulky, expensive, and has high requirements for the environment and operators. This allows the measurement of anaerobic threshold to be carried out conveniently in various scenarios such as bedside, competition field, and gym, significantly reducing the application threshold and providing a practical and efficient technical solution for the fields of exercise physiology research, clinical rehabilitation assessment, and public fitness guidance.

[0019] Therefore, this method enables non-invasive, real-time, and convenient measurement and calculation of the anaerobic threshold using an end-tidal carbon dioxide monitor. Corresponding systems employ the same principle and achieve the same or similar technical effects.

[0020] In this invention, the above-described technical solutions can be combined with each other to achieve more preferred combinations. Other features and advantages of this invention will be set forth in the following description, and some advantages may become apparent from the description or be learned by practicing the invention. The objects and other advantages of this invention can be realized and obtained from what is particularly pointed out in the description and drawings. Attached Figure Description

[0021] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts. Figure 1 The flowchart shows the anaerobic threshold calculation method based on the dynamic change of end-tidal carbon dioxide concentration provided in Embodiment 1 of the present invention. Figure 2 The curve showing the relationship between end-tidal carbon dioxide concentration and exercise power (including derivative) provided in Embodiment 1 of the present invention; Figure 3 The curve showing the relationship between respiratory rate and exercise power (including derivative) provided in Embodiment 1 of the present invention; Figure 4 The curve showing the relationship between motion power, rotational speed, and time provided in Embodiment 1 of the present invention; Figure 5 This is a schematic diagram of the structure of the anaerobic threshold calculation system based on the dynamic change of end-tidal carbon dioxide concentration provided in Embodiment 2 of the present invention. Detailed Implementation

[0022] Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, which form part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not intended to limit the scope of the present invention.

[0023] A specific embodiment 1 of the present invention discloses a method for calculating the anaerobic threshold based on the dynamic changes in end-tidal carbon dioxide concentration using a portable device. The flowchart of this method is shown below. Figure 1 As shown.

[0024] Step S1: Have the subject perform progressively increasing exercise on the exercise bike, and collect the subject's end-tidal carbon dioxide concentration, respiratory rate, and exercise power of the exercise bike in real time.

[0025] To ensure calculation accuracy, subjects can exercise sequentially through a warm-up phase, a formal testing phase, and a rest phase. During the warm-up phase, subjects first undergo 3 minutes of resting observation, followed by 3 minutes of no-load (0W) cycling to establish physiological baseline values ​​such as end-tidal carbon dioxide concentration and respiratory rate. In the formal testing phase, the load is progressively increased until the subject reaches exhaustion and can no longer maintain the prescribed speed. During the rest phase, the power of the stationary bike is reduced to 0W, and the subject continues cycling at approximately 20 rpm for 2 to 3 minutes before ending the exercise.

[0026] In the specific implementation process, during the formal testing phase, the electromagnetic resistance system of a power bike (such as a power bicycle) can be used to enable the subject to perform progressively increasing exercise. The subject's exercise power (P) is set to increase linearly with test time (t), for example, at a constant rate of 20W / min or 0.33W / s. Taking the "Ramp" mode as an example, the CapnoBike software will calculate and automatically adjust the magnetic resistance in real time based on the set rate (e.g., 20W / min), ensuring that the power increases linearly even if the subject's rotation speed fluctuates within a certain range. The system algorithm has the highest accuracy under the Ramp protocol and is also compatible with stepped increasing protocols. Furthermore, the rotation speed of the power bike can be collected during implementation. The rotation speed (RPM) of the power bike refers to the physical frequency (revolutions per minute) of the actual pedal rotation during riding, which is a physiological state quantity maintained autonomously by the subject. Taking the RAMP20W exercise program as an example, the CapnoBike software will send power increase commands to the power bike at a rate of 1W every 3 seconds to ensure a linear increase in exercise power. In terms of connectivity, the power bike connects to the terminal device (PC) of the exercise physiological data acquisition and uploading software (CapnoBike) via physical interfaces such as USB; in terms of protocols, the software sends resistance control commands in real time through standard communication protocols such as standard serial port protocols or power meter standard protocols, while simultaneously uploading back the real-time exercise power (P) and rotational speed (RPM) measured by the sensors.

[0027] In addition, an end-tidal carbon dioxide monitor can be used to collect the subject's end-tidal carbon dioxide concentration (i.e., end-tidal carbon dioxide partial pressure) PETCO2 and respiratory rate RR in real time.

[0028] In the specific implementation process, the collected data on end-tidal carbon dioxide concentration, respiratory rate, and the exercise power and speed of the power vehicle can also be aligned. For example, the software samples the exercise power, speed and other data transmitted by the power vehicle at a frequency of 1Hz (once per second) and timestamps them with physiological signals such as PETCO2 and respiratory rate RR transmitted by the end-tidal carbon dioxide monitor to ensure that each set of physiological indicators corresponds to a precise instantaneous load.

[0029] Step S2: Preprocess the collected end-tidal carbon dioxide concentration and respiratory rate sequences of the subjects to obtain interpolated sequences of end-tidal carbon dioxide concentration and respiratory rate with the power of the dynamometer as the X-axis.

[0030] Preferably, in this embodiment, the preprocessing operations include signal cleaning, signal smoothing, and resampling to eliminate abnormal fluctuations and interference. It is important to emphasize that, due to the inherent physiological fluctuations and equipment noise in respiratory data, direct differentiation will produce significant errors, necessitating rigorous cleaning. Specific details are as follows.

[0031] Step S21: Perform signal cleaning on the sequences of end-tidal carbon dioxide concentration and respiratory rate, respectively.

[0032] Preferably, in this embodiment, a moving median filter is used to clean the signals of the acquired end-tidal carbon dioxide concentration and respiratory rate sequences of the subjects.

[0033] For example, in practical implementation, the window length of the moving median filter can be set to 5 data points (approximately 10 seconds). If the deviation of the end-tidal carbon dioxide concentration (PETCO2) or respiratory rate (RR) at a sampling point from the median in the window exceeds 3σ (standard deviation), then the median is used to replace that point. This method can effectively eliminate abrupt changes caused by coughing or signal loss.

[0034] Step S22: Perform signal smoothing on the sequences of end-tidal carbon dioxide concentration and respiratory rate after cleaning, with the power of the vehicle as the X-axis.

[0035] Preferably, in this embodiment, an SG filter is used to smooth the end-tidal carbon dioxide concentration sequence and respiratory rate sequence after cleaning, with the vehicle's power output as the X-axis. Compared to the ordinary moving average method, the SG filter, based on local polynomial least squares fitting, can better preserve the peak and trough characteristics of the signal. For example, the window length of the SG filter can be set to 15-30 seconds (approximately 7-15 data points). The essence of the SG filter is local polynomial least squares fitting: ordinary moving average calculates the average value through weighted averages, while SG filtering fits a polynomial using least squares within each moving window, and then uses the value of the polynomial at the center point of the window as the smoothed output. The order of the polynomial can be set to 2nd or 3rd order.

[0036] Step S221: Perform motion power mapping on the sequences of end-tidal carbon dioxide concentration and respiratory rate after cleaning to form a sequence of end-tidal carbon dioxide concentration and respiratory rate with the motion power of the power vehicle as the X-axis.

[0037] In this embodiment, since exercise power increases linearly, to facilitate the differentiation of exercise power in step S3, the data based on time t can be mapped to the X-axis representing exercise power (P), and this X-axis can be standardized as a power axis. During the anaerobic threshold test in this embodiment, exercise power increases uniformly with time; therefore, exercise power P can be considered a linear function of time t. Using known timestamps and a linear correspondence, respiratory data (end-expiratory carbon dioxide concentration and respiratory rate) can be marked one by one on the corresponding power scale, thus forming a sequence of end-expiratory carbon dioxide concentration and respiratory rate with the exercise power of the power vehicle as the X-axis. For example, the smallest scale on the power axis can be set to 1W.

[0038] Step S222: Use an SG filter to smooth the signals of the end-tidal carbon dioxide concentration and respiratory rate sequences with the power of the vehicle as the X-axis.

[0039] Specifically, using an SG filter, the following steps are performed: within each local window of the power axis, a k-th order polynomial is used to fit the relationship between the end-tidal carbon dioxide concentration sequence, the respiratory rate sequence, and the exercise power. Subsequently, the end-tidal carbon dioxide concentration sequence within the local window is smoothed using the k-th order polynomial corresponding to the end-tidal carbon dioxide concentration, and the respiratory rate sequence within the local window is smoothed using the k-th order polynomial corresponding to the respiratory rate. In this embodiment, since step S3 requires calculating the second derivative, and calculating the second derivative requires at least second-order continuity, k is typically taken as 2 or 3.

[0040] For example, within a certain local window of the power axis, the k-th order polynomial mathematical model of end-tidal carbon dioxide concentration PETCO2 and exercise power P is as follows: (1) Where P is the independent variable, namely the current power of motion (unit: W). The dependent variable is the end-tidal carbon dioxide concentration PETCO2 corresponding to the exercise power P. - All are parameters to be fitted to a k-th order polynomial. When k is typically 2, This represents the baseline value for local fitting. Indicates the local slope. It represents local curvature (acceleration).

[0041] In practice, the coefficients are obtained by fitting the k-th order polynomial using the least squares method. , …After that, the original data points can be replaced with the value of the fitted polynomial at the center point of the window, thus achieving signal smoothing. Therefore, the smoothed output value is the value of the polynomial at the center point, i.e., PETCO2(Pc) = a0. Since a0 integrates the trend information of all points within the window and eliminates random noise, it plays a smoothing role. It should be noted that the local window length (N) determines the "smoothing strength". The larger N is, the smoother the curve, but the sensitivity to sudden turning points will decrease; the polynomial order k determines the "curve following ability". The higher k is, the better it can fit complex fluctuations, but too high a k will lead to overfitting (misjudging noise as a signal).

[0042] One major advantage of the SG filter is that it can smooth the sequence while directly passing through... and The first and second derivatives of the signal were obtained without the need for additional difference calculations, which provided direct mathematical support for identifying the inflection point in step S3.

[0043] Step S23: Perform linear interpolation on the smoothed end-tidal carbon dioxide concentration and respiratory rate sequences with exercise power as the X-axis to obtain the interpolated sequences of end-tidal carbon dioxide concentration and respiratory rate.

[0044] In the specific implementation process, linear interpolation is performed on the processed end-tidal carbon dioxide concentration (PETCO2) and respiratory rate (RR) to ensure that there are corresponding values ​​at every integer scale of 1W. It should be noted that linear interpolation and coordinate transformation are general signal processing techniques, but applying them to the highly fluctuating respiratory signals collected by portable devices, and combining them with SG filtering for resampling in the power domain to extract specific anaerobic threshold feature points, this combination scheme and parameter selection is an improvement on the application scenario of this invention, and has clear specificity and technical relevance.

[0045] Step S3: Identify the characteristic inflection point of exercise power based on the interpolation sequence of end-tidal carbon dioxide concentration and respiratory rate, and use the characteristic inflection point as the calculation result of the subject's anaerobic threshold.

[0046] Step S31: Calculate the second derivative of the relative motion power of the vehicle within each local window for the interpolation sequences of end-tidal carbon dioxide concentration and respiratory rate.

[0047] In practice, to calculate the characteristic inflection point, it is necessary to calculate the rate of change of the interpolated sequences of end-tidal carbon dioxide concentration and respiratory rate relative to power P.

[0048] (1) Calculation of the first derivative (2) (3) By calculating the first derivatives of end-tidal carbon dioxide (PetCO2) and respiratory rate (RR) as a function of power, the "state evolution" of physiological indicators can be transformed into quantifiable "rate of change signals." This process enables the identification of compensatory inflection points and eliminates individual differences, as explained below. Identifying compensatory inflection points: The anaerobic threshold marks a shift in the body's metabolism from primarily aerobic to anaerobic, at which point the respiratory control system produces a nonlinear ventilation response to balance acid-base levels. The first derivative effectively amplifies this nonlinear characteristic, allowing the algorithm to accurately capture the "abrupt inflection point" where indicators increase with power. Eliminating individual differences: The derivative reflects dynamic trends and can shield against interference from different users' baseline physiological values ​​(such as vital capacity and resting respiratory rate), improving the universality and robustness of the system's automatic calculation of anaerobic thresholds in portable scenarios.

[0049] (2) Calculation of the second derivative (4) (5) In practice, this step can directly perform second derivative on the fitted k-th order polynomial corresponding to the end-tidal carbon dioxide concentration and respiratory rate in each local window, and obtain the second derivative of the interpolation sequence of end-tidal carbon dioxide concentration and respiratory rate relative to the motion power of the power vehicle in the corresponding local window.

[0050] For example, for the fitting function f(P), its second derivative with respect to the motion power P is directly equal to: (6) Where f is the end-tidal carbon dioxide concentration PETCO2 or respiratory rate RR.

[0051] The curves showing the relationship between end-tidal carbon dioxide concentration and exercise power (including derivative), and the curves showing the relationship between respiratory rate and exercise power (including derivative) are as follows: Figure 2 , Figure 3 As shown.

[0052] Step S32: Normalize the data by taking the negative of the second derivative of the relative power of the vehicle within each local window for the interpolated sequence of end-tidal carbon dioxide concentration, and normalize the data by taking the second derivative of the relative power of the vehicle within each local window for the interpolated sequence of respiratory rate.

[0053] Because the end-tidal carbon dioxide concentration (in mmHg) and respiratory rate (in rpm) differ in magnitude and amplitude, they must be normalized for weighting. Furthermore, near the anaerobic threshold (AT), the end-tidal carbon dioxide concentration (PETCO2) typically changes from a steady increase to a decrease (or a significant slowdown in its upward trend), and its curve exhibits a concave shape with a negative maximum second derivative; multiplying by -1 transforms it into a positive maximum.

[0054] In this embodiment, the minimum-maximum normalization method is used to normalize the data, mapping the second derivative or its inverse sequence to the interval [0, 1] or [-1, 1].

[0055] The data normalization formula can be expressed as: (7) 、 These represent the contents of the current local window. The maximum and minimum values. It is the negative of the second derivative of the interpolated sequence of end-tidal carbon dioxide concentration relative to the kinetic power of the vehicle, or the second derivative of the interpolated sequence of respiratory rate relative to the kinetic power of the vehicle.

[0056] Step S33: Within each local window, the data normalization results of the interpolation sequence of end-tidal carbon dioxide concentration relative to the negative of the second derivative of the power of the vehicle and the data normalization results of the interpolation sequence of respiratory rate relative to the second derivative of the power of the vehicle are weighted and fused. The weighted fusion results of all local windows are then concatenated in order of increasing power to construct the fusion objective function.

[0057] Fusion objective function F(P): (8) in, Indicates the weighting coefficient. The result of normalizing the interpolated sequence of end-tidal carbon dioxide concentration relative to the negative of the second derivative of the kinetic power of the vehicle is shown. The interpolated sequence representing the breathing frequency is normalized to the second derivative of the relative power vehicle's kinetic power. Actual testing showed that when... When the value is 0.65, the results are more accurate, but biased towards the end-tidal carbon dioxide concentration PETCO2, because its physiological mechanism is more directly related to lactate buffering.

[0058] Step S34: Take the motion power at which the fusion objective function reaches its maximum value as the feature inflection point.

[0059] It should be noted that in this embodiment, the fusion objective function directly characterizes the change in the anaerobic threshold (AT). The following is a logical explanation of the fusion objective function: Regarding the end-tidal carbon dioxide concentration (PETCO2): near the AT point, the respiratory rate exhibits a nonlinear compensatory surge, resulting in an upward-concave curve with a positive maximum second derivative. In practical applications, the PETCO2 curve near the AT point shows a downward-concave trend (changing from rising to stable or decreasing), and its second derivative is a negative extremum; by adding a negative sign (-1) to it in the fusion function, it can be transformed into a positive characteristic peak. Simultaneously, the RR curve exhibits an upward-concave nonlinear surge, and its second derivative itself is a positive maximum. After weighted fusion, the characteristics of the AT point are effectively amplified, manifesting as a significant positive maximum peak on the objective function F(P). This processing method avoids logical conflicts when judging multiple indicators and improves the robustness of the algorithm's automatic locking.

[0060] In the specific implementation process, after step S1 and before step S2, the following steps can be added: plotting the power curve of the exercise vehicle relative to time, and the speed curve of the exercise vehicle relative to time. If the power curve increases at a constant speed within a preset power deviation range, and the speed curve remains constant within a preset speed deviation range, then step S2 is executed; otherwise, the collected data is invalid, and step S1 needs to be executed again. If the power curve does not increase at a constant speed within the preset power deviation range (meaning that the power fluctuates violently), or the speed curve does not remain constant within the preset speed deviation range (indicating that the subject's speed (RPM) cannot be maintained, such as a sudden drop in speed), the data collected in this round contains serious physiological or mechanical noise, and the anaerobic threshold cannot be accurately calculated based on this data, thereby avoiding misjudgment of the anaerobic threshold. For example, the relationship curves of exercise power, speed, and time are as follows: Figure 4 As shown.

[0061] Furthermore, it is important to emphasize that the power curve maintains a uniform increase within a preset power deviation range. This means that throughout the entire power increase process, the absolute value of the deviation between the actual exercise power and the preset theoretical linear power increase is always less than or equal to a pre-set power threshold. For example, if the preset power increase rate is 30W per minute, then at any given moment, the absolute value of the difference between the actual measured exercise power value and the corresponding theoretical power value (calculated based on the initial power and the increase rate) must not exceed the preset power deviation range, such as ±5W. This condition is set to ensure that the exercise load is applied strictly according to the preset linear pattern, avoiding interference from abnormal power fluctuations (such as sudden jumps, pauses, or excessively rapid / slow increases) on the subsequent changes in respiratory physiological indicators, thereby ensuring the accuracy and reliability of the anaerobic threshold calculation. Only when the power curve meets this uniform increase condition can the exercise load borne by the subject be considered standardized. Only then do the collected data such as end-tidal carbon dioxide concentration and respiratory rate have analytical value and can be used for subsequent derivative calculations and characteristic inflection point identification.

[0062] Maintaining a constant speed within a preset speed deviation range means that throughout the exercise, the absolute value of the deviation between the actual speed and the preset target speed is always less than or equal to a pre-set speed threshold. For example, if the preset target speed is 60 revolutions per minute (RPM), then at any given moment, the absolute value of the difference between the actual measured speed and 60 RPM must not exceed the preset speed deviation range, such as ±3 RPM. This condition is set to ensure that the subject can maintain a stable cadence during exercise on the power bike, avoiding instability in exercise intensity due to large fluctuations in speed (such as sudden increases or decreases, sudden stops, or sharp increases), which in turn affects the regularity of respiratory physiological indicators (such as end-tidal carbon dioxide concentration and respiratory rate) as a function of power. A stable speed is an important prerequisite for ensuring that exercise power accurately reflects physiological load. When the speed remains constant within the preset deviation range, the increase in power can truly be converted into a gradual increase in physiological metabolism, thus enabling the changes in the collected respiratory signals to accurately correspond to changes in exercise intensity. This provides a reliable data basis for subsequent identification of the anaerobic threshold through derivative calculation and fusion of objective functions. If the rotation speed cannot meet this uniform speed condition, it indicates that the subject may be experiencing unstable exercise due to fatigue, improper technique, or other factors. In this case, the collected data will introduce additional interference noise, affecting the accuracy of the anaerobic threshold determination. Therefore, data collection needs to be repeated.

[0063] In the specific implementation process, the system can also display the end-tidal carbon dioxide (PETCO2) curve, the first and second derivative curves of end-tidal carbon dioxide (PETCO2) relative to exercise power, the respiratory rate (RR) curve, the first and second derivative curves of respiratory rate relative to exercise power, as well as the identified inflection point position and the calculated anaerobic threshold value on terminal devices such as computers, tablets, and watches, and output a visual report with a save option.

[0064] In summary, the anaerobic threshold calculation method based on the dynamic change of end-tidal carbon dioxide concentration provided in this embodiment has the following advantages.

[0065] (1) Non-invasive: It only requires the use of respiratory gas analysis, is non-invasive throughout the process, feels comfortable, and is suitable for more people.

[0066] (2) Real-time: During exercise, the end-tidal carbon dioxide concentration PETCO2 curve and inflection point dynamics can be displayed instantly, and the anaerobic threshold AT can be quickly fed back.

[0067] (3) Convenience and low cost: The core component is an infrared carbon dioxide analysis module. The device is lightweight, portable, and far less expensive than large gas metabolism analyzers, making it easy to promote in gyms, sports fields, community hospitals and other places.

[0068] (4) Objective and accurate: The algorithm automatically captures the inflection point, reducing the deviation of human judgment and achieving high accuracy.

[0069] Specific embodiment 2 of the present invention discloses an anaerobic threshold calculation system based on the dynamic change of end-tidal carbon dioxide concentration in a portable device. A schematic diagram of the system is shown below. Figure 5 As shown, the system includes a data acquisition module, a preprocessing module, and an anaerobic threshold calculation module.

[0070] The data acquisition module is used to enable subjects to perform progressively increasing exercise on the exercise vehicle and to collect the subjects' end-tidal carbon dioxide concentration, respiratory rate, and exercise power of the exercise vehicle in real time.

[0071] The preprocessing module is used to preprocess the sequences of end-tidal carbon dioxide concentration and respiratory rate of the subjects, respectively, to obtain the interpolated sequences of end-tidal carbon dioxide concentration and respiratory rate with the power of the dynamometer as the X-axis.

[0072] The anaerobic threshold calculation module is used to identify the characteristic inflection point of exercise power based on the interpolation sequence of end-tidal carbon dioxide concentration and respiratory rate, and use the characteristic inflection point as the calculation result of the subject's anaerobic threshold.

[0073] In this embodiment, the data acquisition module includes an end-tidal carbon dioxide monitor and a feedback sensor of the ergometer; among them, the end-tidal carbon dioxide monitor is used to collect the end-tidal carbon dioxide concentration and respiratory rate of the subject in real time; the feedback sensor of the ergometer is used to collect the exercise power of the ergometer in real time, and the rotation speed of the ergometer during the incremental exercise of the subject can also be used.

[0074] In this embodiment, a portable end-tidal carbon dioxide monitor can be used, which can operate independently of the gas metabolism analysis system. For example, the end-tidal carbon dioxide monitor of Beijing Jinjiaxin Trading Co., Ltd., model: KMI6C, Beijing Medical Device Registration No. 20152070022, its size is about 72 x 155 x 38mm (W x H x D), and the weight is about 380g, and it can be operated by hand.

[0075] In addition, the system may further include a motion detection module, and the motion detection module performs the following operations: draw a power curve of the exercise power of the ergometer relative to time, and a rotation speed curve of the rotation speed of the ergometer relative to time; if the power curve increases uniformly within a preset power deviation range and the rotation speed curve remains uniform within a preset rotation speed deviation range, then perform the preprocessing; otherwise, the collected data is invalid, and the subject is required to restart the incremental exercise on the ergometer.

[0076] For the specific implementation process of the embodiment of the present invention, refer to the above method embodiment, and this embodiment will not be elaborated here.

[0077] Since this embodiment has the same principle as the above method embodiment, this system also has the corresponding technical effects of the above method embodiment.

[0078] Those skilled in the art can understand that all or part of the processes for implementing the above method embodiments can be completed by instructing relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium. Among them, the computer-readable storage medium is a disk, an optical disc, a read-only memory or a random access memory, etc.

[0079] The above is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present invention should be covered by the protection scope of the present invention.

Claims

1. A method for calculating the anaerobic threshold based on dynamic changes in end-tidal carbon dioxide concentration using a portable device, characterized in that, The method includes: Subjects were instructed to perform progressively increasing exercise on a power bike, and their end-tidal carbon dioxide concentration, respiratory rate, and power output of the power bike were collected in real time. The sequences of end-tidal carbon dioxide concentration and respiratory rate of the subjects were preprocessed to obtain interpolated sequences of end-tidal carbon dioxide concentration and respiratory rate with the power of the trolley as the X-axis. The characteristic inflection point of exercise power is identified based on the interpolation sequence of end-tidal carbon dioxide concentration and respiratory rate, and the characteristic inflection point is used as the calculation result of the subject's anaerobic threshold.

2. The method for calculating the anaerobic threshold based on the dynamic change of end-tidal carbon dioxide concentration in a portable device, as described in claim 1, is characterized in that... The preprocessing includes: Signal cleaning was performed on the sequences of end-tidal carbon dioxide concentration and respiratory rate, respectively. The sequences of end-tidal carbon dioxide concentration and respiratory rate after washing were smoothed with the power of the dynamometer as the X-axis. The sequences of end-tidal carbon dioxide concentration and respiratory rate after cleaning were subjected to motion power mapping to form a sequence of end-tidal carbon dioxide concentration and respiratory rate with the motion power of the power vehicle as the X-axis.

3. The method for calculating the anaerobic threshold based on the dynamic change of end-tidal carbon dioxide concentration in a portable device, as described in claim 2, is characterized in that... The signal smoothing process, with the vehicle's motion power as the X-axis, is performed as follows: The sequences of end-tidal carbon dioxide concentration and respiratory rate after cleaning were subjected to motion power mapping to form a sequence of end-tidal carbon dioxide concentration and respiratory rate with the motion power of the power vehicle as the X-axis. The SG filter was used to smooth the sequences of end-tidal carbon dioxide concentration and respiratory rate with the kinetic power of the vehicle as the X-axis.

4. The method for calculating the anaerobic threshold based on the dynamic change of end-tidal carbon dioxide concentration in a portable device, as described in claim 3, is characterized in that... The process involves using an SG filter to smooth the signals of the end-tidal carbon dioxide concentration and respiratory rate sequences, with the vehicle's kinetic power as the X-axis. Within each local window of the power axis, the relationship between the end-tidal carbon dioxide concentration sequence, the respiratory rate sequence and the exercise power is parametrically fitted using a k-order polynomial. The end-tidal carbon dioxide concentration sequence within a local window is smoothed using a fitted k-order polynomial corresponding to the end-tidal carbon dioxide concentration, and the respiratory rate sequence within a local window is smoothed using a fitted k-order polynomial corresponding to the respiratory rate.

5. The method for calculating the anaerobic threshold based on the dynamic change of end-tidal carbon dioxide concentration in a portable device according to any one of claims 1-4, characterized in that, The identification of the characteristic inflection point of motion power is performed as follows: Calculate the second derivative of the relative power of the vehicle's motion within each local window for the interpolation sequences of end-tidal carbon dioxide concentration and respiratory rate. The negative of the second derivative of the relative power of the vehicle within each local window was used to normalize the interpolated sequence of end-tidal carbon dioxide concentration, and the second derivative of the relative power of the vehicle within each local window was used to normalize the interpolated sequence of respiratory rate. Within each local window, the data normalization results of the interpolation sequence of end-tidal carbon dioxide concentration relative to the negative of the second derivative of the power of the vehicle and the data normalization results of the interpolation sequence of respiratory rate relative to the second derivative of the power of the vehicle are weighted and fused. The weighted fusion results of all local windows are then concatenated in order of increasing power to construct the fusion objective function. The motion power at which the fusion objective function reaches its maximum value is taken as the feature inflection point.

6. The method for calculating the anaerobic threshold based on the dynamic change of end-tidal carbon dioxide concentration in a portable device, as described in claim 5, is characterized in that... The method further includes: The rotational speed of the power vehicle was collected during the subjects' exercise with increasing load. Plot the power curve of the vehicle's motion power relative to time, and the speed curve of the vehicle's rotational speed relative to time; If the power curve increases at a constant rate within the preset power deviation range, and the speed curve remains constant within the preset speed deviation range, then the preprocessing is performed.

7. The method for calculating the anaerobic threshold based on the dynamic change of end-tidal carbon dioxide concentration in a portable device, as described in claim 6, is characterized in that... The method further includes: If the power curve does not maintain a constant increase within the preset power deviation range, or the speed curve does not maintain a constant speed within the preset speed deviation range, the collected data is invalid, and the subject must restart the progressively increasing load exercise on the power vehicle.

8. A portable system for calculating the anaerobic threshold based on dynamic changes in end-tidal carbon dioxide concentration, characterized in that, The system includes: The data acquisition module is used to enable subjects to perform progressively increasing exercise on the exercise vehicle and to collect the subjects' end-tidal carbon dioxide concentration, respiratory rate, and exercise power of the exercise vehicle in real time. The preprocessing module is used to preprocess the sequences of end-tidal carbon dioxide concentration and respiratory rate of the subjects respectively to obtain the interpolated sequences of end-tidal carbon dioxide concentration and respiratory rate with the power of the dynamometer as the X-axis. The anaerobic threshold calculation module is used to identify the characteristic inflection point of exercise power based on the interpolation sequence of end-tidal carbon dioxide concentration and respiratory rate, and use the characteristic inflection point as the calculation result of the subject's anaerobic threshold.

9. The anaerobic threshold calculation system based on the dynamic change of end-tidal carbon dioxide concentration in a portable device, as described in claim 8, is characterized in that... The data acquisition module includes: An end-tidal carbon dioxide monitor is used to collect data on the end-tidal carbon dioxide concentration and respiratory rate of subjects in real time. The power vehicle's feedback sensor is used to collect the vehicle's motion power in real time.

10. The anaerobic threshold calculation system based on the dynamic change of end-tidal carbon dioxide concentration in a portable device, as described in claim 8 or 9, is characterized in that... The system also includes a motion detection module; at this time, the preprocessing module also collects the rotational speed of the power vehicle during the subject's exercise with increasing load. The motion detection module performs the following: Plot the power curve of the vehicle's motion power relative to time, and the speed curve of the vehicle's rotational speed relative to time; If the power curve increases at a constant rate within the preset power deviation range, and the speed curve remains constant within the preset speed deviation range, then the preprocessing is performed; otherwise, the collected data is invalid, and the subject restarts the progressively increasing load exercise on the power vehicle.

Images