Method for detecting metabolic energy of exercise

Abstract

This application discloses a method for detecting exercise metabolic energy, belonging to the technical field of exercise physiological parameter detection and indirect calorimetry data processing. It acquires oxygen consumption, measured carbon dioxide excretion, and lactate dynamic parameters during exercise, and synchronizes them over time. Based on the synchronized lactate dynamic parameters, lactate kinetic parameters are obtained, and the contribution of non-oxidizing carbon dioxide in the measured carbon dioxide excretion is estimated and corrected. Aerobic metabolic energy is calculated based on oxygen consumption and the corrected carbon dioxide excretion, and anaerobic glycolysis energy is calculated based on the lactate kinetic parameters, thereby determining the total exercise energy and the proportion of aerobic and anaerobic metabolic contributions. This method has been validated using experimental data from resting, progressively increasing load, and recovery phases.

Description

Technical Field

[0001] This invention relates to the field of exercise physiological parameter detection and indirect calorimetry data processing technology, and in particular to a method for detecting exercise metabolic energy. Background Technology

[0002] Energy supply during human exercise includes aerobic oxidative metabolism and anaerobic metabolism. Aerobic oxidative metabolism primarily provides energy through the oxidation of energy substances such as carbohydrates and fats with the participation of oxygen. As exercise intensity increases or the exercise load changes rapidly, the involvement of anaerobic glycolysis increases, and changes in lactic acid and related metabolism become more pronounced. To evaluate energy consumption and metabolic status during exercise, it is usually necessary to measure respiratory exchange parameters, physiological parameters, or metabolic parameters.

[0003] Indirect calorimetry is a commonly used method in exercise metabolism monitoring. This method calculates energy expenditure by measuring oxygen consumption and carbon dioxide expulsion during exercise, based on respiratory exchange parameters. Under resting conditions, steady-state aerobic exercise, or moderate-to-low intensity continuous exercise, there is a relatively stable correlation between exhaled carbon dioxide and the oxidative metabolism of energy substances; therefore, oxygen consumption and carbon dioxide expulsion can be used for indirect energy calculations.

[0004] In existing technologies, various physiological and biochemical indicators are used to monitor exercise fatigue metabolism. For example, Chinese invention patent application CN 120544900 A (publication date August 26, 2025) discloses a "method and system for monitoring exercise fatigue metabolism based on biomarkers," which can detect lactate concentration and lactate clearance rate, while simultaneously collecting oxygen uptake and carbon dioxide excretion during exercise. It also identifies energy metabolism characteristics and metabolic regulation status based on respiratory exchange parameters, lactate parameters, and exercise endurance indicators. This approach can comprehensively utilize lactate, respiratory exchange parameters, and other biomarkers to evaluate exercise fatigue and recovery status.

[0005] The aforementioned exercise fatigue metabolic monitoring technology can also generate corresponding metabolic analysis results based on data such as lactate, oxygen uptake, carbon dioxide excretion, and exercise endurance, which can then be used for fatigue scoring, recovery trend analysis, or individualized recovery evaluation. Its key technical focus lies in using multiple indicators to identify exercise fatigue and metabolic regulation status.

[0006] Other existing technologies determine the lactate threshold using oxygen ventilation equivalent and carbon dioxide ventilation equivalent. For example, Chinese invention patent application CN 121287071 A (publication date January 9, 2026) discloses a "method and apparatus for determining the lactate threshold, computer program product, and electronic device." The method first obtains discrete sequences of oxygen ventilation equivalent and carbon dioxide ventilation equivalent during exercise, then uses a fitting algorithm to form corresponding curves, and determines the lactate threshold based on the intersection of the two curves. Furthermore, it can determine whether the state is aerobic or anaerobic based on the heart rate, oxygen uptake, carbon dioxide production, and ventilation volume corresponding to the lactate threshold, and adjust the exercise intensity accordingly.

[0007] In this type of lactate threshold detection scheme, oxygen uptake, carbon dioxide production, and ventilation can be obtained either through gas analysis equipment or based on photoplethysmography (PPG) signals collected by wearable sensors and obtained through a pre-trained model. A certain correlation can thus be established between respiratory exchange parameters and lactate metabolism status, providing a reference for exercise intensity control.

[0008] In addition to the methods mentioned above, lactate testing itself is also widely used for evaluating exercise intensity, lactate threshold, fatigue, and recovery status. Traditional lactate testing obtains lactate concentration through fingertip blood, earlobe blood, or venous blood sampling. With the development of sensor technology, lactate-related parameters can also be obtained continuously or quasi-continuously through sweat, tissue fluid, spectral signals, and electrochemical signals. These detection methods make it possible to analyze the dynamic changes in lactate during exercise.

[0009] However, under dynamic conditions such as progressively increasing exercise load, changes in exercise load, or the entry into the recovery phase, the level of anaerobic glycolysis participation and the states of lactate and hydrogen ions will change with the load stage. During the increasing load phase, lactate and hydrogen ion accumulation can increase; during the recovery phase, lactate production and clearance can coexist. To maintain acid-base balance, the bicarbonate buffer system is involved in regulation, which may generate additional carbon dioxide related to anaerobic glycolysis, acid-base buffering, and ventilation regulation. At this time, the actual carbon dioxide expulsion detected at the respiration end may include contributions from both carbon dioxide formed by the oxidation of energy substances and non-oxidizing carbon dioxide. If all measured carbon dioxide expulsion is still directly used as the calculation parameter for traditional indirect calorimetry, it may affect the estimation of aerobic metabolic energy and total exercise energy.

[0010] As can be seen from the aforementioned representative existing technologies, lactate parameters, oxygen uptake, and carbon dioxide excretion can be used together for exercise metabolic evaluation or lactate threshold determination. However, these technologies primarily use lactate as a parameter for evaluating metabolic state, fatigue level, training load, or threshold, or utilize respiratory exchange parameters to determine the lactate threshold. For situations where the exercise load changes progressively and the measured carbon dioxide sources become mixed during recovery, none of the aforementioned representative schemes utilize the dynamic changes in lactate during exercise to further determine the non-oxidative contribution of measured carbon dioxide and accordingly correct the processing relationship for respiratory exchange parameters used in energy calculations.

[0011] On the other hand, respiratory exchange parameters and lactate dynamic parameters typically originate from different detection processes, and the sampling frequency, sensor response time, and physiological response processes of these two types of parameters may differ. Respiratory exchange parameters can respond relatively quickly to changes in exercise load, while lactate signals may experience time delays due to in vivo transmission processes, sampling processes, and sensor response processes. If an appropriate time correspondence is not established for parameters from different sources, the physiological state they represent at the same time point may belong to different exercise stages, thus affecting the results of dynamic exercise metabolism analysis.

[0012] Therefore, how to reduce the impact of mixed sources of measured carbon dioxide and time misalignment of multiple metabolic parameters on exercise energy detection during the gradual change and recovery of exercise load, and to achieve dynamic differentiation between aerobic and anaerobic metabolic energy, has become an urgent technical problem to be solved. Summary of the Invention

[0013] The main objective of this invention is to provide a method for detecting exercise metabolic energy, which reduces the impact of mixed sources of measured carbon dioxide and time misalignment of multiple metabolic parameters on exercise energy detection during the gradual change and recovery of exercise load, and achieves dynamic differentiation between aerobic and anaerobic metabolic energy.

[0014] To achieve the above objectives, this invention proposes a method for detecting exercise metabolic energy, comprising: Acquire respiratory exchange parameters and lactate dynamic parameters during the subject's exercise process. The respiratory exchange parameters include at least oxygen consumption and measured carbon dioxide excretion. The respiratory exchange parameters and lactate dynamic parameters are synchronized in time so that they correspond on the same time axis. Based on the time-synchronized lactic acid dynamic parameters, at least one of the following lactic acid kinetic parameters is obtained: lactic acid change amount, lactic acid change rate, lactic acid accumulation amount, and lactic acid generation flux. Based on the mapping relationship between the lactate kinetic parameters and the contribution of non-oxidative carbon dioxide, the contribution of non-oxidative carbon dioxide related to anaerobic glycolysis, acid-base buffering and ventilation regulation in the measured carbon dioxide emissions is estimated, and the measured carbon dioxide emissions are corrected according to the contribution of non-oxidative carbon dioxide to obtain the corrected carbon dioxide emissions. The aerobic metabolic energy is calculated based on the oxygen consumption and the corrected carbon dioxide output, the anaerobic glycolysis energy is calculated based on the lactate kinetic parameters, and the total exercise energy, as well as the contribution ratios of aerobic metabolism and anaerobic metabolism, are determined based on the aerobic metabolic energy and the anaerobic glycolysis energy.

[0015] Preferably, the method further includes acquiring the heart rate data of the subject during exercise; before performing the time synchronization, outlier removal, filtering and smoothing, unit conversion, sampling frequency unification and baseline calibration are performed on the respiratory exchange parameters, lactate dynamic parameters and heart rate data, wherein the lactate baseline is the resting lactate value before exercise.

[0016] Preferably, the time synchronization includes at least one of timestamp alignment, interpolation resampling, sliding window matching, or threshold event matching, and is based on the time compensation amount of the lactate signal relative to the respiratory exchange signal. Delay compensation is performed to ensure that the synchronized lactate data meets the requirements. The time compensation amount Based on incremental load testing, time to lactate threshold, time to ventilation threshold, and respiratory exchange rate The mutation time or individualized model fitting is used to determine it.

[0017] Preferably, the resting lactate level before exercise is used as the lactate baseline. Based on the synchronized lactate data Determine the change in lactic acid And based on the rate of change of lactic acid Determine the positive accumulation of lactic acid During the lactate decrease or recovery phase, a lactate clearance coefficient is introduced. To distinguish between net change in lactate and lactate production flux.

[0018] Preferably, the contribution of the non-oxidizing carbon dioxide is based on... It is confirmed that, among them, For a moment Change in lactate levels relative to baseline. The rate of change in lactic acid. This is the carbon dioxide conversion factor. For the subject's weight, This is the apparent volume of distribution of lactic acid. This is the buffer coefficient. For a moment The intensity of exercise, This represents the time compensation amount of the lactate signal relative to the respiratory exchange signal. This is the lactic acid kinetic mapping function.

[0019] Preferably, according to Determine the estimated lactate production flux And based on determining the contribution of the non-oxidizing carbon dioxide, wherein, This is the lactate data after time synchronization. This refers to the resting lactate level before exercise. This represents the lactic acid clearance coefficient.

[0020] Preferably, the corrected carbon dioxide emissions are calculated according to... It is confirmed that, among them, To correct carbon dioxide emissions. To measure actual carbon dioxide emissions, Contribution to non-oxidizing carbon dioxide. This refers to oxygen consumption.

[0021] Preferably, when and When using uniform units that match the calculation relationships below, calculate the rate of aerobic metabolic energy expenditure. and according to Calculate cumulative aerobic metabolic energy .

[0022] Preferably, based on the calculated energy consumption rate of anaerobic glycolysis and according to Calculate the cumulative energy from anaerobic glycolysis. ,in, This is the lactic acid-energy conversion factor.

[0023] Preferably, model parameters are determined based on at least one of the subject's weight, age, sex, height, maximum oxygen uptake, historical lactate data, and calibration experimental data, wherein the model parameters include a carbon dioxide conversion factor. Lactic acid-energy conversion factor lactic acid clearance coefficient and the time compensation amount of lactate signal relative to respiratory exchange signal At least one of them.

[0024] The above technical solution has the following advantages: This application synchronizes respiratory exchange parameters and lactate dynamic parameters during exercise, establishing a time correspondence between lactate kinetics and respiratory exchange during the corresponding exercise phase, thus reducing the impact of response delays from different detection signals on subsequent calculations. Based on this, the contribution of non-oxidizing carbon dioxide in measured carbon dioxide emissions is estimated using lactate kinetic parameters, and this contribution is used to correct the measured carbon dioxide emissions, ensuring that subsequent aerobic metabolic energy calculations use the corrected carbon dioxide emissions instead of undifferentiated measured values. The corrected carbon dioxide emissions and oxygen consumption are used to calculate aerobic metabolic energy, while lactate kinetic parameters are used to calculate anaerobic glycolysis energy, establishing both energy results on the corresponding exercise time axis, and further determining the total exercise energy and the proportions of aerobic and anaerobic metabolic contributions. Furthermore, this application uses six sets of incremental load experimental data to numerically verify the above calculation chain, covering resting, and other conditions. Gradually increasing and the load from Increase to During the loading and recovery phases, the proportions of the six glycolysis models were as follows: to Parameter sensitivity analysis showed that changes in lactate conversion parameters could be transmitted to the metabolic contribution ratio output. Therefore, Example 9 provides concrete and recalcifiable experimental data to support the feasibility of carbon dioxide correction and the decomposition of aerobic metabolic energy and anaerobic glycolysis energy during resting, low-intensity progressively increasing load, and recovery phases in this application. Attached Figure Description

[0025] The accompanying drawings of this application are described below: Figure 1 This is a schematic flowchart of the exercise metabolic energy detection method provided in the embodiments of this application. Detailed Implementation

[0026] The following embodiments are used to further illustrate the technical solution of this application, enabling those skilled in the art to understand and implement the exercise metabolic energy detection method described in this application. These embodiments are only used to explain the technical concept of this application and are not intended to limit the scope of protection of this application. Without changing the technical concept of this application, the data acquisition device, data transmission method, and data processing terminal used in each step can be configured according to the actual detection conditions.

[0027] To facilitate understanding and avoid ambiguity in the meaning of mathematical parameters, the time and lactate-related parameters in all mathematical expressions in this paper are uniformly defined as follows: Represents the current moment on a unified timeline. Indicates the time variable of integration; This represents the original lactate data before time compensation. This represents lactate data after time synchronization. This represents the baseline resting lactate level before exercise. This indicates the change in lactate levels relative to the lactate baseline. Indicates the rate of change in lactic acid. This indicates the positive accumulation of lactic acid. Indicates the lactate clearance coefficient. This represents an estimated lactic acid production flux. This indicates the amount of time compensation for the lactate signal relative to the respiratory exchange signal; This represents time-series data on lactic acid in sweat. Indicates the baseline of sweat. This represents the equivalent lactic acid obtained by converting time-series data of sweat lactic acid. This represents the conversion factor from sweat to equivalent lactic acid.

[0028] The parameters related to respiratory exchange and carbon dioxide correction in all mathematical expressions in this paper are uniformly defined as follows: Indicates time oxygen consumption, Indicates time The measured carbon dioxide emissions. This indicates the contribution of carbon dioxide related to oxidative metabolism. This indicates the contribution of non-oxidizing carbon dioxide to anaerobic glycolysis, acid-base buffering, and ventilation regulation. This represents the carbon dioxide emissions after correction for non-oxidizing carbon dioxide contributions. Indicates ventilation volume. Indicates respiratory exchange rate, Indicates the subject's weight. Indicates the apparent volume of distribution of lactic acid. Indicates the buffer coefficient. Indicates time The intensity of exercise, This represents the carbon dioxide conversion factor. This represents the mapping function between lactic acid kinetic parameters and the contribution of non-oxidizing carbon dioxide.

[0029] The energy-related parameters in all mathematical expressions in this paper are uniformly defined as follows: This indicates the rate of energy expenditure during aerobic metabolism. This indicates the cumulative aerobic metabolic energy. This indicates the rate of energy consumption during anaerobic glycolysis. This indicates the cumulative energy from anaerobic glycolysis. This represents the energy of the phosphagen system. Represents the total kinetic energy. This indicates the proportion contributed by aerobic metabolism. Indicates the proportion contributed by anaerobic metabolism. This represents the lactic acid-energy conversion factor; in the formula... and These represent the energy equivalent coefficients for oxygen consumption and carbon dioxide emissions, respectively, in the Weir energy conversion relationship. This indicates a unit conversion factor that matches the gas flow input unit and time scale. This represents the Body Mass Index. Unless otherwise specified, for integral calculations involving different time scales or different energy units, units must be standardized before integration.

[0030] Example 1 This embodiment provides a method for detecting exercise metabolic energy. This method is mainly used to detect aerobic metabolic energy, anaerobic glycolytic energy, and total exercise energy during human exercise, and is particularly suitable for detecting changes in exercise load over time, such as resting, progressively increasing load, and recovery periods.

[0031] This embodiment uses respiratory exchange parameters and lactate dynamic parameters obtained synchronously during exercise as the main inputs. Unlike methods that directly use measured carbon dioxide output for indirect calorimetry calculations, this embodiment first estimates the contribution of non-oxidizing carbon dioxide related to anaerobic glycolysis, acid-base buffering, and ventilation regulation in the measured carbon dioxide output based on lactate kinetics. Then, it uses this non-oxidizing carbon dioxide contribution to correct the measured carbon dioxide output. The corrected respiratory exchange parameters are used for aerobic metabolic energy calculations, and the lactate kinetic parameters are simultaneously used for anaerobic glycolysis energy calculations, thus forming the energy decomposition results during exercise.

[0032] Figure 1 The overall processing flow of the exercise metabolic energy detection method of this application is shown. The system composition, data flow relationships, and specific processing steps for executing this overall processing flow are described below.

[0033] The system used to perform the method of this embodiment may include an individual information input and individualized calibration module, a non-invasive lactate measurement device, an indirect calorimetry device, an exercise intensity acquisition device, a communication and data synchronization module, a lactate kinetics calculation and carbon dioxide correction module, an energy calculation module, and a display output and storage module.

[0034] The non-invasive lactate measurement device, indirect calorimetry device, and exercise intensity acquisition device interact with the communication and data synchronization module. The communication and data synchronization module sends the synchronized data to the lactate kinetics calculation and carbon dioxide correction module. The individual information input and individualized calibration module provides individual information and model parameters to the lactate kinetics calculation and carbon dioxide correction module. The lactate kinetics calculation and carbon dioxide correction module sends the processing results to the energy calculation module. The energy calculation module further sends the exercise metabolic energy calculation results to the display output and storage module.

[0035] The modules described above can consist of independent detection and data processing devices, or they can be integrated from equipment with corresponding data acquisition and processing capabilities. Detection data can be transmitted between the devices via conventional data communication methods. The data processing section can call pre-stored programs to complete parameter preprocessing, time synchronization, kinetic calculations, carbon dioxide correction, and energy calculations according to the following steps.

[0036] Step S1: Input subject information and complete individualized parameter preparation.

[0037] The individual information input and individualized calibration module inputs the subject's weight, age, gender, height, maximum oxygen uptake, and existing historical lactate data or calibration experimental data. Based on the available information, the individualized parameters required for subsequent modeling are determined.

[0038] The individualized parameters may include a carbon dioxide conversion factor. Lactic acid-energy conversion factor lactic acid clearance coefficient and the time compensation of lactate signal relative to respiratory exchange signal One or more of them.

[0039] in, This is used to establish the conversion relationship between lactic acid kinetics and the contribution of non-oxidizing carbon dioxide. Used to convert lactate production flux or lactate accumulation-related parameters into anaerobic glycolysis energy. This is used to describe the effect of lactate clearance on changes in observed lactate concentration during exercise and recovery. It is used to compensate for the response time difference between lactate signals and respiratory exchange signals.

[0040] The model parameters mentioned above can be obtained from the subject's previous data or through specialized calibration tests. Before performing a specific exercise test, the relevant parameters are stored in the data processing section for subsequent calculations.

[0041] Step S2: Obtain respiratory exchange parameters during the subject's exercise process.

[0042] The respiratory gas data of the subject during exercise is collected using an indirect calorimetry device. This device may include a respiratory acquisition interface, a flow rate detection unit, an oxygen concentration detection unit, a carbon dioxide concentration detection unit, a temperature, humidity, and pressure compensation unit, and a data acquisition unit.

[0043] During the subject's exercise, the indirect calorimeter continuously acquired respiratory flow, oxygen concentration, and carbon dioxide concentration, and combined this data with ambient temperature, humidity, and atmospheric pressure for appropriate data processing to obtain the oxygen consumption over time. Actual carbon dioxide emissions and ventilation .

[0044] in, Reflects the subject at time Oxygen intake and utilization per unit time Indicates that the indirect heat measurement device is at time The actual carbon dioxide emissions obtained from the test This represents a moment on a unified timeline. During dynamic phases such as the gradual increase of exercise load or the transition to recovery,... This may include not only carbon dioxide produced by the oxidation of energy substances, but also carbon dioxide contributions related to anaerobic glycolysis, acid-base buffering, and ventilation regulation.

[0045] Step S3: Obtain the dynamic parameters of lactate during the subject's exercise.

[0046] The non-invasive lactate measurement device continuously or quasi-continuously collects detection signals related to the subject's lactate status and converts them into lactate concentration data according to the corresponding calibration relationship.

[0047] Non-invasive lactate measurement devices can employ one of the following methods: sweat electrochemical sensing, tissue fluid sensing, near-infrared spectroscopy, Raman spectroscopy, photoacoustic detection, fluorescence detection, or multi-sensor fusion detection. The non-invasive lactate measurement device can be placed at a suitable body part to obtain lactate-related signals.

[0048] The dynamic data of lactate without time compensation are recorded as raw lactate data. In addition to lactate concentration, dynamic parameters of lactate can be further divided into changes in lactate amount, rate of change in lactate, amount of lactate accumulation, information on the lactate rise phase, and information on the lactate recovery phase.

[0049] Step S4: Obtain exercise intensity data.

[0050] Heart rate data is continuously collected from subjects using an exercise intensity acquisition device to characterize changes in exercise load during the test. Exercise intensity data is used to help identify exercise phases and can serve as input parameters for a lactate kinetic mapping model.

[0051] In this embodiment, the subject may sequentially experience a resting phase, an increasing exercise intensity phase, a higher exercise intensity phase, a decreasing exercise intensity phase, and a recovery phase, depending on the exercise protocol. These exercise phases are only used to illustrate the dynamic relationship between various parameters and load changes during exercise; it is not required that every exercise test use the exact same load variation pattern.

[0052] Step S5: Preprocess the acquired data.

[0053] The communication and data synchronization module receives respiratory exchange parameters output by the indirect calorimetry device, lactate dynamic parameters output by the non-invasive lactate measurement device, and heart rate data output by the exercise intensity acquisition device.

[0054] The data undergoes abnormal data processing, filtering and smoothing, unit conversion, sampling frequency unification, and baseline calibration, enabling data output from different detection devices to enter a unified data processing flow.

[0055] lactate baseline Lactate levels obtained before exercise at rest can be used. Baseline heart rate can be obtained before exercise at rest.

[0056] For respiratory exchange parameters and lactate dynamic parameters with different sampling frequencies, resampling can be performed along a unified time axis. For data with non-overlapping sampling time points, interpolation can be used to obtain the data values ​​at the corresponding time points. The purpose of the above preprocessing is to reduce the impact of data formats, time resolutions, and outlier detection values ​​from different detection devices on subsequent kinetic analysis.

[0057] Step S6: Time synchronization and delay compensation are performed on respiratory exchange parameters and lactate dynamic parameters.

[0058] After the intensity of exercise changes, and The respiratory exchange state can change rapidly, while the non-invasive lactate detection signal may be affected by the physiological transmission process of lactate in the body, the sampling process, and the sensor response process, thus creating a time difference with the respiratory exchange parameters.

[0059] Therefore, the communication and data synchronization module first aligns the timelines of each data sequence according to a unified timestamp, and can further determine the correspondence between the two types of parameters by using interpolation resampling, sliding window matching, or threshold event matching.

[0060] In this embodiment, based on the time compensation amount of the lactate signal relative to the respiratory exchange signal... Time compensation was performed on the raw lactate data to obtain synchronized lactate data. Their relationship is .

[0061] The time compensation amount It can be determined based on incremental load testing, or by the time to lactate threshold, the time to ventilation threshold, and respiratory exchange rate. The time of significant change or the individualized model fit is used to determine this.

[0062] After time compensation is completed, at the same time Used , , Furthermore, the motion intensity data corresponds to the same motion phase after synchronization. This reduces parameter misalignment caused by lactic acid signal lag or different sampling frequencies, allowing subsequent estimations of the contribution of non-oxidizing carbon dioxide to be based on data with a time-corresponding relationship.

[0063] Step S7: Establish lactic acid kinetic parameters.

[0064] The lactate kinetics calculation and carbon dioxide correction module uses time-synchronized lactate data. Calculate lactate kinetic parameters.

[0065] First, based on the lactate baseline Calculation time Changes in lactic acid Their relationship is .

[0066] Further, the rate of change of lactate was determined based on the changes in synchronous lactate data over time. . It is used to reflect how quickly lactic acid concentration increases or decreases during exercise.

[0067] Based on lactic acid change rate Further determine the positive accumulation of lactic acid The positive accumulation of lactic acid can be expressed as... .

[0068] The above treatment enables lactate kinetics to be described not only by the lactate concentration at a certain moment, but also to describe the degree of change, rate of change, and positive accumulation process of lactate relative to the baseline during the movement.

[0069] During the decline or recovery phase of exercise intensity, the observed changes in lactate concentration are influenced by both lactate formation and lactate clearance. Using only the net change in lactate concentration may not adequately reflect the lactate formation process. Therefore, this embodiment further employs the lactate clearance coefficient. Distinguish between net change in lactate and lactate production flux.

[0070] Based on the apparent distribution volume of lactic acid Subject weight lactic acid change rate lactic acid clearance coefficient Synchronous lactate data and lactate baseline Determine the estimated lactate production flux In this embodiment, the following can be used: .

[0071] By introducing When lactic acid is in the decreasing or recovering phase, it can avoid simply interpreting the observed decrease in lactic acid concentration as the absence of lactic acid production, and allow the lactic acid kinetic parameters to further reflect the dynamic state under the combined action of lactic acid production and clearance.

[0072] Step S8: Estimate the contribution of non-oxidizing carbon dioxide.

[0073] During exercise, the measured carbon dioxide emissions obtained by the indirect calorimeter can be expressed as the sum of the contributions from oxidative metabolism and non-oxidative carbon dioxide, i.e. .

[0074] in, This indicates the contribution of carbon dioxide produced by the oxidative metabolism of energy substrates. This indicates the contribution of non-oxidative carbon dioxide to anaerobic glycolysis, acid-base buffering, and ventilation regulation.

[0075] As exercise intensity increases, the level of anaerobic glycolysis can increase, accompanied by the accumulation of lactic acid and hydrogen ions. When the bicarbonate buffer system participates in acid-base balance regulation, an additional contribution related to the buffering process can appear in the carbon dioxide detected at the exhalation end. Therefore, under these exercise conditions, directly consuming all Using carbon dioxide produced by oxidative metabolism for aerobic energy calculations may cause the calculated aerobic metabolic energy to be affected by the contribution of non-oxidative carbon dioxide.

[0076] This embodiment estimates based on lactic acid kinetic parameters. The estimation relationship can be summarized as follows: .

[0077] in, This is the carbon dioxide conversion factor. For the subject's weight, This is the apparent volume of distribution of lactic acid. This is the buffer coefficient. For a moment Corresponding exercise intensity This represents the time compensation amount of the lactate signal relative to the respiratory exchange signal. This is the lactic acid kinetic mapping function.

[0078] In one specific calculation method of this embodiment, the lactic acid production flux is estimated. As an intermediate parameter between lactic acid kinetics and the contribution of non-oxidizing carbon dioxide, and according to the determination of the contribution of non-oxidizing carbon dioxide.

[0079] Based on the above data relationships, lactate detection results are not used as an independent evaluation parameter alongside respiratory exchange results, but rather as a calculation input to determine the non-oxidative contribution to measured carbon dioxide emissions. This establishes a data correlation between lactate kinetics and the correction of respiratory exchange parameters.

[0080] Step S9: Correct the measured carbon dioxide emissions based on the contribution of non-oxidizing carbon dioxide.

[0081] The lactic acid kinetics calculation and carbon dioxide correction module is based on the results obtained in step S8. ,from The carbon dioxide contribution associated with anaerobic glycolysis, acid-base buffering, and ventilation regulation was isolated, thereby obtaining the corrected carbon dioxide output for subsequent aerobic metabolic energy calculations. .

[0082] Under the basic correction relation, the correction process is reflected in subtracting the estimated contribution of non-oxidizing carbon dioxide from the measured carbon dioxide emissions.

[0083] In this embodiment, to improve the robustness of the model in processing correction results during dynamic motion, the following approach is adopted: Determine the revised carbon dioxide emissions.

[0084] After the above processing, the subsequent aerobic metabolic energy calculation uses... Instead of uncorrected source .

[0085] Step S10: Calculate aerobic metabolic energy based on oxygen consumption and corrected carbon dioxide emissions.

[0086] The energy calculation module receives the result from step S9. and on the synchronous timeline .

[0087] when and When using uniform units that match the calculation relationships below, the rate of aerobic metabolic energy expenditure You can proceed as confirmed.

[0088] The aerobic metabolic energy consumption rate is further accumulated over time to obtain the aerobic metabolic energy. ,Right now .

[0089] In the actual calculation process, the data processing section unifies the time scale and energy units before integration, so that the cumulative result corresponds to the selected exercise time interval. Since the carbon dioxide emissions input into the aerobic energy model have been corrected according to lactate kinetics, the estimated non-oxidative carbon dioxide contribution can be avoided from being directly included in the aerobic metabolic energy calculation as oxidative metabolic carbon dioxide during the exercise load change and recovery phases.

[0090] Step S11: Calculate the energy for anaerobic glycolysis based on lactic acid kinetic parameters.

[0091] The energy calculation module further calculates the estimated lactic acid production flux obtained in step S7. Calculate the energy from anaerobic glycolysis.

[0092] In this embodiment, the energy consumption rate of anaerobic glycolysis As determined.

[0093] in, This is the lactic acid-energy conversion factor. It can be determined based on the subject's weight, age, gender, height, maximum oxygen uptake, historical lactate data, or calibration experimental data.

[0094] The energy consumption rate of anaerobic glycolysis was accumulated over time to obtain the energy of anaerobic glycolysis. ,Right now .

[0095] Therefore, the same set of time-synchronized lactate kinetic data is used on the one hand to estimate the contribution of non-oxidative carbon dioxide and correct the aerobic metabolism calculation input, and on the other hand to quantitatively calculate the energy of anaerobic glycolysis. Both calculation paths are based on the same exercise time axis, thus enabling the generation of aerobic and anaerobic metabolic energy results for the corresponding exercise phase.

[0096] Step S12: Determine the total exercise energy and the contribution ratios of aerobic metabolism and anaerobic metabolism.

[0097] Based on the aerobic metabolic energy obtained in step S10 And the energy from anaerobic glycolysis obtained in step S11 Determine the total kinetic energy In this embodiment, the total kinetic energy satisfies .

[0098] exist When the value is not zero, the contribution ratio of aerobic metabolism can be further determined based on the ratio of aerobic metabolic energy to total exercise energy. ,Right now The contribution of anaerobic metabolism is determined based on the ratio of energy from anaerobic glycolysis to total energy from exercise. ,Right now .

[0099] The display output and storage module receives data from the energy calculation module. The output may include aerobic metabolic energy. Anaerobic glycolysis energy Total kinetic energy aerobic metabolism contribution ratio , contribution ratio of anaerobic metabolism The data includes the dynamic trend of lactate levels, the changes in measured and corrected carbon dioxide emissions, and corresponding exercise phase information. This data can also be stored for subsequent exercise metabolic state analysis.

[0100] Throughout the detection process in this embodiment, respiratory exchange parameters, lactate dynamic parameters, and exercise intensity data are all established on a unified time axis. When the exercise load changes, the system first uses time synchronization and delay compensation to make the changes in lactate dynamics correspond to the same exercise stage as the changes in respiratory exchange. Then, it estimates the contribution of non-oxidative carbon dioxide based on the amount of lactate change, the rate of lactate change, the lactate accumulation state, and the lactate generation flux.

[0101] Therefore, the measured carbon dioxide emissions are no longer directly used as part of the carbon dioxide produced by oxidative metabolism in the calculation of aerobic metabolic energy. The system dynamically corrects the measured carbon dioxide emissions using the estimated contribution of non-oxidative carbon dioxide, and then calculates aerobic metabolic energy in conjunction with oxygen consumption. At the same time, the system calculates the energy of anaerobic glycolysis based on lactate production flux, and further determines the total exercise energy and the proportion of aerobic and anaerobic metabolic contributions.

[0102] The above processing establishes a correlation between respiratory exchange data and lactate kinetic data. During the gradual change and recovery of exercise load, this processing uses lactate kinetics to estimate the contribution of non-oxidative carbon dioxide and corrects the calculated input of respiratory exchange energy accordingly. At the same time, it generates dynamic decomposition results of aerobic metabolic energy and anaerobic glycolysis energy during the same detection process.

[0103] Example 2 This embodiment, based on Embodiment 1, further explains the methods for obtaining lactate dynamic parameters, respiratory exchange parameters, and multi-source data synchronization. This embodiment uses the same main workflow for detecting exercise metabolic energy as Embodiment 1, namely, estimating the contribution of non-oxidative carbon dioxide from lactate kinetic parameters, correcting the measured carbon dioxide emissions, and calculating aerobic metabolic energy and anaerobic glycolysis energy separately. The main difference lies in that the data sources and time synchronization processing methods can be selected according to specific detection conditions.

[0104] Lactate dynamic parameters are not limited to obtaining data using a single type of non-invasive lactate detection method. Non-invasive lactate measurement devices can use sweat electrochemical sensing, tissue fluid detection, near-infrared spectroscopy, Raman spectroscopy, fluorescence detection, photoacoustic detection, multi-wavelength spectroscopy, or multimodal sensing methods to obtain data related to lactate concentration or dynamic changes in lactate.

[0105] When using sweat electrochemical sensing, the sensor acquires lactic acid-related electrochemical signals that change with the movement process, and converts them into lactic acid concentration or lactic acid-related characteristic quantities through a pre-established calibration relationship. When using spectroscopic detection, the spectral response that changes with the state of lactic acid can be acquired, and after calibration, dynamic lactic acid data can be obtained. Although the original signal forms of the above different detection methods are different, the data entering the subsequent processing can all be uniformly represented as lactic acid concentration or lactic acid-related characteristic sequences that change over time.

[0106] In experimental calibration or standardization scenarios, lactate reference data can also be obtained through fingertip blood, earlobe blood, venous blood, or tissue fluid. This lactate reference data can be directly used as input for lactate kinetic calculations or to calibrate non-invasive lactate detection results. When using this method, the subsequent data synchronization, non-oxidative carbon dioxide estimation, and exercise metabolic energy calculation processes remain unchanged in this embodiment.

[0107] Indirect calorimetry devices are not limited to a single structure. Depending on the testing environment, various types can be used, including gas-to-gas indirect calorimetry devices, mixing tank indirect calorimetry devices, mask-type respiratory metabolism devices, breathing valve-type detection devices, air bag-type sampling devices, open-loop indirect calorimetry devices, or wearable respiratory metabolism devices. The key is obtaining oxygen consumption for exercise-based indirect calorimetry. and measured carbon dioxide emissions This can then be used to perform subsequent processing in this embodiment.

[0108] In addition to heart rate, exercise intensity data can also be represented by one or more of the following: treadmill speed, treadmill incline, power bike power, exercise power, acceleration, cadence, stride length, electromyography, blood oxygen saturation, body surface temperature, or exercise posture parameters. Exercise intensity parameters are mainly used to characterize changes in exercise load, assist in determining the exercise stage, and participate in lactate kinetic mapping. Therefore, different sports can select parameters that are suitable for that specific sport.

[0109] After receiving data from the different sources, the communication and data synchronization module converts the data to a unified time base. Since different sensors may have different sampling frequencies and response times, the sampling time of the data is unified before directly calculating lactic acid kinetics.

[0110] In one synchronization method, respiratory exchange parameters, lactate dynamic parameters, and exercise intensity parameters are recorded with a unified timestamp according to their respective acquisition times, and then reconstructed into a time series according to selected time intervals. If a data source does not have a sampled value at a corresponding time point, interpolation resampling can be used to generate the corresponding data value.

[0111] In another synchronization method, a sliding time window can be used to match lactate dynamic data with respiratory exchange data. During the continuous movement of the time window, the correspondence is determined based on the characteristic changes of the two types of data before and after changes in exercise intensity, and the time compensation amount of the lactate signal is determined accordingly. .

[0112] Alternatively, cross-correlation matching can be used to determine the time offset between the two types of signals. This method utilizes the temporal characteristics of changes in lactate dynamic parameters and respiratory exchange parameters with exercise load to determine the relative delay, and then performs time compensation based on the obtained relative delay.

[0113] When using motion events for synchronization, a correspondence between different data sequences can be established based on the time points when motion intensity changes significantly. Alternatively, an event correspondence can be used between lactate threshold and ventilation threshold, or between changes in lactate and respiratory exchange rate. The event correspondence between changes determines the amount of time that needs to be compensated for in the dynamic data of lactate.

[0114] Time compensation can be achieved using either fixed-time compensation or individualized dynamic time compensation. When using fixed-time compensation, the time is determined based on pre-completed calibration experiments. And the same compensation relationship is used under the same detection conditions. When using individualized dynamic time compensation, the compensation relationship can be redefined based on changes in lactate, respiratory exchange, and the stage of exercise during the current subject's exercise process. .

[0115] After time compensation is completed, the dynamic data of lactate, , The exercise intensity data is also mapped onto a unified time axis. Subsequent calculations use synchronized data to ensure consistency between lactate dynamics and respiratory exchange at corresponding time points during the exercise phase.

[0116] Example 3 This embodiment, based on Embodiment 1 and Embodiment 2, explains the selection method of lactic acid kinetic parameters and the estimation method of non-oxidizing carbon dioxide contribution.

[0117] The lactate kinetics calculation and carbon dioxide correction module can select one or more lactate kinetic parameters based on the detection conditions. Lactic acid kinetic parameters can include lactate concentration. Change in lactate relative to baseline lactic acid change rate Second-order rate of change of lactate, positive accumulation area of ​​lactate, lactate formation flux lactic acid clearance coefficient lactate baseline lactate peak, lactate threshold occurrence time, lactate recovery rate, indicators of the lactate rising phase, and indicators of the lactate falling phase.

[0118] During periods of sustained increase in exercise intensity and overall rise in lactate concentration, the change in lactate levels can be the primary focus. lactic acid change rate and positive accumulation of lactic acid Describe the kinetic state of lactic acid.

[0119] As exercise enters the recovery phase, lactate concentration may begin to decrease. However, this decrease in lactate concentration does not indicate that lactate production has completely ceased. To differentiate between the combined effects of lactate production and clearance, a lactate clearance coefficient can be introduced. .

[0120] This embodiment can use the relationship described in Example 1 to determine the estimated lactate production flux. After characterizing the lactate production process by lactate production flux, the buffer coefficient is then used to further analyze the results. Conversion factor with carbon dioxide Determine the contribution of non-oxidizing carbon dioxide .

[0121] In addition to the specific relationships mentioned above, the mapping function between lactic acid kinetic parameters and the contribution of non-oxidizing carbon dioxide is also discussed. It can also be established based on calibration data. The input to the mapping function may include... , , , , , , as well as One or more of them.

[0122] Mapping function Linear or nonlinear models can be used. Other models that can be employed include empirical regression models, state-space models, Kalman filter models, Bayesian models, machine learning regression models, neural network models, lookup table models, or individualized parameter models.

[0123] The common objective of different models is to estimate the contribution of non-oxidizing carbon dioxide related to anaerobic glycolysis, acid-base buffering, and ventilation regulation in measured carbon dioxide emissions based on lactate dynamics information. Therefore, regardless of the specific model used, its output is used for subsequent corrections. It is used not only for independent evaluation of lactate status.

[0124] Model parameters can be determined through individual information input and a personalized calibration module. Personalized calibration can be performed using data obtained from resting-state tests, incremental load tests, or fixed-power tests, or using the subject's historical exercise data, weight, training level, VO2 max, or invasive lactate reference values.

[0125] When complete individual calibration data is unavailable, initial parameters can be derived from existing population data, and these parameters can be updated as subsequent test data from the subjects are obtained. The objects of parameter adjustment can include... , , , , as well as .

[0126] Therefore, while keeping the technical principle of lactate dynamics-based correction of respiratory exchange parameters unchanged, appropriate parameter combinations and mapping models can be selected according to different populations, different exercise modes, and different lactate detection methods.

[0127] Example 4 This embodiment further illustrates the correction method for measured carbon dioxide emissions.

[0128] Lactic acid kinetics calculation and carbon dioxide correction module obtained Then, based on this contribution amount... Make corrections.

[0129] In real-time monitoring scenarios, a time-point correction method can be adopted. Each time a set of lactate kinetic parameters and respiratory exchange parameters that have been synchronized with the time limit is obtained, the corresponding time point can be calculated. Subsequently received This method can generate a carbon dioxide correction curve that changes continuously with the motion process.

[0130] When there are short-term fluctuations in lactate detection data or respiratory exchange data, a sliding window averaging correction method can be used. First, representative values ​​for lactate kinetic parameters or the contribution of non-oxidizing carbon dioxide are generated within a preset time window. Then, the values ​​within the corresponding time window are corrected. The specific length of the time window is determined based on the sensor used and the detection task; this embodiment does not set a fixed value for it.

[0131] For exercise tests with clearly defined load phases, a phase-specific correction approach can be used. Based on the exercise intensity data, the testing process is divided into different phases, and then the lactic acid kinetics and the contribution of non-oxidizing carbon dioxide are determined for each phase. For incremental load tests, segmented corrections can be implemented according to the load change phases.

[0132] In addition to real-time correction, offline playback correction can also be performed after the exercise data acquisition is completed. In this method, complete lactate dynamics data, respiratory exchange data, and exercise intensity data are first saved, and then time synchronization, lactate kinetics calculations, and carbon dioxide correction are performed uniformly. This method is suitable for situations requiring post-processing analysis of the complete exercise process.

[0133] When obtaining information on the credibility of different data sources, a confidence-based weighted correction method can be used to ensure that time intervals with higher data quality have higher credibility in model calculations, thereby reducing the impact of anomaly detection status on the correction results.

[0134] The correction result can be directly expressed as the corrected carbon dioxide emissions. This can also be expressed as a carbon dioxide correction or correction ratio during internal processing. Regardless of the form used, the data ultimately used for aerobic metabolic energy calculations should reflect the result after deducting or correcting for the contribution of non-oxidizing carbon dioxide from the measured carbon dioxide emissions.

[0135] In one specific implementation, the relationship described in Embodiment 1 is adopted. The function constrains the correction results to improve the robustness of the model under conditions of rapid changes in motion parameters.

[0136] Example 5 This embodiment explains an alternative calculation method for aerobic metabolic energy and anaerobic glycolysis energy.

[0137] The energy calculation module is based on oxygen consumption. and revised carbon dioxide emissions Calculate aerobic metabolic energy.

[0138] In addition to the calculation method described in Example 1, other energy calculation models based on the principle of indirect calorimetry can also be used. These models can calculate aerobic metabolic energy based on the corrected respiratory exchange rate, respiratory quotient, oxygen calorific value, glucose and lipid oxidation ratio, or other indirect calorimetric parameters.

[0139] Among these alternative calculation methods, the following are still used As an input reflecting the contribution of carbon dioxide from oxidative metabolism, instead of directly using the uncorrected input. Therefore, changing the specific indirect calorimetry calculation model will not alter the relationship between lactate kinetics and the correction of respiratory exchange parameters.

[0140] The energy from anaerobic glycolysis is not limited to the lactate production flux alone. Calculation. Based on the detection conditions, the energy for anaerobic glycolysis can be determined using lactic acid accumulation, lactic acid change rate, the integral of lactic acid concentration over time, and the lactic acid equivalent oxygen consumption or oxygen deficit model.

[0141] Furthermore, anaerobic metabolic energy estimation relationships can be established by combining exercise power information. When the exercise intensity acquisition device can obtain exercise power, exercise power can be used as an auxiliary parameter for anaerobic energy estimation or individualized calibration.

[0142] Regardless of the alternative anaerobic energy calculation method used, the input is related to the synchronized lactate kinetics. Therefore, aerobic metabolic energy and anaerobic glycolysis energy still correspond to the same exercise time axis, which can be further used to determine the total energy and metabolic energy supply ratio during exercise.

[0143] Example 6 This embodiment, as an optional extension, further considers phosphagen system energy in addition to the aerobic metabolic energy and anaerobic glycolysis energy in Embodiment 1. This extension is used for detection tasks that require further subdivision of short-term energy sources, without changing the aforementioned main processing flow of lactate kinetics-corrected carbon dioxide.

[0144] When a detection task requires further evaluation of short-term rapid energy supply, the phosphagen system can be estimated as an additional energy supply component. The phosphagen system energy can be estimated using motion intensity, motion power, or calibration data related to the phosphagen system energy.

[0145] At this point, the total kinetic energy It can be represented as .

[0146] in, This refers to the aerobic metabolic energy calculated based on oxygen consumption and carbon dioxide excretion corrected for lactate kinetics. This refers to the energy of anaerobic glycolysis calculated based on lactate kinetic parameters. This is the energy source for the phosphagen system.

[0147] In this embodiment, increasing the energy of the phosphagen system does not alter the aforementioned carbon dioxide correction process. During the exercise test, the time synchronization of lactate dynamic parameters with respiratory exchange parameters is still completed first, and then lactate kinetics are used for estimation. and form Then, the corresponding aerobic metabolic energy, anaerobic glycolysis energy, and phosphagen system energy were calculated.

[0148] For motion scenarios where further differentiation of phosphagen-powered energy is not required, the method described in Example 1 can continue to be used. Energy calculation method.

[0149] Example 7 This embodiment illustrates different device implementations of a sports metabolic energy detection system.

[0150] In one implementation, the non-invasive lactate measurement device, the indirect calorimetry device, the exercise intensity acquisition device, the communication and data synchronization module, the lactate kinetics calculation and carbon dioxide correction module, the energy calculation module, and the display output and storage module can be set in the same exercise metabolism detection device to form an integrated device.

[0151] During exercise, the non-invasive lactate measurement device and the indirect calorimetry device send the detection data to the communication and data synchronization module. After completing data preprocessing and time synchronization, the communication and data synchronization module sends the data to the lactate kinetics calculation and carbon dioxide correction module. After the lactate kinetics calculation and carbon dioxide correction module generates the corrected carbon dioxide excretion, the energy calculation module completes various metabolic energy calculations and outputs the results through the display output and storage module.

[0152] In another implementation, the non-invasive lactate measurement device and the indirect calorimetry device can be used as independent detection devices, connected to a data processing terminal via wired or wireless communication. The data processing terminal can be a sports metabolism analyzer, mobile phone, tablet computer, personal computer, server, cloud platform, or edge computing terminal.

[0153] In this split-system, the lactate detection device and the respiratory metabolism detection device are only responsible for obtaining the raw or pre-processed detection data. Time synchronization, lactate kinetic calculation, non-oxidizing carbon dioxide estimation, carbon dioxide correction, and energy calculation can be completed by the data processing terminal.

[0154] The non-invasive lactate measuring device and the indirect calorimetry device can also transmit data to the processing terminal via Bluetooth communication, Wi-Fi communication, wired communication, mobile communication or other data communication methods.

[0155] In another implementation, the lactate detection component and some motion parameter detection components can be integrated into a wearable device, while the indirect thermal detection component serves as a separate respiratory metabolism testing device. The computation process is then deployed on a mobile terminal or cloud algorithm platform. All components establish data correspondences using unified time information.

[0156] The system in this embodiment can also be integrated as a functional module into sports rehabilitation equipment, treadmills, exercise bikes, smart fitness equipment, or fitness testing equipment. After integration, the exercise intensity or exercise power information provided by each device can be synchronized and used in metabolic calculations along with lactate dynamic parameters and respiratory exchange parameters.

[0157] Regardless of whether the system adopts an integrated, split, or wearable device combined with a data processing platform, the lactate kinetics calculation and carbon dioxide correction module determines the contribution of non-oxidative carbon dioxide based on lactate dynamic parameters and uses this contribution to correct the measured carbon dioxide emissions. The energy calculation module then determines the energy for aerobic metabolism and anaerobic glycolysis based on the corrected respiratory exchange parameters and lactate kinetic parameters, respectively.

[0158] Example 8 This embodiment illustrates the execution process of the method of this application in different motion detection tasks.

[0159] For incremental exercise, baseline lactate and respiratory exchange levels at rest can be obtained before starting exercise. The exercise load can then be gradually increased, with baseline levels obtained simultaneously. , Lactic acid dynamic parameters and exercise intensity data.

[0160] As exercise load increases, lactate dynamic parameters and respiratory exchange parameters can exhibit different response rates. The system employs time synchronization and delay compensation to ensure that both types of parameters correspond to the same load stage. Subsequently, the contribution of non-oxidative carbon dioxide is estimated based on the lactate change rate, lactate accumulation state, or lactate generation flux, and is dynamically obtained. .

[0161] For the recovery phase following progressively increasing load exercise, data from both the load phase and the recovery phase can be obtained separately. As the load phase changes, lactate production, lactate clearance, and the contribution of non-oxidizing carbon dioxide can vary; during the recovery phase, lactate clearance and lactate production can coexist. This can be achieved by introducing... The lactate production flux can be estimated during the recovery phase, rather than solely relying on the decrease in lactate concentration to determine the state of anaerobic glycolysis. This represents the lactate clearance coefficient.

[0162] For training load monitoring, lactate kinetic parameters, pre- and post-correction carbon dioxide emissions, aerobic metabolic energy, anaerobic glycolytic energy, and total exercise energy obtained from each test can be continuously saved throughout multiple exercise sessions. Historical data can be used for individualized parameter calibration and to observe the metabolic response of the same subject under different exercise load conditions.

[0163] For fitness assessment or exercise rehabilitation load management, the energy supply status of subjects under different loads can be observed based on the aerobic and anaerobic metabolic contributions corresponding to each exercise phase. During exercise rehabilitation, appropriate testing phases can be selected based on the actual exercise load borne by the subject, and metabolic energy results can be obtained using the same data processing methods.

[0164] For proactive health management or exercise energy expenditure assessment, the total exercise energy during the detection period can be output. Including the contribution ratios of aerobic and anaerobic metabolism, the results of exercise energy testing not only include total consumption but also reflect the relative contributions of different metabolic pathways.

[0165] All the aforementioned detection tasks utilize the same core data processing procedure. The system simultaneously acquires lactate dynamic parameters and respiratory exchange parameters, corrects the temporal relationship between the two types of signals, estimates the contribution of non-oxidative carbon dioxide based on lactate kinetics, uses this contribution to correct the measured carbon dioxide excretion, and calculates aerobic metabolic energy based on the corrected respiratory exchange parameters. Simultaneously, it calculates anaerobic glycolysis energy using lactate kinetic parameters. By changing the type of detection equipment, synchronization method, model type, or specific energy calculation model, the system can adapt to different exercise conditions without altering the fundamental data processing relationships for exercise metabolic energy detection.

[0166] Example 9 This embodiment uses incremental load experimental data from 6 test subjects to verify the calculation chain of this application based on lactate kinetics to correct carbon dioxide and to perform aerobic metabolism energy and anaerobic glycolysis energy decomposition. The experiment consists of three phases: rest, incremental load, and recovery. Data is collected for 5 minutes each in the rest and recovery phases; the incremental load phase follows... Incremental load is carried out, from Increase to The basic information of the testers is shown in Table 1.

[0167] Table 1. Basic Information of Testers

[0168] This experiment measures lactic acid through sweat and converts the sweat time-series input into the "equivalent lactic acid" required by the model according to the implementation scheme of this application, before performing lactic acid flux, carbon dioxide correction, and energy calculations. The parameters used for numerical calculations are shown in Table 2. The parameters below are used to determine the recalcible numerical scenarios and do not cover all parameters involved in the scheme of this application, nor do their values ​​represent the optimal parameters after individualized calibration.

[0169] Table 2. Parameter list used for numerical calculation

[0170] Record the sweat time series data as follows The subject's weight was The equivalent lactic acid after conversion is denoted as The model's lactate production flux is denoted as The conversion of sweat data to equivalent lactic acid and the rate of lactic acid change were calculated according to the following relationship:

[0171] Based on this, the lactate production flux, non-oxidative carbon dioxide contribution, corrected carbon dioxide emissions, and energy components of aerobic and anaerobic glycolysis are calculated according to the following relationships:

[0172]

[0173]

[0174]

[0175]

[0176]

[0177] The energy integration interval is from the first to the last observation time of each experimental data set, without extrapolating the first and last missing times; the integration range includes the resting, exercise, and recovery phases, and does not deduct the resting baseline. This embodiment does not account for phosphagen energy supply. The proportion of aerobic activity and glycolysis in the model are respectively calculated according to... and Confirmed. The results of the six calculations are shown in Table 3.

[0178] Table 3 Calculation results of aerobic and glycolysis components in six groups

[0179] Note: All energy units in Table 3 are... Energy supply from phosphate was not taken into account.

[0180] As shown in Table 3, the aerobic ratio of the six test models was as follows: to The proportion of glycolysis models is to Taking number 001 as an example, the uncorrected Weir integral is... After adopting the joint computation chain described in this application, the model has aerobic properties. Anaerobic glycolysis is The two components combined are All six sets of data could be output from the same computational chain as aerobic model components, anaerobic glycolysis model components, and their respective proportions. Furthermore, there were quantifiable differences between the aerobic model results and the uncorrected Weir integral.

[0181] To investigate the effect of the sweat to lactic acid conversion parameter on the output results, under the condition that other parameters remain unchanged, the following parameters are used: By default value Reduce by half respectively and double The proportion of the glycolysis model was recalculated. This analysis is a deterministic parameter sensitivity analysis, not an actual individual difference test. The results are shown in Table 4.

[0182] Table 4 Parameter sensitivity analysis results

[0183] As can be seen from Table 4, the change The proportions of the latter six glycolysis models all changed accordingly. For example, model number 005... , and The proportions of glycolysis models at different times were respectively , and ,in, This represents the conversion factor from sweat to equivalent lactic acid. The above-mentioned incremental load experiments and parameter sensitivity results demonstrate that the calculation chain described in this application—"sweat lactic acid conversion—lactic acid flux calculation—non-oxidizing carbon dioxide estimation and correction—aerobic / anaerobic energy decomposition"—can produce specific and recalcifiable numerical outputs based on data including resting, incremental load, and recovery phases. Furthermore, changes in the lactic acid conversion parameter can be reflected in the output of the metabolic contribution ratio. This provides specific experimental data and verification results for the feasibility of this application in detecting exercise metabolic energy and energy decomposition during resting, low-intensity incremental load, and recovery phases.

[0184] The above embodiments illustrate the implementation of the exercise metabolic energy detection method and system provided in this application. It should be understood that the above embodiments are used to illustrate the technical solution of this application, and not to limit the scope of protection of this application.

[0185] The core technical concept of this application lies in establishing a time-correlation relationship between dynamic lactate parameters and respiratory exchange parameters during exercise. Based on lactate kinetic parameters, the contribution of non-oxidizing carbon dioxide related to anaerobic glycolysis, acid-base buffering, and ventilation regulation in the measured carbon dioxide output is estimated, and the measured carbon dioxide output is corrected based on this non-oxidizing carbon dioxide contribution. Aerobic metabolic energy is calculated based on the corrected respiratory exchange parameters, and anaerobic glycolysis energy is calculated simultaneously based on lactate kinetic parameters, thereby obtaining the total exercise energy and the contribution of different metabolic energy supply methods.

[0186] In implementing this application, the detection methods for lactate dynamic parameters, respiratory exchange parameters, exercise intensity parameters, data communication methods, time synchronization methods, delay compensation methods, selection methods for lactate kinetic parameters, estimation models for non-oxidative carbon dioxide contributions, correction methods for carbon dioxide emissions, and exercise energy calculation methods can all adopt the corresponding implementation methods described in this specification, depending on the specific detection conditions. As long as the corresponding changes are still based on determining the carbon dioxide contribution related to non-oxidative metabolism using lactate kinetic parameters, and using this contribution to correct the respiratory exchange parameters used for aerobic metabolism energy calculation, the technical process described in this application can be implemented.

[0187] For different subjects or different exercise detection conditions, corresponding individualized model parameters can be determined using resting-state tests, incremental load tests, fixed-power tests, historical exercise data, or calibration experimental data. For differences in sampling frequency, sensor response time, or physiological response delay between different detection devices, time-correspondence relationships between data can be established using timestamp alignment, interpolation resampling, sliding time windows, event matching, or delay compensation. The adjustments to the above implementation methods are all for adapting to actual data acquisition conditions and do not change the fundamental processing relationships for lactate kinetic correction of respiratory exchange parameters and further exercise metabolic energy decomposition.

[0188] Based on the disclosure of this application, those skilled in the art can make equivalent substitutions or adaptive adjustments to the specific detection device, data processing model, parameter determination method, and system deployment form without departing from the above-described technical concept. The scope of such equivalent substitutions or adjustments should be determined according to the technical solution defined in the claims.

Claims

1. A method for detecting exercise metabolic energy, characterized in that, include: Acquire respiratory exchange parameters and lactate dynamic parameters during the subject's exercise process. The respiratory exchange parameters include at least oxygen consumption and measured carbon dioxide excretion. The respiratory exchange parameters and lactate dynamic parameters are synchronized in time so that they correspond on the same time axis. Based on the time-synchronized lactic acid dynamic parameters, at least one of the following lactic acid kinetic parameters is obtained: lactic acid change amount, lactic acid change rate, lactic acid accumulation amount, and lactic acid generation flux. Based on the mapping relationship between the lactate kinetic parameters and the contribution of non-oxidative carbon dioxide, the contribution of non-oxidative carbon dioxide related to anaerobic glycolysis, acid-base buffering and ventilation regulation in the measured carbon dioxide emissions is estimated, and the measured carbon dioxide emissions are corrected according to the contribution of non-oxidative carbon dioxide to obtain the corrected carbon dioxide emissions. The aerobic metabolic energy is calculated based on the oxygen consumption and the corrected carbon dioxide output, the anaerobic glycolysis energy is calculated based on the lactate kinetic parameters, and the total exercise energy, as well as the contribution ratios of aerobic metabolism and anaerobic metabolism, are determined based on the aerobic metabolic energy and the anaerobic glycolysis energy.

2. The method for detecting exercise metabolic energy according to claim 1, characterized in that, It also includes acquiring the heart rate data of the subject during exercise; before performing the time synchronization, outlier removal, filtering and smoothing, unit conversion, sampling frequency unification and baseline calibration are performed on the respiratory exchange parameters, lactate dynamic parameters and heart rate data, wherein the lactate baseline is the resting lactate value before exercise.

3. The method for detecting exercise metabolic energy according to claim 1, characterized in that, The time synchronization includes at least one of the following methods: timestamp alignment, interpolation resampling, sliding window matching, or threshold event matching, and is based on the time compensation amount of the lactate signal relative to the respiratory exchange signal. Delay compensation is performed to ensure that the synchronized lactate data meets the requirements. ;in, Representing moments on a unified timeline Synchronized lactate data, This indicates the original lactate data at time [time]. The value of , Representing moments on a unified timeline, This indicates the amount of time compensation for the lactate signal relative to the respiratory exchange signal; Based on incremental load testing, time to lactate threshold, time to ventilation threshold, and respiratory exchange rate The mutation time or individualized model fitting is determined, where, This indicates the respiratory exchange rate.

4. The method for detecting exercise metabolic energy according to claim 1, characterized in that, The resting lactate level before exercise was used as the lactate baseline. Based on the synchronized lactate data according to Determine the change in lactic acid ,according to Determine the rate of change of lactate and in accordance with Determine the positive accumulation of lactic acid During the lactate decrease or recovery phase, a lactate clearance coefficient is introduced. To distinguish between net change in lactate and lactate production flux; among which, Indicates the current moment. Represents the time variable of integration. This indicates the resting lactate level before exercise. Indicates time Synchronized lactate data, Indicates time Change in lactate levels relative to baseline. Indicates the rate of change in lactic acid. Indicates from the start time to time 1 The positive accumulation of lactic acid, This represents the lactate clearance coefficient.

5. The method for detecting exercise metabolic energy according to claim 1, characterized in that, The contribution of non-oxidizing carbon dioxide is based on Determined; among them, Indicates time The contribution of non-oxidizing carbon dioxide. This represents the carbon dioxide conversion factor. This represents the mapping function between lactic acid kinetic parameters and the contribution of non-oxidizing carbon dioxide. Indicates time Change in lactate levels relative to baseline. Indicates the rate of change in lactic acid. Indicates the subject's weight. Indicates the apparent volume of distribution of lactic acid. Indicates the buffer coefficient. Indicates time The intensity of exercise, This indicates the amount of time compensation for the lactate signal relative to the respiratory exchange signal. It represents a moment on a unified timeline.

6. The method for detecting exercise metabolic energy according to claim 5, characterized in that, according to Determine the estimated lactate production flux and according to Determine the contribution of the non-oxidizing carbon dioxide; wherein, Indicates time The estimated lactate production flux, Indicates the apparent volume of distribution of lactic acid. Indicates the subject's weight. Indicates the rate of change in lactic acid. Indicates the lactate clearance coefficient. Indicates time Synchronized lactate data, This indicates the resting lactate level before exercise. This represents the carbon dioxide conversion factor. Indicates the buffer coefficient. Indicates time The contribution of non-oxidizing carbon dioxide.

7. The method for detecting exercise metabolic energy according to claim 5, characterized in that, The corrected carbon dioxide emissions are calculated according to Determined; among them, Indicates time Corrected carbon dioxide emissions Indicates time The measured carbon dioxide emissions. Indicates time The contribution of non-oxidizing carbon dioxide. Indicates time oxygen consumption, It represents a moment on a unified timeline.

8. The method for detecting exercise metabolic energy according to claim 7, characterized in that, when and When using a unified unit that matches the following calculation relationship, according to Calculate the rate of aerobic metabolic energy expenditure, and based on Calculate the cumulative aerobic metabolic energy; among which, Indicates time The rate of aerobic metabolic energy expenditure. Indicates from the start time to time 1 The cumulative aerobic metabolic energy, Indicates time oxygen consumption, Indicates time Corrected carbon dioxide emissions and These represent the energy equivalent coefficients for oxygen consumption and carbon dioxide emissions, respectively, in the Weir energy conversion relationship. This indicates a unit conversion factor that matches the gas flow rate unit and time scale used. This represents the time variable for integration.

9. The method for detecting exercise metabolic energy according to claim 6, characterized in that, according to Calculate the energy consumption rate of anaerobic glycolysis, and based on... Calculate the cumulative energy from anaerobic glycolysis; among which, Indicates time The rate of energy consumption in anaerobic glycolysis. Indicates from the start time to time 1 The cumulative energy from anaerobic glycolysis This represents the lactic acid-energy conversion factor. Indicates time The estimated lactate production flux, This represents the time variable for integration.

10. The method for detecting exercise metabolic energy according to claim 1, characterized in that, The model parameters, including a carbon dioxide conversion factor, are determined based on at least one of the following: the subject's weight, age, sex, height, maximum oxygen uptake, historical lactate data, and calibration experimental data. Lactic acid-energy conversion factor lactic acid clearance coefficient and the time compensation amount of lactate signal relative to respiratory exchange signal At least one of them.

Images