Research On The Relationship Between The Subjective And Objective Load Indexes Of Training Monitoring Of Ergometer

Background:Rowing is a typical physical dominant event.The absolute speed of modern rowing is increasing and the pace strategy is developing towards uniform type,which puts forward new requirements for athletes.Whether the athletes can achieve excellent results in the competition depends on whether they have the physical fitness and skill level required for the competition.To improve the competitive ability,scientific and reasonable training plans should be formulated,and the training load should be measured,evaluated and regulated to optimize the sports performance and reduce the injury rate,which puts forward stricter requirements for the training load monitoring activities.Current studies mainly focus on the monitoring of heart rate and blood lactic acid,and evaluate the load borne by athletes in combination with other physiological and biochemical indicators,or evaluate the training load from the aspects of displacement,acceleration and deceleration with the help of GPS.However,objective indicators cannot fully reflect the subjective cost and neural driving force paid by athletes in training.At the same time,due to the influence of factors such as athlete status,training environment and different training methods,the data collected by different methods have a certain degree of error,and "objective" is not completely equal to science.In recent years,athlete self-report method has been applied in training.It is a subjective evaluation of the bearing load,which can be a useful supplement to the objective index method.It is an important regulatory factor among external load,sports performance and injury risk,and an inevitable choice under the training concept of "athlete as the main body".Objective:The study takes the debugging and application of the subjective load monitoring scale as the starting point to explore the feasibility and reliability of load monitoring in rowing training practice.On this basis,the influence of different training methods and sampling nodes on this index was explored,and the relationship between subjective and objective indicators under different conditions was clarified,so as to serve as the basis for training load regulation,and experimental research was conducted to design different load regulation methods based on the central and peripheral formation mechanism of subjective load,hoping to provide references for the application and implementation of subjective monitoring methods for rowing training load.Methods:Study 1:Firstly,based on the Brislin scale localization process,the sRPE scale was debugged.and the reliability and validity of CR10 and sRPE scales applied to ergometer,strength and function training were determined by retest reliability and calibration validity methods.In study 2,15 rowers were selected as test subjects.First,during the 18-week rowing training load monitoring,objective index data such as PM5,heart rate band and blood lactic acid were obtained,and subjective load was collected by CR10 and sRPE scales at the interval between training classes,post10,post30 and post60 sampling nodes.Secondly,three-factor analysis of variance(training means × adoption node × exercise level)was used to determine the differences of subjective load under different factors.The relationship between CR10,training center rate,blood lactic acid and other objective indicators in training was discussed by binary linear regression.When the linear regression curve was poorly fitted,the nonlinear method was used to deal with it.Multiple linear regression method and structural equation model were used to explore the multivariable relationship between sRPETL and objective indicators,and the regulating variables and mediating variables were mined to determine the path relationship between each factor.The third study was conducted in a randomized controlled trial to study the effects of inspiratory muscle preparation and change training on brain activation level,muscle oxygen saturation and CR10.24 rowers were included as subjects and randomly divided into IMW group,exchange group and blank control group.A total of 2 tests were completed in two training weeks of basically the same stage:all subjects in the first test performed general preparation activities;before the second test,the control group and the exchange group continued to perform general preparation activities;the IMW group increased the inspiratory muscle quasi-activity intervention.In the training,the control group and the IMW group completed the same ergometer training 10min×3 as the first time,while the change group took the lower extremity alternate training(power bicycle).The changes of oxygen content in the lateral femur muscles were continuously monitored with portable Moxy(USA).Oxygenated hemoglobin and deoxygenated hemoglobin were obtained to characterize the changes in oxygen uptake in muscle tissue(μmol/L·cm).The changes of muscle oxygen saturation(SmO2)during exercise and recovery period were monitored in real time from the quiet state,and the average value of every 10s during exercise and recovery period was taken as the recorded value.fNIRS technology used the Brite Lite real-time wireless brain tissue blood oxygen monitoring system of Artinis Medical Systems in the Netherlands to measure the oxygenation state of the prefrontal cortex.In order to explore the influence of different interventionrelated indicators,muscle oxygen,CR10 and HbO2 of the first test were taken as baseline values and included in covariance analysis as covariables to explore the changes of indicators after intervention.Results:(1)The sRPE scale localized by Brislin has high retest reliability(r=0.758,ICC=0.755),and has certain validity when applied to ergometer,strength and function training.Among them,the validity of ergometer 6km×3 training is the best(r=0.811,P<0.001).When applied to functional training,TRIMP had the lowest correlation with sRPETL(r=0.258),which may be caused by the low correlation between functional training center rate and training intensity.The Bland-Altman scatter plot was further drawn to verify the consistency of different methods.The results showed that there was no statistically significant difference between different monitoring methods in the maximum force and ergometer 10min×3 training means(P>0.05),and 93.4%and 98.6%of the observed values were within the consistency limit interval,respectively.There are a few observation points outside the consistency limit,which indicates that there is a strong consistency between the two sets of data.(2)The results of multivariate analysis of variance showed that both training methods and sampling nodes could affect CR10 scores of athletes(P<0.001),and the interaction between the two factors was significant(F=6.074,P<0.001).However,the interaction of training means×movement level,sampling node × movement level,training means × sampling node × movement level is not significant On this basis,simple effect analysis is carried out.Further multiple comparisons with fixed sampling nodes showed that at node 1,the CR10 of the 6km×3 training group was significantly different from that of the 10min×3 training group(P<0.001),while the CR10 of the 45min incremental training group was not significantly different(P=0.211).At node 2,there were significant differences among different training groups(P<0.001).At node 3,the difference in CR10 between the 6km×3 training group and the other two groups was statistically significant.In the three sampling nodes,the CR10 of 10min×3 training was significantly higher than that of other training methods(P<0.001).It is suggested that CR10 in training can sensitively reflect the difference in intensity of different training methods.Bonferroni’s multiple comparison test showed that with fixed training methods,the difference of CR10 at different sampling nodes was statistically significant and showed an increasing trend.At the same time,although the intensity of 6km×3 training and 10min×3 training in a single group was the same,the CR10 rating would increase with the increase of the number of groups,which meant that no matter whether the interval was sufficient or not,the CR10 rating would increase.This strength index is affected by the number of groups and the amount of training.Both training methods and sampling nodes can affect the sRPE scores of athletes(P<0.001).Further multiple comparisons with fixed sampling node factors show that the sRPETL of 10min×3 training is significantly higher than that of other training methods in the three sampling nodes(P<0.001).At the post10 node,the sRPETL difference between the 6km×3 training group and the 45min incremental training group was significant(P<0.001),but at the post30 and 60 nodes,the difference between the 6km×3 training group and the 45min incremental training group was not significant.It was suggested that the sRPE of the 45min increasing group was higher at 10min after training,and tended to be stable at 30min after training,possibly due to the higher intensity at the end of training.(3)The relationship between sRPETL and other objective indicators was determined by multiple linear regression,and a preliminary model 1 was established by incorporating the dependent variable "sRPETL" and the independent variable "training time,paddle frequency,total paddle number,500m segment time,average heart rate,blood lactic acid and TRIMP".The coefficient of determination of the regression equation R2 was 0.606.The regression method was used to eliminate the independent variables,and the model 5 containing three independent variables was obtained as the optimal result.At the same time,the D-W value of 1.721 was close to 2,which can be considered that the residual sequence has no autocorrelation,indicating that the residual satisfies the conditions of multiple linear regression analysis.The adjusted R2 of the final model is 0.522,that is,HRmean,Bla and TRIMP can explain 52.2%of the sRPETL variation.The final regression equation is Y=477.8+3.477HRmean+17.262Bla+2.156TRIMP.According to this method,regression equations of sRPETL of other two training different sampling nodes were calculated.Since the sRPETL difference between post30 and post60 nodes was not statistically significant,only post30 node and post 10 node were included for comparison.The results showed that,The sRPETL fitting of post30 nodes was better than that of post 10 nodes in the three training types,and the SRPETL fitting degree of post30 nodes in the 10×3 training was the highest(R2=0.577,P<0.001).(4)Theoretical hypothesis was made according to Banister’s impulse-response model and Mehrabian’s Stimulus-Organism Response model.It is believed that the Objective measurement of external load is the stimulus factor(OEL)as the ante-dependent variable,and the Objective evaluation of internal load is the objective-internal load(OIL)as the intermediate variable.As a result variable,the Subjective-internal load(SIL)of the final training course is an overall response of the training subject’s behavior.In addition,physical perception during training is a Moderating variable(M)of the relationship between OEL and OIL.According to this,a mediated model with adjustment is established,and the minimum value of standardized factor load of each observed variable is 0.488 and greater than 0.4.The minimum AVE value is 0.509 and greater than 0.5;The minimum value of CR is 0.784 and greater than 0.7,which meets the standard,indicating that the validity of the model meets the standard.Path analysis showed that the four path relationships in the theoretical model were basically confirmed:OEL could significantly predict SIL(P<0.001),and Bootstrap test showed that OIL played a partial mediating role in the path of OEL to SIL.The standardized effect size of the direct effect of OEL on SIL was 0.402,95%CI was[0.3 16,0.495],accounting for 67.4%of the total effect;the standardized effect size of the intermediate effect was 0.194,95%CI was[0.109,0.305],accounting for 32.6%of the total effect;In the training,CR10 plays a regulating role in the first half of this path.When CR10 is at a high level(M+1SD),OEL’s positive prediction ability for OIL decreases.It is further suggested that CR10 should be reduced through some intervention measures in difficult training,so as to promote the internal load of athletes to reach a higher level of training requirements.(5)After intervention with different load regulation methods,the difference of CR10 among different interventions reached a significant level.Multiple comparisons after the intervention showed that CR10 in the exchange group was significantly lower than that in the IMW group and the control group(P<0.001),while CR10 in the IMW group was significantly lower than that in the control group(P=0.022).After controlling the pre-measured baseline values,the bilateral PFC brain oxygenation indexes were significantly different between different interventions.After multiple comparisons,the results showed that there was no difference between the IMW group and HbO2 group,but the HbO2 of the two groups was lower than that of the control group(P=0.019;P=0.035).The level of muscle oxygen saturation measured after IMW group was significantly higher than the pre-measured value(P<0.05),while the difference between the two tests in the control group was not significant,suggesting that IMW intervention can alleviate the"steal"phenomenon of respiratory muscle,reduce the decline of muscle oxygen level during training,and reduce the discomfort of lung and exercise muscles.Conclusion:(1)The Chinese version of sRPE scale has good reliability and validity,and can be used as a tool to monitor the training load of rowing training.Its application in the training of ergometer with low intensity,short interval and slightly longer time is relatively high in timeliness.(2)Both training methods and sampling nodes can affect the subjective load.With fixed training means,CR10 at three sampling nodes showed an increasing trend,indicating that the subjective method can identify accumulated fatigue.The sRPETL of post 10 node is significantly higher than that of post30 and post60 node,and the latter node has the best fitting degree for multiple linear regression,suggesting that 30-60min after training may be a better sampling node.(3)In continuous training,the objective and subjective indexes showed a linear relationship,while in incremental intensity training and interval training,there was a nonlinear relationship between the indexes.After multiple linear regression analysis including objective indicators,there is still a large variance that cannot be explained,suggesting that subjective methods and objective methods are not completely substituted,and subjective indicators may have unique incremental validity in load monitoring,which can supplement the shortcomings of objective methods.(4)The modulated intermediate model built between OEL,OIL and SIL was effectively fitted,and the dose-effect relationship between subjective and objective load was determined.The results further suggested that the athletes’ heart rate could meet the established requirements by reducing CR10 through item change,respiratory muscle training and other measures.IMW and alternative training intervention can effectively reduce subjective load and improve training quality through peripheral and central action paths respectively.