Development and Assessment of a Method to Estimate the Value of a Maximum Voluntary Isometric Contraction Electromyogram from Submaximal Electromyographic Data

in Journal of Applied Biomechanics
View More View Less
  • 1 Iowa State University
  • | 2 Mayo Clinic

The electromyographic (EMG) normalization (often to maximum voluntary isometric contraction [MVIC]) is used to control for interparticipant and day-to-day variations. Repeated MVIC exertions may be inadvisable from participants’ safety perspective. This study developed a technique to predict the MVIC EMG from submaximal isometric voluntary contraction EMG. On day 1, 10 participants executed moment exertions of 100%, 60%, 40%, and 20% of the maximum (biceps brachii, rectus femoris, neck flexors, and neck extensors) as the EMG data were collected. On day 2, the participants replicated the joint moment values from day 1 (60%, 40%, and 20%) and also performed MVIC exertions. Using the ratios between the MVIC EMGs and submaximal isometric voluntary contraction EMG data values established on day 1, and the day 2 submaximal isometric voluntary contraction EMG data values, the day 2 MVIC EMGs were predicted. The average absolute percentage error between the predicted and actual MVIC EMG values for day 2 were calculated: biceps brachii, 45%; rectus femoris, 27%; right and left neck flexors, 27% and 33%, respectively; and right and left neck extensors, both 29%. There will be a trade-off between the required accuracy of the MVIC EMG and the risk of injury due to exerting actual MVIC. Thus, using the developed predictive technique may depend on the study circumstances.

The characteristics of an electromyographic (EMG) signal captured using surface electromyography are influenced by a number of extrinsic and intrinsic factors. De Luca1 discussed several factors that can influence the collected EMG signal, including electrode positioning, configuration and orientation, blood flow in the muscle, and the amount of tissue between the electrode and muscle of interest. Other researchers have further identified perspiration and temperature,2 as well as the cross-talk of surrounding muscles,3 as factors that influence the EMG signal. These factors can vary among participants in an experiment and can also vary from day-to-day in a single participant, making it difficult to accurately compare the experimental EMG signals across days and individuals. To overcome the challenges associated with the day-to-day and person-to-person EMG signal variability, researchers often collect EMG data during reference contraction(s) and then normalize the experimental EMG data relative to the EMG values generated during these reference contractions.

Different types of EMG reference contractions have been employed to create the “denominator” reference value of the normalization process. Some of these reference values include the EMG from the maximum voluntary contraction (MVC) (either isometric or nonisometric contractions),411 the EMG from a submaximal voluntary contraction (SVC) (either isometric or nonisometric contractions),1,12,13 or a value of the EMG derived from the experimental task that is evaluated.5,6,1317 There is no consensus on the best method for EMG normalization that would be effective in all research studies.1821 An MVC during an isometric contraction “maximum voluntary isometric contraction” (MVIC) is the most common reference value to normalize the EMG data22 and is widely used for normalization in EMG-assisted biomechanical models.23,24 While the MVIC technique is widely used to normalize EMG data, it has faced criticism from researchers,6,13,25 including the verity of the maximum contraction, increased injury risk, and potential fatigue effects.

It is often not easy to obtain a true and reliable MVIC for EMG normalization,23 as it can be impacted by participants’ motivation and sincerity,26,27 as well as participants’ level of pain/discomfort.28 Furthermore, utilization of the MVIC method can be restricted or impossible for individuals with musculoskeletal disorders,29 and the MVIC method can also be uncomfortable or cause injuries in healthy participants in regions vulnerable to injury.23,30,31 One example is collecting MVIC values of the neck muscles. Maximal force exerted by the neck musculature can cause injury, particularly when the neck is in a nonneutral position32 or when simply executing MVIC exertions that are not typical for the neck region. To further add to this challenge, the neck MVIC technique can be affected by the pain or the fear of pain in participants with a history of neck pain.33 These challenges could also be troublesome in other regions with a history of injury and susceptibility to reoccurrence of the injury. Furthermore, in some situations, the participant is able to exert MVIC, but the maximum exertion may not be advisable on the day of data collection (eg, surgeons before performing a surgical procedure). The development of a method to generate estimates of day-specific MVIC EMG values with a minimal number of true MVIC exertions would achieve the goals of normalization of experimental EMG data while improving the level of safety of the participants, especially in a typical multiday experiment.

Marras and Davis26 proposed a normalization method for the lumbar musculature that did not require that participants perform an MVIC exertion to generate the MVIC EMG. Their method involved developing regression equations to predict maximum trunk contraction moments based solely on anthropometric measurements. Then, submaximal and maximal EMG data from a new set of participants were used to develop a linear regression model to determine the EMG–moment relationship for each of the lumbar muscles under study. This relationship was extrapolated to the previously predicted maximum contraction moment to get an EMG normalization reference point. This method assumed there was a relationship between anthropometry and trunk moments. The authors reported a large portion of unexplained variability in this model, which indicates that the anthropometry-trunk moment model has limitations.

The purpose of the current study was to develop and assess a regression modeling technique that seeks to predict MVIC EMG based on a previously (different day) established relationship between that muscle’s MVIC EMG and an SVC EMG value. Following the example of Marras and Davis,26 we propose a linear regression model between MVIC EMG and SVC EMG values. Our regression equations will be based on the participant's own submaximal and maximal exertions. It is hypothesized that the relationship between SVC EMG and MVIC EMG on a reference day can be used effectively to estimate MVIC EMG on another day wherein only SVC exertions are performed. If found to be suitably accurate, this technique could be employed in studies that require multiday participation and would result in an increase in participant safety (as MVIC EMG is collected only on the first participation day) while still reaping the benefits of the MVIC EMG normalization technique.

Method

Participants

Ten healthy adults (9 males and 1 female) were recruited for this study. The mean and SD of their age and selected anthropometric measurements were as follows: age 34 (9) years, stature 175.3 (9.2) cm, and body weight 77.5 (23.2) kg. All participants self-identified as right-handed (also considered their dominant side).

Data Collection Instrumentation

The experimental apparatus consisted of 2 main instruments. A Kin Com Isokinetic Dynamometer (125E; Kin Com Isokinetic Physical Therapy, Chattanooga, TN) was used to provide isometric resistance while measuring moments generated by the muscles during MVICs and SVCs. The posture of the joint under study during exertions was standardized and fixed using the dynamometer. Surface electromyography (EMG) was employed to collect the desired muscle’s activities using the DELSYS Trigno Wireless Biofeedback System (Delsys Inc, Natick, MA). Each wireless sensor was 27 × 37 × 13 mm and weighed 14 g. The surface EMG data were collected at a frequency of 2148 Hz. The dynamometer provided visual feedback so that the participant could observe the real-time generated moment (presented as a number on the video monitor) and match the joint moment requirements for a given SVC exertion trial.

Experimental Procedure

Each participant participated 2 times (different days) with at least a 48-hour interval between the 2 data collections to control for fatigue and carryover effects. The basic demographic and anthropometric data were collected on the first day of participation. On each day of data collection, the experiment was explained to the participant, and an informed consent was obtained. The participants were then guided through a series of nonstrenuous warm-up/stretch exercises that focused on the neck, elbow, and knee regions (flexion/extension, rolling and lateral motions of the neck, standing elbow flexion/extension, and standing hamstring curl). Then, the participant’s skin at the location of the EMG electrodes was cleaned using alcohol. One surface EMG electrode was attached to the biceps brachii on the participant’s dominant side, and one electrode was attached to their rectus femoris on the same side. Surface Electromyography for Non-Invasive Assessment of Muscles (SENIAM) recommendations for sensor locations34 were followed for biceps brachii and rectus femoris. Four EMG electrodes were attached to the participant’s neck at right and left anterior locations and right and left posterior locations at the C4/C5 level of the cervical spine. The neck sensors were bilaterally symmetrical and inferiorly–superiorly oriented. The offset from the midline was 3.0 to 3.5 cm for the anterior sensors and 2.0 to 2.5 cm for the posterior sensors, depending on the participant's neck width.

On each day of data collection, the MVIC and SVC exertions performed were (1) elbow flexion (biceps brachii electrode), (2) knee extension (rectus femoris electrode), (3) neck flexion (anterior neck electrodes), and (4) neck extension (posterior neck electrodes). The biceps brachii and rectus femoris were included to explore the generalizability of the technique developed in this study. The experimental procedure followed for each of these muscle groups was the same, so while the full procedure is described here for the biceps brachii, all described experimental procedures were likewise followed for the other 3 muscles/muscle groups. The only difference was the arrangement of the dynamometer and the participant’s position at the dynamometer (Figures 1 and 2). The detailed description of the dynamometer setup and participant’s position for each muscle/muscle group is available online (see Supplementary Material 1 [available online]). For the biceps brachii, the participant was asked to flex their elbow, exerting their MVIC force in the biceps brachii for 2 seconds. The participants were asked to relax their muscle and then get to their maximum exertion in a quasi-linear ramp increase in around 2 seconds overall and hold this maximum torque for 2 seconds as steadily as possible. The participants performed this exertion 2 times with a 1-minute rest between the 2 MVIC exertions. Verbal encouragement was used during all MVIC data collections. The momentary peak generated moment was observed and recorded by the experimenter using the visual feedback of the dynamometer, and the EMG electrodes collected the activity of the biceps brachii muscle. The greatest value of the peak generated moment during the 2 trials was selected as the maximum moment for the biceps brachii. The values of 20%, 40%, and 60% of this moment value were calculated. After a 2-minute rest interval from the completion of the second MVIC exertion, the participants were asked to use the video feedback system to generate 20%, 40%, and 60% of this MVIC moment value with a 1-minute rest between the exertions. The order of presentation of the 3 levels of %SVC moment was randomized independently for each specific participant, muscle, and day of data collection. The participants were asked to exert the corresponding torques for the SVC exertions steadily with an acceptable error of about ±5%. They performed each SVC exertion only once unless the acceptable error was violated. In such cases, the participant was given a 1-minute rest, and then the corresponding SVC exertion was repeated (only 3 occasions [different participants] throughout the study). As they performed these exertions, the EMG data were collected and noted as 20%, 40%, and 60% of biceps brachii MVIC EMG. In each of these exertions, the participant used the video feedback to hold the required moment for 2 seconds, and the muscle activities were collected at this steady state. The MVIC and SVCs of rectus femoris and neck flexor and extensor muscles were recorded following similar procedures. At the completion of the experimental trials, the electrodes were removed, and the participant was led through a 5-minute cooldown procedure; then, they were free to go.

Figure 1
Figure 1

—Apparatus and participant position for (A) biceps brachii and (B) rectus femoris exertions.

Citation: Journal of Applied Biomechanics 38, 2; 10.1123/jab.2021-0229

Figure 2
Figure 2

—Apparatus and participant position for (A) neck flexion and (B) neck extension exertions.

Citation: Journal of Applied Biomechanics 38, 2; 10.1123/jab.2021-0229

On the second day of the data collection, the exact same procedure was repeated except that the SVC moments generated were those that were used on day 1 (those that were calculated based on the MVIC moment of the first day of participation). For example, the SVCs of the biceps brachii on the second day were 20%, 40%, and 60% of the MVIC moment value of biceps brachii on the first day of data collection. New MVIC exertions were also collected per muscle for day 2, independently, following a similar procedure used in day 1. Identical EMG sensor placement procedures were followed on both days to ensure the consistency of the electrode location across the 2 days of the data collection. It was also ensured that on day 2 the moment arm for exerting moments on the dynamometer were the same as day 1 for each muscle and participant.

Data Processing

All EMG data were demeaned and filtered using a fourth-order bandpass Butterworth filter (high pass = 10 Hz and low pass = 400 Hz) and a second-order band-stop Butterworth filter (60 Hz and its closest harmonic alias [120 Hz]) (see Supplementary Material 2 [available online]). Then, the EMG data in the time domain were rectified. The rectified MVIC EMG data were smoothed using a 10-Hz low-pass Butterworth filter of a fourth order to create a linear envelope.35 Moving average filter with a 500-millisecond window size was used on the 2-second MVIC EMG data, and the maximum of these processed MVIC EMG data was found across the 2 MVIC trials performed.36,37 The values of the SVC EMGs were calculated for each muscle and day of data collection following a similar procedure except that the mean values of the EMG data over the 2 seconds of SVC exertions were used instead of the 500-millisecond moving average filter. This was because, during the SVC exertions, the participant was exerting an acceptably steady torque (error of ±5%), while during the MVIC exertions, confirming a steady torque is not easily achievable due to the intense nature of the maximum exertions. The mean of the 2-second SVC EMG data was calculated for 20%, 40%, and 60% SVC EMGs and was denoted as SVC-20%, SVC-40%, and SVC-60%.

Data Analysis

The JMP Pro software package (version 15; SAS, Cary, NC) was used to perform a sequential data analysis, including 3 steps.

Step 1. The Best Predictor

To establish which %SVC value was the best predictor of the value of the MVIC on a given day, multiple linear regression technique was employed. All 7 possible models were evaluated, which included predicting MVIC based on (1) SVC-20%; (2) SVC-40%; (3) SVC-60%; (4) SVC-20% and SVC-40%; (5) SVC-20% and SVC-60%; (6) SVC-40% and SVC-60%; and (7) SVC-20%, SVC-40%, and SVC-60%. Both penalized likelihood criteria of the Akaike information criterion (AIC) and Bayesian information criterion (BIC) were employed as criterion to select the best SVC% EMG for each muscle.

Step 2. Individual Multipliers and Predicted MVIC EMG

Once the best SVC% EMG predictor of MVIC EMG of the muscle was found, the multiplier that related the SVC% to the MVIC was found (per muscle and participant) for the day 1 data. This multiplier was then applied to the appropriate SVC% value for the day 2 data to predict the day 2 MVIC per muscle and participant. Finally, this predicted MVIC EMG value of day 2 was then compared with the actual MVIC EMG of day 2, and the absolute percentage error was calculated for the muscle.

Step 3. Between-Day Reliability Analyses

Finally, the reliability of the repeated peak generated moments and the MVIC EMG and SVC EMG values between days (days 1 and 2) was evaluated per muscle. Then, a paired t test or nonparametric Wilcoxon signed-rank test (if normality was violated using the Shapiro–Wilk test) was used to evaluate if there was a significant difference between days 1 and 2 for each desired variable (significance level of α = .05).

Results

Step 1. The Best Predictor

The results of the multiple linear regression analyses revealed that the consistently effective model for predicting MVIC EMG for the biceps brachii, rectus femoris, and right and left neck flexor muscles utilized the SVC-60% values. These findings were consistent using both AIC and BIC criterion. Out of the 28 models (7 models × 4 muscles), there was only one model (left neck flexor) where combining SVC-20% and SVC-60% resulted in smaller AIC (0.9% decrease) and BIC (0.6% decrease) criterion compared with the model that only included SVC-60% (ranked second). For the right and left neck extensor muscles, the model that only included SVC-20% values led to the best regression based on both the AIC and BIC criterion. Thus, for the sake of simplicity and easier usability, SVC-60% values were chosen for all studied muscles in calculating individual multipliers in the next step. For right and left extensor muscles, the analysis was repeated for SVC-20% values, and the results were compared with the model that employed SVC-60%.

Step 2. Individual Multipliers and Predicted MVIC EMG

The mean (SD) of the 10 individual multipliers for the 6 studied muscles was calculated (Table 1). The average absolute percentage error ranged from 45% (biceps brachii) to 26% (left neck extensor) (Table 2). Figures 3 and 4 compare the predicted and actual MVIC EMG of day 2 for right and left neck extensor muscles using SVC-20% values and SVC-60% values. Ideally, all the data points in these figures would be on the line y = x. Furthermore, to put the average absolute percentage errors of this study in context, these values were compared with the average differences between the 2 muscle-specific MVIC EMG from the exertions performed on the same day (Table 3).

Table 1

The Mean (SD) of the 10 Individual Multipliers That Were Used to Predict MVIC EMG Based on SVC-60%

MuscleMean (SD) of the multipliers
Biceps brachii2.77 (0.98)
Rectus femoris2.49 (0.78)
Right neck flexor2.02 (0.58)
Left neck flexor2.40 (0.62)
Right neck extensor3.41 (1.90) [12.19 (5.19)]
Left neck extensor2.18 (0.58) [5.88 (1.59)]

Abbreviations: EMG, electromyographic; MVIC, maximum voluntary isometric contraction; SVC, submaximal voluntary contraction. Note: The values in the square brackets for right and left neck extensor muscles allocate to the model based on SVC-20%.

Table 2

The Mean (SD) of Absolute Percentage Error for the Studied Muscles Based on SVC-60% Values

MuscleAbsolute percentage error
Biceps brachii45 (41.8)
Rectus femoris27 (13.2)
Right neck flexor27 (13.4)
Left neck flexor33 (20.8)
Right neck extensor29 (24.7) [30 (32.8)]
Left neck extensor29 (25.4) [26 (16.1)]

Abbreviation: SVC, submaximal voluntary contraction. Note: The values in the square brackets for right and left neck extensor muscles allocate to the model based on SVC-20%.

Figure 3
Figure 3

—The predicted and actual MVIC EMG of day 2 for right neck extensor muscle. EMG indicates electromyography; MVIC, maximum voluntary isometric contraction.

Citation: Journal of Applied Biomechanics 38, 2; 10.1123/jab.2021-0229

Figure 4
Figure 4

—The predicted and actual MVIC EMG of day 2 for left neck extensor muscle. EMG indicates electromyography; MVIC, maximum voluntary isometric contraction.

Citation: Journal of Applied Biomechanics 38, 2; 10.1123/jab.2021-0229

Table 3

The Mean (SD) of Absolute Percentage Difference Between the 2 MVIC EMG Exertions Performed on the Same Day

MuscleAbsolute percentage difference
Biceps brachii20 (11.8)
Rectus femoris14 (12.1)
Right neck flexor13 (8.8)
Left neck flexor15 (12.5)
Right neck extensor20 (14.9)
Left neck extensor13 (10.0)

Abbreviations: EMG, electromyographic; MVIC, maximum voluntary isometric contraction.

Step 3. Between-Day Reliability Analyses

The mean (SD) of the SVC and MVIC EMG values in millivolt and the peak generated moments in newton meter during days 1 and 2 were calculated for each muscle (Table 4). The statistical analysis did not show a significant difference between days 1 and 2 for any of these variables except for the peak generated moment of biceps brachii (P = .039). However, this increase was not reflected when comparing the corresponding average MVIC EMG values between days 1 and 2.

Table 4

The Mean (SD) of the SVC and MVIC EMG Values in Millivolt and the Peak Exerted Moment in Newton-Meter for Days 1 and 2

MuscleSVC-20%SVC-40%SVC-60%MVICPeak moment
Biceps brachii
 Day 10.051 (0.026)0.139 (0.096)0.221 (0.102)0.585 (0.281)68.0* (17.3)
 Day 20.044 (0.032)0.114 (0.073)0.188 (0.136)0.432 (0.318)73.7* (19.8)
Rectus femoris
 Day 10.015 (0.006)0.038 (0.052)0.048 (0.033)0.111 (0.055)193.4 (76.5)
 Day 20.012 (0.008)0.029 (0.013)0.048 (0.033)0.122 (0.067)200.3 (74.3)
Right neck flexor
 Day 10.024 (0.015)0.061 (0.041)0.105 (0.064)0.201 (0.122)26.2 (11.3)
 Day 20.028 (0.017)0.063 (0.035)0.118 (0.067)0.220 (0.120)24.4 (11.8)
Left neck flexor
 Day 10.023 (0.015)0.053 (0.031)0.108 (0.075)0.269 (0.217)
 Day 20.029 (0.017)0.065 (0.042)0.120 (0.073)0.224 (0.123)
Right neck extensor
 Day 10.006 (0.003)0.014 (0.013)0.026 (0.024)0.071 (0.045)33.3 (12.2)
 Day 20.006 (0.003)0.012 (0.009)0.023 (0.023)0.084 (0.040)36.3 (13.4)
Left neck extensor
 Day 10.008 (0.004)0.015 (0.011)0.025 (0.018)0.050 (0.029)
 Day 20.008 (0.004)0.013 (0.008)0.022 (0.016)0.060 (0.041)

Abbreviations: EMG, electromyographic; MVIC, maximum voluntary isometric contraction; SVC, submaximal voluntary contraction. Note: Since the neck flexion and neck extension peak moment exertions were bilateral, the peak moments are only shown once.

*Statistically significant difference between days 1 and 2 (P < .05). One value for both right and left sides.

Discussion

The main purpose of this study was to develop and assess a method to predict the MVIC EMG of the muscle based on a system relating a muscle’s SVC EMG to its MVIC EMG on a reference day. The motivation for developing this model was to decrease the risk of musculoskeletal injury that might result from multiple MVIC exertions across multiple days of experimental participation in vulnerable regions (eg, neck) as well as joints with a history of injury. When considering specifically the results of the neck muscles, the average absolute percentage errors were more focused, in the range of 26% to 33%. These data would indicate that there is a trade-off between the required accuracy of the MVIC EMG and the risk of fatigue and injury due to exerting actual MVIC. Table 3 shows that the average absolute percentage difference between the 2 MVIC EMG exertions performed on the same day were between 13% and 20%, indicating that there is a natural variability in these MVIC EMG values.

Comparing the models for predicting MVIC EMG values (Table 2), it is inferred that SVC-60% is the consistently acceptable predictor of MVIC EMG values for all studied muscles. It is worth noting that SVC-80% may have resulted in better predictions of MVIC EMG. However, our pilot experiments revealed that exerting SVC-80% and holding it for 2 seconds is not easily achievable. Such exertions are very uncomfortable and may be prone to injury and discomfort. The proposed method for estimating MVIC EMG is intended to reduce the number of risky and uncomfortable exertions; thus, SVC-80% exertions were not included in this study.

Using MVIC EMG data to normalize task EMG data is a well-established technique.22 The principal benefit of normalizing to EMG obtained during an MVIC exertion is that it provides a clear standard that has physiological meaning relative to a participant’s capacity/capability. For example, Sjøgaard et al38 explored the relationship between muscle blood flow and the magnitude of the muscle exertion, and they used %MVIC to establish the level of muscle exertion. Other methods have been introduced and studied, such as normalizing with respect to the EMG collected during an SVC exertion1,12,13 or normalizing task EMG relative to some EMG value that occurred during the task—normalized to the greatest value observed during the experimental task, for example. These methods do not require that the participants perform MVIC exertion. It is recognized, however, that the interpretation of the resulting normalized EMG is a bit more limited. The method developed in the current study is based on actual MVIC EMG and would be suitable in studies wherein the participant is able to exert muscle MVIC, but they are required to attend on several days, or they are recruited for different studies. In these studies, muscle MVIC EMG could be collected once and be used to predict muscle MVIC EMG on other days.

It is clear that the accuracy and consistency of the magnitude of EMG data collected during MVIC exertions rely on the participant's motivation and willingness to provide true MVICs repeatedly,26,27 and therefore, limiting the number of MVIC exertions performed may limit the error that this might introduce. For example, in the current study, the experimental procedure was explained to the participants, and verbal encouragement was used during MVIC exertions. However, a simple analysis showed that the absolute percentage difference between the 2 MVIC EMG exertions performed on the same day (averaged across 10 participants) was considerable (between 13% and 20%), indicating that there is a natural variability in these MVIC EMG values. Comparing this level of typical MVIC–MVIC variability to the average absolute percentage errors shown in our study (between 27% and 33% for neck flexor/extensor muscles, 27% for rectus femoris, and 45% for biceps brachii; all SVC-60% models) would indicate that the predictive method developed, while not perfect in its predictions, is reasonably accurate, considering the natural variability in MVIC exertions. The between-day reliability analysis revealed that the average peak generated moment of biceps brachii increased significantly from day 1 to day 2, while the average MVIC EMG values for this muscle decreased (not significant) from day 1 to day 2 (Table 4). In addition, the average absolute percentage error for biceps brachii (45%) was abnormally greater than the other studied muscles (27%–33%). This could be related to the experimental setup when exerting the peak moments on the dynamometer (see Supplementary Material 1 [available online]). For the rectus femoris, neck flexion, and neck extension exertions, the participants were sitting with their trunk fixed to the dynamometer seating system (Figures 1B and 2). It could have limited their exertions strictly to the desired muscle/muscle group. However, the standing posture during the biceps brachii exertions (Figure 1A) may have led to employing additional muscles (eg, leaning trunk backward slightly) despite all the instructions and the experimenter’s supervision.

There are a few limitations in this study that require attention before generalizing its results. Ten participants (1 female) provided a limited and imbalanced sample size. It should be mentioned that the results for the female participant were found to be consistent with that of the 9 male participants without revealing any specific trend or behavior. More studies with balanced and larger sample sizes will enhance the generalizability of our findings, especially regarding the potential effects of gender. Only linear models were employed in this study to predict MVIC EMG based on SVC EMG. This decision was made based on the limited number of data points and to be consistent with the previous similar study26 that assumed a linear relationship between EMG values and exerted moments. Nonlinear regression models with larger data sets and more levels of SVC EMG could potentially enhance the accuracy and reliability of predicted MVIC EMG values.

The main purpose of this study was to develop and assess a method to predict the MVIC EMG of the muscle based on a system relating a muscle’s SVC EMG to its MVIC EMG on a reference day. We suggest that, in each study, individual multipliers should be calculated based on SVC-60% EMG values (per participant and muscle) to predict MVIC EMG values. Our results implied that there will be a trade-off between the required accuracy of the MVIC EMG and the risk of fatigue and injury due to exerting actual MVIC. Acceptability of the developed predictive technique depends on the study circumstances. For example, in a study that requires neck MVIC exertions on 2 different days with a limited number of participants while a high accuracy in the EMG analysis is required, using MVIC EMG for normalization of the EMG data may be a better option compared with our predictive model. On the other hand, in another study, where participants are required to attend several times, exerting neck MVIC EMG on each day of participation could question the feasibility of the study (eg, risk of injury and IRB approval concerns). In addition, in a study that investigates the muscle activity of surgeons during surgical procedures, exerting MVIC is not advisable because even a slight muscle cramp could have serious consequences for the surgeon and/or patient. Our developed predictive technique could be a better alternative in such studies where the participants are needed to exert MVIC only once and not necessarily on the day of data collection. The results of the EMG analysis would be affected by the approximations from our model; however, the findings of the study (as it is feasible to conduct the study using our model) could greatly compensate for these approximations.

Acknowledgments

The authors would like to acknowledge the support of the John Ryder Professorship that assisted in the completion of this work.

References

  • 1.

    De Luca CJ. The use of surface electromyography in biomechanics. J Appl Biomech. 1997;13(2):135163. doi:10.1123/jab.13.2.135

  • 2.

    Winkel J, Jorgensen K. Significance of skin temperature-changes in surface electromyography. Eur J Appl Physiol Occup Physiol. 1991;63(5):345348. PubMed ID: 1773810 doi:10.1007/bf00364460

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 3.

    Koh TJ, Grabiner MD. Evaluation of methods to minimize cross-talk in surface electromyography. J Biomech. 1993;26:151157. PubMed ID: 8505349 doi:10.1016/0021-9290(93)90086-t

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 4.

    Mirka G. The quantification of EMG normalization error. Ergonomics. 1991;34(3):343352. PubMed ID: 1829038 doi:10.1080/00140139108967318

  • 5.

    Knutson LM, Soderberg GL, Ballantyne BT, et al. A study of various normalization procedures for within day electromyographic data. J Electromyogr Kinesiol. 1994;4(1):4759. PubMed ID: 20870546 doi:10.1016/1050-6411(94)90026-4

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 6.

    Burden A, Bartlett R. Normalisation of EMG amplitude: an evaluation and comparison of old and new methods. Med Eng Phys. 1999;21(4):247257. PubMed ID: 10514043 doi:10.1016/s1350-4533(99)00054-5

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 7.

    Sorbie GG, Grace FM, Gu YD, et al. Comparison of thoracic and lumbar erector spinae muscle activation before and after a golf practice session. J Appl Biomech. 2017;33(4):288293. PubMed ID: 28290751 doi:10.1123/jab.2016-0209

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 8.

    Roth R, Donath L, Zahner L, et al. Muscle activation and performance during trunk strength testing in high-level female and male football players. J Appl Biomech. 2016;32(3):241247. PubMed ID: 26671894 doi:10.1123/jab.2014-0303

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 9.

    Yaghoubi M, Esfehani MM, Hosseini HA, et al. Comparative electromyography analysis of the upper extremity between inexperienced and elite water polo players during an overhead shot. J Appl Biomech. 2015;31(2):7987. PubMed ID: 25387072 doi:10.1123/jab.2014-0068

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 10.

    Marchetti PH, Uchida MC. Effects of the pullover exercise on the pectoralis major and latissimus dorsi muscles as evaluated by EMG. J Appl Biomech. 2011;27(4):380384. PubMed ID: 21975179 doi:10.1123/jab.27.4.380

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 11.

    Juker D, McGill S, Kropf P. Quantitative intramuscular myoelectric activity of lumbar portions of psoas and the abdominal wall during cycling. J Appl Biomech. 1998;14(4):428438. doi:10.1123/jab.14.4.428

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 12.

    Yang JF, Winter DA. Electromyography reliability in maximal and submaximal isometric contractions. Arch Phys Med Rehabil. 1983;64(9):417420. PubMed ID: 6615179

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 13.

    Burden A. How should we normalize electromyograms obtained from healthy participants? What we have learned from over 25 years of research. J Electromyogr Kinesiol. 2010;20(6):10231035. PubMed ID: 20702112 doi:10.1016/j.jelekin.2010.07.004

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 14.

    Allison GT, Marshall RN, Singer KP. EMG signal amplitude normalization technique in stretch-shortening cycle movements. J Electromyogr Kinesiol. 1993;3(4):236244. PubMed ID: 20870539 doi:10.1016/1050-6411(93)90013-m

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 15.

    Suydam SM, Manal K, Buchanan TS. The advantages of normalizing electromyography to ballistic rather than isometric or isokinetic tasks. J Appl Biomech. 2017;33(3):189196. PubMed ID: 27918690 doi:10.1123/jab.2016-0146

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 16.

    Monteleone BJ, Ronsky JL, Meeuwisse WH, et al. Lateral hop movement assesses ankle dynamics and muscle activity. J Appl Biomech. 2012;28(2):215221. PubMed ID: 22085998 doi:10.1123/jab.28.2.215

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 17.

    Tillman MD, Criss RM, Brunt D, et al. Landing constraints influence ground reaction forces and lower extremity EMG in female volleyball players. J Appl Biomech. 2004;20(1):3850. doi:10.1123/jab.20.1.38

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 18.

    Perry J, Burnfield JM. Gait Analysis: Normal and Pathological Function. 2nd ed. New Jersey: Slack Inc; 2010:1551.

  • 19.

    Bolgla LA, Uhl TL. Reliability of electromyographic normalization methods for evaluating the hip musculature. J Electromyogr Kinesiol. 2007;17(1):102111. PubMed ID: 16423539 doi:10.1016/j.jelekin.2005.11.007

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 20.

    Albertus-Kajee Y, Tucker R, Derman W, et al. Alternative methods of normalising EMG during cycling. J Electromyogr Kinesiol. 2010;20(6):10361043. PubMed ID: 20696597 doi:10.1016/j.jelekin.2010.07.011

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 21.

    Ball N, Scurr J. Electromyography normalization methods for high-velocity muscle actions: review and recommendations. J Appl Biomech. 2013;29(5):600608. PubMed ID: 23270917 doi:10.1123/jab.29.5.600

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 22.

    Sinclair J, Taylor PJ, Hebron J, et al. The reliability of electromyographic normalization methods for cycling analyses. J Hum Kinet. 2015;46(1):1927. doi:10.1515/hukin-2015-0030

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 23.

    Dufour JS, Marras WS, Knapik GG. An EMG-assisted model calibration technique that does not require MVCs. J Electromyogr Kinesiol. 2013;23(3):608613. PubMed ID: 23415699 doi:10.1016/j.jelekin.2013.01.013

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 24.

    McFarland DC, Brynildsen AG, Saul KR. Sensitivity of neuromechanical predictions to choice of glenohumeral stability modeling approach. J Appl Biomech. 2020;36(4):249258. doi:10.1123/jab.2019-0088

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 25.

    Baggen RJ, van Dieen JH, Verschueren SM, et al. Differences in maximum voluntary excitation between isometric and dynamic contractions are age-dependent. J Appl Biomech. 2019;35(3):196201. PubMed ID: 30860419 doi:10.1123/jab.2018-0215

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 26.

    Marras WS, Davis KG. A non-MVC EMG normalization technique for the trunk musculature: Part 1. Method development. J Electromyogr Kinesiol. 2001;11(1):19. PubMed ID: 11166603 doi:10.1016/s1050-6411(00)00039-0

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 27.

    McNair PJ, Depledge J, Brettkelly M, et al. Verbal encouragement: effects on maximum effort voluntary muscle action. Br J Sports Med. 1996;30(3):243245. PubMed ID: 8889120 doi:10.1136/bjsm.30.3.243

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 28.

    Ettinger L, Weiss J, Shapiro M, et al. Normalization to maximal voluntary contraction is influenced by subacromial pain. J Appl Biomech. 2016;32(5):433440. PubMed ID: 27115101 doi:10.1123/jab.2015-0185

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 29.

    Zellers JA, Parker S, Marmon A, et al. Muscle activation during maximum voluntary contraction and m-wave related in healthy but not in injured conditions: implications when normalizing electromyography. Clin Biomech. 2019;69:104108. doi:10.1016/j.clinbiomech.2019.07.007

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 30.

    Zeh J, Hansson T, Bigos S, et al. Isometric strength testing—recommendations based on a statistical-analysis of the procedure. Spine. 1986;11(1):4346. PubMed ID: 2939569 doi:10.1097/00007632-198601000-00011

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 31.

    Battie MC, Bigos SJ, Fisher LD, et al. Isometric lifting strength as a predictor of industrial back pain reports. Spine. 1989;14(8):851856. doi:10.1097/00007632-198908000-00014

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 32.

    Huelke DF, Nusholtz GS. Cervical-spine biomechanics—a review of the literature. J Orthop Res. 1986;4(2):232245. PubMed ID: 3519910 doi:10.1002/jor.1100040212

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 33.

    Lindstroem R, Graven-Nielsen T, Falla D. Current pain and fear of pain contribute to reduced maximum voluntary contraction of neck muscles in patients with chronic neck pain. Arch Phys Med Rehabil. 2012;93(11):20422048. PubMed ID: 22546536 doi:10.1016/j.apmr.2012.04.014

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 34.

    Hermens HJ, Freriks B, Disselhorst-Klug C. Development of recommendations for SEMG sensors and sensor placement procedures. J Electromyogr Kinesiol. 2000;10(5):361374. PubMed ID: 11018445 doi:10.1016/s1050-6411(00)00027-4

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 35.

    Gillette JC, Stephenson ML. Electromyographic assessment of a shoulder support exoskeleton during on-site job tasks. IISE Trans Occup Ergon Hum Factors. 2019;7(3–4):302310.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 36.

    Dahlqvist C, Enquist H, Löfqvist L, Nordander C. The effect of two types of maximal voluntary contraction and two electrode positions in field recordings of forearm extensor muscle activity during hotel room cleaning. Int J Occup Saf Ergon. 2020;26(3):595602. PubMed ID: 30932748 doi:10.1080/10803548.2019.1599572

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 37.

    Jensen C, Vasseljen O Jr, Westgaard RH. Estimating maximal EMG amplitude for the trapezius muscle: on the optimization of experimental procedure and electrode placement for improved reliability and increased signal amplitude. J Electromyogr Kinesiol. 1996;6(1):5158. PubMed ID: 20719662 doi:10.1016/1050-6411(94)00012-3

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 38.

    Sjøgaard G, Savard G, Juel C. Muscle blood-flow during isometric activity and its relation to muscle fatigue. Eur J Appl Physiol Occup Physiol. 1988;57(3):327335. PubMed ID: 3371342 doi:10.1007/bf00635992

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation

Norasi, Koenig, and Mirka are with the Department of Industrial and Manufacturing Systems Engineering, Iowa State University, Ames, IA, USA. Norasi is also with the Robert D. & Patricia E. Kern Center for the Science of Healthcare Delivery, Mayo Clinic, Rochester, MN, USA.

Norasi (Norasi.Hamid@mayo.edu) is corresponding author.
  • View in gallery

    —Apparatus and participant position for (A) biceps brachii and (B) rectus femoris exertions.

  • View in gallery

    —Apparatus and participant position for (A) neck flexion and (B) neck extension exertions.

  • View in gallery

    —The predicted and actual MVIC EMG of day 2 for right neck extensor muscle. EMG indicates electromyography; MVIC, maximum voluntary isometric contraction.

  • View in gallery

    —The predicted and actual MVIC EMG of day 2 for left neck extensor muscle. EMG indicates electromyography; MVIC, maximum voluntary isometric contraction.

  • 1.

    De Luca CJ. The use of surface electromyography in biomechanics. J Appl Biomech. 1997;13(2):135163. doi:10.1123/jab.13.2.135

  • 2.

    Winkel J, Jorgensen K. Significance of skin temperature-changes in surface electromyography. Eur J Appl Physiol Occup Physiol. 1991;63(5):345348. PubMed ID: 1773810 doi:10.1007/bf00364460

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 3.

    Koh TJ, Grabiner MD. Evaluation of methods to minimize cross-talk in surface electromyography. J Biomech. 1993;26:151157. PubMed ID: 8505349 doi:10.1016/0021-9290(93)90086-t

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 4.

    Mirka G. The quantification of EMG normalization error. Ergonomics. 1991;34(3):343352. PubMed ID: 1829038 doi:10.1080/00140139108967318

  • 5.

    Knutson LM, Soderberg GL, Ballantyne BT, et al. A study of various normalization procedures for within day electromyographic data. J Electromyogr Kinesiol. 1994;4(1):4759. PubMed ID: 20870546 doi:10.1016/1050-6411(94)90026-4

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 6.

    Burden A, Bartlett R. Normalisation of EMG amplitude: an evaluation and comparison of old and new methods. Med Eng Phys. 1999;21(4):247257. PubMed ID: 10514043 doi:10.1016/s1350-4533(99)00054-5

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 7.

    Sorbie GG, Grace FM, Gu YD, et al. Comparison of thoracic and lumbar erector spinae muscle activation before and after a golf practice session. J Appl Biomech. 2017;33(4):288293. PubMed ID: 28290751 doi:10.1123/jab.2016-0209

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 8.

    Roth R, Donath L, Zahner L, et al. Muscle activation and performance during trunk strength testing in high-level female and male football players. J Appl Biomech. 2016;32(3):241247. PubMed ID: 26671894 doi:10.1123/jab.2014-0303

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 9.

    Yaghoubi M, Esfehani MM, Hosseini HA, et al. Comparative electromyography analysis of the upper extremity between inexperienced and elite water polo players during an overhead shot. J Appl Biomech. 2015;31(2):7987. PubMed ID: 25387072 doi:10.1123/jab.2014-0068

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 10.

    Marchetti PH, Uchida MC. Effects of the pullover exercise on the pectoralis major and latissimus dorsi muscles as evaluated by EMG. J Appl Biomech. 2011;27(4):380384. PubMed ID: 21975179 doi:10.1123/jab.27.4.380

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 11.

    Juker D, McGill S, Kropf P. Quantitative intramuscular myoelectric activity of lumbar portions of psoas and the abdominal wall during cycling. J Appl Biomech. 1998;14(4):428438. doi:10.1123/jab.14.4.428

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 12.

    Yang JF, Winter DA. Electromyography reliability in maximal and submaximal isometric contractions. Arch Phys Med Rehabil. 1983;64(9):417420. PubMed ID: 6615179

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 13.

    Burden A. How should we normalize electromyograms obtained from healthy participants? What we have learned from over 25 years of research. J Electromyogr Kinesiol. 2010;20(6):10231035. PubMed ID: 20702112 doi:10.1016/j.jelekin.2010.07.004

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 14.

    Allison GT, Marshall RN, Singer KP. EMG signal amplitude normalization technique in stretch-shortening cycle movements. J Electromyogr Kinesiol. 1993;3(4):236244. PubMed ID: 20870539 doi:10.1016/1050-6411(93)90013-m

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 15.

    Suydam SM, Manal K, Buchanan TS. The advantages of normalizing electromyography to ballistic rather than isometric or isokinetic tasks. J Appl Biomech. 2017;33(3):189196. PubMed ID: 27918690 doi:10.1123/jab.2016-0146

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 16.

    Monteleone BJ, Ronsky JL, Meeuwisse WH, et al. Lateral hop movement assesses ankle dynamics and muscle activity. J Appl Biomech. 2012;28(2):215221. PubMed ID: 22085998 doi:10.1123/jab.28.2.215

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 17.

    Tillman MD, Criss RM, Brunt D, et al. Landing constraints influence ground reaction forces and lower extremity EMG in female volleyball players. J Appl Biomech. 2004;20(1):3850. doi:10.1123/jab.20.1.38

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 18.

    Perry J, Burnfield JM. Gait Analysis: Normal and Pathological Function. 2nd ed. New Jersey: Slack Inc; 2010:1551.

  • 19.

    Bolgla LA, Uhl TL. Reliability of electromyographic normalization methods for evaluating the hip musculature. J Electromyogr Kinesiol. 2007;17(1):102111. PubMed ID: 16423539 doi:10.1016/j.jelekin.2005.11.007

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 20.

    Albertus-Kajee Y, Tucker R, Derman W, et al. Alternative methods of normalising EMG during cycling. J Electromyogr Kinesiol. 2010;20(6):10361043. PubMed ID: 20696597 doi:10.1016/j.jelekin.2010.07.011

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 21.

    Ball N, Scurr J. Electromyography normalization methods for high-velocity muscle actions: review and recommendations. J Appl Biomech. 2013;29(5):600608. PubMed ID: 23270917 doi:10.1123/jab.29.5.600

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 22.

    Sinclair J, Taylor PJ, Hebron J, et al. The reliability of electromyographic normalization methods for cycling analyses. J Hum Kinet. 2015;46(1):1927. doi:10.1515/hukin-2015-0030

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 23.

    Dufour JS, Marras WS, Knapik GG. An EMG-assisted model calibration technique that does not require MVCs. J Electromyogr Kinesiol. 2013;23(3):608613. PubMed ID: 23415699 doi:10.1016/j.jelekin.2013.01.013

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 24.

    McFarland DC, Brynildsen AG, Saul KR. Sensitivity of neuromechanical predictions to choice of glenohumeral stability modeling approach. J Appl Biomech. 2020;36(4):249258. doi:10.1123/jab.2019-0088

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 25.

    Baggen RJ, van Dieen JH, Verschueren SM, et al. Differences in maximum voluntary excitation between isometric and dynamic contractions are age-dependent. J Appl Biomech. 2019;35(3):196201. PubMed ID: 30860419 doi:10.1123/jab.2018-0215

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 26.

    Marras WS, Davis KG. A non-MVC EMG normalization technique for the trunk musculature: Part 1. Method development. J Electromyogr Kinesiol. 2001;11(1):19. PubMed ID: 11166603 doi:10.1016/s1050-6411(00)00039-0

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 27.

    McNair PJ, Depledge J, Brettkelly M, et al. Verbal encouragement: effects on maximum effort voluntary muscle action. Br J Sports Med. 1996;30(3):243245. PubMed ID: 8889120 doi:10.1136/bjsm.30.3.243

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 28.

    Ettinger L, Weiss J, Shapiro M, et al. Normalization to maximal voluntary contraction is influenced by subacromial pain. J Appl Biomech. 2016;32(5):433440. PubMed ID: 27115101 doi:10.1123/jab.2015-0185

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 29.

    Zellers JA, Parker S, Marmon A, et al. Muscle activation during maximum voluntary contraction and m-wave related in healthy but not in injured conditions: implications when normalizing electromyography. Clin Biomech. 2019;69:104108. doi:10.1016/j.clinbiomech.2019.07.007

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 30.

    Zeh J, Hansson T, Bigos S, et al. Isometric strength testing—recommendations based on a statistical-analysis of the procedure. Spine. 1986;11(1):4346. PubMed ID: 2939569 doi:10.1097/00007632-198601000-00011

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 31.

    Battie MC, Bigos SJ, Fisher LD, et al. Isometric lifting strength as a predictor of industrial back pain reports. Spine. 1989;14(8):851856. doi:10.1097/00007632-198908000-00014

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 32.

    Huelke DF, Nusholtz GS. Cervical-spine biomechanics—a review of the literature. J Orthop Res. 1986;4(2):232245. PubMed ID: 3519910 doi:10.1002/jor.1100040212

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 33.

    Lindstroem R, Graven-Nielsen T, Falla D. Current pain and fear of pain contribute to reduced maximum voluntary contraction of neck muscles in patients with chronic neck pain. Arch Phys Med Rehabil. 2012;93(11):20422048. PubMed ID: 22546536 doi:10.1016/j.apmr.2012.04.014

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 34.

    Hermens HJ, Freriks B, Disselhorst-Klug C. Development of recommendations for SEMG sensors and sensor placement procedures. J Electromyogr Kinesiol. 2000;10(5):361374. PubMed ID: 11018445 doi:10.1016/s1050-6411(00)00027-4

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 35.

    Gillette JC, Stephenson ML. Electromyographic assessment of a shoulder support exoskeleton during on-site job tasks. IISE Trans Occup Ergon Hum Factors. 2019;7(3–4):302310.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 36.

    Dahlqvist C, Enquist H, Löfqvist L, Nordander C. The effect of two types of maximal voluntary contraction and two electrode positions in field recordings of forearm extensor muscle activity during hotel room cleaning. Int J Occup Saf Ergon. 2020;26(3):595602. PubMed ID: 30932748 doi:10.1080/10803548.2019.1599572

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 37.

    Jensen C, Vasseljen O Jr, Westgaard RH. Estimating maximal EMG amplitude for the trapezius muscle: on the optimization of experimental procedure and electrode placement for improved reliability and increased signal amplitude. J Electromyogr Kinesiol. 1996;6(1):5158. PubMed ID: 20719662 doi:10.1016/1050-6411(94)00012-3

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 38.

    Sjøgaard G, Savard G, Juel C. Muscle blood-flow during isometric activity and its relation to muscle fatigue. Eur J Appl Physiol Occup Physiol. 1988;57(3):327335. PubMed ID: 3371342 doi:10.1007/bf00635992

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
All Time Past Year Past 30 Days
Abstract Views 10 10 0
Full Text Views 1195 1195 349
PDF Downloads 293 293 43