Concurrent Validity and Within-Session Reliability of a Wireless Surface Electromyography Device (MR EMG) Compared to a Criterion Measure During the Leg Extension Exercise.
Ghigiarelli J, Gonzalez A, et al. • Sensors (Basel, Switzerland) • 2026
MR EMG demonstrated excellent within-session reliability but muscle-dependent concurrent validity, with good VMO agreement and moderate VL agreement with the criterion system.
Key Findings
Results
Both the MR EMG and criterion system (Delsys Trigno) demonstrated excellent within-session reliability across all conditions.
ICC values ranged from 0.92 to 0.97 for within-session reliability across both devices.
Coefficients of variation ranged from 7.8% to 11.3% across all conditions.
Reliability was assessed during dynamic leg extension at 30%, 60%, and 90% of predicted 1RM.
Both the vastus lateralis (VL) and vastus medialis obliquus (VMO) were included in reliability assessments.
Thirty-one resistance-trained adults performed four repetitions at each loading intensity.
Results
Concurrent validity between the MR EMG and the criterion system was good for the VMO across all loading intensities.
ICC values for VMO agreement ranged from 0.82 to 0.85 across all loads.
Pearson correlation coefficients for VMO ranged from r = 0.83 to 0.86.
Bland-Altman analyses revealed systematic positive bias for VMO conditions.
Proportional bias was absent for the VMO, indicating consistent agreement across the range of EMG amplitudes.
Results
Concurrent validity between the MR EMG and the criterion system was only moderate for the vastus lateralis across all loading intensities.
ICC values for VL agreement ranged from 0.53 to 0.57 across all loads.
Pearson correlation coefficients for VL ranged from r = 0.54 to 0.64.
Bland-Altman analyses revealed systematic positive bias for VL conditions.
Proportional bias was present for the VL, indicating that agreement varied as a function of EMG amplitude magnitude.
Results
Bland-Altman analyses revealed systematic positive bias across all conditions for both muscles.
Positive bias indicates that the MR EMG consistently recorded higher or lower amplitude values relative to the Delsys Trigno criterion system in a directional manner.
Proportional bias was present for VL but absent for VMO, indicating muscle-location-specific sources of measurement discrepancy.
EMG amplitude was normalized to maximal voluntary isometric contraction (MVIC) for comparison.
Assessments were conducted across three loading intensities: 30%, 60%, and 90% of predicted 1RM.
Methods
The study employed a concurrent measurement design using both wireless EMG systems simultaneously on resistance-trained adults.
Thirty-one resistance-trained adults served as participants.
Both the MR EMG and Delsys Trigno systems recorded simultaneously from the VL and VMO during plate-loaded leg extension.
Root mean square (RMS) amplitude was the primary EMG metric analyzed.
Statistical analyses included ICC, CV%, standard error of measurement, Pearson correlations, and Bland-Altman analyses.
Loading intensities of 30%, 60%, and 90% of predicted one-repetition maximum were used to assess validity across a range of muscle activation levels.
What This Means
This research suggests that a wireless muscle activity sensor called the MR EMG is highly consistent when used repeatedly within the same testing session, matching the reliability of a well-established research-grade system (the Delsys Trigno). The study tested 31 adults who regularly resistance train, measuring electrical activity in two thigh muscles — the vastus lateralis (outer thigh) and vastus medialis obliquus (inner thigh) — during leg extension exercises at light, moderate, and heavy loads. Both devices produced very stable, repeatable readings across all conditions, with reliability scores in the 'excellent' range.
However, when comparing the MR EMG's readings directly against the criterion system to assess accuracy, the results depended on which muscle was being measured. For the inner thigh muscle (VMO), agreement between the two devices was good, meaning the MR EMG and the gold-standard device gave similar readings. For the outer thigh muscle (VL), agreement was only moderate, and the degree of disagreement grew larger as muscle activity increased — a pattern called proportional bias. This suggests the placement location or anatomical characteristics of the VL may introduce greater variability when using the MR EMG.
This research suggests that the MR EMG device may be a reliable tool for tracking changes in muscle activity over time within a session, but users should be cautious when comparing its absolute readings to those from criterion-standard laboratory systems, particularly for the vastus lateralis. Practitioners using this device for training or rehabilitation monitoring should be aware that the device's accuracy may vary depending on the specific muscle being assessed.
Check Your Own Numbers
Upload your bloodwork. We'll cross-reference your results against this study and 4,700 others.
Ghigiarelli J, Gonzalez A, Valdez G, Johannessen T, Calovini A, Hackney J. (2026). Concurrent Validity and Within-Session Reliability of a Wireless Surface Electromyography Device (MR EMG) Compared to a Criterion Measure During the Leg Extension Exercise.. Sensors (Basel, Switzerland). https://doi.org/10.3390/s26165165