The synergy of EMG waveform during bicycle pedaling is related to elemental force vector waveform.


  • Tomoki Kitawaki Kansai Medical University, Department of Mathematics


Background & Purpose:

In recent years, the pedaling force vector can now be accurately measured using pedaling analyzer systems (, Sittard, Netherlands). Using this device, we showed that the pedaling force vector components in the tangential and radial directions can be represented by the sum of two or three elemental waveform components, respectively (Kitawaki et al 2018). Besides, a previous study that analyzed the electromyogram (EMG) signals of the lower limb muscles demonstrated that pedaling is accomplished by combining three similar muscle synergies (Hug et al. 2010).

Therefore, this study aims to clarify the relationship between the elemental components of the force vector and EMG synergies. We performed synergy analysis of the EMG waveform, which was measured simultaneously with the force vector.


Two participants (a former professional and a top-level amateur cyclist) performed pedaling under a variety of conditions (load: 100 W, 200 W, 300 W; cadence: 70 rpm, 90 rpm, 110 rpm; saddle position: back (5 cm), forward (10 cm), up (3 cm), and down (5 cm, 10 cm); pedaling action type: normal, spinning, pulling, and pushing and pulling). Pedaling force vector data was obtained every 15° using a pedaling analyzer system ( Pedaling vector data were expressed as the sum of elemental vectors, as demonstrated in our previous study (Kitawaki et al 2018). The surface EMG was synchronously measured on the dominant leg at eight locations (anterior tibialis: TA, gastrocnemius medialis: GM, soleus: SOL, rectus femoris: RF, vastus medialis: VM, biceps femoris: BF, gluteus maximus (upside: GM1, downside: GM2)). After the EMG waveforms have been rectified and integrated, the unnormalized iEMG waveforms were obtained every 5° using crank position data. A non-negative matrix factorization (NNMF) algorithm was applied to the iEMG waveforms of the pedaling cycles to differentiate muscle synergies. The number of synergies was set to five to accurately express the muscle output exerted according to the variety of pedaling conditions.

Pearson's correlation coefficient between amplitude of the force vector wave and the amplitude of EMG synergy by muscle was calculated for each participant.

Results & Discussion:

The iEMG waveform can be represented by the sum of five synergies, as shown in Figure 1. The synergy waveforms in the figure are normalized for easy understanding, although not normalized in the NNMF calculation. The amplitude of the synergy varies with the pedaling conditions. The analysis of NNMF does not include change in phase, whereas the force vector waveform analysis includes change in phase angle. Moreover, the change in phase angle was not included in the EMG analysis as it was approximately 5°.

Table 1 lists the correlation coefficient between the amplitude of five EMG synergies and three force vector amplitude (A1, A2, A3) of elemental vector waveforms for each participant. A few muscles with less muscular amplitude were removed from the table. The following can be observed from the results:

Subject ID1: In the pushing phase of Synergy 2-3, the magnitude of A1 that related to pedaling power has a positive correlation with most muscular activity. On the other hand, in the recovery phases of Synergy 1 & 5, most muscular activities have negative correlation with the magnitude of A1. Thus, it seems that subject ID1 is performing more integrated pedaling.

Subject ID2: By contrast, in the early pushing phase of Synergy 2-3, certain muscle (RF, VM, GM) activities increase due to the difference in pedaling, whereas some muscular (GC, SOL, BF) activities decrease. In the case of subject ID 2, the combination of various muscles changes and the pedaling action seems to be changing.

These results indicate that the change in the force vector is caused by the difference in pedaling due to the difference in the muscle force activity. However there are only two participants, hence this study is a pilot study. In the future, we will continue to investigate the difference between change in the element waveform and muscle force assessment by increasing the number of participants and we will study the corresponding muscle force activity and pedaling action.


The change in the amplitude of elemental waveform components of the force vector and the amplitude of EMG synergy are interrelated; we clarified that the force vector and the EMG waveform change at the same time due to a variety of pedaling conditions.


Download data is not yet available.



How to Cite

Kitawaki, T. (2019). The synergy of EMG waveform during bicycle pedaling is related to elemental force vector waveform. Journal of Science and Cycling, 8(2), 3-4. Retrieved from