Introduction: The contractile dynamics of peripheral muscles are governed by complex recruitment and relaxation strategies optimized by the central nervous system. These dynamics aim to maximize the efficiency of resulting work and are finely regulated by synergies, intermuscular coordination, and sensory feedback mechanisms. When individuals are affected by injuries, trauma, cognitive impairments, or neurodegenerative diseases, among others, such contractile dynamics are altered and often manifested in the musculature through changes in the frequency content of electromyographic (EMG) signals. During rapid contractions, these changes are challenging to study and detect because the time series comprising the EMG exhibit highly non-stationary processes.
Methods: Here, we have proposed an exploratory analysis of the time–frequency characteristics of EMG signals using three different approaches: spectrograms (SPs), Hilbert transform (HT), and empirical mode decomposition. Specifically, for empirical mode decomposition, we employed the noise-assisted multivariate empirical mode decomposition (NA-MEMD). These methodologies were applied to EMG signals obtained from a Parkinson's disease (PD) lesion model to longitudinally study the muscle function alterations.
Results and Discussion: These approaches allowed for determining and characterizing the contraction phases of the biceps femoris muscle in a free movement protocol. The SP of the EMG revealed changes in frequency content in the initial phase of contraction, depending on the progression of the injury. These initial observations were made under certain limitations of time–frequency resolution. The HT revealed subphases at the onset of contraction with significant differences in the frequency content of the EMG signals obtained across different stages of injury progression. Finally, the NA-MEMD of the signals revealed intrinsic mode functions primarily affected by anatomical–functional changes in the animal model over time.
Conclusions: This study allowed for extracting spectral information contained in non-stationary segments of the EMG, thus characterizing changes in contractile dynamics caused by progressive functional alterations in the animal model of PD.
