Table 1 Extracted features from the acquisitioned EMG database

Amplitude

\(A \left(\mu V\right)=mean(anv\left(EMG\right))\)

Extracted from the envelope or absolute value of the signal. The average value of all maxima values in the EMG signal

It represents the strength of muscle contraction at a given moment, as higher amplitude indicates stronger muscle activity

[26]

Variance

\(\mu =\frac{1}{N}{\sum }_{i=1}^{N}{({x}_{i}-\underline{x})}^{2}\)

A measure of the extent of which the EMG signal deviates from its mean value. It quantifies the variability of the signal

A higher variance indicates greater fluctuations in muscle activity

[27]

Root mean square (RMS)

\(RMS=\sqrt{\frac{1}{N}{\sum }_{i=1}^{N}{{x}_{i}}^{2}}\)

Reflects the energy content of the signal. RMS is often more robust to noise than raw amplitude measures

It is used to assess the overall muscle activation level

[28]

Kurtosis

\(k=\frac{N{\sum }_{i=1}^{N}{({x}_{i}-\underline{x})}^{4}}{{({\sum }_{i=1}^{N}{({x}_{i}-\underline{x})}^{2})}^{2}}\)

A statistical measure to describe the distribution “tailedness” of the EMG signal. High kurtosis indicates the presence of more outliers or sharp peaks in the signal

It helps detect abnormal muscle activities or artifacts

[29]

Median frequency

\({\int }_{0}^{{f}_{\text{median}}}P\left(f\right)df=\frac{1}{2}{\int }_{0}^{{f}_{\text{max}}}P\left(f\right)df\)

It is the frequency at which the power spectrum of the EMG signal is divided into two equal halves

Used to assess muscle fatigue, as shifts in median frequency are often associated with muscle fatigue. Median frequency tends to decrease as fatigue increases

[27]

Mean frequency

\({f}_{\text{mean}}=\frac{{\sum }_{i=1}^{N}{f}_{i}P({f}_{i})}{{\sum }_{i=1}^{N}P({f}_{i})}\)

A weighted average of all frequencies in the EMG power spectrum. It represents the central tendency of the signal’s frequency distribution

Changes in mean frequency can indicate muscle fatigue or recruitment patterns

[27, 28]

Power in low band

\(BPL= {\int }_{0}^{50}P\left(f\right)df\)

Measures the total power of the EMG signal within the lower frequency range, typically below 50 Hz

This range is often associated with baseline muscle tone and slow muscle activity. It provides insight into sustained or tonic muscle contractions

[30, 31]

Power in medium band

\(BPM= {\int }_{50}^{150}P\left(f\right)df\)

Measures the total power of the EMG signal within the medium frequency range, over 50 Hz and below 150 Hz

This range captures moderate muscle contractions and is indicative of voluntary muscle activity. It helps differentiate between slow and fast muscle fibers

[32]

Power in high band

\(BPH= {\int }_{150}^{{f}_{\text{max}}}P\left(f\right)df\)

Measures the total power of the EMG signal within the high frequency range, over 150 Hz

Higher frequency components are often associated with fast muscle contractions or muscle fiber recruitment. This measure is used to analyze rapid and intense muscle activities

[33]

Max power

\({P}_{\text{max}}=max(P\left(f\right))\)

The highest value of power found in the EMG signal’s power spectrum, indicating the most dominant muscle activity at a specific frequency

Shifts in this frequency can provide insights into muscle function and fatigue

[34]

Frequency at max power

\(Maxf=argmax(P\left(f\right))\)

Identifies the frequency at which the highest power occurs in the EMG signal’s power spectrum

It highlights the dominant frequency of muscle activity. Shifts in this frequency can provide insights into muscle function and fatigue

[34]

Total power

\(P= {\int }_{0}^{{f}_{\text{max}}}P\left(f\right)df\)

The sum of all power values across the entire frequency spectrum of the EMG signal. It represents the overall energy content of the muscle activity

This metric is useful for assessing the total work done by the muscle during a specific period of time

[34]