**3.1. Muscular activity: EMG**

Computational Intelligence in Electromyography Analysis – 76 A Perspective on Current Applications and Future Challenges

values contained in a moving window of 50 points.

length of the phases of the cycle for both feet.

each 2% of the normalized length of jogging cycle.

subjects, the Great Ensemble Average (GEAV) was obtained.

The EMG signals and those from the pressure sensors were recorded digitally at a frequency of 3000 samples per second using an analog-to-digital card CODAS (DataQ Instruments, OH, USA). The records were afterwards selected for further processing based on the signal

In order to determine the intensity of the signal, peak amplitude (peak activity), and time of its appearance, it is necessary to use quantitative methods on the EMG signal such as the LE. The steps to obtain it are: (1) complete (full-wave) rectification of the raw EMG, and (2) obtaining the linear envelop window (LE window) by calculating the average amplitude

Further processing was carried out with the signals of six subjects (the records of four subjects were excluded because their signals were not fully valid). From each record, the activity corresponding to 2 (out of 5) cycles were used (5x2 = 10), for 6 subjects (10 x 6 = 60), so 60 cycles were used in each condition. Since there are 3 different conditions, we analyzed a total of 180 (60 x 3) cycles. Therefore, the database consisted of 180 files; each file containing the 51 values of the LE corresponding to each of the 12 muscles and the time

Each file has been then processes in a spreadsheet to obtain time-space parameters and the 51 values of each of the 12 muscles' EMG signals expressed in mV and corresponding to

For each subject, the LE corresponding to the 2% for each consecutive cycle locomotion were averaged across the 10 selected strides resulting in a pattern "ensemble average" of EMG (EAV), which represents an average pattern of the intra-subject LE. Using the EAV of all

Muscle activity, represented by its LE, was expressed between the 0% and the 100% of the duration of the cycle. The maximum amplitude and time of peak onset were obtained from

The influence of footwear on speed, the phases of locomotion, right-left legs symmetry, general effort, and on the maximum amplitude of the EMG activity of the 12 target muscles was evaluated by analysis of variance (ANOVA) for one factor with the statistical package of Matlab (The MathWorks, Massachusetts, USA), using the following factorial model: footwear, with three levels (barefoot, shod with standard shoes, and shod with athletic shoes). The statistic F has been analyzed and the significance level considered was p <0.05.

The aim of our study was to evaluate the influence of footwear on the electrical activity of muscles of both legs when jogging barefoot and jogging with two types of footwear

**2.5. Signals analysis** 

the GEAV.

**3. Results** 

**2.6. Statistical analysis** 

obtained with the pressure sensors.

**Figure 3** shows the EMG signal, processed and expressed as GEAV for the three conditions. To assess the influence of footwear, we have divided the stride into two phases: (1) support phase and (2) non-support phase, each of which exhibited its own characteristics in muscle activity.

In order to check whether there are differences in the muscular effort depending on the locomotion condition, we calculated the area under each muscle's GEAV for each condition and used it as an estimation of that effort. The ANOVA found that there is no statistically significant difference between the three conditions with respect to the general effort required for the locomotion.

Comparison by EMG of Running Barefoot and Running Shod 79

**Figure 3.** Profile of the GEAV for each muscle and for each locomotion condition (BF: barefoot, SS:

standard shoes, and AS: athletic shoes).

**Figure 2.** Space-temporal parameters expressed respect to the percentage of the jogging cycle (JC). Locomotion Conditions: BF –barefoot, AS –wearing athletic shoes, SS –wearing standard shoes.

Computational Intelligence in Electromyography Analysis – 78 A Perspective on Current Applications and Future Challenges

**Figure 2.** Space-temporal parameters expressed respect to the percentage of the jogging cycle (JC). Locomotion Conditions: BF –barefoot, AS –wearing athletic shoes, SS –wearing standard shoes.

**Figure 3.** Profile of the GEAV for each muscle and for each locomotion condition (BF: barefoot, SS: standard shoes, and AS: athletic shoes).

Computational Intelligence in Electromyography Analysis – 80 A Perspective on Current Applications and Future Challenges
