In this study, the kinematic results of the users were divided into two main groups:
- behavior of the ankle kinematics and (2) range of motion (ROM) of the lower limb As an initial approach, ankle kinematics showed no significant changes (p > 0.05) for the two groups (i.e., unassisted and assisted), including the com- plete sample of participants through a Student’s t-test. Nevertheless, diverse as- pects stated in the following section could explain those results. Therefore, this part presented the results individually for each participant.
For the first group, Figure 3 shows the ankle kinematics during a gait cycle for both the healthy pattern and the results of each volunteer, where letters from a to j represent the participant from 1 to 10, respectively. This cycle comprises phases between each heel-strike event for both modalities assessed (i.e., baseline and as- sisted) and the healthy ankle pattern. Moreover, the vertical line included in the figure highlights the toe-off phase for each case.
Concerning to the toe-off phase (TO), 40% of the participants showed differences of more than 5% in the time of occurrence of this event during the gait cycle (see Figure 3), when they used the T-FLEX actuation system. Likewise, 30% of the subjects brought this event to the estimated percentage in a healthy pattern. The other volunteers did not show changes in this aspect. On the other hand, the ankle angle shape had variations due to the effect of using the device. Specifically, subject 5 registered an increase of 15 degrees in the dorsiflexion movement during the swing phase. However, participants 1 and 9 reduced this movement at 10 degrees, although this reduction was within the healthy range.
For the other group, Table 2 summarizes the range of motion (ROM) for the ankle dorsi-plantarflexion (A-F), knee flexo-extension (K-F), hip flexo-extension (H-F), and hip ab-adduction (H-A) in both modalities. The second part of the table shows the percental variation of the joints when the participant used the T-FLEX orthosis. Positive values in this variation indicate an increase in the ROM of the joint, and by contrast, negative values represent a decrease in this parameter. For this part, the highlighted values represent increases greater than 10% on the joint concerning the baseline state.
From the variation table, 70% of the volunteers exhibited significant changes in the paretic (P) ankle ROM using the device, whether increases or decreases. Likewise, the changes in the ROM for the paretic ankle also tend to variate for the non- paretic (N-P) joint. On the other hand, the number of altered joints was directly proportional to the change presented on the ankle, where values of paretic ankle ROM with variation above 50% registered changes in at least half of the analyzed joints. In general, the changes did not show a common tendency in terms of increases or decreases. Furthermore, the larger values corresponded to alterations on the A-F, although subjects 4, 5, and 7 showed the H-A value as the maximum variation.
According to the variations on the ROM of the lower limb joints (see Table 2), it is essential to determine whether this change represents a positive o negative effect in the joint of the participant (see Fig. 4). For that, the ROM obtained was compared with the mean value in a healthy gait . In this context, 60% of the volunteers showed improvement in the A-F using the device. Among this, subjects 2, 5, and 7
achieved values whose errors, regarding the ROM in healthy people, were less than 2%. This way, positive changes in the paretic ankle joint improved the ranges for the non-paretic joints, especially in the ankle joint. For 30% of the participants, the variations in the A-F did not represent significant improvements, and additionally, one volunteer exhibited a negative effect in this ROM related to a reduction of 33% in its value.
Thus, Figure 4 summarizes the consequences of using the T-FLEX system ac- tuation on the analyzed joints for each participant. The positive effects indicate improvement in the ROM of the corresponding joint, approaching this value to healthy ranges. Negative impacts indicate a pattern disruption, and hence a dis- tancing of the movement with a healthy pattern. Undetermined conditions grouped changes where, although the variation is significant (i.e., above 10%), this value does not improve or impair the ROM. Lastly, the no-changes group integrates the differences between both scenarios of less than 10%.
Bearing in mind the classification of variations for each subject (see Figure 4), 70% of the volunteers showed a positive effect on at least one joint, where the paretic ankle was the more prevalent. The exhibited negative impacts were mainly related to a reduction in the ROM, so only two joints reflected increases that did not represent a risk for the participant.
On the other part, Table 3 contains the values of the Gait Deviation Index (GDI) for each participant. The GDI showed a significant difference for 30% of the paretic limbs of the participants, wherein 20% manifested a reduction below 14%, and one volunteer registered an increase of 14%. For the non-paretic, 40% of the participants exhibited decreases by less than 30% for this index. Reduction in GDI is related to a higher difference between the participant kinematics and a healthy pattern. In contrast, an improvement in the gait kinematics depends on an increase in this index. The mean value of GDI for the participants (see Table 5) did not present a significant difference between the scenarios, and both limbs remained the no healthy condition as the GDI percentage was less than 90%.
Lastly, Figure 5 illustrates the Movement Analysis Profile (MAP) for the paretic (Fig. 5a) and non-paretic (Fig. 5b) limbs between baseline and assisted gait. The most affected joints in the scenarios were the ankle, the knee, and the hip (i.e., ankle rotation, hip rotation, and knee flexo-extension). The ankle dorsi-plantarflexion did not show significant changes in both the paretic and non-paretic. The Gait Profile Score (GPS) significantly increased its value between unassisted and assisted conditions of the non-paretic limb, although this change moved away from the value of healthy people. Nevertheless, this value did not exhibit significant changes for the paretic side.
Considering the variation in ROM presented above, the second part of this study analyzes the changes in spatio-temporal values. For this purpose, Table 4 shows the percentages of variation for the parameters in each participant. The parameters include mean values for the paretic (P) and non-paretic limbs (N-P) in aspects such as the percentage of duration for the stance phase (SP) on the gait cycle, as well as the step length (SL) and the step width (SW). Likewise, walking speed (S), stride length (ST), and cadence (C) also are part of this table.
In general terms, the spatio-temporal parameters did not show significant changes using the T-FLEX actuation system to either of the participants’ limbs. Neverthe- less, the cadence exhibited a reduction in 70% of the volunteers. This parameter registered decreases below 24% of the baseline state, although, subject 8 presented an increase in the cadence of 20% for the assisted gait.
On the other hand, Table 5 contains the mean values for the participants’ pa- rameters. This table summarizes the results aforementioned, showing a decrease of cadence in 14% (i.e., from 99 to 85 steps per minute). Additionally, this table also exhibits other Spatio-temporal values without significant changes.
For the users’ perception, Figure 6 shows the relevant aspects selected by the participants through the QUEST survey. The most selected parameter was the device’s comfort with 70% of recurrence. Other important aspects for the users were safety, weight, and dimensions. Finally, the level of satisfaction of the user was between satisfied and very satisfied in 60% and 40% of the users, respectively.
To understand the participants’ effects on the gait cycle, statistical analysis aimed to identify differences between assisted and baseline conditions. In terms of the an- kle kinematics, the results revealed statistically significant changes for 70% of the subjects in at least one gait phase for the angle. Specifically, this joint showed statis- tical differences in the stance and swing phase for 60% and 70% of the participants, respectively (see Table 6). Moreover, 40% of them exhibited variations in the entire gait cycle.
In the spatio-temporal context, the parameters showed a statistically significant decrease in the cadence (p=0.0002) and speed (p=0.03) concerning the assisted gait. The parameters of long stride, step length, step width, and stance phase did not show statistically significant changes.