Fig 1.
The overall system and experimental environment.
(A) and (B) represent the 2- and 1-DOF PAFOs, respectively. (C) shows the experimental setup. The subject wore an in-show pressure sensor and IMU on the trunk and walked on the treadmill. The figure also shows the flow of data. The signals detected by the PAFO sensors are measured by micro controller unit 1 (MCU1; STM Nucleo board) mounted on it and transferred to MCU2 (STM Discovery board) via CAN communication. MCU2 generates a PWM signal for solenoid valve control and enables serial debugging in the desktop through simultaneous serial communication. And the individual in this figure has given written informed consent (as outlined in PLOS consent form) to publish these case details.
Fig 2.
Contraction of PAM2 for simultaneous eversion and plantar flexion by rotation of the subtalar and talocrural joints.
Table 1.
Types of assistance technique for PAFOs.
Fig 3.
Characteristics of the assistance techniques.
This figure describes the characteristics of P-bcs such as assistance timing, peak force, and onset of positive power. The illustrations of the foot posture on the ground during gait cycle helps the understanding of the process of assistance.
Fig 4.
Methods for calculating the average COP position and examples with 1- and 2- DOF PAFOs.
The trajectory was interpolated to increase the data to 100 points during the stance phase.
Fig 5.
Stabilogram of the trunk angular acceleration using a density plot.
Converting to a Gaussian ellipsoid through covariance calculation makes the comparison easier, as shown on the right. In the figure, δ1 and δ2 represent 2 standard deviations of covariance and are used as vertices of the Gaussian ellipsoid.
Table 2.
Experimental results of all the participants.
Fig 6.
P-table of the center of pressure.
The values with blue and red colors represent significant differences of p < 0.05 and p < 0.0333 for before and after correction, respectively. The significance level of 0.0333 was determined on the basis of FDR. The shadows colored in yellow, pink, and green represent comparisons between 1 DOF − 1 DOF, 1 DOF − 2 DOF, and 2 DOF − 2 DOF, respectively.
Fig 7.
The values with blue and red colors represent significant differences of p<0.05 and p<0.0326 for before and after correction, respectively. A significance level of 0.0326 was determined on the basis of FDR. The shadows colored in yellow, pink, and green represent comparisons between 1 DOF − 1 DOF, 1 DOF − 2 DOF, and 2 DOF − 2 DOF, respectively.
Table 3.
Average correlation coefficient.
Fig 8.
Experimental results of the stabilogram for subject 2.
(A) and (B) The results of the experiments that used the assistance techniques demonstrated in Table 1 with 1- and 2-DOF PAFOs, respectively. (C) The stabilogram of the unpowered and maximum output states of 1- and 2-DOF PAFO experiments for comparison. The circles represent 1δ of covariance, which comprised 39% of the data. 2δ (68%) and 3δ (99%) are not shown for clarity of comparison, as they are simple magnifications.
Fig 9.
PCC plot of the 4 cases for subject 2 (COPAvg−δRMS, PAvg−θAvg, θAvg−COPAvg, and PAvg−COPAvg).
The blue and gray points in (A) represent the mean data of 1- and 2-DOF PAFO experiments, respectively. (B) and (C) The 2- and (D) 1-DOF PAFO experiments.