A Pilot Study of Muscle Force between Normal Shoes and Bionic Shoes during Men Walking and Running Stance Phase Using Opensim

he original idea for bionic shoes (BSs) involves combining the function of unstable foot conditions and the structure of the human plantar. The purpose of this study was to investigate the differences between the normal shoes (NS) and the BS during the stance phases of walking and running. A total of 15 Chinese males from Ningbo University were recruited for this study (age: 24.3 2.01 years ; height: 176.25 7.11 cm, body weight (BW): 75.75 8.35 kg). The participants were asked to perform a walking and running task. Statistical parametric mapping (SPM) analysis was used to investigate any differences between NSs and BSs during the walking and running stance

total of 15 Chinese males from Ningbo University were recruited for this study (age: 24.3 2.01 years ; height: 176.25 7.11 cm, body weight (BW): 75.75 8.35 kg). The participants were asked to perform a walking and running task. Statistical parametric mapping (SPM) analysis was used to investigate any differences between NSs and BSs during the walking and running stance phases. The results demonstrated that there were signi cant differences found (21.23–28.24%, p= 0.040; 84.47–100%,p= 0.017) in hip extension and exion between the NS and the BS during the walking stance phase. There were no signi cant differences found in ankle and moment during the running stance phase. Signi cant differences were found in the rectus femoris (5.29–6.21%;p= 0.047 ), tibialis anterior (14.37–16.40%;p= 0.038), and medial gastrocnemius (25.55–46.86%;p< 0.001) between the NS and the BS during the walking stance phase. Signi cant differences were found in rectus femoris (12.83–13.10%,p= 0.049; 15.89–80.19%,p< 0.001), tibialis anterior (15.85–18.31%,p= 0.039; 21.14–24.71%,p= 0.030), medial gastrocnemius (80.70–90.44%;p= 0.007), and lateral gastrocnemius (11.16–27.93%,p< 0.001; 62.20–65.63%,p= 0.032; 77.56–93.45%,p< 0.001) between the NS and the BS during the running stance phase. These ndings indicate that BSs are more ef cient for muscle control than unstable shoes and maybe suitable for rehabilitation training. Keywords:walking; running; muscle force; bionic shoes; rehabilitation 1. Introduction The primary function of footwear design is to protect human feet and provide posture stability during daily activities [1]. However, some scholars have proven that the function of conventional shoes could lead to overprotection in humans [2,3]. These researches indicated that the function of conventional shoes may degrade the function of human lower limb muscle groups, which means conventional shoes could lead to potential risks of injury. During the early stages of human evolution, there was no footwear available that humans could use to protect their feet [4,5]. To date, there are still certain indigenous populations that walk or run barefoot, without using any footwear. This suggests that from a necessity viewpoint, shoes are not essential for human survival or deemed to be a priority. Actuators2021,10, 274.

available that humans could use to protect their feet [4,5]. To date, there are still certain indigenous populations that walk or run barefoot, without using any footwear. This suggests that from a necessity viewpoint, shoes are not essential for human survival or deemed to be a priority. Actuators2021,10, 274.

Actuators2021,10, 274 2 of 12 Currently, studies have demonstrated that traditional shoes are impacting on lower limb function, causing gradual degeneration. With the evolution of human beings and the use of shoes over many years, the cuticle of the foot has not enabled humans to return to the state of barefoot walking. Based on considerations of unstable barefoot conditions, re ex control (RC) shoes and the Masai Barefoot Technology (MBT) have stimulated considerable research interest [6]. According to previous studies, the main functions of unstable shoes can be summarized as follows: (1) the mechanism of the unstable sole is used to activate the muscle passively, so that more muscles can be engaged in the movement [7]; (2) lower limb muscles are strengthened [7]; (3) the ability of the lower limbs is increased to dissipate the impact of ground reaction forces [8]; (4) muscle control is enhanced, and movement ability is improved [9]. However, the research by Nigg proved that after six weeks of training in normal shoes (NSs) and MBT shoes, the MBT shoes showed no signi cant improvements in balance [10]. From the perspective of the sole structure of the MBT, the instability conditions are close to the instability of ankle plantar exion and dorsi exion. On the other hand, from the human plantar structure, this unstable structure is more prone to medial and lateral problems. Thus, we can conclude that perhaps unstable MBT and RC shoes did achieve the effect of instability but did not achieve the desired increase in balance even after six weeks of training. This may be because they did not consider the unstable condition of the human body when standing. Therefore, some of the studies have focused on bionic shoes (BSs) [11–13], and the original idea of BS involves combining the function of unstable conditions and the structure of the human plantar. They further mentioned that BS may actually reduce injury risk in the lower limbs. The most interesting aspect of BS is the combination of the human condition and the design of a shoe, which may be more suitable for training or rehabilitation

original idea of BS involves combining the function of unstable conditions and the structure of the human plantar. They further mentioned that BS may actually reduce injury risk in the lower limbs. The most interesting aspect of BS is the combination of the human condition and the design of a shoe, which may be more suitable for training or rehabilitation than MBT and RC shoes. Walking and running are the most basic daily movements of humans. From a biome- chanics point of view, there are many similarities between walking and running [14,15]. On the other hand, the biggest difference between walking and running is the strike pattern. Running has a swing phase, whereas walking has two support phases. This means that the impact force and load on the lower limbs are greater when running on the ground or during the support period. Thus, although walking and running do not appear to be fundamentally different from each other, their novelty can be found through kinematic and kinetic analysis. Kotaro et al. have published a paper that investigated the difference between walking and running [16], and they found muscle force and activation vary with different strike patterns. This also shows that on the basis of the analysis of the movement, combined with the analysis of the muscles, we can better and fully understand the internal mechanism characteristics in the process of movement. OpenSim is free, open-source software. It can be used in many elds, such as walking dynamics analysis, motion performance research, surgical process simulation, and medical device design. In OpenSim, a musculoskeletal model consists of multiple bones connected by joints, where muscles attach to the bones and force the joints to move [17,18]. Combined with the BS mentioned above, the range of motion of muscle force can be truly simulated through this simulation software, which is of great signi cance to our in-depth discussion of BSs. Many of the previous studies believed that the analysis of muscle force can re ect the internal mechanism of injury, nd prevention methods and make rehabilitation plans [17,18]. The concept of applying simple basic walking

of motion of muscle force can be truly simulated through this simulation software, which is of great signi cance to our in-depth discussion of BSs. Many of the previous studies believed that the analysis of muscle force can re ect the internal mechanism of injury, nd prevention methods and make rehabilitation plans [17,18]. The concept of applying simple basic walking and running principles to explore the inner muscle changes often can more truly re ect the difference between the NS and the BS. To our knowledge, there are no studies that have investigated the differences in muscle force between the NS and the BS during the walking and running stance phases. Therefore, the purpose of this study was to investigate the differences between the NS and the BS during the stance phases of walking and running. We hypothesized that the muscle force of the BS will be bigger than that of the NS during the stance phases of walking and running. We further hypothesized the muscle force of the BS will be bigger than that of the NS during the toe-off phase.

Actuators2021,10, 274 3 of 12 2. Materials and Methods 2.1. Participants A total of 15 Chinese males from Ningbo University were recruited for this study (age: 24.3 2.01 years ; height: 176.25 7.11 cm; body weight (BW): 75.75 8.35 kg). They all had sports three times a week at least for one hour at a time. No surgical injuries were found in the past six months prior to this study, and none of the participants had any kind of lower limb medical issues that could impact the results of this study. Written informed consent was obtained from all participants, when they were informed about the purpose, procedures, conditions, and requirements of this study. The Ethics Committee of Ningbo University approved this study (protocol code: RAGH 20210106). 2.2. Shoes Figurea shows two kinds of shoes used in the walking and running experiment. These two shoes were produced by Ningbo Jiangbei Feibu Sports goods Co., Ltd. (Ningbo, China). The BS was designed based on the foot shape of each participant, and the materials and stiffness of both shoe types, i.e., BS and NS, were the same [11–13]. The shoes details are present in Table.Actuators 2021, 10, x FOR PEER REVIEW 4 of 13 Figure 1. (a) Illustration of shoe-making procedures. (b) Illustration of experiment design for col- lecting the kinematics and dynamics data during the walking and running between the NS and the BS. 2.3. Experimental Protocol and Equipment Ningbo University Research Academy of Grand Health is a sports biomechanics la- boratory with adequate facilities. All tests were performed the biomechanics laboratory. A Vicon motion capture system (Oxford Metrics Ltd., Oxford, UK) and a force platform (Kistler, Switzerland) were used for collecting kinematics and dynamics data. Kinematics and dynamics data were captured at frequencies of 200 and 1000 Hz, respectively. An EMG system (Delsys, Boston, MA, USA) set to a frequency of 1000 Hz was used to collect surface muscle activations and forces, including medial gastrocnemius, lateral gas- trocnemius, vastus medialis, vastus lateralis, rectus femoris, and tibialis anterior (Figure 2a). The data from all the equipment were captured synchronously. Tight shorts and

at frequencies of 200 and 1000 Hz, respectively. An EMG system (Delsys, Boston, MA, USA) set to a frequency of 1000 Hz was used to collect surface muscle activations and forces, including medial gastrocnemius, lateral gas- trocnemius, vastus medialis, vastus lateralis, rectus femoris, and tibialis anterior (Figure 2a). The data from all the equipment were captured synchronously. Tight shorts and pants were asked to be worn for each test by all participants. Thirty-nine (12.5 mm in diameter) reflective markers were secured onto each participant. Figure 2b shows the placement of each marker. Figure 1. (a) Illustration of shoe-making procedures. (b) Illustration of experiment design for collecting the kinematics and dynamics data during the walking and running between the NS and the BS.

Actuators2021,10, 274 4 of 12 Table 1.Illustration of the speci c details between bionic shoes (BS) and normal shoes (NS). BS NS Heel height (mm) 23.0 (1.0) 27.0 (1.0) Weight (g) 271.0 (2.0) 294.5 (2.3) Sole hardness (Asker C) 50.0 (0.9) 49.6 (0.6) Bending stiffness (N/mm) 14.2 (0.5) 13.6 (0.4) Shoe upper material PVC PVC Shoe sole material EVA EVA Note: PVC, nylon (polyamide) polyvinyl chloride; EVA, ethylene-vinyl acetate. 2.3. Experimental Protocol and Equipment Ningbo University Research Academy of Grand Health is a sports biomechanics laboratory with adequate facilities. All tests were performed the biomechanics laboratory. A Vicon motion capture system (Oxford Metrics Ltd., Oxford, UK) and a force platform (Kistler, Switzerland) were used for collecting kinematics and dynamics data. Kinematics and dynamics data were captured at frequencies of 200 and 1000 Hz, respectively. An EMG system (Delsys, Boston, MA, USA) set to a frequency of 1000 Hz was used to collect surface muscle activations and forces, including medial gastrocnemius, lateral gastrocnemius, vastus medialis, vastus lateralis, rectus femoris, and tibialis anterior (Figurea). The data from all the equipment were captured synchronously. Tight shorts and pants were asked to be worn for each test by all participants. Thirty-nine (12.5 mm in diameter) re ective markers were secured onto each participant. Figureb shows the placement of each marker.Actuators 2021, 10, x FOR PEER REVIEW 5 of 13 Figure 2. (a) Illustration of the placements of the EMG sensor on three different sides. (b) Illustra- tion of the placements of the marker on three different sides. 2.4. Procedure Each participant was required to complete a warm-up session that included the fol- lowing: (a) running on a treadmill at a speed of 8 km/h for 10 min; and (b) lower limb stretching exercises. All participants were permitted 3 trials to familiarize themselves with test movements prior to formal experiments. In order to reduce the impedance of the in- terface between the skin and the electrode, the skin was prepared by hair removal from the tested area, as well as skin abrasion and alcohol cleaning. After all the participants were fully familiarized with procedures and

participants were permitted 3 trials to familiarize themselves with test movements prior to formal experiments. In order to reduce the impedance of the in- terface between the skin and the electrode, the skin was prepared by hair removal from the tested area, as well as skin abrasion and alcohol cleaning. After all the participants were fully familiarized with procedures and experimental conditions, the markers and the sensors were attached to the participants as the maximal voluntary contraction (MVC) of each muscle was collected from six muscles. After recording the MVC, the participants were required to stand on a force platform to collect static coordinates. Each participant was asked to stand parallel to the Y-axis of the force platform and cross their arms on the shoulder, with eyes looking forward, until complete static coordi- nates were captured. For the kinematics and dynamics data collection, all participants performed walking and running tasks at a self-selected speed along a 10 m walkway (Fig- ure 1b). The data collection was only performed on a dominant leg, which was defined as the preferred leg to kick a ball. The dominant leg was used to gather 5 successful data sets, and there was a 1 min break time between each task. The initial contact was defined as exceeding 10 N on a ground reaction force [19]. The walking and running tasks were di- vided into two different days for data collection and were performed at the same time on each day. This was important for avoiding inaccuracies in data collection caused by fa- tigue. 2.5. Data Collection and Processing Kinematics and ground reaction force data were identified and acquired by Viocn Nexus 1.8.6 software and exported into a c3d format file. MATLAB R2019a (The Math- Works, Natick, MA, USA) was used to perform coordinate system conversion, low-pass filtering, data extraction, and format conversion for kinematics and ground reaction force data. Specifically, the following steps were followed: (1) The coordinate system of the kin- ematics and ground reaction force data was converted to the coordinate system used in subsequent simulations. That is, the forward direction of the

Natick, MA, USA) was used to perform coordinate system conversion, low-pass filtering, data extraction, and format conversion for kinematics and ground reaction force data. Specifically, the following steps were followed: (1) The coordinate system of the kin- ematics and ground reaction force data was converted to the coordinate system used in subsequent simulations. That is, the forward direction of the human body was the positive direction of the X-axis, and the upward direction perpendicular to the ground was the Y- axis. The positive direction and the direction of the human body to the right were the positive directions of the Z-axis. (2) The biomechanical data for the marker trajectory and the ground reaction force were filtered by 6 and 30 Hz fourth-order zero-phase lag But- terworth low-pass filters. (3) The kinematics and ground reaction force data were ex- Figure 2. (a) Illustration of the placements of the EMG sensor on three different sides. (b) Illustration of the placements of the marker on three different sides. 2.4. Procedure Each participant was required to complete a warm-up session that included the following: (a) running on a treadmill at a speed of 8 km/h for 10 min; and (b) lower limb stretching exercises. All participants were permitted 3 trials to familiarize themselves with test movements prior to formal experiments. In order to reduce the impedance of the interface between the skin and the electrode, the skin was prepared by hair removal from the tested area, as well as skin abrasion and alcohol cleaning. After all the participants were fully familiarized with procedures and experimental conditions, the markers and the sensors were attached to the participants as the maximal voluntary contraction (MVC) of each muscle was collected from six muscles. After recording the MVC, the participants were required to stand on a force platform to collect static coordinates.

Actuators2021,10, 274 5 of 12 Each participant was asked to stand parallel to theY-axis of the force platform and cross their arms on the shoulder, with eyes looking forward, until complete static coor- dinates were captured. For the kinematics and dynamics data collection, all participants performed walking and running tasks at a self-selected speed along a 10 m walkway (Figureb). The data collection was only performed on a dominant leg, which was de ned as the preferred leg to kick a ball. The dominant leg was used to gather 5 successful data sets, and there was a 1 min break time between each task. The initial contact was de ned as exceeding 10 N on a ground reaction force [19]. The walking and running tasks were divided into two different days for data collection and were performed at the same time on each day. This was important for avoiding inaccuracies in data collection caused by fatigue. 2.5. Data Collection and Processing Kinematics and ground reaction force data were identi ed and acquired by Viocn Nexus 1.8.6 software and exported into a c3d format le. MATLAB R2019a (The MathWorks, Natick, MA, USA) was used to perform coordinate system conversion, low-pass ltering, data extraction, and format conversion for kinematics and ground reaction force data. Speci cally, the following steps were followed: (1) The coordinate system of the kinematics and ground reaction force data was converted to the coordinate system used in subsequent simulations. That is, the forward direction of the human body was the positive direction of theX-axis, and the upward direction perpendicular to the ground was theY-axis. The positive direction and the direction of the human body to the right were the positive directions of theZ-axis. (2) The biomechanical data for the marker trajectory and the ground reaction force were ltered by 6 and 30 Hz fourth-order zero-phase lag Butterworth low-pass lters. (3) The kinematics and ground reaction force data were extracted during the stance phases of running and walking, and the data format was converted to the trc. (marker trajectory) and mot. (force plate data) formats required by the OpenSim

This study compares muscle force in normal and bionic shoes during walking and running.