Effects of Technique Asymmetry on 500 m Speed Skating Performance

s study aimed to determine the effects of technique asymmetry on 500 m straight-track speed skating performance. We analyzed 20 elite skaters, measuring their joint angles, center of mass shift, and times and speeds during the gliding and push-off phases. The technique asymmetry index (ASI) was calculated for each parameter, and pairedt-tests were used to compare bilateral asymmetry. Spearman correlation coefficients assessed the relationship between the ASI and both the average straight track speed and overall performance. Significant bilateral asymmetries in the knee, push-off, trunk, and hip angles were found in both male and female participants (p< 0.05). The male participants demonstrated a higher right push-off speed (p= 0.029) and a longer left gliding time (p= 0.048). Significant asymmetry was also observed in the lateral shift of the center of mass during each phase of the straight-track skating gait cycle (p< 0.001). No significant correlation was found between the ASIs and the overall performance (p≥0.067). These findings indicate that while elite speed skaters demonstrated

right push-off speed (p= 0.029) and a longer left gliding time (p= 0.048). Significant asymmetry was also observed in the lateral shift of the center of mass during each phase of the straight-track skating gait cycle (p< 0.001). No significant correlation was found between the ASIs and the overall performance (p≥0.067). These findings indicate that while elite speed skaters demonstrated significant bilateral technique asymmetry in straight track skating, these asymmetries did not significantly impact their overall performance. Keywords:technique symmetry; speed skating performance; speed skating training 1. Introduction Bilateral asymmetries of strength, flexibility, and movements in athletes affect their performance in sports. Running and swimming are sports in which techniques ought to be bilaterally symmetrical. Studies, however, showed that athletes in these sports have bilateral asymmetry in strength and flexibility, which commonly leads to technique asymmetry in these sports [1–3]. In running, technique asymmetry not only reduces performance [4] but also increases the risk of hamstring muscle strain [5] and tibial fatigue fracture [6]. In swimming, technique asymmetry affects the effective propulsion force generated by the left and right limbs, which has a significant negative effect on swimmers’ performance [7,8]. The adaptations to asymmetric movements lead to asymmetry in bilateral limb functions [9]. Speed skating is a sport in which skaters show apparent asymmetry in strength. A study demonstrated that the maximum joint torques that professional skaters could generate at the left hip and ankle joints were significantly greater than those generated at the right hip and ankle joints [10]. These hip and ankle strength asymmetries were most likely due to the skaters’ adaptations to their skating on the curve track. On the curve track, speed skaters had to push both their legs to the right so they could turn left quickly [11]. Compared to the right leg, the left knee flexion angle is smaller when pushing the ice, so greater strength is needed from the left hip and knee joint extensors [10]. Consequently, the Bioengineering2024,11, 899.

quickly [11]. Compared to the right leg, the left knee flexion angle is smaller when pushing the ice, so greater strength is needed from the left hip and knee joint extensors [10]. Consequently, the Bioengineering2024,11, 899.

Bioengineering2024,11, 899 2 of 14 left semitendinosus, rectus femoris, and lateral gastrocnemius have significantly greater strength compared to their right side counterparts [10]. Although bilateral strength and technique asymmetries are necessary for the perfor- mance of curve skating, they may have negative effects on the performance of straight track skating and, thus, the overall performance of skating. Studies showed that the average curve speed as well as the average straight track speed of speed skating are significantly correlated with performance [12,13]. Studies also showed that the gliding distance, the trunk flexion angle, the knee flexion angle, the push-off angle, and the shift of center of mass (COM) are important technique factors affecting the average straight track speed [12,14,15]. Understanding the effects of the asymmetry of these technique factors on the performance of speed skating is important for speed skaters to adjust their training programs to maxi- mize their performance. Technical asymmetry in speed skating not only affects performance on the straight track but is also associated with an increased risk of injury. Studies have shown that a skater’s left lower limb is overloaded [16], and significant asymmetry in bilateral glu- teus maximus fatigue [17] may increase the risk of lower limb injury [16,17]. Studying asymmetry in speed skating and adjusting training accordingly are important not only for improving performance but also for preventing possible injuries in speed skaters. The purposes of this study were to determine (1) the bilateral technique asymmetry of elite speed skaters in straight track skating and (2) the effects of bilateral technique asym- metry on skating performance. We hypothesized that in straight track skating, (1) speed skaters could have bilateral asymmetry in their trunk flexion angle, knee flexion angle, and push-off angle; (2) speed skaters could have bilateral asymmetry in the shift of COM; (3) speed skaters may exhibit bilateral asymmetry in speed and time during the gliding and push-off stages; and (4) technique asymmetry could be negatively correlated with the average straight track speed and overall performance of skaters. 2. Methods 2.1. Participants A total of 10 male and 10 female elite speed skaters were

have bilateral asymmetry in the shift of COM; (3) speed skaters may exhibit bilateral asymmetry in speed and time during the gliding and push-off stages; and (4) technique asymmetry could be negatively correlated with the average straight track speed and overall performance of skaters. 2. Methods 2.1. Participants A total of 10 male and 10 female elite speed skaters were selected and agreed to partici- pate in this study. They were the top 10 finishers in the men’s and women’s 500 m speed skat- ing competition of the 2020–2021 China National Speed Skating Championships (Harbin, 28 March 2021). For the male participants, the overall performance was35.99±0.33 s , with the best performance being 35.64 s and the worst performance being 36.42 s. For the female participants, the overall performance was 39.46±0.60 s, with the best performance being 38.34 s and the worst performance being 40.45 s. The best performance of each skater in two 500 m speed skating competitions was used for analyses of performance on the straight track during the first lap. 2.2. Data Collection Skaters’ straight-track skating movements were recorded using 4 time-synchronized video cameras (SONY FDR-AX700, SONY, Tokyo, Japan) with a resolution of 1920×1080 at a sampling frequency of 60 frames/s [18]. Considering camera positions relative to movement direction, a shutter speed of 1/500 s was used to obtain videographic images with a blur error of less than 10 mm. The four video cameras were set as two groups with two cameras in each group. One camera was set on the side of the track and the other on the front side of the track in each group. The calibration space of each group of cameras was 14 m long×6 m wide×2.5 m high with an overlap area of 4 m long. The total calibrated space was 24 m long×6 m wide×2.5 m high. The calibration space was 76 m in front of the starting line. A suspended calibration system was used to define each calibration space before the competition. The calibration system included four calibration poles with six calibration points suspended on each pole [18]. Five

area of 4 m long. The total calibrated space was 24 m long×6 m wide×2.5 m high. The calibration space was 76 m in front of the starting line. A suspended calibration system was used to define each calibration space before the competition. The calibration system included four calibration poles with six calibration points suspended on each pole [18]. Five markers were placed in the overlap area of two calibration spaces for defining the reference frame in data processing. The ice surface

Bioengineering2024,11, 899 3 of 14 reference frame was defined with theX-axis pointing in the forward direction of skating, theY-axis pointing to the left side of the track, and theZ-axis pointing upward. The skaters’ skating movements in the calibration space were recorded in the competitions. 2.3. Data Process The video records of skating were digitized using an artificial intelligence markerless motion capture system (FastMove Inc., Dalian, China) to obtain two-dimensional (2D) coordinates of 21 critical body landmarks [18,19]. Digitized 2D coordinates from two cameras were synchronized using multiple critical events [20,21]. The critical events included blades landing on the ice and blades taking off from the ice. Three-dimensional (3D) coordinates of 21 body landmarks were obtained from synchronized 2D coordinates using the Direct Linear Transformation method [22]. The mean calibration error was less than 0.005 m. The raw 3D coordinates of body landmarks were filtered through the Butterworth low-pass filter with an estimated optimal frequency of 10 Hz [23]. The artificial intelligence markerless motion capture system was used to collect 3D coordinates of body landmarks in this study. The reproducibility of the 3D coordinates of body landmarks collected using this system was rigorously evaluated. The multiple correlation coefficients of repeatedly collected coordinate data were no less than 0.90 [18,19], which indicated that the data obtained using this system were highly reproducible. 2.4. Data Reduction Displacements and the speeds of whole body COM, movement time, push-off angle, knee and hip flexion angles and range of flexion–extension motions (ROMs), trunk flexion angle, and flexion ROMs were calculated from smoothed 3D coordinates of body landmarks. A speed skating gait cycle on the straight track was defined as the duration between the two consecutive push-offs of the same side and consisted of a gliding phase and a push-off phase for each side (Figure). Taking the left side as an example, the left leg gliding phase started when the right blade was taken off the ice and ended when the right blade was placed on the ice. The left leg push-off phase began when the right blade was placed on the ice and ended when the

and a push-off phase for each side (Figure). Taking the left side as an example, the left leg gliding phase started when the right blade was taken off the ice and ended when the right blade was placed on the ice. The left leg push-off phase began when the right blade was placed on the ice and ended when the left blade was taken off the ice (Figure). We calculated the gliding distance, time, and speed in the gliding phase and push-off phase on each side. We also calculated the COM movement distances in the forward–backward, left–right, and up–down directions during the gliding phase and push-off phase.Bioengineering 2024, 11, x FOR PEER REVIEW 3 of 15 points suspended on each pole [18]. Five markers were placed in the overlap area of two calibration spaces for defining the reference frame in data processing. The ice surface ref- erence frame was defined with the X-axis pointing in the forward direction of skating, the Y-axis pointing to the left side of the track, and the Z-axis pointing upward. The skaters’ skating movements in the calibration space were recorded in the competitions. 2.3. Data Process The video records of skating were digitized using an artificial intelligence markerless motion capture system (FastMove Inc., Dalian, China) to obtain two-dimensional (2D) co- ordinates of 21 critical body landmarks [18,19]. Digitized 2D coordinates from two cam- eras were synchronized using multiple critical events [20,21]. The critical events included blades landing on the ice and blades taking off from the ice. Three-dimensional (3D) co- ordinates of 21 body landmarks were obtained from synchronized 2D coordinates using the Direct Linear Transformation method [22]. The mean calibration error was less than 0.005 m. The raw 3D coordinates of body landmarks were filtered through the Butterworth low-pass filter with an estimated optimal frequency of 10 Hz [23]. The artificial intelligence markerless motion capture system was used to collect 3D coordinates of body landmarks in this study. The reproducibility of the 3D coordinates of body landmarks collected using this system was rigorously evaluated. The multiple cor- relation coefficients of repeatedly collected coordinate data

filtered through the Butterworth low-pass filter with an estimated optimal frequency of 10 Hz [23]. The artificial intelligence markerless motion capture system was used to collect 3D coordinates of body landmarks in this study. The reproducibility of the 3D coordinates of body landmarks collected using this system was rigorously evaluated. The multiple cor- relation coefficients of repeatedly collected coordinate data were no less than 0.90 [18,19], which indicated that the data obtained using this system were highly reproducible. 2.4. Data Reduction Displacements and the speeds of whole body COM, movement time, push-off angle, knee and hip flexion angles and range of flexion–extension motions (ROMs), trunk flexion angle, and flexion ROMs were calculated from smoothed 3D coordinates of body land- marks. A speed skating gait cycle on the straight track was defined as the duration be- tween the two consecutive push-offs of the same side and consisted of a gliding phase and a push-off phase for each side (Figure 1). Taking the left side as an example, the left leg gliding phase started when the right blade was taken off the ice and ended when the right blade was placed on the ice. The left leg push-off phase began when the right blade was placed on the ice and ended when the left blade was taken off the ice (Figure 1). We calcu- lated the gliding distance, time, and speed in the gliding phase and push-off phase on each side. We also calculated the COM movement distances in the forward–backward, left–right, and up–down directions during the gliding phase and push-off phase. Figure 1. Movement-stage division diagram. Trunk flexion angle was referred to as the angle between the longitudinal axis of the trunk and the horizontal plane. Hip flexion angle was calculated as the angle between the longitudinal axes of the trunk and the thigh. Knee flexion angle was defined as the angle between the longitudinal axes of the thigh and the calf. Push-off angle was defined as the Figure 1.Movement-stage division diagram. Trunk flexion angle was referred to as the angle between the longitudinal axis of the trunk and the horizontal

as the angle between the longitudinal axes of the trunk and the thigh. Knee flexion angle was defined as the angle between the longitudinal axes of the thigh and the calf. Push-off angle was defined as the Figure 1.Movement-stage division diagram. Trunk flexion angle was referred to as the angle between the longitudinal axis of the trunk and the horizontal plane. Hip flexion angle was calculated as the angle between the longitudinal axes of the trunk and the thigh. Knee flexion angle was defined as the angle between the longitudinal axes of the thigh and the calf. Push-off angle was defined as the angle between the longitudinal axis of the calf and the horizontal plane. The range of motion of joints was defined as the difference between the maximum and minimum values of the joint angles in a certain movement phase.

Bioengineering2024,11, 899 4 of 14 The bilateral asymmetry indexes (ASIs) [24] were calculated as follows: ASI= |XR−XL| 0.5(XR+XL) ×100% (1) where XRrefers to the right leg parameter and XLrefers to the left leg parameter. We calculated the ASIs of joint angles at landing, push-off, and off-ice; ranges of joint motions; COM movement; gliding distance; speed; and time during the push-off phase and gliding phase. The greater the ASI, the greater the bilateral asymmetry of the given technique parameter. 2.5. Data Analyses To test the first three hypotheses, pairedt-tests were performed to compare bilateral technique parameters for male and female participants, and there were no multiple com- parisons in this study. To test the fourth hypothesis, Spearman correlation coefficients were calculated between the ASIs of the participants’ technical parameters and both the average straight track speed and the finishing time of 500 m skating. Statistical testing results with a Type I error rate less than 0.05 were considered statistically significant. All data analyses were performed using SPSS Computer Program Package Version 26.0 (SPSS Science, Chicago, IL, USA). Effect sizes were calculated using Cohen’s d. An effect size less than 0.5 was considered small. An effect size greater than 0.5 but not greater than 0.8 was considered medium. An effect size greater than 0.8 was considered large. 3. Results 3.1. Basic Performance Data by Gender The average straight track speed was 13.53±0.21 m/s for male participants and 12.45±0.25 m/s for female participants. The average straight track speed was significantly correlated to the overall performance for male and female participants (R 2 = 0.442,p= 0.036 for male participants; R 2 = 0.479,p= 0.026 for female participants). 3.2. Asymmetry in Joint Angles and COM Shift For male participants, the right knee flexion angle at the right blade landing was significantly greater than the left knee flexion angle at the left blade landing (p= 0.004) (Table). The trunk flexion angle at the left blade landing was significantly greater than that at the right blade landing (p= 0.028) (Table). The left leg push-off angle was significantly greater compared to the right leg push-off angle (takeoff:p= 0.010;

the right blade landing was significantly greater than the left knee flexion angle at the left blade landing (p= 0.004) (Table). The trunk flexion angle at the left blade landing was significantly greater than that at the right blade landing (p= 0.028) (Table). The left leg push-off angle was significantly greater compared to the right leg push-off angle (takeoff:p= 0.010; opposite-side landing: p= 0.020) (Table). No significant differences were observed in other bilateral joint angles and joint ROMs (Tables). For female participants, the left hip flexion angle at the left blade landing was signifi- cantly greater than the right hip flexion angle at the right blade landing (p= 0.040) (Table). The trunk flexion angle at the right blade landing was significantly greater than that at the left blade landing (p= 0.009) (Table). The trunk flexion angle at the left blade takeoff was significantly greater compared to that at the right blade takeoff (p= 0.048) (Table). The right knee flexion angle of the right leg at the right blade takeoff was significantly greater than that at the left blade takeoff (p= 0.032) (Table). The range of flexion–extension of the right knee during the right blade push-off phase was significantly greater compared to the left knee during the left blade push-off phase (p= 0.045) (Table). No significant differences were observed in other bilateral joint angles and joint ROMs (Tables). The results of this study also showed that for both male and female speed participants, the left shift of COM during the left gliding phase was significantly greater than the right shift of COM during the right gliding phase (p= 0.001) (Table). Similarly, during the right push-off phase, the left shift of COM was significantly greater than the right shift during the left push-off phase (p= 0.001) (Table). Figure asymmetric left–right shift of the participants’ COM.

The study analyzes the impact of technique asymmetry on elite speed skaters' performance.