@article{QuarmbyKhajooeiEngeletal.2019,
  author    = {Quarmby, Andrew James and Khajooei, Mina and Engel, Tilman and Kaplick, Hannes and Mayer, Frank},
  title     = {The feasibility of a split-belt instrumented treadmill running protocol with perturbations},
  series = {Journal of biomechanics},
  volume    = {98},
  journal   = {Journal of biomechanics},
  publisher = {Elsevier},
  address   = {Oxford},
  issn      = {0021-9290},
  doi       = {10.1016/j.jbiomech.2019.109493},
  pages     = {5},
  year      = {2019},
  abstract  = {Unexpected perturbations during locomotion can occur during daily life or sports performance. Adequate compensation for such perturbations is crucial in maintaining effective postural control. Studies utilising instrumented treadmills have previously validated perturbed walking protocols, however responses to perturbed running protocols remain less investigated. Therefore, the purpose of this study was to investigate the feasibility of a new instrumented treadmill-perturbed running protocol. <br /> Fifteen participants (age = 2 8 +/- 3 years; height = 172 +/- 9 cm; weight = 69 +/- 10 kg; 60\% female) completed an 8-minute running protocol at baseline velocity of 2.5 m/s (9 km/h), whilst 15 one-sided belt perturbations were applied (pre-set perturbation characteristics: 150 ms delay (post-heel contact); 2.0 m/s amplitude; 100 ms duration). Perturbation characteristics and EMG responses were recorded. Bland-Altman analysis (BLA) was employed (bias +/- limits of agreement (LOA; bias +/- 1.96*SD)) and intra-individual variability of repeated perturbations was assessed via Coefficients of Variation (CV) (mean +/- SD). <br /> On average, 9.4 +/- 2.2 of 15 intended perturbations were successful. Perturbation delay was 143 +/- 10 ms, amplitude was 1.7 +/- 0.2 m/s and duration was 69 +/- 10 ms. BLA showed -7 +/- 13 ms for delay, -0.3 +/- 0.1 m/s for amplitude and -30 +/- 10 ms for duration. CV showed variability of 19 +/- 4.5\% for delay, 58 +/- 12\% for amplitude and 30 +/- 7\% for duration. EMG RMS amplitudes of the legs and trunk ranged from 113 +/- 25\% to 332 +/- 305\% when compared to unperturbed gait. This study showed that the application of sudden perturbations during running can be achieved, though with increased variability across individuals. The perturbations with the above characteristics appear to have elicited a neuromuscular response during running.},
  language  = {en}
}
@article{MuellerMartinezValdesStolletal.2018,
  author    = {Mueller, Juliane and Martinez-Valdes, Eduardo Andr{\´e}s and Stoll, Josefine and Mueller, Steffen and Engel, Tilman and Mayer, Frank},
  title     = {Differences in neuromuscular activity of ankle stabilizing muscles during postural disturbances},
  series = {Gait \& posture},
  volume    = {61},
  journal   = {Gait \& posture},
  publisher = {Elsevier},
  address   = {Clare},
  issn      = {0966-6362},
  doi       = {10.1016/j.gaitpost.2018.01.023},
  pages     = {226 -- 231},
  year      = {2018},
  abstract  = {The purpose was to examine gender differences in ankle stabilizing muscle activation during postural disturbances. Seventeen participants (9 females: 27 +/- 2yrs., 1.69 +/- 0.1 m, 63 +/- 7 kg; 8 males: 29 +/- 2yrs., 1.81 +/- 0.1 m; 83 +/- 7 kg) were included in the study. After familiarization on a split-belt-treadmill, participants walked (1 m/s) while 15 right-sided perturbations were randomly applied 200 ms after initial heel contact. Muscle activity of M. tibialis anterior (TA), peroneus longus (PL) and gastrocnemius medialis (GM) was recorded during unperturbed and perturbed walking. The root mean square (RMS; [\%]) was analyzed within 200 ms after perturbation. Co-activation was quantified as ratio of antagonist (GM)/agonist (TA) EMG-RMS during unperturbed and perturbed walking. Time to onset was calculated (ms). Data were analyzed descriptively (mean +/- SD) followed by three-way-ANOVA (gender/condition/muscle; alpha= 0.05). Perturbed walking elicited higher EMG activity compared to normal walking for TA and PL in both genders (p < 0.000). RMS amplitude gender comparisons revealed an interaction between gender and condition (F = 4.6, p = 0.049) and, a triple interaction among gender, condition and muscle (F = 4.7, p = 0.02). Women presented significantly higher EMG-RMS [\%] PL amplitude than men during perturbed walking (mean difference = 209.6\%, 95\% confidence interval = -367.0 to -52.2\%, p < 0.000). Co-activation showed significant lower values for perturbed compared to normal walking (p < 0.000), without significant gender differences for both walking conditions. GM activated significantly earlier than TA and PL (p < 0.01) without significant differences between the muscle activation onsets of men and women (p = 0.7). The results reflect that activation strategies of the ankle encompassing muscles differ between genders. In provoked stumbling, higher PL EMG activity in women compared to men is present. Future studies should aim to elucidate if this specific behavior has any relationship with ankle injury occurrence between genders.},
  language  = {en}
}
@article{MuellerHadzicMugeleetal.2017,
  author    = {M{\"u}ller, Juliane and Hadzic, Miralem and Mugele, Hendrik and Stoll, Josefine and M{\"u}ller, Steffen and Mayer, Frank},
  title     = {Effect of high-intensity perturbations during core-specific sensorimotor exercises on trunk muscle activation},
  series = {Journal of biomechanics},
  volume    = {70},
  journal   = {Journal of biomechanics},
  publisher = {Elsevier},
  address   = {Oxford},
  issn      = {0021-9290},
  doi       = {10.1016/j.jbiomech.2017.12.013},
  pages     = {212 -- 218},
  year      = {2017},
  abstract  = {Core-specific sensorimotor exercises are proven to enhance neuromuscular activity of the trunk. However, the influence of high-intensity perturbations on training efficiency is unclear within this context. Sixteen participants (29 +/- 2 yrs; 175 +/- 8 cm; 69 +/- 13 kg) were prepared with a 12-lead bilateral trunk EMG. Warm-up on a dynamometer was followed by maximum voluntary isometric trunk (flex/ext) contraction (MVC). Next, participants performed four conditions for a one-legged stance with hip abduction on a stable surface (HA) repeated randomly on an unstable surface (HAP), on a stable surface with perturbation (HA + P), and on an unstable surface with perturbation (HAP + P). Afterwards, bird dog (BD) was performed under the same conditions (BD, BDP, BD + P, BDP + P). A foam pad under the foot (HA) or the knee (BD) was used as an unstable surface. Exercises were conducted on a moveable platform. Perturbations (ACC 50 m/sec(2);100 ms duration;10rep.) were randomly applied in the anterior-posterior direction. The root mean square (RMS) normalized to MVC (\%) was calculated (whole movement cycle). Muscles were grouped into ventral right and left (VR;VL), and dorsal right and left (DR;DL). Ventral Dorsal and right-left ratios were calculated (two way repeated-measures ANOVA;alpha = 0,05). Amplitudes of all muscle groups in bird dog were higher compared to hip abduction (p <= 0.0001; Range: BD: 14 +/- 3\% (BD;VR) to 53 +/- 4\%; HA: 7 +/- 2\% (HA;DR) to 16 +/- 4\% (HA;DR)). EMG-RMS showed significant differences (p < 0.001) between conditions and muscle groups per exercise. Interaction effects were only significant for HA (p = 0.02). No significant differences were present in EMG ratios (p > 0.05). Additional high-intensity perturbations during core-specific sensorimotor exercises lead to increased neuromuscular activity and therefore higher exercise intensities. However, the beneficial effects on trunk function remain unclear. Nevertheless, BD is more suitable to address trunk muscles.},
  language  = {en}
}