Towards Subject-Specific Strength Training Design through Predictive Use of Musculoskeletal Models

Michael Plüss; Florian Schellenberg; William R. Taylor; Silvio Lorenzetti

doi:10.1155/2018/9721079

Towards Subject-Specific Strength Training Design through Predictive Use of Musculoskeletal Models

Plüss, Michael;Schellenberg, Florian;Taylor, William R.;Lorenzetti, Silvio 2018-03-19 00:00:00 Hindawi Applied Bionics and Biomechanics Volume 2018, Article ID 9721079, 10 pages https://doi.org/10.1155/2018/9721079 Research Article Towards Subject-Specific Strength Training Design through Predictive Use of Musculoskeletal Models Michael Plüss, Florian Schellenberg, William R. Taylor , and Silvio Lorenzetti Institute for Biomechanics, ETH Zürich, Zürich, Switzerland Correspondence should be addressed to Silvio Lorenzetti; slorenzetti@ethz.ch Received 12 September 2017; Revised 5 January 2018; Accepted 28 January 2018; Published 19 March 2018 Academic Editor: Justin Keogh Copyright © 2018 Michael Plüss et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Lower extremity dysfunction is often associated with hip muscle strength deﬁciencies. Detailed knowledge of the muscle forces generated in the hip under speciﬁc external loading conditions enables speciﬁc structures to be trained. The aim of this study was to ﬁnd the most eﬀective movement type and loading direction to enable the training of speciﬁc parts of the hip muscles using a standing posture and a pulley system. In a novel approach to release the predictive power of musculoskeletal modelling techniques based on inverse dynamics, ﬂexion/extension and ab-/adduction movements were virtually created. To demonstrate the eﬀectiveness of this approach, three hip orientations and an external loading force that was systematically rotated around the body were simulated using a state-of-the art OpenSim model in order to establish ideal designs for training of the anterior and posterior parts of the M. gluteus medius (GM). The external force direction as well as the hip orientation greatly inﬂuenced the muscle forces in the diﬀerent parts of the GM. No setting was found for simultaneous training of the anterior and posterior parts with a muscle force higher than 50% of the maximum. Importantly, this study has demonstrated the use of musculoskeletal models as an approach to predict muscle force variations for diﬀerent strength and rehabilitation exercise variations. 1. Introduction exercises [7–10], inverse dynamics approaches have been used in a subject-speciﬁc manner to enable a comparison between diﬀerent exercise variations [2, 11]. In more sophis- Detailed knowledge of the generated forces within the human ticated analyses, these approaches have been combined with musculoskeletal system provides an important step towards muscle optimization techniques in order to compare forces understanding the conditions that are required to eﬀectively in the diﬀerent parts of the hamstring and quadriceps train for speciﬁc sports or undertaking targeted rehabilitation muscles between training exercises, including consideration after injury or during therapy. Ideally, direct measurements of execution form and joint angles [10]. Performing muscu- of the internal loading conditions such as muscle and joint contact forces would guide such training approaches, but loskeletal modelling requires assumptions regarding the anthropometry of the segments, shape and degree of freedom these are diﬃcult to access [1, 2]. Here, while detailed data- of the joints, muscular properties, and optimization criteria sets of kinematics and kinetics are becoming more widely [12] Importantly, an in-depth understanding of the condi- available [3], such approaches are currently limited both to tions under which these models are valid and able to cor- small populations with artiﬁcial joints as well as to only very speciﬁc sites in the human body [4–6]. As a result, musculo- rectly predict the internal loading conditions during squatting exercises has already been performed [13]. In an skeletal simulation is the primary tool for estimating internal analysis using videoﬂuoroscopy and instrumented implants, loading conditions throughout the human body, albeit indi- we have been able to demonstrate a ﬂexion-dependent error rectly, by means of inverse dynamics and numerical optimi- in the predicted joint contact forces, but a good estimation zation processes [2]. In the ﬁrst steps towards understanding the interactions (e.g., within 20%) over the range of 25–65 knee ﬂexion [13]. However, despite their ability to calculate internal between kinematics and kinetics during strength training 2 Applied Bionics and Biomechanics In combination with cable exercises to provide targeted loading conditions throughout the musculoskeletal system, one issue that has limited the applicability of musculoskeletal force application, it therefore seems entirely plausible that modelling techniques for predicting the outcome of new musculoskeletal modelling approaches based on systemati- cally altered kinematic and kinetic data could provide a exercise design is the requirement that inverse dynamics approaches are provided with known segment kinematics powerful tool for designing targeted strength and rehabili- as a modelling input. By systematically modifying the exter- tation training exercises. Therefore, the aim of this study nal loading conditions, the use of these models could provide was to evaluate the forces of hip muscles with respect to range a basis for designing or improving training and rehabilitation of motion and their lengths during sagittal and frontal simulated hip strength exercises, using a musculoskeletal programs for targeting speciﬁc musculoskeletal structures, thus opening a predictive capability of the approaches that model driven by a systematic modiﬁcation of the external has not yet been exploited. force direction. One area that could beneﬁt from the power of such pre- dictive options is the focused training of hip musculature, 2. Materials and Methods strength deﬁciencies, and muscular imbalance, which, until now, has generally been investigated with respect to injury. 2.1. Description of the Exercise. Speciﬁc strength training An example of the association between adductor injury and exercises for the hip muscles on the cable machine were sim- hip strength can be seen in the frequency of adductor strains ulated (Figure 1). For these exercises, the cable is usually ﬁxed in ice hockey players, with injured players exhibiting an 18% with a strap to the shank slightly above the ankle joint and the lower hip adduction strength [14]. Importantly, the risk of pulley position is set as low as required in order to ensure a adductor strain injury was shown to be almost 17 times horizontal force vector. These exercises are single-joint and higher in players where the adductor strength was below free-leg exercises. By varying the body orientation relative 80% of the abductor strength. Furthermore, recovery of the to the cable machine and the movement in the diﬀerent ana- iliotibial band syndrome in long-distance runners [15] and tomical planes, muscle activation changes and thus targeted pain [16] in subjects with retropatellar pain syndrome was muscles for strengthening can be chosen appropriately. The improved with a gain in the strength of the hip abductor hip ﬂexor and extensor muscles are then targeted by posi- muscles. However, current strength training instructions tioning the body backwards and forwards relative to the cable are mostly based on the experience of the coach or physio- machine, respectively. A lateral orientation of the body and therapist and are rarely evidence based. This is possibly due movement in the frontal plane will target the hip adductor to the complexity of the hip muscles, which include large and abductor muscles. cross-sectional areas with diﬀerent parts of the same muscle active for diﬀerent functional tasks, as well as diﬀerent lines 2.2. Musculoskeletal Model. The open source software Open- of action and moment arms around the joint that vary with Sim (OpenSim SimTK 3.2, Stanford, USA) was used to joint angles and muscle activation. As a result, speciﬁc guide- perform the simulation [18]. All the ﬁles required for the lines on how to strengthen speciﬁc parts of the hip muscles, simulation, including motion and external force ﬁles, were including the direction of the external force and the joint created in Matlab (R2015a MathWorks, Natick, Massachu- motion, are missing in the literature. It is therefore clear that setts, USA). For the OpenSim simulation, the Arnold Lower detailed knowledge of the interaction between the form of Limb Model 2010 [19] was used. For the hip joint contact rehabilitation/strength exercise and the internal forces gener- force, this model has been validated using an instrumented ated in diﬀerent parts of the hip muscles could lead to an hip implant [20]. To apply the external loading force, a cylin- evidence-based design of training exercises for prevention der was attached rigidly to the right leg of the model to and rehabilitation programs that focus on either muscular represent the ankle strap used in the strength exercises with weakness or imbalance. a cable machine. The cylinder was characterized by the fol- Compared to strength exercises for the hip muscles that lowing dimensions: radius was set to 0.05 m, thickness include multijoint motion such as squatting, cable exercises 0.001 m, height 0.04 m, and mass 0.078 kg. The attachment enable an isolated movement of the hip joint as well as a spe- location in the Lower Limb Model 2010 was at 0.339 m in ciﬁc force magnitude and direction to be applied. In addition, the distal direction of the tibia coordinate system. cable exercises enable preferential muscle force that does not aﬀect forces and movements in other joints and is therefore a 2.3. Kinematics. For kinematic inputs into the model, two simple exercise to be simulated. Using such a pulley system, diﬀerent motions were created at a frequency of 110 Hz. prone hip extension and straight leg raises were used in com- For each, a sine-shaped movement velocity time curve was bination with musculoskeletal models to investigate the mag- used, with a maximum movement speed of 40 degrees per sec- nitude and direction of hip muscle forces [17]. Their results ond. One motion represented a hip ﬂexion/extension (F/E) showed that the hip joint forces were aﬀected by hip joint movement and was performed in the sagittal plane, while the position and partially by alternations in muscle force contri- second one characterized hip abduction and adduction bution. Such studies demonstrate the importance of muscu- (Abd/Add) and was executed in the frontal plane (Figure 1, loskeletal modelling approaches to provide science-based top). The start and ﬁnish points of the F/E movement were evidence for understanding the internal muscle and force both set at −20 -extended hip, since the Lower Limb Model interactions towards guiding training and rehabilitation and 2010 wasvalidated within thisextension range only.The rever- hence positive adaptation of the tissues. sal point of the movement was set at 60 hip ﬂexion, enabling a Applied Bionics and Biomechanics 3 Flexion/extension Abduction/adduction 90° 0° 15° 15° 180° 90° 270° 0° 270° 180° Internal Internal External External hip rotation (−40°) hip rotation (−40°) hip rotation (40°) hip rotation (40°) Neutral Neutral (b) (a) Figure 1: Simulation illustration (top row) of the model performing the ﬂexion/extension (a) and the abduction/adduction (b) movements, including the 0 position of the external force (green arrow) applied to the right leg of the model. Schematic representations of the diﬀerent loading conditions used in the simulation are shown in the bottom row, including rotational external force (green), which was rotated incrementally in 15 steps and three diﬀerent hip rotation conﬁgurations; externally rotated (blue), neutrally rotated (grey), and internally rotated (orange). total range of motion (RoM) of 80 . The Abd/Add movement (force direction), to estimate the internal muscle force mag- started with a −35 abducted hip position, where the reversal nitudes, in which the sum of the squared muscle activation point of the movement was deﬁned at 5 hip adduction, result- was minimized. This combination led to 138 individual sim- ing in a 40 RoM in the frontal plane. Each limb movement was ulations. Some simulations were run without the individual then simulated with the hip rotated at one of the three follow- wrapping surfaces to enable successful simulation: for the ° ° ing conﬁgurations: neutral (0 ), internally rotated (40 ), or F/E movement in the neutral hip position, the wrapping sur- externally rotated (−40 ) (Figure 1, bottom). In some cases, face of the M. pectineus (PECT_at_femshaft_r) and, in the the eccentric phase and in some the concentric phase, depen- externally rotated hip position, the wrapping surface of the dent on the actual direction of the force, were at the start of M. adductor brevis (AB_aft_femshaft_r) and the proximal themotion. Thetimewas6.28 sforF/E and3.14 sforAbd/Add. part of the M. adductor magnus (AMprox_at_femshaft_r) were disabled due to simulation errors. 2.4. Kinetics. An external force with a magnitude of 100 N 2.6. Evaluation of the Data. The muscle activations A of M. was applied to the centre of the attached cylinder at the shank adductor longus, M. adductor magnus, M. gluteus medius, of the model. This force represents a typical load used in a M. rectus femoris, and M. semimembranosus were calculated health-oriented strength training including the here-used for all hip rotation conﬁgurations and external force direc- cable exercises. In all diﬀerent movements and throughout tions as follows: the whole cycle, the external force remained parallel to the ground. For each movement conﬁguration, diﬀerent external force directions were used to examine the inﬂuence of the act A = , 1 position of the cable machine to the body. Starting in a dorsal F max direction for F/E and medially for Abd/Add simulations, the where F is the acting muscle force and F is the maximal external force was then rotated incrementally by 15 degrees act max possible muscle force of the speciﬁc part of the muscle. For in a counter-clockwise direction until a complete rotation concentric contractions, the activation lies between 0 and 1. of the force was obtained, leading to 23 individual simula- To properly model the anatomical characteristics, the M. tions (Figure 1, bottom). adductor magnus and M. gluteus medius were included with 2.5. Musculoskeletal Simulation. A quasistatic optimization diﬀerent parts in the Lower Limb Model 2010, which were was performed for all movements (F/E and Abd/Add), all also maintained in the analysis of the parameters. 3D surface ° ° ° hip rotations (0 ,40 , and −40 ), and all kinetic parameters plots were then used to visualise the muscular activation, Loading Movement 4 Applied Bionics and Biomechanics For the anterior GM, the largest muscle length changes were which was dependent on the joint angle as well as on the angle of the external force. Additionally, for all three hip rota- observed during the Abd/Add movement, while for the pos- tion positions, the maximal activations for each external terior GM, the externally rotated position caused similar acti- force angle were calculated and displayed in spider plots. Fur- vations and changes in muscle lengths in both the F/E and thermore, the muscle lengths and the corresponding muscle Abd/Add movements. activations for all three hip rotation positions were analysed at the angle of the external force where the highest activation 4. Discussion level occurred. All data evaluation and plot generation was performed in Matlab (R2014b, MathWorks, Inc.). After ini- In order to further improve rehabilitation exercises and to tial review of the simulation data, only the muscles M. adduc- estimate the internal mechanical load of the speciﬁc parts tor longus, M. rectus femoris, M. semimembranosus, and M. of the targeted muscles, it is essential that their activation is rectus femoris were evaluated. Furthermore, the anterior and known, with respect to the chosen movement and external posterior parts of the M. gluteus medius (GM) were chosen loading conditions. In this study, classic hip strength and for in-depth analysis, due to the fact that his muscle repre- rehabilitation exercises with a F/E and an Abd/Add move- sents one of the major target structures of this type of cable ment using a cable machine were simulated by means of exercises and the medial part had an activation lower than 0.5. whole-body musculoskeletal simulation with the aim to quantify muscle activation and lengths during diﬀerent kine- matic and kinetic conﬁgurations. To simulate the strength 3. Results exercises, loading and movement patterns were generated The adductor muscles’ activation remained low for all load- and analysed using diﬀerent directions of the cable with respect to the body, as well as using two movements with ing conditions; except for M. adductor longus, the two rotated hip positions showed higher activities in the F/E movement three diﬀerent hip rotation positions, internally, neutrally, than in Abd/Add. As expected, the M. rectus femoris exhib- and externally rotated. In order to quantify the activation of ited a higher activation for F/E movement than for Abd/ the individual hip muscles and their parts, muscle activation Add movement. Similar results, but in the opposite direction was estimated by means of static optimization using a full of the rotating external force, were observed for the M. semi- body musculoskeletal model as well as targeted kinetic and membranosus, an antagonist of M. rectus femoris. Addition- kinematic conditions. ally, activation in the hamstrings muscles were reduced Although previous models have attempted to modify the when the hip was rotated externally. An agonist/antagonist kinematics of a joint for use in inverse dynamics modelling relationship was clearly visible between the anterior and pos- [21], the external forces imposed on such models are gener- terior parts of the GM during the F/E movement with the hip ally known (from, e.g., ground reaction force plates) and in a neutral position (Figure 2). In this position, the F/E not altered. To our knowledge, this approach, where the movements led to higher muscular activation compared to external loading conditions were systematically varied, was Abd/Add movements. The activation levels for the abductors used for the ﬁrst time in an approach that seems to lend itself versus the adductors remained rather equal, but this was nicely towards the design of targeted training strategies somewhat diﬀerent during F/E, where the activation seemed through identiﬁcation of the optimal movement and loading to increase exponentially towards a dominant maximum condition to speciﬁcally train a certain musculature. Here, level. Furthermore, the activation of the anterior GM part the use of a purposefully designed hip-strengthening pro- was considerably larger within the extension range (negative gram can be beneﬁcial for patients as well as athletes. angles) of the movement than within the ﬂexion range (pos- Whereas it is well known that the direction of the force itive, Figure 2(a)), where the posterior part of the muscle deﬁnes the muscle activation pattern, this work aimed also increased in activation (Figure 2(c)). to show the importance of the hip rotation position. As an With the hip rotated externally, the posterior part of the example, by including strength training exercises for abduc- GM achieved a maximum activation during the ﬂexion and tor muscles and internal rotation in the hip, Khayambashi abduction movements (Figures 2–4). On the contrary, an and coworkers [22] showed an improvement of pain and internally rotated hip position led to maximum activation health status in women with patellofemoral pain syndrome levels for the anterior part of the muscle, compared to neutral compared to a no-exercise control group. Whole-body simu- and external rotated hip positions for both movements. lation, similar to that performed in our study, might help in For the anterior part of the GM, maximum activation was the future to speciﬁcally design an eﬃcient subject-speciﬁc achieved in the internally rotated hip position for external workout program. ° ° forces from 180–300 Overall, the relatively small magnitude of 100 N of the during F/E and from 0–45 as well as 240–360 for the Abd/Add movement. On the other hand, external force did not cause high activations for the M. rectus the posterior part of the GM exhibited maximum activation femoris, M. adductor longus, M. semimembranosus, and M. adductor magnus. Interestingly, the M. adductor longus in the externally rotated conﬁguration with an external force ° ° direction of 240–315 and 300–315 for the F/E and Abd/Add showed higher activations for the F/E movement than for the Abd/Add. Here, this speciﬁc behaviour, together with movements, respectively (Figure 5). While changing the rota- tion position of the hip had an inﬂuence on muscle length, increased loading, could be of interest for this muscle, since changing the external force angles within one movement as an adductor muscle, a higher activation in the Abd/Add could be expected. Contrary to the low activation levels of conﬁguration had no eﬀect on the muscle length (Figure 6). Applied Bionics and Biomechanics 5 M. gluteus medius neutral hip position Flexion/extension movement Abduction/adduction movement 1 1 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 50 50 0 0 345 345 40 40 −5 −5 30 30 270 270 270 270 −10 −10 0 20 20 −15 −15 180 180 180 180 10 10 −20 −20 −25 25 0 0 90 90 90 90 −10 10 −30 −20 0 −35 Rotating force angle (°) Rotating force angle (°) Movement angle (°) Movement angle (°) (a) (b) 1 1 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0 0 60 5 0 0 50 50 345 3 345 3 −5 −5 40 40 −10 −10 10 30 30 270 270 270 270 −15 −15 20 20 180 180 180 180 −20 −20 10 10 −25 25 0 0 90 90 90 −30 −10 0 0 −35 −20 Rotating force angle (°) Rotating force angle (°) Movement angle (°) Movement angle (°) (c) (d) Figure 2: Muscle activations [0→ 1] as a function of the movement angle and all external force orientations in the neutral hip position. The anterior (a, b) and the posterior (c, d) parts of the M. gluteus medius (GM) are displayed for ﬂexion/extension (a, c) and abduction/adduction (b, d) movements. the other muscles, GM showed high and alternative activa- length is required [23]. For optimal training, the movement tions with changed kinematic and kinetic conﬁgurations. with the largest change in muscle length, together with an Therefore, the diﬀerent parts of the GM were further ana- external force direction that causes the highest muscle force over the whole movement, should be chosen. Please note lysed. Since the middle part of the GM did not achieve activa- tions larger than 0.5, which would lead to a more eﬃcient that, in this work, the muscle activation was calculated as training stimuli, only the anterior and posterior parts were the actual muscle force normalized by the maximum isomet- included in the in-depth analysis of the GM muscle. ric muscle force. As an example, for the anterior GM part, In all three examined hip rotations, activation patterns of both exercise movements with an internally rotated hip either the anterior or the posterior part was examined showed a high activation (Figure 4), but the Abd/Add move- (Figures 2–4). However, external or internal rotation of the ment also caused a greater change in muscle length hip resulted in a higher muscular activation level compared (Figure 6). Therefore, based on our results, it could be recom- to the neutral position, which can be explained by the sup- mended to train the anterior part of the GM with an inter- portive function of these muscle parts for the hip rotation nally rotated hip position using the direction of the external itself. Rotating the hip also inﬂuenced the length of the M. force in the range of the maximum activation at about 45– gluteus medius during the exercise. In order to provide the 240 . For the posterior part, similar maximum activities and most eﬀective training impulse to the target muscle, large changes in muscle length were achieved with an externally muscle forces over the maximum possible change in muscle rotated hip. Regarding the muscle length and the posterior Posterior muslce part Anterior muslce part Activation Activation Activation Activation 6 Applied Bionics and Biomechanics M. gluteus medius external rotation hip position Flexion/extension movement Abduction/adduction movement 1 1 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0 0 60 5 50 0 40 345 −5 345 30 −10 270 270 20 −15 10 180 −20 180 0 −25 90 90 −10 −30 Rotating force angle (°) Rotating force angle (°) −20 −35 0 0 Movement angle (°) Movement angle (°) (a) (b) 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0 0 50 0 40 345 −5 345 30 −10 270 270 20 −15 10 180 −20 180 0 −25 90 90 −10 −30 Rotating force angle (°) Rotating force angle (°) −20 −35 0 0 Movement angle (°) Movement angle (°) (c) (d) Figure 3: Muscle activations [0→ 1] as a function of the movement angle and external force orientations in the externally rotated hip position. The anterior (a, b) and the posterior (c, d) parts of the M. gluteus medius (GM) are displayed for ﬂexion/extension (a, c) and abduction/adduction (b, d) movements. part, two aspects should be considered. Firstly, the highest the anterior part, an internally rotated hip is recommended, activation in the F/E movement occurred at large muscle while the highest loading for the posterior part can be length (around 0.14 m, Figure 6(a)) whereas during Abd/ achieved using an externally rotated hip. Muscle force is known to be highly dependent on force- Add, the highest activation was observed at the shortest mus- cle length (around 0.08 m, Figure 6(b)). Secondly, a greater velocity and force-length relationships (Hill-type muscles muscle length change was observed during F/E movement [24, 25]). As a result, muscle activation is directly linked to than during Add/Abb. Taking these two factors into account, the maximal isometric force capacity of the muscle, the rele- training the posterior part should be performed in an exter- vant lever arm of the speciﬁc muscle, and the external loading nally rotated hip position with a F/E movement to achieve conditions. For the sake of completeness, the force-velocity an eﬀective training regime. In summary, our ﬁndings show relationship will only play a minor role, since the velocity of the importance of properly choosing a suitable hip position, the movement during strength training is 1.7% of the maxi- movement direction, and external force direction, in order mal shortening velocity and furthermore, the external load- to achieve the desired training goals. To eﬀectively train the ing changes only a few percent between the acceleration anterior and posterior parts of the GM, F/E movement seems and deceleration phases [26]. to be preferable but using two diﬀerent loading conﬁgura- Several limitations arising in this study need to be men- tions due to the supportive functions of the two parts of the tioned. Firstly, within the chosen model conﬁguration and muscle in opposite hip rotation directions: for training of the kinematic pathways investigated in this study, only 5 of Posterior muslce part Anterior muslce part Activation Activation Activation Activation Applied Bionics and Biomechanics 7 M. gluteus medius internal rotated hip position Flexion/extension movement Abduction/adduction movement 1 1 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0 0 345 −5 345 −10 270 270 −15 180 −20 180 −25 90 90 −10 −30 −20 −35 0 0 Rotating force angle (°) Rotating force angle (°) Movement angle (°) Movement angle (°) (a) (b) 1 1 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0 0 345 −5 345 −10 270 270 −15 180 −20 180 −25 90 90 −10 −30 Rotating force angle (°) Rotating force angle (°) −20 −35 0 0 Movement angle (°) Movement angle (°) (c) (d) Figure 4: Muscle activations [0→ 1] as a function of the movement angle and external force orientations in the internally rotated hip position. The anterior (a, b) and the posterior (c, d) parts of the M. gluteus medius (GM) are displayed for ﬂexion/extension (a, c) and abduction/adduction (b, d) movements. adduction in the hip could be tolerated to avoid changes in little is known about the large muscles and their changing other joint angles such as the knee angle of the opposite leg. lever arms during dynamic activities. As a result, the accuracy Changes in muscle activation patterns might also occur if of the muscle force and activations estimated in this study larger adduction angles are taken into account. Secondly, cannot be guaranteed. It is therefore imperative that further investigations using subject cohorts and, for example, elec- the force was applied parallel to the ground and by assuming a cylinder around the ankle representing an ankle strap. This tromyography (EMG) measurements, within their possibili- conﬁguration would not represent the loading direction of a ties and limitations [34] are undertaken to test the validity real cable exercise machine and could inﬂuence the results in of these approaches in vivo, followed by training studies to a complex manner. Thirdly and possibly most importantly, assess the relationships between these acute measures and the results of this study are based on the use of a reference longitudinal outcomes [35]. Further enhancement to the and nonscaled musculoskeletal model. It is well known that accuracy and reliability of musculoskeletal models would the exact musculoskeletal conﬁguration, including, for exam- therefore only further improve the estimations of muscle ple, anatomical insertion sites [27], the deﬁnition of joint forces and the speciﬁc training regime strategies. However, centres and axes [28–30], physiological cross-sectional areas despite these limitations, the approaches used in this study of the muscle [31], the scaling procedure [32], and the lever do open perspectives for providing targeted exercise plans arms of the muscles around the joints [33] all play critical on a subject-speciﬁc basis or comparing muscle activations roles on the estimation of internal loading conditions using within diﬀerent strength exercises by systematically varying musculoskeletal models. Particularly in the hip, relatively, the external force and the movement. As a result, speciﬁc Posterior muslce part Anterior muslce part Activation Activation Activation Activation 8 Applied Bionics and Biomechanics Flexion/extension Abduction/adduction 0° 90° 330° 30° 120° 60° 150° 300° 60° 30° 90° 270° 180° 0° 120° 210° 330° 240° 150° 300° 210° 240° 180° 270° Neutral Neutral Ant. part GM Ant. part GM Ext. rot Ext. rot Post. part GM Post. part GM Int. rot Int. rot (a) (b) Figure 5: M. gluteus medius (GM) maximum muscle activations of the anterior (solid line) and posterior (dashed line) parts for the ﬂexion/ extension (a) and abduction/adduction (b) movements and the diﬀerent hip rotation conﬁgurations: neutral (black), externally rotated (red), and internally rotated (blue). The initial external force direction (0 ) of the movement is indicated by the green arrow. Flexion/extension Adduction/abduction 1 285° 1 0.9 0.9 270° 0.8 0.8 0° 0.7 0.7 0.6 0.6 0.5 0.5 240° 330° 0.4 0.4 330° 0.3 0.3 0.2 0.2 300° 0.1 0.1 255° 255° 240° 240° 285° 0 0 0.08 0.1 0.12 0.14 0.16 0.08 0.1 0.12 0.14 0.16 Muscle length (m) Muscle length (m) Neutral Neutral Ant. part GM Ant. part GM Ext. rot Ext. rot Post. part GM Post. part GM Int. rot Int. rot (a) (b) Figure 6: Progression of the M. gluteus medius (GM) muscle activations [0→ 1] with external force direction. The overall maximum activations of all force directions are shown for the anterior (solid line) and posterior (dashed line) part of the muscle as a function of the muscle length (m) for the ﬂexion/extension (a) and abduction/adduction (b) movements, as well as for the diﬀerent hip rotation conﬁgurations: neutral (black), externally rotated (red), and internally rotated (blue). exercises could be identiﬁed in order to achieve optimal load- 5. Perspectives ing patterns. Furthermore, by including subject-speciﬁcdeﬁ- cits, eﬃcient rehabilitation regimes could be designed and Using musculoskeletal simulation and systematic variation of even updated to follow the rehabilitation progress, if the loading conditions, this study opens perspectives for the adaptation of muscle status could be quantiﬁed. identiﬁcation of optimal training exercises for speciﬁc muscle 0.75 0.5 0.25 0.75 0.5 0.25 Activation Activation Applied Bionics and Biomechanics 9 subjects,” Journal of Biomechanics, vol. 43, no. 11, pp. 2164– strengthening in rehabilitation and sports medicine. By 2173, 2010. selecting hip rotation and body positions relative to the cable exercise machine, higher muscle activation and large changes [7] R. List, T. Gülay, M. Stoop, and S. Lorenzetti, “Kinematics of the trunk and the lower extremities during restricted and unre- in muscle lengths can be achieved. To eﬀectively train the stricted squats,” Journal of Strength and Conditioning anterior and posterior parts of GM muscle, two diﬀerent Research, vol. 27, no. 6, pp. 1529–1538, 2013. exercises likely need to be performed. While an internally [8] S. Lorenzetti, T. Gülay, M. Stoop et al., “Comparison of the rotated hip is recommended to train the anterior part, the pos- angles and corresponding moments in the knee and hip during terior part should be trained using an externally rotated hip. restricted and unrestricted squats,” Journal of Strength and The application of these approaches in this relevant example Conditioning Research, vol. 26, no. 10, pp. 2829–2836, 2012. demonstrates that precisely validated models, fed with kine- [9] F. Schellenberg, J. Lindorfer, R. List, W. R. Taylor, and matics of diﬀerent exercises, could provide a powerful option S. Lorenzetti, “Kinetic and kinematic diﬀerences between to compare the eﬀectiveness of exercises that target speciﬁc deadlifts and goodmornings,” BMC Sports Science Medicine muscles. For more complex and dynamic exercises, a similar & Rehabilitation, vol. 5, no. 1, p. 27, 2013. approach might be possible in the future. Here, the general [10] F. Schellenberg, N. Schmid, R. Häberle, N. Hörterer, W. R. procedure is similar, but the musculoskeletal model needs to Taylor, and S. Lorenzetti, “Loading conditions in the spine, enable large and extreme joint angles, individual muscle, hip and knee during diﬀerent executions of back extension segmental and joint properties, and subject-speciﬁc scaling. exercises,” BMC Sports Science Medicine & Rehabilitation, vol. 9, no. 1, p. 10, 2017. Conflicts of Interest [11] P. Schütz, R. List, R. Zemp, F. Schellenberg, W. R. Taylor, and S. Lorenzetti, “Joint angles of the ankle, knee, and hip and The authors declare no conﬂict of interests. loading conditions during split squats,” Journal of Applied Bio- mechanics, vol. 30, no. 3, pp. 373–380, 2014. [12] A. Erdemir, S. McLean, W. Herzog, and A. J. van den Bogert, Authors’ Contributions “Model-based estimation of muscle forces exerted during Michael Plüss performed the study, analysed the results, and movements,” Clinical Biomechanics, vol. 22, no. 2, pp. 131– 154, 2007. wrote the manuscript. Florian Schellenberg and Silvio Loren- zetti designed and supervised the study, analysed the data, [13] F. Schellenberg et al., “Musculoskeletal squat simulation evalu- and helped in drafting the manuscript. William R. Taylor ation by means of an instrumented total knee arthroplasty,” in 34th International Conference on Biomechanics in Sports, Tsu- provided the original concept and helped in drafting and kuba, Japan, July 2016. editing the manuscript. [14] T. F. Tyler, S. J. Nicholas, R. J. Campbell, and M. P. McHugh, “The association of hip strength and ﬂexibility with the inci- Supplementary Materials dence of adductor muscle strains in professional ice hockey players,” The American Journal of Sports Medicine, vol. 29, The MATLAB code to generate the velocity proﬁle as well as no. 2, pp. 124–128, 2001. all the OpenSim ﬁles is available here http://www.movement. [15] M. Fredericson, C. L. Cookingham, A. M. Chaudhari, B. C. ethz.ch/data-repository.html. (Supplementary Materials) Dowdell, N. Oestreicher, and S. A. Sahrmann, “Hip abductor weakness in distance runners with iliotibial band syndrome,” References Clinical Journal of Sport Medicine, vol. 10, no. 3, pp. 169– 175, 2000. [1] M. G. Pandy, “Computer modeling and simulation of human [16] T. R. T. Santos, B. A. Oliveira, J. M. Ocarino, K. G. Holt, and S. T. movement,” Annual Review of Biomedical Engineering, vol. 3, Fonseca, “Eﬀectiveness of hip muscle strengthening in patello- no. 1, pp. 245–273, 2001. femoral pain syndrome patients: a systematic review,” Brazilian [2] F. Schellenberg, K. Oberhofer, W. R. Taylor, and S. Lorenzetti, Journal of Physical Therapy, vol. 19, no. 3, pp. 167–176, 2015. “Review of modelling techniques for in vivo muscle force esti- [17] C. L. Lewis, S. A. Sahrmann, and D. W. Moran, “Eﬀect of posi- mation in the lower extremities during strength training,” tion and alteration in synergist muscle force contribution on Computational and Mathematical Methods in Medicine, hip forces when performing hip strengthening exercises,” Clin- vol. 2015, Article ID 483921, 12 pages, 2015. ical Biomechanics, vol. 24, no. 1, pp. 35–42, 2009. [3] W. R. Taylor, P. Schütz, G. Bergmann et al., “A comprehensive [18] S. L. Delp, F. C. Anderson, A. S. Arnold et al., “OpenSim: open- assessment of the musculoskeletal system: the CAMS-knee source software to create and analyze dynamic simulations of data set,” Journal of Biomechanics, vol. 65, pp. 32–39, 2017. movement,” IEEE Transactions on Biomedical Engineering, [4] D. D. D'Lima, N. Steklov, B. J. Fregly, S. A. Banks, and C. W. vol. 54, no. 11, pp. 1940–1950, 2007. Colwell Jr, “In vivo contact stresses during activities of daily [19] E. M. Arnold, S. R. Ward, R. L. Lieber, and S. L. Delp, “A model living after knee arthroplasty,” Journal of Orthopaedic of the lower limb for analysis of human movement,” Annals of Research, vol. 26, no. 12, pp. 1549–1555, 2008. Biomedical Engineering, vol. 38, no. 2, pp. 269–279, 2010. [5] P. Damm, I. Kutzner, G. Bergmann, A. Rohlmann, and H. Schmidt, “Comparison of in vivo measured loads in knee, [20] L. Modenese, A. T. M. Phillips, and A. M. J. Bull, “An open source lower limb model: hip joint validation,” Journal of Bio- hip and spinal implants during level walking,” Journal of Bio- mechanics, vol. 51, pp. 128–132, 2017. mechanics, vol. 44, no. 12, pp. 2185–2193, 2011. [6] I. Kutzner, B. Heinlein, F. Graichen et al., “Loading of the knee [21] M. O. Heller, W. R. Taylor, C. Perka, and G. N. Duda, “The joint during activities of daily living measured in vivo in ﬁve inﬂuence of alignment on the musculo-skeletal loading 10 Applied Bionics and Biomechanics conditions at the knee,” Langenbeck's Archives of Surgery, vol. 388, no. 5, pp. 291–297, 2003. [22] K. Khayambashi, Z. Mohammadkhani, K. Ghaznavi, M. A. Lyle, and C. M. Powers, “The eﬀects of isolated hip abductor and external rotator muscle strengthening on pain, health sta- tus, and hip strength in females with patellofemoral pain: a randomized controlled trial,” Journal of Orthopaedic & Sports Physical Therapy, vol. 42, no. 1, pp. 22–29, 2012. [23] B. J. Schoenfeld, “The mechanisms of muscle hypertrophy and their application to resistance training,” The Journal of Strength and Conditioning Research, vol. 24, no. 10, pp. 2857–2872, 2010. [24] D. G. Thelen, “Adjustment of muscle mechanics model parameters to simulate dynamic contractions in older adults,” Journal of Biomechanical Engineering, vol. 125, no. 1, pp. 70– 77, 2003. [25] G. T. Yamaguchi and F. E. Zajac, “A planar model of the knee joint to characterize the knee extensor mechanism,” Journal of Biomechanics, vol. 22, no. 1, pp. 1–10, 1989. [26] T. Gülay, R. List, and S. Lorenzetti, “Moments in the knee and hip during descent and ascent of squats,” Portuguese Journal of Sport Sciences, vol. 11, Supplement 2, 2011. [27] S. Pal, J. E. Langenderfer, J. Q. Stowe, P. J. Laz, A. J. Petrella, and P. J. Rullkoetter, “Probabilistic modeling of knee muscle moment arms: eﬀects of methods, origin-insertion, and kine- matic variability,” Annals of Biomedical Engineering, vol. 35, no. 9, pp. 1632–1642, 2007. [28] R. M. Ehrig, M. O. Heller, S. Kratzenstein, G. N. Duda, A. Trepczynski, and W. R. Taylor, “The SCoRE residual: a quality index to assess the accuracy of joint estimations,” Jour- nal of Biomechanics, vol. 44, no. 7, pp. 1400–1404, 2011. [29] W. R. Taylor, E. I. Kornaropoulos, G. N. Duda et al., “Repeat- ability and reproducibility of OSSCA, a functional approach for assessing the kinematics of the lower limb,” Gait & Posture, vol. 32, no. 2, pp. 231–236, 2010. [30] S. L. Delp and W. Maloney, “Eﬀects of hip center location on the moment-generating capacity of the muscles,” Journal of Biomechanics, vol. 26, no. 4-5, pp. 485–499, 1993. [31] G. Massey, P. Evangelidis, and J. Folland, “Inﬂuence of con- tractile force on the architecture and morphology of the quad- riceps femoris,” Experimental Physiology, vol. 100, no. 11, pp. 1342–1351, 2015. [32] F. Schellenberg, W. R. Taylor, I. Jonkers, and S. Lorenzetti, “Robustness of kinematic weighting and scaling concepts for musculoskeletal simulation,” Computer Methods in Biome- chanics and Biomedical Engineering, vol. 20, no. 7, pp. 720– 729, 2017. [33] A. Navacchia, V. Kefala, and K. B. Shelburne, “Dependence of muscle moment arms on in vivo three-dimensional kinematics of the knee,” Annals of Biomedical Engineering, vol. 45, no. 3, pp. 789–798, 2017. [34] A. D. Vigotsky, I. Halperin, G. J. Lehman, G. S. Trajano, and T. M. Vieira, “Interpreting signal amplitudes in surface elec- tromyography studies in sport and rehabilitation sciences,” Frontiers in Physiology, vol. 8, p. 985, 2018. [35] I. Halperin, A. D. Vigotsky, C. Foster, and D. B. Pyne, “Strengthening the practice of exercise and sport science,” International Journal of Sports Physiology and Performance, pp. 1–8, 2018. International Journal of Advances in Rotating Machinery Multimedia Journal of The Scientific Journal of Engineering World Journal Sensors Hindawi Hindawi Publishing Corporation Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 http://www www.hindawi.com .hindawi.com V Volume 2018 olume 2013 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 Journal of Control Science and Engineering Advances in Civil Engineering Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 Submit your manuscripts at www.hindawi.com Journal of Journal of Electrical and Computer Robotics Engineering Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 VLSI Design Advances in OptoElectronics International Journal of Modelling & Aerospace International Journal of Simulation Navigation and in Engineering Engineering Observation Hindawi Hindawi Hindawi Hindawi Volume 2018 Volume 2018 Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com www.hindawi.com www.hindawi.com Volume 2018 International Journal of Active and Passive International Journal of Antennas and Advances in Chemical Engineering Propagation Electronic Components Shock and Vibration Acoustics and Vibration Hindawi Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Applied Bionics and Biomechanics Hindawi Publishing Corporation http://www.deepdyve.com/lp/hindawi-publishing-corporation/towards-subject-specific-strength-training-design-through-predictive-IMT532rZfR

Loading next page...

References (37)

G. Yamaguchi, G. Yamaguchi, F. Zajac, F. Zajac (1989)
A planar model of the knee joint to characterize the knee extensor mechanism.
Journal of biomechanics, 22 1
W. Taylor, E. Kornaropoulos, Georg Duda, S. Kratzenstein, R. Ehrig, Adamantios Arampatzis, M. Heller (2010)
Repeatability and reproducibility of OSSCA, a functional approach for assessing the kinematics of the lower limb.
Gait & posture, 32 2
A. Erdemir, S. McLean, W. Herzog, A. Bogert (2007)
Model-based estimation of muscle forces exerted during movements.
Clinical biomechanics, 22 2
I. Kutzner, B. Heinlein, B. Heinlein, F. Graichen, A. Bender, A. Rohlmann, A. Halder, A. Beier, G. Bergmann (2010)
Loading of the knee joint during activities of daily living measured in vivo in five subjects.
Journal of biomechanics, 43 11
Florian Schellenberg, Nicole Schmid, Ramona Häberle, Nicole Hörterer, W. Taylor, S. Lorenzetti (2017)
Loading conditions in the spine, hip and knee during different executions of back extension exercises
BMC Sports Science, Medicine and Rehabilitation, 9
D. D’Lima, N. Steklov, B. Fregly, S. Banks, C. Colwell (2008)
In vivo contact stresses during activities of daily living after knee arthroplasty
Journal of Orthopaedic Research, 26
K. Khayambashi, Zeynab Mohammadkhani, K. Ghaznavi, Mark Lyle, C. Powers (2012)
The effects of isolated hip abductor and external rotator muscle strengthening on pain, health status, and hip strength in females with patellofemoral pain: a randomized controlled trial.
The Journal of orthopaedic and sports physical therapy, 42 1
P. Schütz, R. List, R. Zemp, Florian Schellenberg, W. Taylor, S. Lorenzetti (2014)
Joint angles of the ankle, knee, and hip and loading conditions during split squats.
Journal of applied biomechanics, 30 3
A. Navacchia, V. Kefala, K. Shelburne (2016)
Dependence of Muscle Moment Arms on In Vivo Three-Dimensional Kinematics of the Knee
Annals of Biomedical Engineering, 45
M. Pandy (2001)
Computer modeling and simulation of human movement.
Annual review of biomedical engineering, 3
P. Damm, I. Kutzner, G. Bergmann, A. Rohlmann, H. Schmidt (2017)
Comparison of in vivo measured loads in knee, hip and spinal implants during level walking.
Journal of biomechanics, 51
Florian Schellenberg, Julia Lindorfer, R. List, W. Taylor, S. Lorenzetti (2013)
Kinetic and kinematic differences between deadlifts and goodmornings
BMC Sports Science, Medicine and Rehabilitation, 5
W. Taylor, P. Schütz, G. Bergmann, R. List, B. Postolka, Marco Hitz, J. Dymke, P. Damm, G. Duda, H. Gerber, V. Schwachmeyer, S. Nasab, A. Trepczynski, I. Kutzner (2017)
A comprehensive assessment of the musculoskeletal system: The CAMS-Knee data set.
Journal of biomechanics, 65
G. Massey, Pavlos Evangelidis, J. Folland (2015)
Influence of contractile force on the architecture and morphology of the quadriceps femoris
Experimental Physiology, 100
T. Tyler, S. Nicholas, R. Campbell, M. McHugh (2001)
The Association of Hip Strength and Flexibility with the Incidence of Adductor Muscle Strains in Professional Ice Hockey Players *
The American Journal of Sports Medicine, 29
C. Lewis, S. Sahrmann, D. Moran (2009)
Effect of position and alteration in synergist muscle force contribution on hip forces when performing hip strengthening exercises
Clinical Biomechanics, 24
L. Modenese, A. Phillips, A. Bull (2011)
An open source lower limb model: Hip joint validation.
Journal of biomechanics, 44 12
B. Schoenfeld (2010)
The mechanisms of muscle hypertrophy and their application to resistance training.
Journal of strength and conditioning research, 24 10
M. Heller, W. Taylor, C. Perka, G. Duda (2003)
The influence of alignment on the musculo-skeletal loading conditions at the knee
Langenbeck's Archives of Surgery, 388
Florian Schellenberg (2016)
MUSCULOSKELETAL SQUAT SIMULATION EVALUATION BY MEANS OF AN INSTRUMENTED TOTAL KNEE ARTHROPLASTY
, 34
A. Vigotsky, I. Halperin, G. Lehman, G. Trajano, T. Vieira
TORINO Repository ISTITUZIONALE
T. Santos, B. Oliveira, J. Ocarino, K. Holt, S. Fonseca (2015)
Effectiveness of hip muscle strengthening in patellofemoral pain syndrome patients: a systematic review
Brazilian Journal of Physical Therapy, 19
S. Pal, J. Langenderfer, J. Stowe, P. Laz, A. Petrella, P. Rullkoetter (2007)
Probabilistic Modeling of Knee Muscle Moment Arms: Effects of Methods, Origin–Insertion, and Kinematic Variability
Annals of Biomedical Engineering, 35
D. Thelen (2003)
Adjustment of muscle mechanics model parameters to simulate dynamic contractions in older adults.
Journal of biomechanical engineering, 125 1
S. Delp, S. Delp, William Maloney (1993)
Effects of hip center location on the moment-generating capacity of the muscles.
Journal of biomechanics, 26 4-5
Florian Schellenberg, K. Oberhofer, W. Taylor, S. Lorenzetti (2015)
Review of Modelling Techniques for In Vivo Muscle Force Estimation in the Lower Extremities during Strength Training
Computational and Mathematical Methods in Medicine, 2015
R. List, Turgut Gülay, M. Stoop, S. Lorenzetti (2013)
Kinematics of the Trunk and the Lower Extremities During Restricted and Unrestricted Squats
Journal of Strength and Conditioning Research, 27
S. Delp, F. Anderson, A. Arnold, P. Loan, Ayman Habib, Chand John, Eran Guendelman, D. Thelen (2007)
OpenSim: Open-Source Software to Create and Analyze Dynamic Simulations of Movement
IEEE Transactions on Biomedical Engineering, 54
S. Lorenzetti, Turgut Gülay, M. Stoop, R. List, H. Gerber, Florian Schellenberg, E. Stüssi (2012)
Comparison of the Angles and Corresponding Moments in the Knee and Hip During Restricted and Unrestricted Squats
Journal of Strength and Conditioning Research, 26
I. Halperin, A. Vigotsky, C. Foster, D. Pyne (2017)
Strengthening the Practice of Exercise and Sport-Science Research.
International journal of sports physiology and performance, 13 2
Turgut Gülay, R. List, S. Lorenzetti (2011)
MOMENTS IN THE KNEE AND HIP DURING DESCENT AND ASCENT OF SQUATS
, 11
C. Lewis, S. Sahrmann, Fapta, D. Moran
Effect of Position and Alteration in Synergist Muscle Force Contribution on Hip Forces When Performing Hip Strengthening Exercises Nih Public Access
Florian Schellenberg, W. Taylor, I. Jonkers, S. Lorenzetti (2017)
Robustness of kinematic weighting and scaling concepts for musculoskeletal simulation
Computer Methods in Biomechanics and Biomedical Engineering, 20
E. Arnold, S. Ward, R. Lieber, S. Delp (2010)
A Model of the Lower Limb for Analysis of Human Movement
Annals of Biomedical Engineering, 38
M. Fredericson, Curtis Cookingham, A. Chaudhari, Brian Dowdell, N. Oestreicher, S. Sahrmann (2000)
Hip Abductor Weakness in Distance Runners with Iliotibial Band Syndrome
Clinical Journal of Sport Medicine, 10
A. Vigotsky, I. Halperin, G. Lehman, G. Trajano, T. Vieira (2018)
Interpreting signal amplitudes in surface electromyography studies in sport and rehabilitation sciences
Frontiers in Physiology, 8
R. Ehrig, M. Heller, S. Kratzenstein, G. Duda, A. Trepczynski, W. Taylor (2011)
The SCoRE residual: a quality index to assess the accuracy of joint estimations.
Journal of biomechanics, 44 7

Publisher: Hindawi Publishing Corporation
Copyright: Copyright © 2018 Michael Plüss et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
ISSN: 1176-2322
eISSN: 1754-2103
DOI: 10.1155/2018/9721079
Publisher site: See Article on Publisher Site

Abstract

Hindawi Applied Bionics and Biomechanics Volume 2018, Article ID 9721079, 10 pages https://doi.org/10.1155/2018/9721079 Research Article Towards Subject-Specific Strength Training Design through Predictive Use of Musculoskeletal Models Michael Plüss, Florian Schellenberg, William R. Taylor , and Silvio Lorenzetti Institute for Biomechanics, ETH Zürich, Zürich, Switzerland Correspondence should be addressed to Silvio Lorenzetti; slorenzetti@ethz.ch Received 12 September 2017; Revised 5 January 2018; Accepted 28 January 2018; Published 19 March 2018 Academic Editor: Justin Keogh Copyright © 2018 Michael Plüss et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Lower extremity dysfunction is often associated with hip muscle strength deﬁciencies. Detailed knowledge of the muscle forces generated in the hip under speciﬁc external loading conditions enables speciﬁc structures to be trained. The aim of this study was to ﬁnd the most eﬀective movement type and loading direction to enable the training of speciﬁc parts of the hip muscles using a standing posture and a pulley system. In a novel approach to release the predictive power of musculoskeletal modelling techniques based on inverse dynamics, ﬂexion/extension and ab-/adduction movements were virtually created. To demonstrate the eﬀectiveness of this approach, three hip orientations and an external loading force that was systematically rotated around the body were simulated using a state-of-the art OpenSim model in order to establish ideal designs for training of the anterior and posterior parts of the M. gluteus medius (GM). The external force direction as well as the hip orientation greatly inﬂuenced the muscle forces in the diﬀerent parts of the GM. No setting was found for simultaneous training of the anterior and posterior parts with a muscle force higher than 50% of the maximum. Importantly, this study has demonstrated the use of musculoskeletal models as an approach to predict muscle force variations for diﬀerent strength and rehabilitation exercise variations. 1. Introduction exercises [7–10], inverse dynamics approaches have been used in a subject-speciﬁc manner to enable a comparison between diﬀerent exercise variations [2, 11]. In more sophis- Detailed knowledge of the generated forces within the human ticated analyses, these approaches have been combined with musculoskeletal system provides an important step towards muscle optimization techniques in order to compare forces understanding the conditions that are required to eﬀectively in the diﬀerent parts of the hamstring and quadriceps train for speciﬁc sports or undertaking targeted rehabilitation muscles between training exercises, including consideration after injury or during therapy. Ideally, direct measurements of execution form and joint angles [10]. Performing muscu- of the internal loading conditions such as muscle and joint contact forces would guide such training approaches, but loskeletal modelling requires assumptions regarding the anthropometry of the segments, shape and degree of freedom these are diﬃcult to access [1, 2]. Here, while detailed data- of the joints, muscular properties, and optimization criteria sets of kinematics and kinetics are becoming more widely [12] Importantly, an in-depth understanding of the condi- available [3], such approaches are currently limited both to tions under which these models are valid and able to cor- small populations with artiﬁcial joints as well as to only very speciﬁc sites in the human body [4–6]. As a result, musculo- rectly predict the internal loading conditions during squatting exercises has already been performed [13]. In an skeletal simulation is the primary tool for estimating internal analysis using videoﬂuoroscopy and instrumented implants, loading conditions throughout the human body, albeit indi- we have been able to demonstrate a ﬂexion-dependent error rectly, by means of inverse dynamics and numerical optimi- in the predicted joint contact forces, but a good estimation zation processes [2]. In the ﬁrst steps towards understanding the interactions (e.g., within 20%) over the range of 25–65 knee ﬂexion [13]. However, despite their ability to calculate internal between kinematics and kinetics during strength training 2 Applied Bionics and Biomechanics In combination with cable exercises to provide targeted loading conditions throughout the musculoskeletal system, one issue that has limited the applicability of musculoskeletal force application, it therefore seems entirely plausible that modelling techniques for predicting the outcome of new musculoskeletal modelling approaches based on systemati- cally altered kinematic and kinetic data could provide a exercise design is the requirement that inverse dynamics approaches are provided with known segment kinematics powerful tool for designing targeted strength and rehabili- as a modelling input. By systematically modifying the exter- tation training exercises. Therefore, the aim of this study nal loading conditions, the use of these models could provide was to evaluate the forces of hip muscles with respect to range a basis for designing or improving training and rehabilitation of motion and their lengths during sagittal and frontal simulated hip strength exercises, using a musculoskeletal programs for targeting speciﬁc musculoskeletal structures, thus opening a predictive capability of the approaches that model driven by a systematic modiﬁcation of the external has not yet been exploited. force direction. One area that could beneﬁt from the power of such pre- dictive options is the focused training of hip musculature, 2. Materials and Methods strength deﬁciencies, and muscular imbalance, which, until now, has generally been investigated with respect to injury. 2.1. Description of the Exercise. Speciﬁc strength training An example of the association between adductor injury and exercises for the hip muscles on the cable machine were sim- hip strength can be seen in the frequency of adductor strains ulated (Figure 1). For these exercises, the cable is usually ﬁxed in ice hockey players, with injured players exhibiting an 18% with a strap to the shank slightly above the ankle joint and the lower hip adduction strength [14]. Importantly, the risk of pulley position is set as low as required in order to ensure a adductor strain injury was shown to be almost 17 times horizontal force vector. These exercises are single-joint and higher in players where the adductor strength was below free-leg exercises. By varying the body orientation relative 80% of the abductor strength. Furthermore, recovery of the to the cable machine and the movement in the diﬀerent ana- iliotibial band syndrome in long-distance runners [15] and tomical planes, muscle activation changes and thus targeted pain [16] in subjects with retropatellar pain syndrome was muscles for strengthening can be chosen appropriately. The improved with a gain in the strength of the hip abductor hip ﬂexor and extensor muscles are then targeted by posi- muscles. However, current strength training instructions tioning the body backwards and forwards relative to the cable are mostly based on the experience of the coach or physio- machine, respectively. A lateral orientation of the body and therapist and are rarely evidence based. This is possibly due movement in the frontal plane will target the hip adductor to the complexity of the hip muscles, which include large and abductor muscles. cross-sectional areas with diﬀerent parts of the same muscle active for diﬀerent functional tasks, as well as diﬀerent lines 2.2. Musculoskeletal Model. The open source software Open- of action and moment arms around the joint that vary with Sim (OpenSim SimTK 3.2, Stanford, USA) was used to joint angles and muscle activation. As a result, speciﬁc guide- perform the simulation [18]. All the ﬁles required for the lines on how to strengthen speciﬁc parts of the hip muscles, simulation, including motion and external force ﬁles, were including the direction of the external force and the joint created in Matlab (R2015a MathWorks, Natick, Massachu- motion, are missing in the literature. It is therefore clear that setts, USA). For the OpenSim simulation, the Arnold Lower detailed knowledge of the interaction between the form of Limb Model 2010 [19] was used. For the hip joint contact rehabilitation/strength exercise and the internal forces gener- force, this model has been validated using an instrumented ated in diﬀerent parts of the hip muscles could lead to an hip implant [20]. To apply the external loading force, a cylin- evidence-based design of training exercises for prevention der was attached rigidly to the right leg of the model to and rehabilitation programs that focus on either muscular represent the ankle strap used in the strength exercises with weakness or imbalance. a cable machine. The cylinder was characterized by the fol- Compared to strength exercises for the hip muscles that lowing dimensions: radius was set to 0.05 m, thickness include multijoint motion such as squatting, cable exercises 0.001 m, height 0.04 m, and mass 0.078 kg. The attachment enable an isolated movement of the hip joint as well as a spe- location in the Lower Limb Model 2010 was at 0.339 m in ciﬁc force magnitude and direction to be applied. In addition, the distal direction of the tibia coordinate system. cable exercises enable preferential muscle force that does not aﬀect forces and movements in other joints and is therefore a 2.3. Kinematics. For kinematic inputs into the model, two simple exercise to be simulated. Using such a pulley system, diﬀerent motions were created at a frequency of 110 Hz. prone hip extension and straight leg raises were used in com- For each, a sine-shaped movement velocity time curve was bination with musculoskeletal models to investigate the mag- used, with a maximum movement speed of 40 degrees per sec- nitude and direction of hip muscle forces [17]. Their results ond. One motion represented a hip ﬂexion/extension (F/E) showed that the hip joint forces were aﬀected by hip joint movement and was performed in the sagittal plane, while the position and partially by alternations in muscle force contri- second one characterized hip abduction and adduction bution. Such studies demonstrate the importance of muscu- (Abd/Add) and was executed in the frontal plane (Figure 1, loskeletal modelling approaches to provide science-based top). The start and ﬁnish points of the F/E movement were evidence for understanding the internal muscle and force both set at −20 -extended hip, since the Lower Limb Model interactions towards guiding training and rehabilitation and 2010 wasvalidated within thisextension range only.The rever- hence positive adaptation of the tissues. sal point of the movement was set at 60 hip ﬂexion, enabling a Applied Bionics and Biomechanics 3 Flexion/extension Abduction/adduction 90° 0° 15° 15° 180° 90° 270° 0° 270° 180° Internal Internal External External hip rotation (−40°) hip rotation (−40°) hip rotation (40°) hip rotation (40°) Neutral Neutral (b) (a) Figure 1: Simulation illustration (top row) of the model performing the ﬂexion/extension (a) and the abduction/adduction (b) movements, including the 0 position of the external force (green arrow) applied to the right leg of the model. Schematic representations of the diﬀerent loading conditions used in the simulation are shown in the bottom row, including rotational external force (green), which was rotated incrementally in 15 steps and three diﬀerent hip rotation conﬁgurations; externally rotated (blue), neutrally rotated (grey), and internally rotated (orange). total range of motion (RoM) of 80 . The Abd/Add movement (force direction), to estimate the internal muscle force mag- started with a −35 abducted hip position, where the reversal nitudes, in which the sum of the squared muscle activation point of the movement was deﬁned at 5 hip adduction, result- was minimized. This combination led to 138 individual sim- ing in a 40 RoM in the frontal plane. Each limb movement was ulations. Some simulations were run without the individual then simulated with the hip rotated at one of the three follow- wrapping surfaces to enable successful simulation: for the ° ° ing conﬁgurations: neutral (0 ), internally rotated (40 ), or F/E movement in the neutral hip position, the wrapping sur- externally rotated (−40 ) (Figure 1, bottom). In some cases, face of the M. pectineus (PECT_at_femshaft_r) and, in the the eccentric phase and in some the concentric phase, depen- externally rotated hip position, the wrapping surface of the dent on the actual direction of the force, were at the start of M. adductor brevis (AB_aft_femshaft_r) and the proximal themotion. Thetimewas6.28 sforF/E and3.14 sforAbd/Add. part of the M. adductor magnus (AMprox_at_femshaft_r) were disabled due to simulation errors. 2.4. Kinetics. An external force with a magnitude of 100 N 2.6. Evaluation of the Data. The muscle activations A of M. was applied to the centre of the attached cylinder at the shank adductor longus, M. adductor magnus, M. gluteus medius, of the model. This force represents a typical load used in a M. rectus femoris, and M. semimembranosus were calculated health-oriented strength training including the here-used for all hip rotation conﬁgurations and external force direc- cable exercises. In all diﬀerent movements and throughout tions as follows: the whole cycle, the external force remained parallel to the ground. For each movement conﬁguration, diﬀerent external force directions were used to examine the inﬂuence of the act A = , 1 position of the cable machine to the body. Starting in a dorsal F max direction for F/E and medially for Abd/Add simulations, the where F is the acting muscle force and F is the maximal external force was then rotated incrementally by 15 degrees act max possible muscle force of the speciﬁc part of the muscle. For in a counter-clockwise direction until a complete rotation concentric contractions, the activation lies between 0 and 1. of the force was obtained, leading to 23 individual simula- To properly model the anatomical characteristics, the M. tions (Figure 1, bottom). adductor magnus and M. gluteus medius were included with 2.5. Musculoskeletal Simulation. A quasistatic optimization diﬀerent parts in the Lower Limb Model 2010, which were was performed for all movements (F/E and Abd/Add), all also maintained in the analysis of the parameters. 3D surface ° ° ° hip rotations (0 ,40 , and −40 ), and all kinetic parameters plots were then used to visualise the muscular activation, Loading Movement 4 Applied Bionics and Biomechanics For the anterior GM, the largest muscle length changes were which was dependent on the joint angle as well as on the angle of the external force. Additionally, for all three hip rota- observed during the Abd/Add movement, while for the pos- tion positions, the maximal activations for each external terior GM, the externally rotated position caused similar acti- force angle were calculated and displayed in spider plots. Fur- vations and changes in muscle lengths in both the F/E and thermore, the muscle lengths and the corresponding muscle Abd/Add movements. activations for all three hip rotation positions were analysed at the angle of the external force where the highest activation 4. Discussion level occurred. All data evaluation and plot generation was performed in Matlab (R2014b, MathWorks, Inc.). After ini- In order to further improve rehabilitation exercises and to tial review of the simulation data, only the muscles M. adduc- estimate the internal mechanical load of the speciﬁc parts tor longus, M. rectus femoris, M. semimembranosus, and M. of the targeted muscles, it is essential that their activation is rectus femoris were evaluated. Furthermore, the anterior and known, with respect to the chosen movement and external posterior parts of the M. gluteus medius (GM) were chosen loading conditions. In this study, classic hip strength and for in-depth analysis, due to the fact that his muscle repre- rehabilitation exercises with a F/E and an Abd/Add move- sents one of the major target structures of this type of cable ment using a cable machine were simulated by means of exercises and the medial part had an activation lower than 0.5. whole-body musculoskeletal simulation with the aim to quantify muscle activation and lengths during diﬀerent kine- matic and kinetic conﬁgurations. To simulate the strength 3. Results exercises, loading and movement patterns were generated The adductor muscles’ activation remained low for all load- and analysed using diﬀerent directions of the cable with respect to the body, as well as using two movements with ing conditions; except for M. adductor longus, the two rotated hip positions showed higher activities in the F/E movement three diﬀerent hip rotation positions, internally, neutrally, than in Abd/Add. As expected, the M. rectus femoris exhib- and externally rotated. In order to quantify the activation of ited a higher activation for F/E movement than for Abd/ the individual hip muscles and their parts, muscle activation Add movement. Similar results, but in the opposite direction was estimated by means of static optimization using a full of the rotating external force, were observed for the M. semi- body musculoskeletal model as well as targeted kinetic and membranosus, an antagonist of M. rectus femoris. Addition- kinematic conditions. ally, activation in the hamstrings muscles were reduced Although previous models have attempted to modify the when the hip was rotated externally. An agonist/antagonist kinematics of a joint for use in inverse dynamics modelling relationship was clearly visible between the anterior and pos- [21], the external forces imposed on such models are gener- terior parts of the GM during the F/E movement with the hip ally known (from, e.g., ground reaction force plates) and in a neutral position (Figure 2). In this position, the F/E not altered. To our knowledge, this approach, where the movements led to higher muscular activation compared to external loading conditions were systematically varied, was Abd/Add movements. The activation levels for the abductors used for the ﬁrst time in an approach that seems to lend itself versus the adductors remained rather equal, but this was nicely towards the design of targeted training strategies somewhat diﬀerent during F/E, where the activation seemed through identiﬁcation of the optimal movement and loading to increase exponentially towards a dominant maximum condition to speciﬁcally train a certain musculature. Here, level. Furthermore, the activation of the anterior GM part the use of a purposefully designed hip-strengthening pro- was considerably larger within the extension range (negative gram can be beneﬁcial for patients as well as athletes. angles) of the movement than within the ﬂexion range (pos- Whereas it is well known that the direction of the force itive, Figure 2(a)), where the posterior part of the muscle deﬁnes the muscle activation pattern, this work aimed also increased in activation (Figure 2(c)). to show the importance of the hip rotation position. As an With the hip rotated externally, the posterior part of the example, by including strength training exercises for abduc- GM achieved a maximum activation during the ﬂexion and tor muscles and internal rotation in the hip, Khayambashi abduction movements (Figures 2–4). On the contrary, an and coworkers [22] showed an improvement of pain and internally rotated hip position led to maximum activation health status in women with patellofemoral pain syndrome levels for the anterior part of the muscle, compared to neutral compared to a no-exercise control group. Whole-body simu- and external rotated hip positions for both movements. lation, similar to that performed in our study, might help in For the anterior part of the GM, maximum activation was the future to speciﬁcally design an eﬃcient subject-speciﬁc achieved in the internally rotated hip position for external workout program. ° ° forces from 180–300 Overall, the relatively small magnitude of 100 N of the during F/E and from 0–45 as well as 240–360 for the Abd/Add movement. On the other hand, external force did not cause high activations for the M. rectus the posterior part of the GM exhibited maximum activation femoris, M. adductor longus, M. semimembranosus, and M. adductor magnus. Interestingly, the M. adductor longus in the externally rotated conﬁguration with an external force ° ° direction of 240–315 and 300–315 for the F/E and Abd/Add showed higher activations for the F/E movement than for the Abd/Add. Here, this speciﬁc behaviour, together with movements, respectively (Figure 5). While changing the rota- tion position of the hip had an inﬂuence on muscle length, increased loading, could be of interest for this muscle, since changing the external force angles within one movement as an adductor muscle, a higher activation in the Abd/Add could be expected. Contrary to the low activation levels of conﬁguration had no eﬀect on the muscle length (Figure 6). Applied Bionics and Biomechanics 5 M. gluteus medius neutral hip position Flexion/extension movement Abduction/adduction movement 1 1 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 50 50 0 0 345 345 40 40 −5 −5 30 30 270 270 270 270 −10 −10 0 20 20 −15 −15 180 180 180 180 10 10 −20 −20 −25 25 0 0 90 90 90 90 −10 10 −30 −20 0 −35 Rotating force angle (°) Rotating force angle (°) Movement angle (°) Movement angle (°) (a) (b) 1 1 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0 0 60 5 0 0 50 50 345 3 345 3 −5 −5 40 40 −10 −10 10 30 30 270 270 270 270 −15 −15 20 20 180 180 180 180 −20 −20 10 10 −25 25 0 0 90 90 90 −30 −10 0 0 −35 −20 Rotating force angle (°) Rotating force angle (°) Movement angle (°) Movement angle (°) (c) (d) Figure 2: Muscle activations [0→ 1] as a function of the movement angle and all external force orientations in the neutral hip position. The anterior (a, b) and the posterior (c, d) parts of the M. gluteus medius (GM) are displayed for ﬂexion/extension (a, c) and abduction/adduction (b, d) movements. the other muscles, GM showed high and alternative activa- length is required [23]. For optimal training, the movement tions with changed kinematic and kinetic conﬁgurations. with the largest change in muscle length, together with an Therefore, the diﬀerent parts of the GM were further ana- external force direction that causes the highest muscle force over the whole movement, should be chosen. Please note lysed. Since the middle part of the GM did not achieve activa- tions larger than 0.5, which would lead to a more eﬃcient that, in this work, the muscle activation was calculated as training stimuli, only the anterior and posterior parts were the actual muscle force normalized by the maximum isomet- included in the in-depth analysis of the GM muscle. ric muscle force. As an example, for the anterior GM part, In all three examined hip rotations, activation patterns of both exercise movements with an internally rotated hip either the anterior or the posterior part was examined showed a high activation (Figure 4), but the Abd/Add move- (Figures 2–4). However, external or internal rotation of the ment also caused a greater change in muscle length hip resulted in a higher muscular activation level compared (Figure 6). Therefore, based on our results, it could be recom- to the neutral position, which can be explained by the sup- mended to train the anterior part of the GM with an inter- portive function of these muscle parts for the hip rotation nally rotated hip position using the direction of the external itself. Rotating the hip also inﬂuenced the length of the M. force in the range of the maximum activation at about 45– gluteus medius during the exercise. In order to provide the 240 . For the posterior part, similar maximum activities and most eﬀective training impulse to the target muscle, large changes in muscle length were achieved with an externally muscle forces over the maximum possible change in muscle rotated hip. Regarding the muscle length and the posterior Posterior muslce part Anterior muslce part Activation Activation Activation Activation 6 Applied Bionics and Biomechanics M. gluteus medius external rotation hip position Flexion/extension movement Abduction/adduction movement 1 1 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0 0 60 5 50 0 40 345 −5 345 30 −10 270 270 20 −15 10 180 −20 180 0 −25 90 90 −10 −30 Rotating force angle (°) Rotating force angle (°) −20 −35 0 0 Movement angle (°) Movement angle (°) (a) (b) 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0 0 50 0 40 345 −5 345 30 −10 270 270 20 −15 10 180 −20 180 0 −25 90 90 −10 −30 Rotating force angle (°) Rotating force angle (°) −20 −35 0 0 Movement angle (°) Movement angle (°) (c) (d) Figure 3: Muscle activations [0→ 1] as a function of the movement angle and external force orientations in the externally rotated hip position. The anterior (a, b) and the posterior (c, d) parts of the M. gluteus medius (GM) are displayed for ﬂexion/extension (a, c) and abduction/adduction (b, d) movements. part, two aspects should be considered. Firstly, the highest the anterior part, an internally rotated hip is recommended, activation in the F/E movement occurred at large muscle while the highest loading for the posterior part can be length (around 0.14 m, Figure 6(a)) whereas during Abd/ achieved using an externally rotated hip. Muscle force is known to be highly dependent on force- Add, the highest activation was observed at the shortest mus- cle length (around 0.08 m, Figure 6(b)). Secondly, a greater velocity and force-length relationships (Hill-type muscles muscle length change was observed during F/E movement [24, 25]). As a result, muscle activation is directly linked to than during Add/Abb. Taking these two factors into account, the maximal isometric force capacity of the muscle, the rele- training the posterior part should be performed in an exter- vant lever arm of the speciﬁc muscle, and the external loading nally rotated hip position with a F/E movement to achieve conditions. For the sake of completeness, the force-velocity an eﬀective training regime. In summary, our ﬁndings show relationship will only play a minor role, since the velocity of the importance of properly choosing a suitable hip position, the movement during strength training is 1.7% of the maxi- movement direction, and external force direction, in order mal shortening velocity and furthermore, the external load- to achieve the desired training goals. To eﬀectively train the ing changes only a few percent between the acceleration anterior and posterior parts of the GM, F/E movement seems and deceleration phases [26]. to be preferable but using two diﬀerent loading conﬁgura- Several limitations arising in this study need to be men- tions due to the supportive functions of the two parts of the tioned. Firstly, within the chosen model conﬁguration and muscle in opposite hip rotation directions: for training of the kinematic pathways investigated in this study, only 5 of Posterior muslce part Anterior muslce part Activation Activation Activation Activation Applied Bionics and Biomechanics 7 M. gluteus medius internal rotated hip position Flexion/extension movement Abduction/adduction movement 1 1 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0 0 345 −5 345 −10 270 270 −15 180 −20 180 −25 90 90 −10 −30 −20 −35 0 0 Rotating force angle (°) Rotating force angle (°) Movement angle (°) Movement angle (°) (a) (b) 1 1 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0 0 345 −5 345 −10 270 270 −15 180 −20 180 −25 90 90 −10 −30 Rotating force angle (°) Rotating force angle (°) −20 −35 0 0 Movement angle (°) Movement angle (°) (c) (d) Figure 4: Muscle activations [0→ 1] as a function of the movement angle and external force orientations in the internally rotated hip position. The anterior (a, b) and the posterior (c, d) parts of the M. gluteus medius (GM) are displayed for ﬂexion/extension (a, c) and abduction/adduction (b, d) movements. adduction in the hip could be tolerated to avoid changes in little is known about the large muscles and their changing other joint angles such as the knee angle of the opposite leg. lever arms during dynamic activities. As a result, the accuracy Changes in muscle activation patterns might also occur if of the muscle force and activations estimated in this study larger adduction angles are taken into account. Secondly, cannot be guaranteed. It is therefore imperative that further investigations using subject cohorts and, for example, elec- the force was applied parallel to the ground and by assuming a cylinder around the ankle representing an ankle strap. This tromyography (EMG) measurements, within their possibili- conﬁguration would not represent the loading direction of a ties and limitations [34] are undertaken to test the validity real cable exercise machine and could inﬂuence the results in of these approaches in vivo, followed by training studies to a complex manner. Thirdly and possibly most importantly, assess the relationships between these acute measures and the results of this study are based on the use of a reference longitudinal outcomes [35]. Further enhancement to the and nonscaled musculoskeletal model. It is well known that accuracy and reliability of musculoskeletal models would the exact musculoskeletal conﬁguration, including, for exam- therefore only further improve the estimations of muscle ple, anatomical insertion sites [27], the deﬁnition of joint forces and the speciﬁc training regime strategies. However, centres and axes [28–30], physiological cross-sectional areas despite these limitations, the approaches used in this study of the muscle [31], the scaling procedure [32], and the lever do open perspectives for providing targeted exercise plans arms of the muscles around the joints [33] all play critical on a subject-speciﬁc basis or comparing muscle activations roles on the estimation of internal loading conditions using within diﬀerent strength exercises by systematically varying musculoskeletal models. Particularly in the hip, relatively, the external force and the movement. As a result, speciﬁc Posterior muslce part Anterior muslce part Activation Activation Activation Activation 8 Applied Bionics and Biomechanics Flexion/extension Abduction/adduction 0° 90° 330° 30° 120° 60° 150° 300° 60° 30° 90° 270° 180° 0° 120° 210° 330° 240° 150° 300° 210° 240° 180° 270° Neutral Neutral Ant. part GM Ant. part GM Ext. rot Ext. rot Post. part GM Post. part GM Int. rot Int. rot (a) (b) Figure 5: M. gluteus medius (GM) maximum muscle activations of the anterior (solid line) and posterior (dashed line) parts for the ﬂexion/ extension (a) and abduction/adduction (b) movements and the diﬀerent hip rotation conﬁgurations: neutral (black), externally rotated (red), and internally rotated (blue). The initial external force direction (0 ) of the movement is indicated by the green arrow. Flexion/extension Adduction/abduction 1 285° 1 0.9 0.9 270° 0.8 0.8 0° 0.7 0.7 0.6 0.6 0.5 0.5 240° 330° 0.4 0.4 330° 0.3 0.3 0.2 0.2 300° 0.1 0.1 255° 255° 240° 240° 285° 0 0 0.08 0.1 0.12 0.14 0.16 0.08 0.1 0.12 0.14 0.16 Muscle length (m) Muscle length (m) Neutral Neutral Ant. part GM Ant. part GM Ext. rot Ext. rot Post. part GM Post. part GM Int. rot Int. rot (a) (b) Figure 6: Progression of the M. gluteus medius (GM) muscle activations [0→ 1] with external force direction. The overall maximum activations of all force directions are shown for the anterior (solid line) and posterior (dashed line) part of the muscle as a function of the muscle length (m) for the ﬂexion/extension (a) and abduction/adduction (b) movements, as well as for the diﬀerent hip rotation conﬁgurations: neutral (black), externally rotated (red), and internally rotated (blue). exercises could be identiﬁed in order to achieve optimal load- 5. Perspectives ing patterns. Furthermore, by including subject-speciﬁcdeﬁ- cits, eﬃcient rehabilitation regimes could be designed and Using musculoskeletal simulation and systematic variation of even updated to follow the rehabilitation progress, if the loading conditions, this study opens perspectives for the adaptation of muscle status could be quantiﬁed. identiﬁcation of optimal training exercises for speciﬁc muscle 0.75 0.5 0.25 0.75 0.5 0.25 Activation Activation Applied Bionics and Biomechanics 9 subjects,” Journal of Biomechanics, vol. 43, no. 11, pp. 2164– strengthening in rehabilitation and sports medicine. By 2173, 2010. selecting hip rotation and body positions relative to the cable exercise machine, higher muscle activation and large changes [7] R. List, T. Gülay, M. Stoop, and S. Lorenzetti, “Kinematics of the trunk and the lower extremities during restricted and unre- in muscle lengths can be achieved. To eﬀectively train the stricted squats,” Journal of Strength and Conditioning anterior and posterior parts of GM muscle, two diﬀerent Research, vol. 27, no. 6, pp. 1529–1538, 2013. exercises likely need to be performed. While an internally [8] S. Lorenzetti, T. Gülay, M. Stoop et al., “Comparison of the rotated hip is recommended to train the anterior part, the pos- angles and corresponding moments in the knee and hip during terior part should be trained using an externally rotated hip. restricted and unrestricted squats,” Journal of Strength and The application of these approaches in this relevant example Conditioning Research, vol. 26, no. 10, pp. 2829–2836, 2012. demonstrates that precisely validated models, fed with kine- [9] F. Schellenberg, J. Lindorfer, R. List, W. R. Taylor, and matics of diﬀerent exercises, could provide a powerful option S. Lorenzetti, “Kinetic and kinematic diﬀerences between to compare the eﬀectiveness of exercises that target speciﬁc deadlifts and goodmornings,” BMC Sports Science Medicine muscles. For more complex and dynamic exercises, a similar & Rehabilitation, vol. 5, no. 1, p. 27, 2013. approach might be possible in the future. Here, the general [10] F. Schellenberg, N. Schmid, R. Häberle, N. Hörterer, W. R. procedure is similar, but the musculoskeletal model needs to Taylor, and S. Lorenzetti, “Loading conditions in the spine, enable large and extreme joint angles, individual muscle, hip and knee during diﬀerent executions of back extension segmental and joint properties, and subject-speciﬁc scaling. exercises,” BMC Sports Science Medicine & Rehabilitation, vol. 9, no. 1, p. 10, 2017. Conflicts of Interest [11] P. Schütz, R. List, R. Zemp, F. Schellenberg, W. R. Taylor, and S. Lorenzetti, “Joint angles of the ankle, knee, and hip and The authors declare no conﬂict of interests. loading conditions during split squats,” Journal of Applied Bio- mechanics, vol. 30, no. 3, pp. 373–380, 2014. [12] A. Erdemir, S. McLean, W. Herzog, and A. J. van den Bogert, Authors’ Contributions “Model-based estimation of muscle forces exerted during Michael Plüss performed the study, analysed the results, and movements,” Clinical Biomechanics, vol. 22, no. 2, pp. 131– 154, 2007. wrote the manuscript. Florian Schellenberg and Silvio Loren- zetti designed and supervised the study, analysed the data, [13] F. Schellenberg et al., “Musculoskeletal squat simulation evalu- and helped in drafting the manuscript. William R. Taylor ation by means of an instrumented total knee arthroplasty,” in 34th International Conference on Biomechanics in Sports, Tsu- provided the original concept and helped in drafting and kuba, Japan, July 2016. editing the manuscript. [14] T. F. Tyler, S. J. Nicholas, R. J. Campbell, and M. P. McHugh, “The association of hip strength and ﬂexibility with the inci- Supplementary Materials dence of adductor muscle strains in professional ice hockey players,” The American Journal of Sports Medicine, vol. 29, The MATLAB code to generate the velocity proﬁle as well as no. 2, pp. 124–128, 2001. all the OpenSim ﬁles is available here http://www.movement. [15] M. Fredericson, C. L. Cookingham, A. M. Chaudhari, B. C. ethz.ch/data-repository.html. (Supplementary Materials) Dowdell, N. Oestreicher, and S. A. Sahrmann, “Hip abductor weakness in distance runners with iliotibial band syndrome,” References Clinical Journal of Sport Medicine, vol. 10, no. 3, pp. 169– 175, 2000. [1] M. G. Pandy, “Computer modeling and simulation of human [16] T. R. T. Santos, B. A. Oliveira, J. M. Ocarino, K. G. Holt, and S. T. movement,” Annual Review of Biomedical Engineering, vol. 3, Fonseca, “Eﬀectiveness of hip muscle strengthening in patello- no. 1, pp. 245–273, 2001. femoral pain syndrome patients: a systematic review,” Brazilian [2] F. Schellenberg, K. Oberhofer, W. R. Taylor, and S. Lorenzetti, Journal of Physical Therapy, vol. 19, no. 3, pp. 167–176, 2015. “Review of modelling techniques for in vivo muscle force esti- [17] C. L. Lewis, S. A. Sahrmann, and D. W. Moran, “Eﬀect of posi- mation in the lower extremities during strength training,” tion and alteration in synergist muscle force contribution on Computational and Mathematical Methods in Medicine, hip forces when performing hip strengthening exercises,” Clin- vol. 2015, Article ID 483921, 12 pages, 2015. ical Biomechanics, vol. 24, no. 1, pp. 35–42, 2009. [3] W. R. Taylor, P. Schütz, G. Bergmann et al., “A comprehensive [18] S. L. Delp, F. C. Anderson, A. S. Arnold et al., “OpenSim: open- assessment of the musculoskeletal system: the CAMS-knee source software to create and analyze dynamic simulations of data set,” Journal of Biomechanics, vol. 65, pp. 32–39, 2017. movement,” IEEE Transactions on Biomedical Engineering, [4] D. D. D'Lima, N. Steklov, B. J. Fregly, S. A. Banks, and C. W. vol. 54, no. 11, pp. 1940–1950, 2007. Colwell Jr, “In vivo contact stresses during activities of daily [19] E. M. Arnold, S. R. Ward, R. L. Lieber, and S. L. Delp, “A model living after knee arthroplasty,” Journal of Orthopaedic of the lower limb for analysis of human movement,” Annals of Research, vol. 26, no. 12, pp. 1549–1555, 2008. Biomedical Engineering, vol. 38, no. 2, pp. 269–279, 2010. [5] P. Damm, I. Kutzner, G. Bergmann, A. Rohlmann, and H. Schmidt, “Comparison of in vivo measured loads in knee, [20] L. Modenese, A. T. M. Phillips, and A. M. J. Bull, “An open source lower limb model: hip joint validation,” Journal of Bio- hip and spinal implants during level walking,” Journal of Bio- mechanics, vol. 51, pp. 128–132, 2017. mechanics, vol. 44, no. 12, pp. 2185–2193, 2011. [6] I. Kutzner, B. Heinlein, F. Graichen et al., “Loading of the knee [21] M. O. Heller, W. R. Taylor, C. Perka, and G. N. Duda, “The joint during activities of daily living measured in vivo in ﬁve inﬂuence of alignment on the musculo-skeletal loading 10 Applied Bionics and Biomechanics conditions at the knee,” Langenbeck's Archives of Surgery, vol. 388, no. 5, pp. 291–297, 2003. [22] K. Khayambashi, Z. Mohammadkhani, K. Ghaznavi, M. A. Lyle, and C. M. Powers, “The eﬀects of isolated hip abductor and external rotator muscle strengthening on pain, health sta- tus, and hip strength in females with patellofemoral pain: a randomized controlled trial,” Journal of Orthopaedic & Sports Physical Therapy, vol. 42, no. 1, pp. 22–29, 2012. [23] B. J. Schoenfeld, “The mechanisms of muscle hypertrophy and their application to resistance training,” The Journal of Strength and Conditioning Research, vol. 24, no. 10, pp. 2857–2872, 2010. [24] D. G. Thelen, “Adjustment of muscle mechanics model parameters to simulate dynamic contractions in older adults,” Journal of Biomechanical Engineering, vol. 125, no. 1, pp. 70– 77, 2003. [25] G. T. Yamaguchi and F. E. Zajac, “A planar model of the knee joint to characterize the knee extensor mechanism,” Journal of Biomechanics, vol. 22, no. 1, pp. 1–10, 1989. [26] T. Gülay, R. List, and S. Lorenzetti, “Moments in the knee and hip during descent and ascent of squats,” Portuguese Journal of Sport Sciences, vol. 11, Supplement 2, 2011. [27] S. Pal, J. E. Langenderfer, J. Q. Stowe, P. J. Laz, A. J. Petrella, and P. J. Rullkoetter, “Probabilistic modeling of knee muscle moment arms: eﬀects of methods, origin-insertion, and kine- matic variability,” Annals of Biomedical Engineering, vol. 35, no. 9, pp. 1632–1642, 2007. [28] R. M. Ehrig, M. O. Heller, S. Kratzenstein, G. N. Duda, A. Trepczynski, and W. R. Taylor, “The SCoRE residual: a quality index to assess the accuracy of joint estimations,” Jour- nal of Biomechanics, vol. 44, no. 7, pp. 1400–1404, 2011. [29] W. R. Taylor, E. I. Kornaropoulos, G. N. Duda et al., “Repeat- ability and reproducibility of OSSCA, a functional approach for assessing the kinematics of the lower limb,” Gait & Posture, vol. 32, no. 2, pp. 231–236, 2010. [30] S. L. Delp and W. Maloney, “Eﬀects of hip center location on the moment-generating capacity of the muscles,” Journal of Biomechanics, vol. 26, no. 4-5, pp. 485–499, 1993. [31] G. Massey, P. Evangelidis, and J. Folland, “Inﬂuence of con- tractile force on the architecture and morphology of the quad- riceps femoris,” Experimental Physiology, vol. 100, no. 11, pp. 1342–1351, 2015. [32] F. Schellenberg, W. R. Taylor, I. Jonkers, and S. Lorenzetti, “Robustness of kinematic weighting and scaling concepts for musculoskeletal simulation,” Computer Methods in Biome- chanics and Biomedical Engineering, vol. 20, no. 7, pp. 720– 729, 2017. [33] A. Navacchia, V. Kefala, and K. B. Shelburne, “Dependence of muscle moment arms on in vivo three-dimensional kinematics of the knee,” Annals of Biomedical Engineering, vol. 45, no. 3, pp. 789–798, 2017. [34] A. D. Vigotsky, I. Halperin, G. J. Lehman, G. S. Trajano, and T. M. Vieira, “Interpreting signal amplitudes in surface elec- tromyography studies in sport and rehabilitation sciences,” Frontiers in Physiology, vol. 8, p. 985, 2018. [35] I. Halperin, A. D. Vigotsky, C. Foster, and D. B. Pyne, “Strengthening the practice of exercise and sport science,” International Journal of Sports Physiology and Performance, pp. 1–8, 2018. International Journal of Advances in Rotating Machinery Multimedia Journal of The Scientific Journal of Engineering World Journal Sensors Hindawi Hindawi Publishing Corporation Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 http://www www.hindawi.com .hindawi.com V Volume 2018 olume 2013 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 Journal of Control Science and Engineering Advances in Civil Engineering Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 Submit your manuscripts at www.hindawi.com Journal of Journal of Electrical and Computer Robotics Engineering Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 VLSI Design Advances in OptoElectronics International Journal of Modelling & Aerospace International Journal of Simulation Navigation and in Engineering Engineering Observation Hindawi Hindawi Hindawi Hindawi Volume 2018 Volume 2018 Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com www.hindawi.com www.hindawi.com Volume 2018 International Journal of Active and Passive International Journal of Antennas and Advances in Chemical Engineering Propagation Electronic Components Shock and Vibration Acoustics and Vibration Hindawi Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018

Journal

Applied Bionics and Biomechanics – Hindawi Publishing Corporation

Published: Mar 19, 2018

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Towards Subject-Specific Strength Training Design through Predictive Use of Musculoskeletal Models

Towards Subject-Specific Strength Training Design through Predictive Use of Musculoskeletal Models

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Towards Subject-Specific Strength Training Design through Predictive Use of Musculoskeletal Models

Towards Subject-Specific Strength Training Design through Predictive Use of Musculoskeletal Models

References (37)

Abstract

Journal

Recommended Articles

There are no references for this article.

Our policy towards the use of cookies