1,721,039 research outputs found
Pedestrian whole body ground contact mechanisms and head injury assessment following vehicle impact
No full textAccording to the WHO, there are an estimated 1.35 million road-traffic related deaths each year, with pedestrians constituting approximately 22% of this figure, which justifies the necessity of research into vehicle-pedestrian collisions. Previous researchers have primarily focused on the injuries that pedestrians experience as a result of contact with the vehicle, however, ground related injuries and the mechanisms have been largely neglected. Accordingly, the purpose of this study is to significantly strengthen the understanding of pedestrian ground contact, to investigate what factors influence pedestrian ground contact injury severity, and to determine injury prevention strategies.
An analysis of GIDAS vehicle-pedestrian crash data showed that head, thorax, and upper/lower extremities injuries are the most frequent pedestrian ground related injuries. The severity of ground related injuries is greatly affected by vehicle speed and pedestrian age. Older pedestrians are more at risk of suffering thorax injuries. Logistical analysis indicates that normalized bonnet leading edge height (NBLEH) is a predictor of the risk of AIS2 + ground related injuries. Prevention of all ground related pedestrian injuries for vehicle impact speeds below 40 km/h would bring very substantial injury cost reductions.
An analysis of real-world vehicle-pedestrian collision videos from Youtube has been done in Chapter 5 to provide a basic understanding of pedestrian ground contact mechanism. The study consisted of 29 videos and examined the influencing factors that affect the mechanisms of pedestrian ground contact. It was observed that pedestrian projection increases with the vehicle speed, while smaller NBLEH resulted in larger pedestrian rotations, which indicates the potential effects that vehicle front shape has on the resulting pedestrian ground contact injuries.
Six cadaver tests were conducted in LBA, IFSTTAR, Aix-Marseille University, France, which provided data relating to the pedestrian\u27s kinematics during ground contact. It was observed that there is approximately 500ms of continued interaction between the pedestrian and the vehicle until separation occurs, which is followed by a flight period of around 200ms, finally terminating during ground contact. The linear accelerations in ground contact for vehicle impact speeds of 20 and 30 km/h are generally higher than the acceleration in the vehicle contact The predicted risk of rotationally induced brain injury caused by ground contact is high for the 20 km/h test, highlighting the risk of pedestrian injuries from ground contact even at very low speeds.
Validation of both the MB and FE pedestrian models is yet to be completed in ground contact, therefore, a robust comparison of the pedestrian\u27s motion in MADYMO environment with the pedestrian\u27s motion in the cadaver test footage was conducted, thus revealing the competency of the MB model to predict the pedestrian\u27s trajectory during a collision. It shows that contact characteristics of vehicle front-end greatly influence pedestrian post-impact kinematics and the induced injury predictions. In two of these reconstructed simulations, the MB pedestrian model bounded off the vehicle in a dissimilar motion to the motion observed in the staged tests. Although the pedestrian model failed to represent all the cadaver tests with exact kinematics, the model is partially suitable for use in a virtual test system (VTS) under low speed impact configurations.
An inverse method based on a Virtual Test System (VTS) was used to correlate the distribution of impact parameters (vehicle speed, pedestrian height and pedestrian gait) with the predicted injuries, thus allowing the weighting of each parameter (Weighted Injury Costs) with its predicted injury to be determined. VTS showed that there was no significant difference in the WIC scores for the two shapes (\u27Good shape\u27 and \u27poor Shape\u27) in each category of vehicle. Although for the van and SUV categories, the differences become significantly large under test samples of 120. The good shape vehicles are at least not worse in pedestrian-friendly than the poor shape vehicles.
Together these studies provide significant new insights into pedestrian ground contact kinematics and injuries
A Biomechanical Assessment of Direct and Inertial Head Loading in Rugby Union
No full textRugby union is a territorial, dynamic and high-impact collision sport. Unfortunately, due to its physical and high impact nature, the incidence of concussion is high. There is mounting evidence that repeatedly sustaining concussion injuries can lead to long-term brain health issues. Furthermore, the adverse effects of repeated sub-concussive impacts in contact sport are an emerging concept. Despite this, little research has been conducted on the regular head loading environment associated with rugby union. In particular, the magnitude and influencing factors associated with direct head impacts and inertial head loading are poorly understood. Accordingly, the aim of this thesis is to biomechanically assess direct and inertial head loading in rugby union to identify prevention strategies. The thesis is split into two main areas: direct head impacts and inertial head loading. A combination of general video analysis, 3D motion analysis, multibody modelling and a novel approach known as Model-Based Image Matching (MBIM) was utilised. For direct head impacts, an initial aim was to understand how head impacts were occurring in rugby union. A general video analysis review of elite level competitions discovered that the tackle accounted for 60% of direct head impacts. The tackler was much more likely to receive a direct head impact than the ball carrier. Additional video analysis identified tackle characteristics that have a lower propensity to result in a tackler Head Injury Assessment (HIA) and a positive influence on tackle success. Specific tackler proficiency variables were identified such as identify/track ball carrier onto shoulder, head up and forward/face up, shortening steps and head placement on correct side of ball carrier. For the ball carrier, much fewer tackle characteristics were identified, however incorrect fending was identified as a risk factor for upper body front-on tackles. A large majority (81%) of tackle related direct head impacts occurred in the second half of games. A disproportionate number of direct head impacts from upper body tackles (63%) occurred in the final quarter. However, tackling proficiency was found to remain relatively constant throughout the game. Instead, more tackles occur in the final quarter of a game. Further video analysis identified that tackling at the upper trunk accounted for nearly half (47%) of all tackler HIAs and had no greater propensity to result in tackler success outcomes. Tackling at the upper trunk and upper legs had a greater propensity to result in a tackler HIA.
MBIM is a novel approach for measuring six degree of freedom head kinematics from uncalibrated multiple camera view video footage of sporting head impacts. An assessment was conducted on the accuracy of the MBIM method. A vehicle-cadaver head-windscreen impact case was utilised. Reflective marker-based motion capture system head kinematic time-histories were available as an independent measure. The method exhibited Root Mean Square Errors (RMSE) between 10-20 mm for linear displacement and 0.01-0.03 rad for rotational displacement for reconstructing 6 degree of freedom head motion. However, the MBIM method was deemed unsuitable for measuring componential angular velocity during direct head impacts (RMSE up to 5.61 rad/s). For inertial head loading, MBIM was utilised to measure the head kinematics of a visually unaware ball carrier during an active shoulder tackle to the upper trunk. The componential head angular velocities were similar to the average values previously reported for concussive direct head impacts. This is a potentially concern. It was postulated that lower tackle heights may reduce inertial head kinematics for the ball carrier.
Staged tackles in a motion analysis laboratory and multibody modelling simulations indicated that higher tackle heights cause greater ball carrier inertial head kinematics. By tackling below the upper trunk, the multibody simulations suggest that average ball carrier peak head linear acceleration, angular acceleration and change in angular velocity values could be reduced in the tackle by 35%, 61% and 40%, respectively. Based on the staged tackles, median ball carrier peak head linear acceleration, angular acceleration and change in angular velocity values could be reduced in the tackle by 44%, 55% and 57%, respectively. The MADYMO ellipsoid human body model was assessed for reconstructing head kinematics during the abovementioned staged tackles. The results indicated that the model is currently unsuitable for detailed reconstruction of head kinematics on an individual case basis. However, the model identified the kinematic trend that upper trunk tackles cause greater ball carrier inertial head kinematics than mid/lower trunk tackles, even with significant variations in initial player-to-player configurations and speeds.
The findings from this thesis provide an evidence base, at the elite level, for coaches to develop and implement technical based concussion prevention strategies. Focus should be placed on safe and proficient tackle technique. Upper trunk tackles were identified as a risk factor for direct head impacts for tacklers and high inertial head kinematics for ball carriers. Tackling at the upper trunk of the ball carrier should be discouraged. Instead, coaching strategies should place emphasis on tackling at lower HIA risk body regions such as the mid and lower trunk
The mechanics of wound closure for laparoscopic surgery
No full textTHESIS 10725The increase in popularity of laparoscopic surgery over the past 25 years has led to a greater importance of reducing the incidence of complications associated with the surgery. Development of a hernia at a trocar port site is a serious complication with a 1-3% incidence that often results in additional surgery for the patient. There is evidence that the occurrence of a hernia is related to the quality of the wound closure, with an absence of closure being a significant risk factor for the development of a hernia. There is also evidence that a mesh-based, tension-free repair of herniae results in fewer complications than suture repair but there has been limited study on the nature of closure failure or methods to improve it.
Methodical design of a novel closure method, using a tension-free mesh approach that will reduce the incidence of hernia formation requires a fundamental understanding of the abdominal environment and requires facilities to test concepts and prototypes.
To achieve this, an experimental rig representing the abdomen has been developed that incorporates an ability to generate intra-abdominal pressure. This rig can hold either real porcine small intestine or a surrogate material and a real porcine abdominal wall or a surrogate. Simulating an intra-abdominal pressure in the rig causes the intestine to extrude through the abdominal wall, similar to the formation of a hernia following laparoscopic surgery. A surrogate small intestine material has been developed by examining the extrusion properties of porcine small intestine and a number of potential surrogate materials. Reconstituted powdered potato was selected as the most suitable surrogate material that accurately replicates the extrusion properties of small intestine and can be used in the surrogate abdomen rig. A fundamental mathematical and analytical understanding of the mechanics and physiology of hernia formation has also been developed and provides a clear understanding of the root cause of the problem of trocar site herniae and of intra-abdominal pressure. There is limited literature on the structural properties of the abdominal wall, particularly the rectus sheath, which has been shown to be implicated in ventral incisional herniae. A detailed analysis of the uniaxial and biaxial structural properties of the rectus sheath is presented. It was found that the response of porcine rectus sheath to uniaxial loading is similar to human rectus sheath, permitting the use of a porcine model in herniae investigations. Comprehensive stress-stretch plots are also presented which could allow future development of a surrogate abdominal wall for surgical device testing.
The surrogate abdomen rig was partially validated against data from a small observational study of surgical patients. The rig was found to perform well, with hernia generation at pressures similar to those realised physiologically and RPP again proving to be a suitable surrogate for real small intestine. Additionally, a quadratic relationship was established between the pressure required to initiate a hernia and the diameter of the defect.
This quadratic relationship was further developed in an examination of mesh overlap requirements for defect closure. With limited literature on the ideal mesh overlap, and current practice likely to be over-estimating the mesh size required, the surrogate abdomen model was employed in a preliminary study to develop a mesh and defect size relationship. Similar to the findings of the rig validation study, it was found that the relationship between mesh diameter, defect diameter and hernia generation pressure is quadratic, explained by the relationship between pressure, force and area. A mathematical formula developed to predict the required mesh size was found to under-perform due to the complexity of the interaction with tacks used to secure the mesh, however a novel empirical model recommended a mesh twice the hole diameter plus an additional 25mm.
In conclusion, a detailed understanding of the intra-abdominal pressure environment has been developed through fundamental analysis of the physiological processes and tissues involved. This understanding has been used to develop a novel surrogate abdomen environment in which new abdominal surgery devices and techniques can be tested with the aim of reducing the incidence of post-operative complications
Contributions of Muscle, Skin, and Adipose Tissue to Indentation Response, Assessed with Computational Arm Model Under Quasi Static Conditions
Full text linkComputational Human Body Models (HBMs) enable the assessment of human response in potentially injurious impact scenarios; however, to do so HBMs require biofidelic material representations to predict the potential for injury risk. HBMs are recently seeing a widespread application in modelling vehicle occupant response for car crash scenarios. Although the soft tissue deformations in these scenarios occur at high deformation rates, a biofidelic model requires quasi static properties of the tissues prior to the incorporation of deformation rate effects. Specifically, soft tissues such as skeletal muscle, adipose tissue, and skin play an important role in impact response including supporting and protecting the internal organs.
At present, the mechanical responses of muscle, skin, and adipose tissue are measured using excised tissue experiments and assessed at the individual tissue level, providing a valuable source of information on the behaviour of a particular tissue. Current computational models use these experimental results to define individual tissue response. However, the literature review revealed that there is a paucity of experimental and numerical studies assessing the cooperation of the soft tissues and their joint contribution to a loading scenario. The current study addresses this paucity by creating a Simplified Arm Model (SAM) comprising skeletal muscle, adipose tissue, and skin assessed using a quasi static upper arm indentation test. In this study, material properties for the skeletal muscle, adipose tissue, and skin were collected from individual experimental tissue tests reported in the literature. This data set was called New Soft Tissues (NST), and the properties were implemented in the SAM. Finally, the arm model was assessed using a quasi static indentation test and the predicted force displacement response was compared to a published human volunteer data set.
The model predicted the response of the upper arm to the indentation in agreement with the shape and the magnitude of the experimental data. The work performed by the indenter differed by 2% to 65% from the experimental response. To address the variability within the human population, a series of parametric studies were performed assessing: skin thickness, skin age, and the circumference of the arm. The response of the SAM to indentation was close to the experimental average and demonstrated sensitivity to arm diameter. Moreover, the current study showed that the main contributor to the indentation response was the compressive properties of the muscle tissue followed by the adipose tissue, and to a lesser extent the skin. Future research should consider deformation rate effects and the importance of muscle activation on response
A biomechanical analysis of the trunk and spine during paediatric cerebral palsy gait
No full textTHESIS 11497Movement pathologies in the lower limbs of children with cerebral palsy are well established in the literature. However, movement pathologies of the upper body, in particular the trunk, are less well defined. Mechanical loads at the spine and the surrounding areas are influenced by gravity, inertia and externally applied loads and pathological motion patterns could result in mechanical changes in structural tissue surrounding the spine over time. Consequently, this thesis reports an investigation into the relationship between pathological movement of the trunk and loading at the lower lumbar spine. The specific impact of Trendelenburg and Duchenne type movement patterns on loading at the lower lumbar spine was also examined. Prior to assessment, the role of gait analysis in the assessment of trunk and lower lumbar spinal loading was considered. This highlighted a number of issues relating to the kinematic and kinetic models that warranted further investigation. Specifically, those issues related to (1) the choice of body segment parameter set, (2) the choice of hip joint centre regression equation set, (3) the thorax kinematic protocol and (4) the lumbar segment kinematic protocol. Before data were collected to address the primary goals of the thesis, a number of preparatory studies were therefore conducted. The first two preparatory studies identified the most suitable choice of body segment parameter and hip joint centre regression equation sets for the purposes of this thesis. Next, a thorax kinematic protocol was proposed to address some of the practical issues experienced when using other protocols during the assessment procedure. As a preliminary evaluation before use, the protocol was compared to two reference protocols from the literature and performed well and was later used in the thesis. A separate investigation into choice of lumbar segment kinematic protocol identified a skin surface protocol as suitable for studies where axial rotation may be a consideration and so was used in this thesis. With the kinematic and kinetic models in place, 3-dimensional thoracic, lumbar and L5/S1 kinetics were measured in 52 children with cerebral palsy and 26 typically developed children. Differences were present for cerebral palsy children, most notably in the coronal plane for thorax kinematics and L5/S1 kinetics. Peak L5/S1 moment data were up to 21% higher than normal for GMFCS level one children and up to 90% higher than normal for GMFCS level two children. Moderate to strong correlations were evident between movement of the thorax and L5/S1 loading (r = 0.52). However, other factors may have contributed to this loading and further investigation was suggested, perhaps by means of forward dynamics, to determine other underlying factors that may contribute to this loading. The final investigation of this thesis examined the impact of Duchenne and Trendelenburg type gait on loading at the lower lumbar spine in children with cerebral palsy. The same cohort of subjects was divided according to clinical presentation of each pattern. Trendelenburg gait had little impact on L5/S1 loading. However, increased loading was evident where Duchenne type movements were present. To conclude, increased loading was evident at the lower lumbar spine during cerebral palsy gait. This loading was related to the position of the thorax. It is important that interventions relating to movement of the trunk during cerebral palsy gait, or specifically related to Duchenne or complex Trendelenburg-Duchenne type gait, are aimed at reducing trunk motion specifically in the coronal plane in order to reduce abnormal loading which could, in turn, impact the health of the spine in this population
The relationship between micro-structure and mechanical behaviour in passive skeletal muscle
No full textTHESIS 10800This thesis presents and discusses investigative work performed on characterising the behavioural response of skeletal muscle tissue that was subjected to large deformations. The passive mechanical properties of muscle tissue are important for many biomechanics applications, including impact biomechanics, tissue engineering and rehabilitation engineering. However, significant gaps remain in our understanding of the passive three-dimensional tensile and compressive response of skeletal muscle tissue. In particular, the tensile quasi-static soft tissue anisotropy remains unclear and the responses to loading at intermediate fibre directions, as well as the asymmetrical behavioural response of muscle tissue have not been investigated before
An improved framework for the inverse analysis of skeletal muscle tissue in-vivo
No full textTHESIS 9682This thesis focusses on the development of an experimental and computational framework for the non-invasive analysis of the passive mechanical properties of living human skeletal muscle tissue. This is relevant to many areas of research including impact biomechanics and rehabilitation engineering. Although constitutive models have been proposed for muscle tissue these have been insufficiently validated for human tissue which requires non-invasive methods. Non-invasive analysis of the mechanical properties of soft tissue requires non-invasive mechanical exciting and inverse analysis of non-invasively measured experimental boundary conditions such as tissue deformation and applied load. Magnetic Resonance Imaging (MRI) provides excellent soft tissue contrast without ionizing radiation. In addition it allows for the measurement of soft tissue anatomy, architecture and deformation boundary conditions. Hence for mechanical excitation a novel MRI compatible and computer controllable soft tissue indentation device was developed and implemented with an accurate high acquisition rate (100Hz) optical force sensor capable of viscoelastic force registration. In order to measure the resultant deformation SPAtial Modulation of the Magnetisation (SPAMM) tagged MRI was used. Traditional SPAMM tagging methods require large numbers of repetitions of motion cycles causing repeatability difficulties and volunteer discomfort. However for this thesis a unique set of high speed SPAMM tagged MRI techniques, and fully automatic post-processing methods based on Gabor wavelet filtering, were developed allowing for the measurement of complex dynamic 3D deformation following the combination of just 3 motion cycles. The SPAMM tagged MRI techniques were validated using marker tracking in a silicone gel phantom and underwent in-vivo evaluation whereby sub-voxel accuracy and precision levels were reported. Constitutive models for passive skeletal muscle tissue were evaluated using inverse Finite Element (FE) Analysis (FEA) based fitting to experimental data from the literature. It was shown that current models do not allow appropriate modelling of anisotropy. A new constitutive law was proposed which formed a close match to the data and was based on Gaussian weighting of transverse and longitudinal direction contributions of a spherical fibre distribution model. This thesis focusses on the development of an experimental and computational framework for the non-invasive analysis of the passive mechanical properties of living human skeletal muscle tissue. This is relevant to many areas of research including impact biomechanics and rehabilitation engineering. Although constitutive models have been proposed for muscle tissue these have been insufficiently validated for human tissue which requires non-invasive methods. Non-invasive analysis of the mechanical properties of soft tissue requires non-invasive mechanical exciting and inverse analysis of non-invasively measured experimental boundary conditions such as tissue deformation and applied load. Magnetic Resonance Imaging (MRI) provides excellent soft tissue contrast without ionizing radiation. In addition it allows for the measurement of soft tissue anatomy, architecture and deformation boundary conditions. Hence for mechanical excitation a novel MRI compatible and computer controllable soft tissue indentation device was developed and implemented with an accurate high acquisition rate (100Hz) optical force sensor capable of viscoelastic force registration. In order to measure the resultant deformation SPAtial Modulation of the Magnetisation (SPAMM) tagged MRI was used. Traditional SPAMM tagging methods require large numbers of repetitions of motion cycles causing repeatability difficulties and volunteer discomfort. However for this thesis a unique set of high speed SPAMM tagged MRI techniques, and fully automatic post-processing methods based on Gabor wavelet filtering, were developed allowing for the measurement of complex dynamic 3D deformation following the combination of just 3 motion cycles. The SPAMM tagged MRI techniques were validated using marker tracking in a silicone gel phantom and underwent in-vivo evaluation whereby sub-voxel accuracy and precision levels were reported. Constitutive models for passive skeletal muscle tissue were evaluated using inverse Finite Element (FE) Analysis (FEA) based fitting to experimental data from the literature. It was shown that current models do not allow appropriate modelling of anisotropy. A new constitutive law was proposed which formed a close match to the data and was based on Gaussian weighting of transverse and longitudinal direction contributions of a spherical fibre distribution model
Tension and compression stress-strain asymmetry in passive skeletal muscle
No full textTHESIS 11341The general aim of this study is to advance the knowledge of the relationship between the skeletal muscle passive compressive and tensile behaviour, and the microstructure of the muscle through combined experimental, microstructural and theoretical approaches. The mechanics of passive skeletal muscle are important in many biomechanical applications. Existing data from porcine tissue has shown a significant tension/compression asymmetry, which is not captured by current constitutive modelling approaches using a single set of material parameters, and an adequate explanation for this effect remains elusive. In this thesis, the passive elastic deformation properties of chicken pectoralis muscle are assessed for the first time, to provide deformation data on a skeletal muscle which is very different to porcine tissue. Uniaxial, quasi-static compression and tensile tests were performed on fresh chicken pectoralis muscle in the fibre and cross-fibre directions, and at 45? to the fibre direction. Results show that chicken muscle elastic behaviour is nonlinear and anisotropic. The tensile stress?stretch response is two orders of magnitude larger than in compression for all directions tested, which reflects the tension/compression asymmetry previously observed in porcine tissue. In compression the tissue is stiffest in the cross-fibre direction. However, tensile deformation applied at 45? gives the stiffest response, and this is different to previous findings relating to porcine tissue. Chicken muscle tissue is most compliant in the fibre direction for both tensile and compressive applied deformation. Generally, a small percentage of fluid exudation was observed in the compressive samples. Since collagen is the main structural protein in animal connective tissues, it is believed to be primarily responsible for their passive load-bearing properties. The direct detection and visualisation of collagen using fluorescently tagged CNA35 binding protein (fused to EGFP or tdTomato) is also reported for the first time on fixed skeletal muscle tissue. A working protocol is then established by examining tissue preparation, dilution factor, exposure time etc. for sensitivity and specificity. Penetration of the binding protein into intact mature skeletal muscle was found to be very limited, but detection works well on tissue sections with higher sensitivity on wax embedded sections compared to frozen sections. CNA35 fused to tdTomato has a higher sensitivity than CNA35 fused to EGFP but both show specific detection. Best results were obtained with 15 ?m wax embedded sections, with blocking of non-specific binding in 1% BSA and antigen retrieval in Sodium Citrate. There was a play-off between dilution of the binding protein and time of incubation but both CNA35-tdTomato and CNA35-EGFP worked well with approximately 100 ?g/ml of purified protein with overnight incubation, while CNA35- tdTomato could be utilized at 5 fold less concentration. The tension/compression asymmetry observed in the stress-strain response of skeletal muscle is not well understood. The optimised protocol is then applied to report qualitatively on skeletal muscle ECM reorganization during applied deformation using a combination of CNA35 binding protein and confocal imaging of tensile and compressive deformation of porcine and chicken muscle samples applied in both the fibre and cross-fibre directions. Results show the overall three-dimensional structure of collagen in perimysium visible in planes perpendicular (w1) and parallel (w2) to the muscle fibres in both porcine and chicken skeletal muscle. Furthermore, there is clear evidence of the reorganization of these structures under compression and tension applied in both the muscle fibre and cross-fibre directions, which generally explains anisotropy observed in the stress-strain response of skeletal muscle both in tension and compression for chicken and porcine tissues. These observations improve our understanding of how perimysium responds to three-dimensional deformations.
The proposed three-dimensional illustration of perimysium structure is then used as a basis to create a microstructural-geometrical model to predict the passive mechanical stress-strain response observed in skeletal muscle. The current model represents the whole muscle response as a combination of both a group of muscle fibres (fascicle) response and the perimysium (ECM) response. It shows that although perimysium was believed to be a key element in the muscle stress response, the muscle fibres (in Tension-Fibre and Compression-XFibre deformations) also contribute to stress-stretch response since the order of magnitude for the stress in muscle fibres is similar to that of perimysium. The model shows more asymmetric response than previously published micromechanical model (Gindre et al., 2013). The model yields a good prediction of the whole muscle behaviour in Tension-Fibre and Compression-Fibre deformations using the optimum values for the model parameters obtained from the conducted sensitivity studies; connective tissue percentage of pc=1.75 , Elast modulus of Ec=300 MPa, and perimysium sheet waviness of w=1.25. However, the model overestimates the Compression-XFibre deformation and underestimates the Tension-XFibre deformations even by using the optimum parameters. The current model attempts to relate the mechanical stress-stretch response observed in muscle to the collagen reorganization in the muscle microstructure under load application, which further help develop better constitutive models for finite element modelling purposes.
The general aim of this study is to advance the knowledge of the relationship between the skeletal muscle passive compressive and tensile behaviour, and the microstructure of the muscle through combined experimental, microstructural and theoretical approaches. The mechanics of passive skeletal muscle are important in many biomechanical applications. Existing data from porcine tissue has shown a significant tension/compression asymmetry, which is not captured by current constitutive modelling approaches using a single set of material parameters, and an adequate explanation for this effect remains elusive. In this thesis, the passive elastic deformation properties of chicken pectoralis muscle are assessed for the first time, to provide deformation data on a skeletal muscle which is very different to porcine tissue. Uniaxial, quasi-static compression and tensile tests were performed on fresh chicken pectoralis muscle in the fibre and cross-fibre directions, and at 45? to the fibre direction. Results show that chicken muscle elastic behaviour is nonlinear and anisotropic. The tensile stress?stretch response is two orders of magnitude larger than in compression for all directions tested, which reflects the tension/compression asymmetry previously observed in porcine tissue. In compression the tissue is stiffest in the cross-fibre direction. However, tensile deformation applied at 45? gives the stiffest response, and this is different to previous findings relating to porcine tissue. Chicken muscle tissue is most compliant in the fibre direction for both tensile and compressive applied deformation. Generally, a small percentage of fluid exudation was observed in the compressive samples. Since collagen is the main structural protein in animal connective tissues, it is believed to be primarily responsible for their passive load-bearing properties. The direct detection and visualisation of collagen using fluorescently tagged CNA35 binding protein (fused to EGFP or tdTomato) is also reported for the first time on fixed skeletal muscle tissue. A working protocol is then established by examining tissue preparation, dilution factor, exposure time etc. for sensitivity and specificity. Penetration of the binding protein into intact mature skeletal muscle was found to be very limited, but detection works well on tissue sections with higher sensitivity on wax embedded sections compared to frozen sections. CNA35 fused to tdTomato has a higher sensitivity than CNA35 fused to EGFP but both show specific detection. Best results were obtained with 15 ?m wax embedded sections, with blocking of non-specific binding in 1% BSA and antigen retrieval in Sodium Citrate. There was a play-off between dilution of the binding protein and time of incubation but both CNA35-tdTomato and CNA35-EGFP worked well with approximately 100 ?g/ml of purified protein with overnight incubation, while CNA35- tdTomato could be utilized at 5 fold less concentration. The tension/compression asymmetry observed in the stress-strain response of skeletal muscle is not well understood. The optimised protocol is then applied to report qualitatively on skeletal muscle ECM reorganization during applied deformation using a combination of CNA35 binding protein and confocal imaging of tensile and compressive deformation of porcine and chicken muscle samples applied in both the fibre and cross-fibre directions. Results show the overall three-dimensional structure of collagen in perimysium visible in planes perpendicular (w1) and parallel (w2) to the muscle fibres in both porcine and chicken skeletal muscle. Furthermore, there is clear evidence of the reorganization of these structures under compression and tension applied in both the muscle fibre and cross-fibre directions, which generally explains anisotropy observed in the stress-strain response of skeletal muscle both in tension and compression for chicken and porcine tissues. These observations improve our understanding of how perimysium responds to three-dimensional deformations.
The proposed three-dimensional illustration of perimysium structure is then used as a basis to create a microstructural-geometrical model to predict the passive mechanical stress-strain response observed in skeletal muscle. The current model represents the whole muscle response as a combination of both a group of muscle fibres (fascicle) response and the perimysium (ECM) response. It shows that although perimysium was believed to be a key element in the muscle stress response, the muscle fibres (in Tension-Fibre and Compression-XFibre deformations) also contribute to stress-stretch response since the order of magnitude for the stress in muscle fibres is similar to that of perimysium. The model shows more asymmetric response than previously published micromechanical model (Gindre et al., 2013). The model yields a good prediction of the whole muscle behaviour in Tension-Fibre and Compression-Fibre deformations using the optimum values for the model parameters obtained from the conducted sensitivity studies; connective tissue percentage of pc=1.75 , Elast modulus of Ec=300 MPa, and perimysium sheet waviness of w=1.25. However, the model overestimates the Compression-XFibre deformation and underestimates the Tension-XFibre deformations even by using the optimum parameters. The current model attempts to relate the mechanical stress-stretch response observed in muscle to the collagen reorganization in the muscle microstructure under load application, which further help develop better constitutive models for finite element modelling purposes
- …
