Tobias Siebert is passionate about understanding the contraction mechanisms of a muscle at the smallest level as well as seeing what happens to the muscle fibers when the muscle contracts or stretches and how it can store its energy. The Professor of the Institute of Sport and Movement Science at the University of Stuttgart wants to find out how movement occurs and how proteins such as myofilaments, myosin, actin, and titin – the smallest building blocks of muscles – interact. Together with Prof. Oliver Röhrle from the Institute for Modeling and Simulation of Biomechanical Systems, he is investigating the human tibialis anterior muscle. “In these experiments, we investigate superficial muscles that have a long end tendon, which makes them easier to study,” he explains. The tibialis anterior also has another advantage: When the ankle joint moves and the foot is pulled up, it is the only muscle that is activated. This means that measurements can be clearly assigned to this muscle during a movement.
Muscle fibers are cellular structures that are about the width of a single human hair. They are contained in skeletal muscles, which are important for locomotion. Among other things, they consist of myofilaments, which are thread-like proteins, and contain mainly myosin (a motor protein) and actin. Titin, the largest protein in the human body, is another myofilament important for muscle structure, mechanosensing (response to mechanical stimuli), and force generation. If the muscle fibers are not aligned in the line of action between the muscle insertion and the origin but rather at an angle, they are referred to as pinnate muscles. The pennation angle is the angle at which the muscle fibers are oriented in relation to this line of action. This angle increases or decreases with movement. The greater the pennation angle, the lower the force transferred to the tendon and therefore to the skeleton.
Innovative 3D ultrasound scanner
In order to examine the inner workings of the muscle, the scientists in Siebert’s team are using a new type of 3D ultrasound scanner developed by doctoral candidate Annika Sahrmann. This allows them to visualize the fascicles (bundles of muscle fibers) not only when the muscle is at rest but also when it is moving. The aim is to measure as many properties of the muscle as possible in order to determine the muscle architecture based on this data. They first extracted the muscle properties from 2D ultrasound images. They were then able to obtain general information about skeletal muscles. For example how the 'force and length' or the 'contraction velocity and force' of the muscle are related as well as the thickness of the muscles, the length of the muscle fibers, and their pennation angle. This information can be used to make statements about force generation and architectural changes.
Muscle contraction
What happens inside the muscle when it is contracted? The light spots are the collagen or fascia tissue from which the movement of the muscle fibers can be inferred. The muscle fibers are enveloped by collagen tissue but are not visible because of their small diameter (they have the same width as a human hair). The thicker white line in the center is the aponeurosis, a flat connective tissue structure that originates from the end of the tendon. The angle of the fibers to the tendon or aponeurosis is referred to as the pennation angle.
Function of the 3D ultrasound scanner
The properties of the muscle fibers are then determined using 3D ultrasound images. The scientists use an ultrasound scanning system specially developed for this purpose. In order to generate the 3D images, the ultrasound probe is placed transversely (i.e. perpendicular to the leg axis). This produces many images of the cross-section of the muscle. Using a special algorithm, these images are then stitched together slice by slice to create a three-dimensional image of the muscle.
“The 3D ultrasound images are generated from the 2D images. You take the various cross-sections and recognize the position of the respective cross-section image in relation to a global coordinate system. This can then be used to create a 3D image on which the fascia movement can be observed during dynamic movements,” explains Lukas Vosse, who is carrying out this study as part of his doctoral thesis. The aim is to determine the muscle architecture from these 3D ultrasound images during dynamic movements.
Many processes during movement have not yet been researched
Understanding these mechanisms of muscle contraction at the smallest level is important in order to be able to model a muscle. “In our SimTech project (PN 2-8), we aim to create a finite element model of a muscle,” says Siebert. The scientists then want to use this model to complete the digital human model. “There are already many human models. But they are usually not detailed enough to depict all the mechanisms.” In addition to muscle geometry, lateral forces are an important aspect. When the muscle contracts and stiffens, it generates forces not only in the direction of pull but also in the lateral direction.
The finite element method is a tool in which a muscle is subdivided into a finite number of tiny components. The behavior of these individual particles or elements can then be used to draw conclusions about the overall behavior of the muscle. This method can then be used to carry out simulations, among other things.
This lateral force is used by the body to push something away or stabilize it. “These lateral forces are extremely important but are neglected in 99% of all models,” says Siebert. Many processes that take place during movement have not yet been researched. “For example, we don’t know exactly how the force is generated during eccentric movement; this is important when running, jumping, and slowing down. Muscles are stretched, and energy is stored as in a spring.
That’s why we are so effective and could walk for 24h continuously (i.e. around 100 km) if our lives were at stake. No robot can do that off-road at the moment. This has to do with energy storage – how you store energy during an eccentric impact and can then reuse it for the next step,” says Siebert. In the eccentric phase, the activated muscles are stretched and can generate high forces.
Personalized models for medical diagnostics
This is why the SimTech researchers are interested not only in how muscles function on a small scale but also how they work together on a large scale to perform a stable and robust movement. Another goal is to then personalize these models because every human system functions individually and therefore differently. To this end, they want to carry out another study with test subjects who have as many different characteristics as possible in terms of age, sex, weight, and physical condition.
“At the moment, most models used for simulations are standard or mean value models,” says Siebert. “They usually depict a young man aged 25 who is around 1.75 to 1.80 m tall. But not an older woman or an obese person.” If you want to calculate the risk of injury using the muscle models, you can’t simply transfer it to someone else. “You first have to know the basic mechanisms that are the same for everyone and then find out which parameters in the model need to be changed in order to personalize it.”
Siebert is particularly interested in the neuromuscular system (i.e. the interplay between muscle control and force generation). By capturing data in 3D, the scientists in the SimTech project PN 2-8 are the first to determine a 3D muscle architecture during active dynamic contractions. Individualized models could be used for medical diagnostics in neuromuscular diseases such as Parkinson’s.
They could be helpful for fitting prostheses or predicting how a body will react in the event of an accident and making recommendations accordingly in order to minimize the risk of injury. In the field of competitive sports, training plans could be tailored even more effectively to athletes. It would also be possible to predict the influence of chemicals or drugs on muscle fibers.
Processing data with machine learning and artificial intelligence
From the measurement of the tibialis anterior, Siebert would then like to draw conclusions about the other muscles in our body: the heart muscle or the smooth muscles that occur in our hollow organs such as the bladder, stomach, or intestines. The procedure is always the same. First, the properties of the respective muscle are determined. These are then incorporated into a model until the human model is complete. The only remaining problem is the abundance of data. The 3D models of the muscle are extremely computationally intensive. It would be impossible to process this amount of data for all the muscles in a body. The vision is to solve this using machine learning and artificial intelligence methods. The scientists will continue to research this.
Read more
Sahrmann, A. S., Vosse, L., Siebert, T., Handsfield, G. G., & Röhrle, O. (2024). 3D ultrasound-based determination of skeletal muscle fascicle orientations. Biomech Model Mechanobiol. DOI: https://doi.org/10.1007/s10237-024-01837-3
Oliver Röhrle and Annika Sahrmann developed and validated the system for the automated and controlled acquisition of 3D ultrasound images with an integrated force control mechanism to ensure consistent tissue deformation as part of the SimTech project PN 2-8, among others. The 3D ultrasound scanning system has been patented.
About the scientists
Tobias Siebert has been working with muscles for over 20 years. He initially studied sport and biology as a teacher at the University of Jena and was offered a doctoral position in muscle physiology by his mentor Professor Reinhard Blickhan. After completing his doctoral degree studies, he completed his internship and second state examination before working on microstructural muscle models as a postdoctoral fellow. He did his habilitation and was offered a professorship in Stuttgart, where he has held the Sport and Movement Science chair since 2013.
He has done research with spiders, cockroaches, frogs, rats, and cats and is fascinated by how perfectly the locomotor system has developed over hundreds of millions of years across all living creatures. He is involved in several projects at SimTech and develops microstructural muscle fiber models that are intended to contribute to a better understanding of force development.
Lukas Vosse studied cybernetics, biomechanics, and movement sciences at the University of Stuttgart and worked as a research assistant at various university institutes and at the Fraunhofer Institute for Manufacturing Engineering and Automation IPA in Stuttgart. There he conducted basic research into human physiology, created CAD models of hard and soft tissue, worked on the segmentation of the muscle-tendon complex, and carried out ultrasound experiments. He is now doing his doctorate in the SimTech project “Modeling of architectural-informed and activation-driven contractions of the human M. tibialis anterior” and is investigating the properties of the tibialis anterior muscle in order to create a 3D model of it. He also works as a taekwondo martial arts trainer.
