Project Details
Projekt
Modelling, Simulation and Optimisation for Agonist-Antagonist Myoneural Interface Surgeries
Applicants
Professor Dr. Dominik Göddeke; Professor Oliver Röhrle, Ph.D.; Professorin Dr. Miriam Schulte
Subject Area
Applied Mechanics, Statics and Dynamics
Term
since 2021
Project identifier
Deutsche Forschungsgemeinschaft (DFG) - Project number 465243391
This continuation proposal is centred around the assumptions and goals of our original proposal, i.e., to design, implement and evaluate an in-silico simulation and optimisation framework that enables a deeper understanding of the neuromuscular system. It will include matching workflows and strategies for (clinical and simulation) data handling. We build upon a unique opportunity offered through a collaboration started in the first funding period: By teaming up with a trauma surgeon and focusing on the novel agonist-antagonist myoneural interface (AMI) limb amputation technique, we work towards a concrete pilot application. However, we keep the applicability of our approaches and solutions for a wider context as an explicit goal of the project. In particular, our framework will contribute to enabling surgeons, medical doctors in general, but also modellers to improve their understanding of adaptation mechanisms, signal pathways and feedback mechanisms within the neuromuscular system. The AMI is a novel limb amputation technique currently performed by very few groups worldwide that aims to preserve signal pathways and neural feedback loops. In surgery practice, expert decisions on how to truncate agonist and antagonist muscles are needed, and equally importantly, how to connect them through an (artificial) tendon. One important parameter for the quality of the surgery outcome is the resulting pre-stretch in both muscles. The pre-stretch describes the length difference between a (theoretical) completely relaxed muscle and the `most relaxed' position in the actual physical setting with attachment of the muscles via tendons to bones and to each other, respectively. We have already contributed to understanding the role of pre-stretch in the first funding period by providing a first prototype for a detailed coupled simulation of two-muscle-one-tendon systems and by developing a Bayesian inference approach for pre-stretch as a parameter for a separate simplified model. The proposed project aims at advances in terms of (i) modelling the feedback mechanisms in realistic skeletal muscle pairs, (ii) their efficient simulation utilising HPC resources, (iii) efficient and accurate surrogate models, and (iv) the development of Bayesian techniques for parameter inference and optimisation using the surrogate models. The entire project entails a close interaction and combination of expertise, in the form of co-design, between the pilot application and the modelling-simulation-optimisation cycle, involving shared expertise brought in by the unique composition of the group of applicants and the context of the priority programme. We envision outreach of our methodological and factual findings beyond the hypothesis-driven AMI-related questions we explicitly target in this proposal. Not only for the pilot application, we consider FAIR data handling and its extension to FAIR software to be crucial to achieving our work's intended impact and reproducibility.
DFG Programme
Priority Programmes
