Prof. Dr. Anne Koelewijn
Chair of Machine Learning and Data Analytics Lab

Prof. Dr. Anne Koelewijn
Chair of Machine Learning and Data Analytics Lab

It is my goal to improve movement for people, e.g. those with a disability or athlete. To do so, I aim to better understand human motion, and design better devices, such as prostheses, exoskeletons, and running shoes, as well as prevent injuries, such as knee osteoarthritis. I focus on wearable technologies and the combination of physics-based models with machine learning methods.

Research Projects

Personalized musculoskeletal models and gait simulations: using imaging techniques such as diffusion tensor imaging (DTI), as well as machine learning methods, we aim to personalize musculoskeletal models and gait simulations.

  • FossilGaitSim: FossilGaitSim: Exploring the Major Postural Transition in the Evolution from Synapsids to Mammals through Movement Simulations


    (Third Party Funds Single)
    Project leader:
    Term: 1. May 2026 - 30. April 2029
    Acronym: FossilGaitSim
    Funding source: DFG-Einzelförderung / Sachbeihilfe (EIN-SBH)

    We will investigate the evolutionary trajectory from sprawling, lizard-like, gait to cursorial, erect, gait in mammals and their ancestors using movement simulations and fossil trackways. We have selected five ichnotaxa, fossil records of movements, spanning from the early Permian to the Miocene, that cover the transition from sprawling to cursorial gait. We will investigate likely gaits and postures used to create these trackways, as well as the trackmaker’s morphometry, using movement simulations which are created by solving optimal control problems. The optimal control problem replicates the optimization that is solved by the central nervous system when planning and executing movements. Such simulations can successfully predict different human movements, by finding the movement that is most energy efficient. Similarly, we will explore optimal control problems as a tool to find the gait and posture used to create fossil trackway patterns. The dynamics model and its morphometric parameters (or parameter ranges) will be based on different information sources, from trackway measurements as well as from modern analogues and (incomplete) body fossils, if they are available. First, we will develop an adaptable full-body dynamics model that will be used to create the simulations. The adaptability ensures that different animals can be investigated by adapting the morphometric model parameters, which overcomes the limitation of current approaches, in which dynamics models of extinct animals are created from time-consuming scanning of body fossils. We will validate this model and our simulation approach using trackway and video data of extant animals, reptiles as basal relatives and mammals as derived relatives. Second, we will investigate the relationship between morphometry, gait, and posture. Using a sensitivity analysis, we aim to gain more insight into which morphometric parameters affect gait and posture, and especially the resulting trackway pattern. This analysis will inform us on the likely range of the morphometric parameters. We will also use phylogenetic trees and evolutionary optimization to explore the evolutionary trajectory of gait and morphometry using these simulations and the trackway information. Third, we will investigate the five selected trackways and find the likely gait and posture used to create them, as well as trackmaker morphometry, to explore the evolutionary trajectory from sprawling to cursorial gait. To do so, we will first gather and select trackways that are suitable for simulations for the five ichnotaxa, and then develop dynamics models based on different information source. Using these, we can create simulations and evaluate which gait and posture likely created these trackways.

  • FAME: Bridging the gap in ACL injury prevention with FAME: Field-based Athlete Motion Evaluation and simulation


    (Third Party Funds Single)
    Project leader:
    Term: since 15. January 2024
    Acronym: FAME
    Funding source: Deutsche Forschungsgemeinschaft (DFG)

2026

2025

2024

2023

2022

2021

Related Research Fields