114. Computer-aided personalized diagnostics in rhinology

Moritz Waldmann – RWTH Aachen University, Chair of Fluid Mechanics and Institute of Aerodynamics. W. Schröder – Jülich Aachen Research Alliance Center for Simulation and Data Science (JARA-CSD), RWTH Aachen University and Forschungszentrum Jülich GmbH, Aachen, Germany


Background: Due to the increasing computational power of High-Performance Computing (HPC) systems, numerical simulations have become more demanding in recent years. They enable to investigate increasingly complex fluid flows as, for example, the respiration in the upper airways based on realistic, patient-specific surface geometries obtained from CT data sets. This opens the possibility to develop new diagnostic methods. For instance, in rhinology, numerical simulations can be employed for surgery planning and evaluation1.

Methods: The presented approach is based on a thermal lattice-Boltzmann method (LB)2. The solver is part of the simulation framework multiphysics – Aerodynamisches Institut Aachen (m-AIA)3. Since the LB method is well suited to efficiently simulate incompressible flows in intricated geometries, it is the ideal approach to investigate the respiratory flow numerically. Furthermore, it can be coupled with other solvers implemented in the m-AIA framework such as a level-set (LS) method. Such a coupling allows to track geometry modifications during a simulation run enabling the user to perform virtual surgeries on patient-specific nose geometries.

Results: The presented approach is tested by simulating different nasal cavities. Additionally, a study on possible error sources and the reliability of the simulation results is presented. The results show that a comparison with in-vivo measurements is challenging. Nevertheless, the impact of the shape of the cavities on respiration can be investigated. Hence, a virtual surgery of a deviated septum and a bone spur shows promising results.

Conclusion/Outlook: Numerical methods can be used to for surgery planning and evaluation. The presented simulation results show that numerical methods can be used to improve a surgery planning. To further automatize the presented method, the solvers are currently coupled with a machine learning algorithm, which is trained to find the optimal geometry in the given limits based on fluid mechanical properties of respiration4.


[1] A. Lintermann, M. Meinke, W. Schröder: „Fluid mechanics based classification of the respiratory efficiency of several nasal cavities”, Computers in Biology and Medicine 43 (11), 2013, pp. 1833-1852, doi: 10.1016/j.compbiomed.2013.09.003

[2] M. Waldmann, M. Rüttgers, A. Lintermann, W. Schröder: „Virtual Surgeries of Nasal Cavities Using a Coupled Lattice-Boltzmann-Level-Set Approach”, ASME Journal of Engineering and Science in Medical Diagnostics and Therapy 5 (3), 2022, pp. 031104, doi: 10.1115/1.4054042

[3] A. Lintermann, M. Meinke, W. Schröder: „Zonal Flow Solver (ZFS): a highly efficient multi-physical simulation framework”, International Journal of Computational Fluid Dynamics 34 (7-8), 2020, pp. 458-485, doi: 10.1080/10618562.2020.1742328

[4] M. Rüttgers, M. Waldmann, W. Schröder, A. Lintermann: „Machine-Learning-Based Control of Perturbed and Heated Channel Flows”, In: Jagode, H., Anzt, H. Ltaief, H., Luszczek, P. (eds) High Performance Computing, ISC High Performance 2021, Springer, Cham, doi: 10.1007/978-3-030-90539-2_1

About the author

Moritz Waldmann is a researcher at the Chair of Fluid Mechanics and Institute of Aerodynamics of RWTH Aachen University. He finalized his Ph.D. studies in September 2023. During his time as a Ph.D. student, his research was focused on the fluid mechanical analysis of respiration using the lattice-Boltzmann method.