Thesis Project Form

Due to failure rates of up to 40% in certain surgical procedures, such as septoplasty and turbinoplasty, multidisciplinary approaches incorporating computational fluid dynamics (CFD) and heat transfer analysis are increasingly being adopted to support surgical decision-making. In this work, we focus on modeling the mucus layer to improve the prediction of convective heat transfer and temperature distributions within the nasal cavity. All investigations are conducted on patient-specific geometries reconstructed from computed tomography (CT) scans.

The nasal mucus layer is a thin liquid film covering the epithelial surface of the nasal cavity. Besides its biological functions, it plays an important role in the thermofluid dynamics of nasal airflow by mediating heat and moisture transfer between the mucosal tissue and the inspired air. From a modelling and computational point of view the mucus layer is not trivial. The nasal mucus layer is a thin, viscoelastic fluid coating the nasal epithelium. It humidifies and warms inhaled air, traps particles and pathogens, and protects the underlying tissue. Composed mainly of water, mucins, salts, and proteins, it is continuously transported by ciliary motion toward the nasopharynx. Moreover its thickness is much smaller than the characteristic width of the nasal cavity. To avoid discretising the mucus layer an equivalent boundary condition has been developed by the group that models, perfectly, the presence of the finite mucus layer. This model will be used in this work to investigate, parametrically, the influence of the mucus layer thickness distribution and heat transfer properties on the wall temperature, and temperature distribution in the nasal cavity. Different geometries and external conditions will be investigated.

The student will learn how to setup and run computational fluid dynamics, and heat transfer analysis using an industrial opensource software. All steps in the work flow, such as creating the geometry from CT scans, creating the computational mesh, setting up the numerical framework, running simulations and post-processing the results, will be covered.