Cochlear Aqueduct Advection and Diffusion Inferred from Computed Tomography Imaging with a Bayesian Approach

Area of Science:

• Biomedical Engineering
• Medical Imaging
• Computational Biology

Background:

• Studying fluid flow and transport within the cochlea is challenging due to its small size and bony encasement.
• Non-invasive imaging techniques are crucial for understanding transport mechanisms in this complex anatomical region.

Purpose of the Study:

• To infer cochlear aqueduct transport parameters using non-invasive Computed Tomography (CT) imaging.
• To apply a Bayesian approach to a 1-dimensional advection-diffusion model for analyzing fluid and solute transport.

Main Methods:

• In vivo CT scans of mice injected with a tracer (iohexol) were performed.
• The advection-diffusion equation was solved using CT data for boundary conditions and inferred transport parameters.
• Bayes' formula and the CUQIpy library were employed for parameter inference, with validation using synthetic data.

Main Results:

• The advection-diffusion model accurately predicted tracer concentration evolution in the cochlear aqueduct.
• Tracer diffusion was generally close to free diffusion, suggesting minimal influence from membranes or flow.
• Inferred advection was near zero in most cases, indicating diffusion dominance.

Conclusions:

• Free diffusion is the primary mechanism for small molecule transport in the cochlear aqueduct.
• Simple 1-dimensional diffusion models can effectively approximate transport in this region.