Abstract

The usage of 1D and 3D models to simulate drug transport through the inner ear is a prominent method in cochlear fluid pharmacokinetics. However, the data used to create these models, is often based on invasive sampling methods that limit the spatial resolution given the size of the cochlear compartments within which solute can be measured. In this work by leveraging 3-D registered micro-Computed Tomography (μCT) scans of the murine cochlea that have been taken as iodinated contrast agent is delivered to it, we extract transport parameters and simulate a forward 1D model that allows variable and pulsatile delivery profiles over time and can be extended to the use of other drugs.

Our 1D model may be used to simulate transport of a compound within the primary scalae or compartments of the cochlea namely: scala tympani (ST), scala vestibuli (SV) and scala media (SM). We investigate extracting transport parameters of the 1D model for the iodinated contrast agent (Iopamidol), such as the concentration dependent diffusion coefficient, along with permeabilities across membranes that represent transfer between the primary scalae and clearance out to blood. Flow rates that change over time are also learned, to account for leakage due to experimental set up. Dimensions of cochlear structures considered in the model, and empirical concentration profiles are extracted non-invasively over regional cross sections of a set of registered μCT scans of the mouse cochlea while an Iodinated contrast agent (Iopamidol) is delivered to it. Given initial estimates of the transport parameters, we use a simple iterative gradient descent approach to minimize the mean-squared error between our predicted concentrations from the 1D model and those derived empirically. We put forth a method to illustrate that once these parameters are learned for the contrast agent, they can be adjusted to simulate the delivery of other compounds, and can also be used to study various infusion paradigms to maintain a suitable therapeutic window for optimal, effective and safe administration of a drug. The results are important in the development of such paradigms for the prevention and treatment of acute and chronic types of hearing loss.

Library of Congress Subject Headings

Drug delivery systems--Computer simulation; Cochlea--Imaging; Veterinary tomography

Publication Date

12-5-2018

Document Type

Thesis

Student Type

Graduate

Degree Name

Electrical Engineering (MS)

Department, Program, or Center

Electrical Engineering (KGCOE)

Advisor

David A. Borkholder

Advisor/Committee Member

Nathan Cahill

Advisor/Committee Member

Majid Rabbani

Campus

RIT – Main Campus

Plan Codes

EEEE-MS

Share

COinS