Date of Award
Spring 5-2025
Degree Name
Master of Science (MS)
Degree Type
Thesis
Abstract
Endovascular devices like catheters feature non-uniform flexural rigidity (EI) profiles (flexi- ble tip, stiff shaft) crucial for navigation and pushability. Standard three-point bend analysis (assuming uniform stiffness) yields inaccurate apparent rigidity (B0) for these graded devices, especially in transition zones. This inaccuracy hinders device comparison, clinical selection, and predictive modeling. This thesis develops and validates a correction model for three-point bend tests aimed at improving local flexural rigidity accuracy for devices with varying stiffness, particularly those exhibiting transition zones similar to logistic profiles. The model assumes a linear gradient in rigidity (B(x) = Bm(1 + 2mx/L)) across the test span L. Solving the governing equation yields a correction factor W (m) relating true midpoint rigidity (Bm) to apparent rigidity via Bm = B0W (m). The gradient parameter m is estimated numerically via finite differences on sequential B0 measurements. The model’s performance was evaluated through numerical simulations using known true stiffness profiles. For ideal linear profiles, the correction accurately recovered the true stiffness (Bm ≈ Btrue), confirming the model’s theoretical validity. For non-linear logistic profiles, chosen as representative examples of catheter transitions, the correction significantly reduced viii percentage error compared to uncorrected B0, especially in high-gradient regions. Sensitivity analyses showed accuracy improves with smaller test spacing (better m estimation) but decreases for steeper profiles or profiles deviating significantly from the underlying linear assumption. Application to experimental catheter data demonstrated feasibility, yielding quantifiable corrections and a more representative local stiffness profile. In conclusion, the linear gradient model offers a computationally inexpensive method to potentially improve flexural rigidity characterization for devices whose stiffness profiles are reasonably approximated by the model’s assumptions (e.g., smooth, logistic-like transitions). While it can provide more reliable mechanical data under these conditions, aiding clinical decisions, device design, and modeling, its necessity and effectiveness may be limited for devices with more complex or abrupt stiffness changes.
Language
English (en)
Chair
Guy M. Genin
Committee Members
Matthew R. Bersi, Mohamed A. Zayed