0
Research Papers

Computationally Efficient Particle Release Map Determination for Direct Tumor-Targeting in a Representative Hepatic Artery System

[+] Author and Article Information
E. M. Childress

Department of Mechanical
and Aerospace Engineering,
North Carolina State University,
Raleigh, NC 27695

C. Kleinstreuer

Department of Mechanical
and Aerospace Engineering,
North Carolina State University,
Raleigh, NC 27695
Joint Department of Biomedical Engineering,
North Carolina State University,
Raleigh, NC 27695
University of North Carolina at Chapel Hill,
Chapel Hill, NC 27514
e-mail: ck@ncsu.edu

1Corresponding author.

Contributed by the Bioengineering Division of ASME for publication in the JOURNAL OF BIOMECHANICAL ENGINEERING. Manuscript received June 21, 2013; final manuscript received October 21, 2013; accepted manuscript posted October 31, 2013; published online December 4, 2013. Assoc. Editor: Ender A. Finol.

J Biomech Eng 136(1), 011012 (Dec 04, 2013) (8 pages) Paper No: BIO-13-1275; doi: 10.1115/1.4025881 History: Received June 21, 2013; Revised October 21, 2013; Accepted October 31, 2013

Implementation of a novel direct tumor-targeting technique requires a computer modeling stage to generate particle release maps (PRMs) which allow for optimal catheter positioning and selection of best injection intervals for drug-particles. This simulation task for a patient-specific PRM may require excessive computational resources and a relatively long turn-around time for a fully transient analysis. Hence, steady-state conditions were sought which generates PRMs equivalent to the pulsatile arterial flow environment. Fluid-particle transport in a representative hepatic artery system was simulated under fully transient and steady-state flow conditions and their corresponding PRMs were analyzed and compared. Comparisons of the transient PRMs from ten equal intervals of the cardiac pulse revealed that the diastolic phase produced relatively constant PRMs due to its semisteady flow conditions. Furthermore, steady-state PRMs, which best matched the transient particle release maps, were found for each interval and over the entire cardiac pulse. From these comparisons, the flow rate and outlet pressure differences proved to be important parameters for estimating the PRMs. The computational times of the fully transient and steady simulations differed greatly, i.e., about 10 days versus 0.5 to 1 h, respectively. The time-averaged scenario may provide the best steady conditions for estimating the transient particle release maps. However, given the considerable changes in the PRMs due to the accelerating and decelerating phases of the cardiac cycle, it may be better to model several steady scenarios, which encompass the wide range of flows and pressures experienced by the arterial system in order to observe how the PRMs may change throughout the pulse. While adding more computation time, this method is still significantly faster than running the full transient case. Finally, while the best steady PRMs provide a qualitative guide for best catheter placement, the final injection position could be adjusted in vivo using biodegradable mock-spheres to ensure that patient-specific optimal tumor-targeting is achieved. In general, the methodology described could generate computationally very efficient and sufficiently accurate solutions for the transient fluid-particle dynamics problem. However, future work should test this methodology in patient-specific geometries subject to various flow waveforms.

FIGURES IN THIS ARTICLE
<>
Copyright © 2014 by ASME
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Fig. 2

Geometry and boundary conditions for steady and transient scenarios: (a) representative hepatic artery system, (b) CHA inlet flow rate, (c) daughter vessel outlet pressure, and (d) GDA outlet pressure. The steady conditions for case 26 were determined by taking the time-averaged values for the inflow, daughter outlet pressure, and pressure difference between the GDA and daughter outlets, while for case 27 the midpoints between the maximum and minimum values for these conditions were used. The colors at each outlet in (a) correspond to their injection region in the particle release map.

Grahic Jump Location
Fig. 1

Illustration of the direct tumor-targeting methodology. The colors at each outlet correspond to their injection region in the particle release map.

Grahic Jump Location
Fig. 3

Particle release map (PRM) comparison. This is accomplished by dividing the PRM into square subsections and computing, within each subsection, the fraction of particles exiting each branch (i.e., GDA and D1–D4), and comparing these values between two PRMs (e.g., the PRM of transient interval 1 (left) and steady case 1 (right)).

Grahic Jump Location
Fig. 7

Comparison of steady cases which best approximate each transient interval. The matching subsections exclude those in which multiple branches were targeted. See Fig. 2 for the PRM color code and the Fig. 4 caption for the matching subsection color code.

Grahic Jump Location
Fig. 10

Matching subsections for targeting individual branches (excluding subsections where multiple branches were targeted) for intervals 4 through 8 (left) and when comparing cases 17 (middle) and 26 (right) to intervals 4 through 8. The approximate injection region area retention is also provided for the comparisons with Cases 17 and 26.

Grahic Jump Location
Fig. 4

Single daughter branch targeting. Transient interval 1 particle release map (left). Transient interval 1 subsections which predominantly target (αb,subi > 0.9) one branch ((right) D1 = red, D2 = pale blue, D3 = aqua, D4 = orange, GDA = yellow, none/multiple branches = dark blue). Arrows indicate regions where multiple branches were targeted.

Grahic Jump Location
Fig. 8

RMSDavg for each steady case. The filled bar signifies the best match.

Grahic Jump Location
Fig. 9

Matching subsections for targeting individual branches (excluding subsections where multiple branches were targeted) between the steady and transient intervals (top) and RMSDavg,intvli (bottom) for steady cases 17, 26, and 27. See Fig. 4 caption for the matching subsection color code.

Grahic Jump Location
Fig. 5

Subsections in which one daughter branch is predominantly targeted (αb,subi > 0.9) for each transient interval (top two rows) and for combined intervals (bottom row). See Fig. 4 caption for the color code.

Grahic Jump Location
Fig. 6

RMSDavg,intvli for each steady case compared to each transient interval. The filled bar signifies the best match.

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In