Computer Simulation of Prostate Cryoablation—Fast and Accurate Approximation of the Exact Solution

Abstract

Objective. Cryosurgery is a minimally invasive cancer treatment that uses liquid nitrogen or supercooled argon to freeze and destroy tumors. To achieve complete ablation of the prostate, we have developed a computer program that can determine treatment effects by calculating iceball formation. This is based on a three-dimensional (3D) model of the patient's prostate constructed from ultrasound images. The program predicts the best set of cryoablation parameters and cryoprobe spatial positions, then displays these parameters in graphical or numerical form. The objective of this work was to improve our prostate cryoablation modeling software by making its partial differential equation (PDE) solver more accurate and faster.

Materials and Methods: CryoSim, our software package, accepts a set of acquired and processed 3D ultrasound images of the prostate, then models heat diffusion using numerical approximations of the heat equation.

Results: We describe the latest version of the CryoSim software, which models cryoablation therapy. Solving the problems of low spatial resolution (now down to a fraction of a millimeter, as compared to 5 mm in the old version) and modeling cryosurgery in a short time (down to few minutes versus hours) provides a platform for proper planning of cryosurgery and a tool for the training of surgeons.

Conclusion. Changes in the PDE solver algorithm produce more accurate results, leading to an improved visualization of the iceball, with a precision of a few mm and a significant decrease in computation time.

Key Words:

• prostate cancer
• cryosurgery
• computer modeling
• cryosimulation