Parallel Implementation of Vascular Network Modeling

Abstract

Abstract. The paper presents modeling of the vascular system in a parallel environment. The aim of this approach is to accelerate the simulation of vascular network growth and make it closer to analogous real life processes. We concentrated on the perfusion process and made an attempt to parallelize the process of connecting ischemic macroscopic functional units to existing vascular systems. The proposed method was implemented on a computing cluster with the use of the MPI standard. The results show that it is possible to gain a significant speedup that allows us to make simulations for a greater number of macroscopic functional units and vessels in a reasonable time, which increases the possibility to create more complex and more precise virtual organs.

Vascular Network Modeling - Improved Parallel Implementation on Computing Cluster

“…Therefore, in order to accelerate this most time consuming part of vascular development, an algorithm to spread the computations of the perfusion process over processors was introduced (see Fig. 1) [3]. Moreover, our intention was to bring the solution closer to reality in which analogous perfusion phenomena can also occur in a parallel way.…”

“…The previously presented parallel algorithm of vascular network growth can be treated as the first step of bringing the discussed model closer to reality [3]. Moreover, in spite of the fact that in this solution the efficiency of computational nodes was strongly dependent on the operation of the managing node that splitted the tasks step by step, it allowed us to gain a significant speedup (about 8 for 16 processors).…”

“…Initially, we made use of a sequential algorithm to generate a virtual liver [7], CT simulator and MRI virtual scanner. Subsequently, we introduced a parallel solution of vascular network growth [3], which significantly reduced the computation time. In this paper, we wish to present an improved parallel algorithm of physiological vascular modeling whose aim is to further accelerate the simulation process.…”