Prof. Dr. Matej Praprotnik (KI, FMF): Particle-based models of encapsulated microbubbles and gas vesicles for biomedical ultrasound simulations

Ultrasound-guided drug and gene delivery (USDG) enables controlled and spatially precise delivery of drugs, encapsulated in microbubbles (eMBs) and submicron gas vesicles (GVs), to target areas such as cancer tumors. It is a non-invasive, low toxicity process with drastically reduced drug dosage. Physical properties of GVs and eMBs critically affect the outcome of USDG. Detailed understanding and modeling of their physical properties is thus essential for ultrasound therapeutic applications. State-of-the-art continuum models of shelled bodies cannot incorporate details such as varying thickness of the shell. In this talk, I will introduce a general particle-based modeling framework for encapsulated bodies that accurately captures elastic and rheological properties of GVs and eMBs [1]. We use dissipative particle dynamics to model the solvent, the gaseous phase, and the triangulated surfaces of immersed objects. Their elastic behavior is studied and validated through stretching and buckling simulations, eigenmode analysis, shear flow simulations, and comparison of predicted buckling pressure with published experimental data. Furthermore, I will present a particle-based virtual ultrasound machine capable of simulating immersed structures under the influence of ultrasound in the medically relevant frequency regime [2,3]. Existing computational approaches typically decouple acoustic wave propagation from fluid dynamics, solving the wave equation and hydrodynamic equations separately, which limits accuracy and generality of the system under study. We address this limitation with a novel particle-based fluid simulation method that captures both wave propagation and fluid flow, thereby improving physical fidelity while reducing modeling complexity [3]. This is achieved through an implicit compressible pressure solver, together with pressure stabilization schemes that robustly handle negative pressure regimes encountered in ultrasound. The presented computational framework (the new mesoscopic models of eMBs and GVs and virtual ultrasound machine) allows for controlled testing, data-driven quantification of uncertainties in design parameters and a rational optimization of experimental ultrasound parameters.
[1] N. Ntarakas, M. Lah, D. Svenšek, T. Potisk, M. Praprotnik, ACS Appl. Nano Mater. 8, 16053 (2025)
[2] M. Lah, N. Ntarakas, T. Potisk, P. Papež, M. Praprotnik, J. Chem. Phys. 162, 024103 (2025)
[3] U. Čoko, T. Potisk, M. Praprotnik, arXiv:2602.15442 [cond-mat.soft]