arXiv subjects

J. Lowengrub

Publications and source records attributed to J. Lowengrub.

4 recordsLinked to original sources

arXiv2025

PREDICT-GBM: A multi-center platform to advance personalized glioblastoma radiotherapy planning

Glioblastoma recurrence is largely driven by diffuse infiltration beyond radiologically visible tumor margins, yet standard radiotherapy, the mainstay of glioblastoma treatment, relies on uniform expansions that ignore patient-specific biological and anatomical factors. While computational models promise to map this invisible growth and guide personalized treatment planning, their clinical translation is hindered by the lack of standardized, large-scale benchmarking and reproducible validation workflows. To bridge this gap, we present PREDICT-GBM, a comprehensive open-source platform that integrates a curated, longitudinal, multi-center dataset of 243 patients with a standardized evaluation pipeline, and fuels model development and validation. We demonstrate PREDICT-GBM's potential by training and benchmarking a novel U-Net-based recurrence prediction model against state-of-the-art biophysical and data-driven methods. Our results show that both biophysical and deep-learning approaches significantly outperform standard-of-care protocols in predicting future recurrence sites while maintaining iso-volumetric treatment constraints. Notably, our U-Net model achieved a superior coverage of enhancing recurrence (79.37 +/- 2.08 %), markedly surpassing the standard-of-care (paired Wilcoxon signed-rank test, p = 0.0000057). Furthermore, the biophysical model GliODIL reached 78.91 +/- 2.08 % (p = 0.00045), validating the platform's ability to compare diverse modeling paradigms. By providing the first rigorous, reproducible ecosystem for model training and validation, PREDICT-GBM eliminates a major bottleneck for personalized, computationally guided radiotherapy. This work establishes a new standard for developing computationally guided, personalized radiotherapy, with the platform, models, and data openly available at github.com/BrainLesion/PredictGBM

eess.IV↗

arXiv2021

Controlling fingering instabilities in Hele-Shaw flows in the presence of wetting film effects

In this paper, the interfacial motion between two immiscible viscous fluids in the confined geometry of a Hele-Shaw cell is studied. We consider the influence of a thin wetting film trailing behind the displaced fluid, which dynamically affects the pressure drop at the fluid-fluid interface by introducing a nonlinear dependence on the interfacial velocity. In this framework, two cases of interest are analyzed: The injection-driven flow (expanding evolution), and the lifting plate flow (shrinking evolution). In particular, we investigate the possibility of controlling the development of fingering instabilities in these two different Hele-Shaw setups when wetting effects are taken into account. By employing linear stability theory, we find the proper time-dependent injection rate $Q(t)$ and the time-dependent lifting speed ${\dot b}(t)$ required to control the number of emerging fingers during the expanding and shrinking evolution, respectively. Our results indicate that the consideration of wetting leads to an increase in the magnitude of $Q(t)$ [and ${\dot b}(t)$] in comparison to the non-wetting strategy. Moreover, a spectrally accurate boundary integral approach is utilized to examine the validity and effectiveness of the controlling protocols at the fully nonlinear regime of the dynamics and confirms that the proposed injection and lifting schemes are feasible strategies to prescribe the morphologies of the resulting patterns in the presence of the wetting film.

physics.flu-dyn↗

arXiv2020

Pattern formation of the three-layer Saffman-Taylor problem in a radial Hele-Shaw cell

The three-layer Saffman-Taylor problem introduces two coupled moving interfaces separating the three fluids. A very recent weakly nonlinear analysis of this problem in a radial Hele-Shaw cell setup has shown that the morphologies of the emerging fingering patterns strongly depend on the initial thickness of the intermediate layer connecting the two interfaces. Here we go beyond the weakly nonlinear regime and explore full nonlinear interfacial dynamics using a spectrally accurate boundary integral method. We quantify the nonlinear instability of both interfaces as the relevant physical parameters (e.g., viscosities and surface tensions) are varied and show that our nonlinear computations are in good agreement with the experimental observations and the weakly nonlinear analysis. Nonlinear simulations reveal that due to the existence of a second interface, the classical highly branched morphologies are replaced by less unstable structures in which finger tip-splitting and finger competition phenomena are evidently restrained as the initial annulus' thickness is reduced. In addition, these novels patterns develop fingers with a series of low-amplitude bumps at their tips, associated with the enhanced growth of high-frequency modes promoted by the increasing coupling strength of interfaces.

physics.flu-dyn↗

arXiv2015

Consistent Hydrodynamics for Phase Field Crystals

We use the amplitude expansion in the phase field crystal framework to formulate an approach where the fields describing the microscopic structure of the material are coupled to a hydrodynamic velocity field. The model is shown to reduce to the well known macroscopic theories in appropriate limits, including compressible Navier-Stokes and wave equations. Moreover, we show that the dynamics proposed allows for long wavelength phonon modes and demonstrate the theory numerically showing that the elastic excitations in the system are relaxed through phonon emission.

cond-mat.stat-mech↗