Skip to main content


Using High Performance Computing for Biomechanical Finite Element Models

Thursday, September 23, 2021

12:00 pm - 1:00 pm

About the Event

Brain-shift during neurosurgery compromises the validity of preoperative images to identify tumor and vital structures’ location in neuronavigation platforms. Although several studies have used various finite element models (FEMs) to predict successfully inward brain-shift, they are significantly time consuming and thus their CPU performance is not clinically compatible (a result in less than one minute is required). There are two bottlenecks for performance improvement. The first bottleneck is the building of a finite element patient-specific model including segmentation and mesh editing. The second is the CPU time required by the simulation on a high-performance computer. We have developed a model to describe inward brain-shift with viscoelastic biomechanical modeling varying three surgery parameters: craniotomy position, cerebrospinal fluid drainage level and intraoperative head position. We propose to develop a real-time estimation of the inward brain-shift displacement with deep learning from only patient-specific MRI testing without using time-consuming Finite Element building. In this study, we demonstrate that the combination of HPC generated FEM training data and a U-net approach is a promising solution for clinically compatible performance.

About the Speaker

Anne-Cecile Lesage is a research scientist at MDACC. She has 17 years of experience in the field of applied mathematics (numerical analysis, finite elements, finite difference, inverse methods) and applied physics (hydraulics, hydrodynamics, solid mechanics, geophysics). She joined MDACC in 2018, more specifically the Morfeus Lab supervised by Professor Kristy Brock in the Imaging Physics department. Her current research at MDACC focuses on modeling soft tissue mechanics for medical applications. She is developing patient-specific finite element modeling workflows to predict tissue deformations for applications in brain and lung cancer surgery. She is part of the image-guided cancer therapy program.


Martin Huarte Espinosa
Associate Director
HPE Data Science Institute