Computational Modeling and Simulation Lab

At the Informatics Institute in Istanbul Technical University, we specialize in advancing the fields of biomechanics and cardiovascular medicine through cutting-edge computational research. Our team, which includes experts in computational fluid dynamics, cardiovascular mechanics, biomedical engineering and advanced simulation technologies, works to understand how vascular system operates. We collaborate closely with doctors and other medical professionals to ensure our research translates into real-world solutions that improve human health and medical technologies. Join us in exploring how our innovative work is shaping the future of biomedical engineering and cardiovascular biomechanics.

Interested in becoming part of our interdisciplinary research team? Join us to explore academic and research opportunities in computational biomechanics, AI-CFD, developing VR applications, modelling and simulation.

We are also currently seeking interns and volunteer undergraduate students who are eager to contribute to ongoing projects and gain hands-on research experience in a dynamic academic environment.

HiPC 2025: 32nd IEEE International Conference on High Performance Computing, Data, and Analytics - Afrah Farea

by Computational Modeling and Simulation Lab | Jan 09, 2026

At the 32nd IEEE International Conference on High Performance Computing, Data, and Analytics, held in India, Afrah Farea delivered a presentation on improving the accuracy of Physics-Informed Neural Networks (PINNs) for fluid flow applications.

IMG-20251218-WA0001 IMG-20251218-WA0035

Title: Multi-Objective Loss Balancing in Physics-Informed Neural Networks for Fluid Flow Applications

At the 32nd IEEE International Conference on High Performance Computing, Data, and Analytics, held in India, Afrah Farea delivered a presentation on improving the accuracy of Physics-Informed Neural Networks (PINNs) for fluid flow applications.

Paper Abstract
Physics-Informed Neural Networks (PINNs) have emerged as a promising machine learning approach for solving partial differential equations (PDEs). However, PINNs face significant challenges in balancing multi-objective losses, as multiple competing loss terms such as physics residuals, boundary conditions, and initial conditions must be appropriately weighted. While various loss balancing schemes have been proposed, they have been implemented within neural network architectures with fixed activation functions, and their effectiveness has been assessed using simpler PDEs. We hypothesize that the effectiveness of loss balancing schemes depends not only on the balancing strategy itself, but also on the loss function design and the neural network's inherent function approximation capabilities, which are influenced by the choice of activation function. In this paper, we extend existing solutions by incorporating trainable activation functions within the neural network architecture and evaluate the proposed approach on complex fluid flow applications modeled by the Navier-Stokes equations. Our evaluation across diverse Navier-Stokes problems demonstrates that this proposed solution achieves root mean square error (RMSE) improvements ranging from 7.4% to 95.2% across different scenarios. These findings highlight the importance of carefully designing the loss function and selecting activation functions for effective loss balancing.

The publication can be accessed via the link: https://doi.org/10.48550/arXiv.2509.14437

Show oldest on top

Events

01 Dec

Introduction to OpenFOAM®, Hands-on Workshop for Computational Fluid Dynamics, Informatics Institute Building 4th Floor, Computer Room, Istanbul Technical University

Under the supervision of Istanbul Technical University (ITU) Computational Science and Engineering Department faculty member Prof. Dr. Mustafa Serdar Çelebi, a free face-to-face, applied workshop "Introduction to OpenFOAM®" was held for students and academy members on Friday, December 1, 2023.

03 Apr

Seminar: Cardiovascular Blood Flow Simulation: Aorta and Coronary Artery

A hybrid seminar titled "Cardiovascular Blood Flow Simulation - Aorta & Coronary Artery" was held on November 3, 2023, under the supervision of Istanbul Technical University (ITU) Computational Science and Engineering Department faculty member Prof. Dr. Mustafa Serdar Çelebi.

16 May

Seminar: Hemodynamic Simulations and FSI Coupling for Bioprosthetic Valves

An online seminar titled "Hemodynamic Simulations and FSI Coupling for Bioprosthetic Valves" will be held on Monday, May 16, 2022, under the supervision of Prof. Dr. M. Serdar Çelebi, faculty member of the Istanbul Technical University (ITU) Computational Science and Engineering Department.

26 Apr

Workshop: Fluid-Structure Interaction Modeling with Immersed Boundary Method

An online workshop titled "Immersed Boundary Method (IBM) for Fluid-Structure Interaction (FSI) Modeling" was held on April 26, 2022. This one-day hands-on training session was designed to introduce the fundamentals of IBM and its practical application using the IBAMR software.

Show All

ITU Informatics Institute - Phone +90(212) 285 71 94

ITU Informatics Institute - Fax +90 (212) 285 70 73

ITU Informatics Institute, Room: 204 - Address Istanbul Technical University, Ayazağa Campus, Maslak Sarıyer, Istanbul

Slider 1 Team

Slider 2 Coronary

Slider 3 Aort

Slider 4 rbc

Slider 5 aort fem