Contact

Fatemeh Chegini, PhD

Scientific Computing and HPC for Cardiac Electrophysiology

Postdoctoral researcher at Zuse Institute Berlin · MICROCARD · openCARP contributor

I develop scalable numerical methods and scientific software for cardiac electrophysiology, with a focus on cellular-scale EMI simulations, higher-order finite-element methods, domain-decomposition preconditioning, adaptive methods, and communication-efficient high-performance computing.

  • Scientific Computing
  • High-Performance Computing
  • Cardiac Electrophysiology
  • EMI Model
  • openCARP
  • Finite Elements
  • Domain Decomposition
  • BDDC
  • Spectral Deferred Correction
  • Algebraic Adaptivity
  • PETSc
  • Ginkgo
  • C++
  • MPI
  • Inverse Problems
  • PDE-Constrained Optimization
  • Cardiac Digital Twins
Research Publications CV

Fatemeh Chegini, PhD

EMI simulation in openCARP

Cell-resolved EMI simulation generated by Joachim Greiner using openCARP.

This simulation was generated by Joachim Greiner using the EMI model in openCARP. It shows cardiac electrophysiology at cellular resolution, with individual cells, membranes, gap junctions, and extracellular space represented explicitly.

This input EMI mesh was generated by Mark Potse for cellular-scale simulations in openCARP. The mesh-generation strategy is described in the associated paper. It represents individual cardiac myocytes, extracellular space, membranes, and gap junctions explicitly. My work turns this detailed geometry into scalable simulation software through distributed mesh processing, face-based membrane evaluation, robust solver backends, and algorithmic acceleration.

Research Focus

I am a postdoctoral researcher at the Zuse Institute Berlin (ZIB) in the Modeling and Simulation of Complex Processes department. Within the MICROCARD project, I work on numerical algorithms and scalable software for high-fidelity cardiac electrophysiology. My research connects finite-element discretization, parallel algorithms, numerical linear algebra, adaptive methods, and scientific software engineering.

My current work centers on the Extracellular–Membrane–Intracellular (EMI) model, where individual cardiac cells and their membranes are represented explicitly. This cellular resolution creates challenging problems in mesh processing, distributed-memory data structures, linear solvers, time integration, and computational efficiency.

Research topics

  • Distributed-memory EMI in openCARP

    A scalable computational framework for cellular-scale cardiac electrophysiology: distributed mesh construction, membrane-interface handling, transfer operators, parallel assembly, solvers, and validation.

  • Higher-order FEM & algebraic adaptivity

    Hierarchical second-order finite elements combined with selective activation of higher-order degrees of freedom to improve accuracy while controlling computational cost.

  • Spectral Deferred Correction

    Higher-order time integration for cardiac electrophysiology and its combination with adaptivity and scalable preconditioning.

  • BDDC & scalable preconditioning

    Domain-decomposition methods for the large linear systems arising from EMI simulations, with a focus on scalability and communication efficiency.

  • Communication compression

    Compression strategies for reducing data exchange in large-scale parallel solvers and improving communication efficiency on HPC systems.

  • Inverse cardiac electrophysiology

    PhD research on PDE-constrained inverse problems, multilevel optimization, conductivity estimation, and scar identification for patient-specific cardiac models.

From inverse problems to cellular-scale simulation

My research path began with inverse problems in cardiac electrophysiology during my PhD at the Università della Svizzera italiana (USI). I developed multilevel and multifidelity optimization methods for estimating electrophysiological parameters and identifying cardiac scar regions. At ZIB, my focus moved toward high-performance simulation at cellular resolution: first through EMI prototypes in Kaskade and then through the design and development of a distributed-memory EMI framework in openCARP.

Today I am especially interested in the next generation of cardiac digital twins, where scalable PDE simulation, inverse problems, high-performance computing, and physics-based machine learning come together.

Research and projects

Education

PhD in Computational Science, Università della Svizzera italiana, 2022
Thesis: Multilevel Optimization Algorithms for Inverse Problems in Electrocardiography
Collaborated with the Center for Computational Medicine in Cardiology (CCMC) during my PhD.

MSc in Intelligent Systems, Università della Svizzera italiana, 2015
Thesis: A Spectral Approach to Cross-Modal Information Retrieval

BSc in Computer Science, Università della Svizzera italiana, 2013
Thesis: A GLR Parser Generator for Arbitrary Context-Free Grammars

BSc in Electronic Engineering, Azad University, Karaj Branch, 2002
Thesis: Soft Starter for Induction Motors

Current affiliations

Zuse Institute Berlin (ZIB)
Postdoctoral Researcher, Modeling and Simulation of Complex Processes

MICROCARD
Numerical methods and software for cellular-resolution cardiac electrophysiology on advanced HPC systems.

openCARP
Open-source cardiac electrophysiology simulation software. I contribute to EMI-model functionality and distributed-memory workflows.

Contact ZIB profile openCARP Git