# Open-Source Software Tools for Rapid Radiofrequency Coil Modeling and Simulation in MRI > **NIH NIH K99** · NEW YORK UNIVERSITY SCHOOL OF MEDICINE · 2024 · $117,507 ## Abstract Project Summary The aim of this project is to develop a novel, rapid, and memory efficient open-source software for radiofrequency (RF) coil modeling in magnetic resonance imaging (MRI). Ultra-high field (UHF) scanners (> 3T) can reveal remarkable anatomical details to diagnose diseases much earlier than high field scanners. Nevertheless, the operating wavelength at UHF is shorter than the dimensions of the body, therefore constructive and destructive interference patterns of the coil-generated RF fields can occur in the torso and head. This can lead to local RF energy deposition hotspots, which is a patient’s safety concern, and signal dropouts that can deteriorate the diagnostic quality of the MRI. To address this, electromagnetic (EM) simulations are used for the careful design and evaluation of RF coils arrays before prototyping. Commercial software can perform such simulations but are restricted by expensive licenses and can be prohibitively slow for the optimization of complex setups, for example, two months of simulation time for a 64-channel receive coil at 7T. A novel, highly accurate and memory-efficient EM simulation tool for MRI modeling is proposed in this project. Aim 1 will focus on the development of a domain decomposition method for modeling the EM interactions between RF coils and biological tissue, using wire, surface, and volume integral equation methods. Tensor decompositions will be used to assemble low-memory formats of these interactions for faster simulations, which can be run on a GPU for hundred times speedup compared to commercial packages. Aim 2 will focus on devel- oping an RF circuit simulator for coil tuning, matching, and decoupling using particle swarm optimization. These methods will form a co-simulator for comprehensive RF coil modeling in MRI. In Aim 3 we will focus on the validation of the novel simulation software, comparing its accuracy, memory, and speed to commercial pack- ages and experimentally measured fields at 7T. The software will be used to design, optimize, and construct a novel 7T coil for prostate imaging. This project will utilize, for the first time, tensor decompositions to create a powerful and innovative EM mod- eling software tool for MRI. The software will be compatible with any MRI frequency and anatomy without the need for expensive licenses. This “Pathway to Independence” award proposal also includes a mentored career development plan to help the candidate, Dr. Ilias Giannakopoulos, becoming an independent investigator. His primary mentor, Dr. Riccardo Lattanzi, and co-mentors Dr. Daniel Sodickson and Dr. Ryan Brown, are leading experts in the field of EM interactions with biological tissue and coil design. The diversified mentoring plan and the complementary background of these mentors will provide valuable exposure to coil design and prototyping, UHF MRI, and interdisciplinary collaborations to help the candidate transition to an independent investigator posi- tion. Ultimately, Dr. Giannak... ## Key facts - **NIH application ID:** 10983544 - **Project number:** 1K99EB035163-01A1 - **Recipient organization:** NEW YORK UNIVERSITY SCHOOL OF MEDICINE - **Principal Investigator:** Ilias Giannakopoulos - **Activity code:** K99 (R01, R21, SBIR, etc.) - **Funding institute:** NIH - **Fiscal year:** 2024 - **Award amount:** $117,507 - **Award type:** 1 - **Project period:** 2024-08-08 → 2026-05-31 ## Primary source NIH RePORTER: https://reporter.nih.gov/project-details/10983544 ## Citation > US National Institutes of Health, RePORTER application 10983544, Open-Source Software Tools for Rapid Radiofrequency Coil Modeling and Simulation in MRI (1K99EB035163-01A1). Retrieved via AI Analytics 2026-06-12 from https://api.ai-analytics.org/grant/nih/10983544. Licensed CC0. --- *[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*