Chinmoy Saayujya

EECS Department, University of California, Berkeley

Technical Report No. UCB/

May 1, 2025

Magnetic Particle Imaging (MPI) is an emerging tracer-based imaging modality that maps the spatial distribution of superparamagnetic iron oxide nanoparticles (SPIOs) by harnessing their nonlinear magnetic response. While MPI provides exceptional sensitivity, contrast, and depth of penetration, its spatial resolution remains limited in clinical applications due to constraints on gradient field strength imposed by safety and hardware costs. This dissertation examines the fundamental physical and magnetic factors governing MPI spatial resolution and presents theoretical and experimental methodologies for the design of next-generation, high-resolution nanoparticle tracers.

In Part I, we examine the Brownian and Néel relaxation mechanisms that underpin magnetization reversal in SPIOs, as well as their effect on the MPI signal and point spread function (PSF). Using pulsed magnetic field relaxometry, we characterize the transient magnetic behaviour of nanoparticles and experimentally validate theoretical predictions alongside long-standing closed-form approximations, demonstrating their accuracy and practical utility for modelling SPIO dynamics. We also establish pulsed relaxometry as a potential mechanism for contrast generation in applications including in vivo viscometry and binding assays.

Part II explores a novel super-resolution MPI tracer platform that leverages strong dipolar coupling in chain-like nanoparticle assemblies. These assemblies exhibit superferromagnetism, a phenomenon characterized by transient ferromagnetic behaviour in otherwise superparamagnetic systems. This collective magnetic behaviour enables unprecedented order-of-magnitude improvements in both spatial resolution and signal-to-noise ratio. We develop and validate a positive feedback model for these assemblies that predicts key magnetic characteristics such as coercivity, and delineate the physical and magnetic conditions required to observe this phenomenon. We also demonstrate controlled thermal decomposition-based synthesis of iron oxide nanoparticles with tunable core size and oxidation state, enabling experimental validation of model predictions and reproducible fabrication of superferromagnetic iron oxide (SFMIO) MPI tracers .

Finally, Part III presents the development of a comprehensive modular MPI simulation framework designed to bridge nanoparticle magnetic behaviour with system-level imaging performance. The simulator incorporates MPI pulse sequence characteristics, field-dependent relaxation dynamics, and both SPIO and SFMIO tracer properties, enabling accurate prediction of the MPI PSF under realistic imaging conditions. This tool provides a unified platform for the co-design of tracers and MPI pulse sequences, facilitating optimization of resolution, contrast, and safety metrics in x-space MPI.

Collectively, this dissertation advances the fundamental understanding of nanoparticle magnetization dynamics and introduces novel theoretical frameworks and experimental tools for optimizing MPI tracers. These insights engender the rational design of next-generation tracers, pulse sequences, and reconstruction methods, paving the way toward higher-resolution and clinically translatable MPI systems.

Advisors: Steven Conolly

---

BibTeX citation:

@phdthesis{Saayujya:31659,
    Author= {Saayujya, Chinmoy},
    Title= {Fundamental Resolution Limits of Magnetic Particle Imaging},
    School= {EECS Department, University of California, Berkeley},
    Year= {2025},
    Number= {UCB/},
    Abstract= {Magnetic Particle Imaging (MPI) is an emerging tracer-based imaging modality that maps the spatial distribution of superparamagnetic iron oxide nanoparticles (SPIOs) by harnessing their nonlinear magnetic response. While MPI provides exceptional sensitivity, contrast, and depth of penetration, its spatial resolution remains limited in clinical applications due to constraints on gradient field strength imposed by safety and hardware costs. This dissertation examines the fundamental physical and magnetic factors governing MPI spatial resolution and presents theoretical and experimental methodologies for the design of next-generation, high-resolution nanoparticle tracers.

In Part I, we examine the Brownian and Néel relaxation mechanisms that underpin magnetization reversal in SPIOs, as well as their effect on the MPI signal and point spread function (PSF). Using pulsed magnetic field relaxometry, we characterize the transient magnetic behaviour of nanoparticles and experimentally validate theoretical predictions alongside long-standing closed-form approximations, demonstrating their accuracy and practical utility for modelling SPIO dynamics. We also establish pulsed relaxometry as a potential mechanism for contrast generation in applications including in vivo viscometry and binding assays.

Part II explores a novel super-resolution MPI tracer platform that leverages strong dipolar coupling in chain-like nanoparticle assemblies. These assemblies exhibit superferromagnetism, a phenomenon characterized by transient ferromagnetic behaviour in otherwise superparamagnetic systems. This collective magnetic behaviour enables unprecedented order-of-magnitude improvements in both spatial resolution and signal-to-noise ratio. We develop and validate a positive feedback model for these assemblies that predicts key magnetic characteristics such as coercivity, and delineate the physical and magnetic conditions required to observe this phenomenon. We also demonstrate controlled thermal decomposition-based synthesis of iron oxide nanoparticles with tunable core size and oxidation state, enabling experimental validation of model predictions and reproducible fabrication of superferromagnetic iron oxide (SFMIO) MPI tracers .

Finally, Part III presents the development of a comprehensive modular MPI simulation framework designed to bridge nanoparticle magnetic behaviour with system-level imaging performance. The simulator incorporates MPI pulse sequence characteristics, field-dependent relaxation dynamics, and both SPIO and SFMIO tracer properties, enabling accurate prediction of the MPI PSF under realistic imaging conditions. This tool provides a unified platform for the co-design of tracers and MPI pulse sequences, facilitating optimization of resolution, contrast, and safety metrics in x-space MPI.

Collectively, this dissertation advances the fundamental understanding of nanoparticle magnetization dynamics and introduces novel theoretical frameworks and experimental tools for optimizing MPI tracers. These insights engender the rational design of next-generation tracers, pulse sequences, and reconstruction methods, paving the way toward higher-resolution and clinically translatable MPI systems.},
}

EndNote citation:

%0 Thesis
%A Saayujya, Chinmoy 
%T Fundamental Resolution Limits of Magnetic Particle Imaging
%I EECS Department, University of California, Berkeley
%D 2025
%8 May 1
%@ UCB/
%F Saayujya:31659