Conference Agenda

This paper describes the generation mechanism of biomagnetic signals detected by OPM-MEG, and proposes a Bayesian network construction method based on biomagnetic signals to realize the network analysis of biomagnetic signals. Due to its weak characteristics, it is difficult to make full use of prior information for analysis. An MVAR model is constructed based on functional biomagnetic signals, and the parameters of the MVAR model are obtained using an improved Bayesian algorithm. The effectiveness of the biomagnetic signal network connectivity is verified through simulation and actual experiments.

This study investigates electroporation to enhance nanoparticle delivery into cancer cells, combined with magnetic hyperthermia (MH) to induce apoptosis. Optimizing electroporation (EP) pulses increases membrane pore size, reducing cancer cell activity while minimizing normal cell damage. Alternating magnetic fields (AMF) generate a magnetothermal effect from internalized nanoparticles (NPs), triggering apoptosis. Finite element simulations optimize electroporation and MH for enhanced therapeutic efficacy.

EP simulations assessed pulse effects on membrane pore size and potential difference, while MH simulations analyzed NP heating under varying concentrations and fields. Results show that optimized EP enhances NP delivery and MH performance.

This study presents the design of a cylindrical radial non-uniform coil for compensating specific 50 Hz power frequency noise, based on the target field method. The single-ended open Magnetic Shielding Cylinder (MSC) structure suffers from uneven magnetic field distribution and residual magnetism affected by the frequency f after shielding. To address these issues, the distribution of the magnetic field in a 50 Hz power frequency environment is analyzed in the simulation to set the target magnetic field for the coil design. The final design of the 50Hz magnetic compensation coil, compared to the static magnetic compensation coil, reduces the magnetic field error from 59.06% to 5.09%. This research is of great help to eliminate the noise of 50Hz power frequency signal and improve the signal-to-noise ratio in magnetocardiogram (MCG).

Breast cancer significantly impacts public health, with standard mammography facing limitations such as harmful radiationand challenges in dense breasts. This study assesses the safety of microwave imaging, a promising alternative using non-ionizing radiation, through numerical computing of temperature increases in tissues during an electromagnetic field exposure. Electromagnetic simulations were conducted using the finite difference time domain method. Hardware configuration is derived from a microwave imaging device used in clinical studies. Five digital anatomical breast models were considered, varying in presence and type of tumor, and breast density. Specific absorption rate distributions have been used to obtained temperature increases, with a peak lower than 60 µK mostly found in fibroglandular tissue and skin depending on the frequency. No harmful effects are expected during the microwave examination.

A nonlinear parametric Model Order Reduction approach is proposed and applied for the computation of the induced electric field in a subject’s head due to Transcranial Magnetic Stimulation. The resulting reduced order model dramatically decreases computational times when simulating several positions and orientations of the excitation coil in a chosen region of interest. In particular the proposed model allows to obtain field solutions in a faster way with respect to classical solvers as Finite Element Methods without losing accuracy.

We present a finite element modeling of magnetoelectric properties of a myelinated axon. Results are compared against an analytical solution for the case of an axon exposed to a homogeneous electric field. We achieve a mean relative error below 0.3% and a normalized root mean square deviation below 0.5%.

In this paper, a numerical model to predict the electric field distribution due to nanoparticles close to the cell membrane during electrochemotherapy treatment is presented. The effect of a rectangular voltage pulse on the electric field distribution in a region of interest close to the nanoparticle is studied, considering nanoparticles made of several different materials. To solve the electromagnetic problem in transient conditions, Finite Element Analyses are performed. Specifically, both the rise and the constant period of the pulse are investigated in terms of field distribution. Moreover, different rise times are considered.

This paper presents an effective methodology for constructing nonspherical models of the female torso being pregnant with twins and calculating the electric field when it is subjected to radiation from an external point source. The proposed model is described by spherical shapes for the representation of the various tissues, whereas the electric field is derived via a hybrid computational scheme, which involves dyadic Green’s functions as well as the implementation of the appropriate surface boundary conditions and the superposition principle. Numerical results provide instructive deductions for a realistic model-antenna interaction and particularly for the precise evaluation of the electric-field intensity at every point in the domain with regard to diverse polarizations.

The interaction of light with the human eye is a complex phenomenon that involves both visual and non-visual pathways, serving essential roles in various physiological and psychological processes. In this study, we employ the Finite-Difference Time-Domain (FDTD) method, an advanced numerical technique for solving Maxwell's equations across both temporal and spatial domains, facilitating a precise simulation of the interaction between electromagnetic waves and the heterogeneous and anisotropic media of the eye. The FDTD method relies on the discretization of the Maxwell equations in both spatial and temporal domains. The results are anticipated to offer significant insights into the variations in energy distribution patterns and their potential implications for the safety and efficacy of optical applications.

Electrical impedance tomography (EIT) is an imaging technique that reconstructs conductivity distributions to infer internal tissues within the human body. Despite its potential, EIT's ill-posed inverse problem is challenging for conventional inverse solvers, often leading to distortions and artifacts in reconstructions. In this study, a deep neural network (DNN)--based EIT inverse solver is proposed. Using EIT simulation data with targets with randomized positions and radius to train DNN, a DNN-based inverse solver enhances the accuracy of reconstructions and reduces artifacts compared to traditional inverse solvers. Clinical EIT data from stroke will be tested next.

This work aims to model a wireless power transfer (WPT) system to analyze the effectiveness of nonlinear (NL) metamaterials (MMs) with Duffing-resonator (DR) based unit cells. Its purpose is to boost the inductive power transfer (IPT) in near-field (NF) simultaneous wireless information and power transmission (SWIPT) systems applied to biomedical implanted devices. Firstly, a simplified model is presented, and then numerical simulations are carried out to validate the adopted approaches. The performance of the IPT systems with both MM configurations, DR and linear, are finally analyzed with the aid of 3D numerical simulations in a biomedical application.

The Poisson-Boltzmann equation (PBE) is a model commonly used to estimate the electrostatic energy of biomolecular systems immersed in electrolytic solutions.

It is often applied to very large molecules, such as proteins or viruses, comprised of thousands or even millions of atoms, with a complex surface geometry.

Due the extreme simulation sizes, the implementation of efficient yet accurate solvers for this model is a dunting task. Most existing state of the art software packages for solving the PBE are based on structured cartesian grids that reduce computational costs and memory footprint by removing the need for an explicit data structure for mesh representation.

Achieving high accuracy in structured grid-based methods, though, may lead to a huge number of degrees of freedom computational costs.

We present NextGenPB a new parallel solver for the PBE, that exploits properties of

cartesian grids with hierarchycal local refinement in order to maximize

accuracy-to-cost efficiency.

In addition, it applies analytically corrected quadrature rules near interface discontinuities and optimized boundary conditions.

NextGenPB is designed to achieve precise electrostatic potential and energy

estimates at reduced cost. It is validated on dielectric spheres, biomolecular benchmarks, and large-scale systems with millions of atoms,

outperforming state-of-the-art solvers in accuracy and scalability. This framework is extendable to other PDEs with discontinuous

coefficients.

This paper presents a design method for a biplanar coil to achieve a spatially nonlinear magnetic field (SNMF) distribution. The higher-order components of the magnetic field produced by the biplanar coil can increase the discrepancy between simulated and actual values, thereby reducing magnetic field uniformity in the target area. To address this problem, the Taylor expansion method (TEM) is integrated into the traditional target field method (TFM), and an improved particle swarm optimization algorithm is employed to determine the coil parameters and shape. The simulation results show that the magnetic field deviation generated by the asymmetric coils in the -200 mm to 200 mm range is -2.32%~1.08% compared with the preset field, which demonstrates that the proposed method effectively generates the desired magnetic field distribution.