Microwave focusing with temporal interference for non-invasive deep brain stimulation

TL;DR

This paper introduces a novel non-invasive microwave focusing method combining iterative time-reversal and genetic algorithm optimization to precisely target deep brain regions, potentially enabling safer brain stimulation therapies.

Contribution

It presents a new combined approach using iTR and TI optimization with realistic head models for non-invasive deep brain microwave focusing, advancing prior methods.

Findings

Effective focusing within predefined brain regions demonstrated.

Optimization improves convergence and focusing accuracy.

Proof-of-principle shown in realistic head models.

Abstract

Deep Brain Stimulation (DBS) is effective in treating neurological disorders but involves invasive surgery. Non-invasive DBS aims to overcome surgical risks by means of externally applied electromagnetic fields. In this work, we present a method for non-invasive microwave focusing of amplitude-modulated electric fields in the human brain using an external antenna array. The method combines iterative time-reversal (iTR) and genetic-algorithm-based temporal interference (TI) optimization to determine antenna positions, orientations, operating frequency, amplitudes, and phases of each element. Magnetic point dipoles are used as idealized radiative sources, providing a design-independent benchmark for optimization studies. Furthermore, this work incorporates realistic tissue properties from voxelized anatomical head models, accounting for variations in permittivity and conductivity. The iTR algorithm provides an initial antenna configuration that serves as a reliable starting point for the subsequent TI optimization, improving convergence and focusing precision. The results demonstrate that amplitude-modulated electromagnetic fields can be selectively and precisely shaped within predefined focal regions. Thereby, this study demonstrates a proof-of-principle of non-invasive microwave field focusing using iTR and TI in a realistic human head model.