Investigating the safety of fast neural electrical impedance tomography in the rat brain.

Physiol Meas

Department of Medical Physics and Biomedical Engineering, University College London, London, United Kingdom.

Published: April 2019

Objective: Electrical impedance tomography (EIT) can be used to image impedance changes which arise due to fast electrical activity during neuronal depolarisation and so holds therapeutic potential for improving the localisation of epileptic seizure foci in patients with treatment-resistant epilepsy to aid surgical resection of epileptogenic tissue. Prolonged cortical stimulation may, however, induce neural injury through excitotoxicity and electrochemical reactions at the tissue-electrode interface. The purpose of this work was to assess whether current levels used in fast neural EIT studies induce histologically detectable tissue damage when applied continuously to the rat cerebral cortex.

Approach: A 57-electrode epicortical array was placed on one or both hemispheres of adult Sprague Dawley rats anaesthetised with isoflurane. In an initial series of experiments, current was injected simultaneously at 10, 25, 50, 75 and 100 µA for 1 h at 1.725 kHz through five electrodes across two epicortical arrays to provide a preliminary indication of the safety of these current levels. Since no obvious cortical damage was observed in these rats, the current level chosen for further investigation was 100 µA, the upper-bound of the range of interest. In a separate series of experiments, 100 µA was applied through a single electrode for 1 h at 1.725 kHz to verify its safety. Following termination of stimulation, brain samples were fixed in formalin and histologically processed with Haematoxylin and Eosin (H&E) and Nissl stains.

Main Results: Histological analysis revealed that continuous injection of 100 µA current, equating to a current density of 354 Am, into the rat cortex at 1.725 kHz does not cause cortical tissue damage or any alterations to neuronal morphology.

Significance: The safety of current injections during typical EIT protocols for imaging fast neural activity have been validated. The current density established to be safe for continuous application to the cortex, 354 Am, exceeds the present safety limit of 250 Am which has been complied with to date, and thus encourages the application of more intensified fast neural EIT protocols. These findings will aid protocol design for future clinical and in vivo EIT investigations aimed at imaging fast neural activity, particularly in situations where the signal-to-noise ratio is considerably reduced.

Download full-text PDF

Source
http://dx.doi.org/10.1088/1361-6579/ab0d53DOI Listing

Publication Analysis

Top Keywords

fast neural
20
100 µa
16
electrical impedance
8
impedance tomography
8
current
8
current levels
8
neural eit
8
tissue damage
8
series experiments
8
1 h 1725 khz
8

Similar Publications

Faster Acquisition and Improved Image Quality of T2-Weighted Dixon Breast MRI at 3T Using Deep Learning: A Prospective Study.

Korean J Radiol

January 2025

Department of Diagnostic and Interventional Radiology, University Medical Center Freiburg, Faculty of Medicine, University of Freiburg, Freiburg, Germany.

Objective: The aim of this study was to compare image quality features and lesion characteristics between a faster deep learning (DL) reconstructed T2-weighted (T2-w) fast spin-echo (FSE) Dixon sequence with super-resolution (T2) and a conventional T2-w FSE Dixon sequence (T2) for breast magnetic resonance imaging (MRI).

Materials And Methods: This prospective study was conducted between November 2022 and April 2023 using a 3T scanner. Both T2 and T2 sequences were acquired for each patient.

View Article and Find Full Text PDF

The vision transformer (ViT) architecture, with its attention mechanism based on multi-head attention layers, has been widely adopted in various computer-aided diagnosis tasks due to its effectiveness in processing medical image information. ViTs are notably recognized for their complex architecture, which requires high-performance GPUs or CPUs for efficient model training and deployment in real-world medical diagnostic devices. This renders them more intricate than convolutional neural networks (CNNs).

View Article and Find Full Text PDF

Adversarial training has become a primary method for enhancing the robustness of deep learning models. In recent years, fast adversarial training methods have gained widespread attention due to their lower computational cost. However, since fast adversarial training uses single-step adversarial attacks instead of multi-step attacks, the generated adversarial examples lack diversity, making models prone to catastrophic overfitting and loss of robustness.

View Article and Find Full Text PDF

In situ training of an in-sensor artificial neural network based on ferroelectric photosensors.

Nat Commun

January 2025

Institute for Advanced Materials and Guangdong Provincial Key Laboratory of Optical Information Materials and Technology, South China Academy of Advanced Optoelectronics, South China Normal University, Guangzhou, China.

In-sensor computing has emerged as an ultrafast and low-power technique for next-generation machine vision. However, in situ training of in-sensor computing systems remains challenging due to the demands for both high-performance devices and efficient programming schemes. Here, we experimentally demonstrate the in situ training of an in-sensor artificial neural network (ANN) based on ferroelectric photosensors (FE-PSs).

View Article and Find Full Text PDF

Multiband (MB) optical transmission targets increasing the capacity of operators' optical transport networks. However, nonlinear impairments (NLI) affect each optical channel in the C+L+S bands differently, and, therefore, the routing and spectrum assignment (RSA) problem needs to be complemented with fast and accurate tools to consider the quality of transmission (QoT) within the provisioning process. This paper proposes a digital twin-assisted approach for lightpath provisioning to provide a complete solution for the RSA problem that ensures the required QoT in MB optical networks.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!