Multiple microelectrode-recording system for human intracortical applications

https://doi.org/10.1016/S0165-0270(01)00330-2Get rights and content

Abstract

The human brain is dominated by the neocortex, a large folded surface, whose cellular and synaptic elements are arranged in layers. Since cortical structure is relatively constant across its surface, local information processing can be inferred from multiple laminar recordings of its electrical activity along a line perpendicular to its surface. Such recordings need to be spaced at least as close together as the cortical layers, and need to be wideband in order to sample both low frequency synaptic currents as well as high-frequency action potentials. Finally, any device used in the human brain must comply with strict safety standards. The current paper presents details of a system meeting these criteria, together with sample results obtained from epileptic subjects undergoing acute or chronic intracranial monitoring for definition of the epileptogenic region.

Introduction

In recent years, the functional topography of the human cerebral cortex has been mapped with increasing detail using first Positron Emission Tomography (PET) and now functional Magnetic Resonance Imaging (fMRI). However, these techniques measure hemodynamic responses, and thus do not have sufficient spatiotemporal resolution to distinguish the elementary neuronal processes whose interaction constitutes neuronal information processing (Dale et al., 2000). Traditionally the input and output of single neurons has been measured with intracellular or extracellular microelectrodes, respectively. Unfortunately, the technical demands of the intracellular recordings are difficult to meet in behaving animals, and quite impractical in behaving humans. Furthermore, although such recordings indicate a great deal about the activity in the soma of a single neuron, they do not reveal the activity in distant dendrites of the same cell, much less the overall activity of the cortical column that may be argued to be the effective computational unit.

Single and multiple microelectrode techniques have been developed for human and animal research in an attempt to reveal the neuronal and neuronal network processing occurring within identified cortical functional areas (Salcman and Bak, 1976, Krüger, 1982, Verloop and Holsheimer, 1984, Howard et al., 1997, Fried et al., 1999, Ojemann and Schoenfield-McNeill, 1999). In ‘laminar’ recordings, simultaneous measurements of extracellular action potentials (AP) and field potentials (FP) are made from multiple, equally spaced locations spanning the cortical layers (Wise and Starr, 1970, Prohaska et al., 1979, Barna et al., 1981, Kuperstein and Wittington, 1981, Karmos et al., 1982). Since the cortex is topologically a thin sheet with most of its variation occurring perpendicular to its surface, this one-dimensional sampling can lead to a surprisingly detailed view of intracortical functional interactions.

Until recently, laminar multiple microelectrode (multielectrode) recordings have been performed only in animals. However, human cognition is quite different from animal cognition, and language gives us a much more direct access to it in other humans. In particular, cognitive event-related potential (ERP) studies are difficult or impossible to be performed adequately in animals because of the complexity of the tasks where they are evoked and manipulated. Furthermore, it is obviously superior to characterize in detail the pathophysiology of an actual human disease, as opposed to an animal model of that disease. The pathophysiology of epilepsy in particular has to date been primarily investigated in animal models. Additional information from human studies could help improve diagnostic techniques for the surgical resection of epilepsy by better defining tissue for resection, and regions of eloquent cortex (through increased knowledge of local ERP generators and the represented cognitive function). Furthermore, the work may lead to improved medical management of epilepsy by providing insight into the neural mechanisms underlying epileptic propagation.

In the current paper we describe a working system for chronic laminar recordings in humans, and document the performance of that system with results from cognition, spontaneous activity and epileptiform events. These appear to be the first intracortical multielectrode recordings performed in humans. The system can record multiple unit activity (MUA) as well as FPs and APs. The locally generated synaptic currents are computed with one-dimensional current source density (CSD) analysis. The described implementation can record 48 channels of intracortical electrical activity at a sampling rate of 20 kSample/second/channel in the time domain, with a spatial resolution of 150–200 μm. The system's maximal recording capacity is 256 channels. The multielectrode is MRI- and biocompatible. The multielectrode system was used in humans undergoing acute or chronic intracerebral monitoring for epileptogenic activity prior to surgical removal of the seizure focus.

The main concepts of the recording technique will be discussed such as the multielectrode and its implantation procedure, preamplifier, main amplifier, personal computer (PC) based acquisition system and the outlines of the data analysis. We will concentrate on the special features of the human experimental setup, which are different from conventional animal experimental ones and sample data presented.

Section snippets

Electrode fabrication

We designed a thumbtack like 22–24 contact linear array multielectrode (Fig. 1). Intercontact distance can be adjusted from 75 to 200 μm in 25 μm steps, contact diameter of 40 μm. Contacts are formed from polyimide isolated platinum–iridium (90/10%) (California Fine Wire Co.) wires, which act as output leads as well. The entire array was embedded in a polyimide-epoxy substrate with a total outer diameter of 350 μm and a variable length (Fig. 2). The electrodes designed with a rounded profile

Evaluation of the electrode in vitro

CSD analysis requires that the differences in potential between contacts be due to physiological sources, rather than to errors in the spacing of contacts, or the consistency of amplification and filtering. This was tested by comparing the potentials recorded across individual electrode contacts using an in vitro setup with precisely predictable potentials. In theory, if two point-like current sources are located far from each other in a volume conductor, there are parallel equipotential planes

Basic concepts of the electrode-preamplifier system

In order to infer the laminar distribution of intracortical synaptic processes in humans or in animals it is essential to use a sensor, which records FPs, MUA, and APs at the same time from different and well defined cortical depths. Although various kinds of penetrating multielectrodes are reported in the literature, and some of them are commercially available, the laminar probe described in this paper has capabilities and characteristics that make it particularly appropriate for CSD/MUA

Conclusions

We present here a methodology that can reliably record field potentials, multiple unit and single unit activity in behaving and anaesthetized humans. The capacity of the system is shown in acute and chronic recordings, with cognitive stimulation as well as spontaneous activity. This probe should be useful for improving our understanding of the information processing in human brain as well as a tool to improve our understanding in epileptic processes in the future.

Acknowledgements

We thank L. Papp for excellent technical support, and L. Shuer and M. Reisinger for clinical collaboration. Supported by the Human Frontiers Science Program Organization (RG0025/1996), National Institutes of Health (NS 18741), National Foundation for Functional Brain Imaging, and Office of Naval Research.

References (51)

  • D.T. Kewley et al.

    Plasma-etched neural probes

    Sens. Actuators

    (1997)
  • J. Krüger

    A 12-fold microelectrode for recording from vertically aligned cortical neurons

    J. Neurosci. Methods

    (1982)
  • M. Kuperstein et al.

    Unit activity, evoked potentials and slow waves in the rat hippocampus and olfactory bulb

    Neuroscience

    (1985)
  • M.R. Nuwer et al.

    Neuromonitoring during surgery. Report of an IFCN committee

    Electroencephalogr. Clin. Neurophysiol.

    (1993)
  • M.R. Nuwer et al.

    IFCN standards for digital recording of clinical EEG

    Electroencephalogr. Clin. Neurophysiol.

    (1998)
  • H. Pockberger et al.

    Penicillin-induced epileptic phenomena in the rabbit's neocortex I. The development of interictal spikes after epicortical application of penicillin

    Brain Res.

    (1984)
  • O. Prohaska et al.

    A 16-fold semi-microelectrode for intracortical recording of field potentials

    Electroencephalogr. Clin. Neurophysiol.

    (1979)
  • C.E. Schroeder et al.

    Timing and distribution of flash-evoked activity in the lateral geniculate nucleus of the alert monkey

    Brain Res.

    (1989)
  • I. Szabó et al.

    A low-noise preamplifier for multisite recording of brain multi-unit activity in freely moving animals

    J. Neurosci. Methods

    (1993)
  • G. Vaknin et al.

    A method for calculating current source density (CSD) analysis without restoring sites outside the sampling volume

    J. Neurosci. Methods

    (1988)
  • A.J. Verloop et al.

    Simple method for the construction of electrode arrays

    J. Neurosci. Methods

    (1984)
  • S.L. BeMent et al.

    Solid-state electrodes for multichannel multiplexed intracortical neuronal recording

    IEEE Trans. Biomed. Eng.

    (1985)
  • H. Eichenbaum et al.

    Extracellular neural recording with multichannel microelectrodes

    J. Electrophysiol. Tech.

    (1986)
  • J.A. Freeman et al.

    Experimental optimization of current source-density analysis technique for Anuran cerebellum

    J. Neurophysiol.

    (1975)
  • I. Fried et al.

    Cerebral microdialysis combined with single-neuron and electroencephalographic recording in neurosurgical patients

    J. Neurosurg.

    (1999)
  • Cited by (137)

    • Electrode Development for Epilepsy Diagnosis and Treatment

      2024, Neurosurgery Clinics of North America
    • The laminar profile of sleep spindles in humans

      2021, NeuroImage
      Citation Excerpt :

      The asymmetrical co-occurrence patterns between IME and ECoG is reminiscent of the observation that spindle oscillations in electrophysiological channels with larger recording areas (EEG, especially monopolar EEG) are usually also seen in channels with smaller recording areas (MEG, especially gradiometers) but the reverse is not true (Dehghani et al., 2010; Dehghani et al., 2011): that is, a large degree of sleep spindles are extremely local events and monopolar EEG derivations only detect their largest, most synchronized instances. IME data are essentially bipolar EEG recordings from inside the cerebral cortex, with each channel having an extremely small recording area measurable in hundreds of micrometers at most (Ulbert et al., 2001). Therefore, IME-exclusive spindles likely represent the extreme left tail of the distribution of the spatial extent of sleep spindles, encompassing the most local instances.

    View all citing articles on Scopus
    View full text