Subscriber access provided by UNIV AUTONOMA DE COAHUILA UADEC
Mini Review
Soft high-resolution neural interfacing probes: materials and design approaches Mincheol Lee, Hyung Joon Shim, Changsoon Choi, and Dae-Hyeong Kim Nano Lett., Just Accepted Manuscript • DOI: 10.1021/acs.nanolett.8b04895 • Publication Date (Web): 19 Apr 2019 Downloaded from http://pubs.acs.org on April 19, 2019
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Nano Letters
Soft high-resolution neural interfacing probes: materials and design approaches
Mincheol Lee1,2, Hyung Joon Shim1,2, Changsoon Choi1,2, and Dae-Hyeong Kim1,2
1Center
for Nanoparticle Research, Institute for Basic Science (IBS), Seoul 08826, Republic of
Korea 2School
of Chemical and Biological Engineering, Institute of Chemical Processes, Seoul
National University, Seoul 08826, Republic of Korea
M.
Lee, H.J. Shim, C. Choi contributed equally
*To whom correspondence should be addressed. E-mail:
[email protected] 1
ACS Paragon Plus Environment
Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Abstract
Neural interfacing probes are located between the nervous system and the implanted electronic device in order to acquire information of the complex neuronal activity and to reconstruct impaired neural connectivity. Despite remarkable advancement in recent years, conventional neural interfacing is still unable to completely accomplish these goals, especially in long-term brain interfacing. The major limitation arises from physical and mechanical differences between neural interfacing probes and neural tissues that cause local immune responses and production of scar cells near the interface. Therefore, neural interfaces should ideally be extremely soft and have the physical scale of cells to mitigate the boundary between biotic and abiotic systems. Soft materials for neural interfaces have been intensively investigated to improve both interfacing and long-term signal transmission. The design and fabrication of micro and nanoscale devices have drastically decreased the stiffness of probes and enabled single-neuron measurement. In this mini review, we discuss materials and design approaches for developing soft high-resolution neural probes intended for long-term brain interfacing, and outline existent challenges for achieving next-generation neural interfacing probes.
Keywords: Neural probes, neural chronic recording, soft electronics, bioelectronics
2
ACS Paragon Plus Environment
Page 2 of 29
Page 3 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Nano Letters
Recent advances in neuroscience have enabled the exploration of subtle brain activity and allowed understanding of its mechanisms down to the microscopic level. For instance, advances in neuroimaging technologies including functional magnetic resonance imaging (fMRI) have dramatically extended our understanding of neural circuits and their connectivity associated with specific brain activity1. However, it is difficult to apply neuroimaging techniques for analyzing the brain function during daily living, because they depend on bulky equipment, thus lacking portability2. Low spatiotemporal resolution is another key limitation of fMRI in brain research. On the other hand, scalp electroencephalography (EEG) allows the continuous monitoring of macroscopic brain activity even for long-term bio-integration3. However, the low spatial resolution of EEG undermines the measurement accuracy and is not compatible with neuroscience research on topics such as single-neuron level studies. Meanwhile, invasive neural interfacing methods, such as the use of either epicortical multielectrode arrays or intracortical multichannel probes, have enabled the continuous monitoring of brain signals with high spatiotemporal resolution4. Devices based on invasive neural interfacing have been used in various clinical applications including cochlear5 and retinal implants6, spinal7 and peripheral nerve interfaces8, epilepsy monitoring devices9, and deep brain stimulators10. However, it is still challenging to achieve high-quality neural interfaces for both high-resolution recording of action potentials and low-current electrical stimulation, particularly over the long term. The limitations mainly originate from physical and mechanical differences between the neural interfacing devices and biological tissues11–13. In fact, conventional electrodes for brain interfacing are fabricated with metals or doped silicon, which are much stiffer than biological tissues. Consequently, scar formation and neuroinflammatory responses are often induced by the mechanical mismatch between probes and tissue, especially under brain micromotions, causing gradual signal loss 3
ACS Paragon Plus Environment
Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
over time11. Moreover, a probe that is much larger than a neuron cannot acquire microscopic electrophysiology and neural activity, which are essential to understand the brain function and mechanisms. Therefore, new approaches are required to solve these mechanical and physical mismatches without affecting spatiotemporal resolution and electrode performance. Recent developments on soft electronics and micro and nanoelectronics have paved the way to overcome these limitations. Here, we present a mini review regarding novel material and device design approaches for developing high-resolution soft neural interfaces, which exhibits mechanical and physical compatibility with brain tissues.
Towards electrophysiology of deeper brain regions and single neurons To start discussion of neural interfacing probes, we categorize the types of signals that can be measured from the brain. Neural signals can be classified into EEG, electrocorticogram (ECoG), local field potentials (LFPs), and cellular signals from single neurons, depending on the origin of the acquired signal14. Among them, EEG, which is noninvasively acquired from the scalp, is the easiest electrophysiological signal to obtain (fig. 1a, left and middle)15. However, it cannot provide local information regarding a specific brain region of interest given the numerous interfering LFPs (fig. 1a, right). In addition, the EEG signal reflects noise from other internal and external sources, leading to a poor signal-to-noise ratio. In contrast, ECoG signals (epidural/subdural recordings) do not reflect noise from sources between the skull and skin by direct measurement from the cerebral cortex (fig. 1b, left and middle)16. Still, however, ECoG only collects activity from superficial regions of the brain. Therefore, it can be used to identify simple human cognitive and sensory information like EEG (fig. 1b, right), but it cannot reach the spiking resolution present in signals from individual neurons. For more profound research on human cognitive and sensory systems, it is important 4
ACS Paragon Plus Environment
Page 4 of 29
Page 5 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Nano Letters
to collect signals from individual neurons over a specific region of the brain with high spatiotemporal resolution. To this end, the wideband measurement of LFPs reflects local oscillatory signals acquired from small brain volumes (fig. 1c, left). Given that the wideband measurement of LFPs contains both action potentials and other membrane potential fluctuations from a small neuronal volume, it offers valuable information about the evaluated brain region14. This measurement approach usually targets deep brain regions below the cortex surface, and hence electrodes should be invasively inserted into the brain (fig. 1c, middle)17. To prevent inflammatory responses in the long term due to mechanical mismatch with brain tissues, the neural probe should be biocompatible and soft, which become critical properties when considering brain micromotions. When properly implanted, this type of probe can collect neural signals from a small volume of neurons around the insertion site4,11. High-pass filtered (>300 Hz) wideband LFP signals can provide information about nearby neurons’ spikes excluding interference from other membrane potentials (fig. 1c, right)17, and their spatiotemporal resolution can be further enhanced by introducing neural interfaces with the cellular and subcellular size (fig. 1d, left). Ultimately, micro and nanoelectrodes, whose size is comparable to that of single neurons, can detect brain activity down to the cellular level (fig. 1d, middle)18. The microelectrode array integrated with a cultured neuron cell can either record or stimulate the intracellular activity of a single neuron (fig. 1d, right)18. In summary, despite all signals coming from the same underlying brain activity, the quality of the measured brain signal is determined by the recoding location and the recording method. As the neural probe approaches deeper brain regions and the scale of a single neuron, the neural interface can obtain detailed and specific information about the specific brain region or the electrophysiological response of interest. To obtain such accurate information, the neural probes must be tissue-like, i.e., soft and miniaturized, to match the physical dimension and 5
ACS Paragon Plus Environment
Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 6 of 29
mechanical properties of neurons, while providing long-term biocompatibility. In the following section, we discuss current approaches to develop such high-resolution soft probes and to achieve tissue-like neural interfacing.
Materials and design approaches for mechanically compliant neural probes The strategy to improve the mechanical and physical compatibility of neural interfacing probes with brain tissues involves design form factor and materials engineering19. A key factor is bending stiffness D of materials20: 𝐷=
𝐸ℎ3 12(1 ― 𝑣2)
,
where E is the elastic modulus of the material, h is its thickness, and v is its Poisson’s ratio. As the bending stiffness scales cubically with the thickness of the material and linearly with its elastic modulus, either reducing the dimension of the device or using materials with low modulus can drastically reduce the device stiffness. Several studies have demonstrated that ultrathin polymeric layers, such as polyimide21– 23,
parylene16,24,25, epoxy26, and polyethylene naphthalate thin films27, offer high mechanical
compliance (fig. 2a, left). Despite their elastic modulus in the GPa range, ultrathin devices enable conformal contacts on the convoluted surfaces of target organs such as skin, heart, and brain because of their low flexural rigidity and device stiffness16,23. Similarly, in neural interfacing, ultrathin devices can be conformally attached on neural structures for recording and stimulation. For example, figure 2a (middle) shows a parylene-C-based electrode array using gold interconnects and a conducting polymer (poly(3,4-ethylenedioxythiophene) doped with poly(styrenesulfonate)—PEDOT:PSS)16. The ultrathin chemically vapor deposited parylene C (thickness, 4 m) allows the device to intimately adhere to complex curved surfaces. 6
ACS Paragon Plus Environment
Page 7 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Nano Letters
Based on this strategy, reliable ECoG signals have been acquired from the pial surface of human subjects, thus verifying the stable electrical and mechanical contact with the cortical surface. Figure 2a (right) shows another representative example of electrodes for brain interfacing based on polymeric substrates23. Polyimide layers and epoxy layers with thicknesses of approximately 10 m are used for encapsulation. By reducing the bending stiffness with thinned polymeric layers, multiplexed electrodes exhibit extreme flexibility allowing insertion into the interhemispheric fissure (fig. 2a, right inset). Fabricating devices with the shape of thin fibers is another good example for reducing device dimensions (fig. 2b, left). The small diameter of fiber-shaped neural probes offers very low bending stiffness and reduces the electrode footprint28–30. Kozai et al.28 developed a thinfiber electrode based on carbon fibers that are mechanically robust and flexible (fig. 2b, middle). The composite electrode consists of a carbon fiber core, polymeric dielectric coating, and polymeric recording pads. In addition, its small diameter (
