2d and 3D neural electrodes and methods thereof

Abstract

The invention generally relates to 2D and 3D neural electrodes and methods thereof. The invention comprises a needle array comprising an optically transparent substrate; and a plurality of needles operably coupled to the optically transparent substrate; an optical system, wherein the optical system is arranged with respect to the needle array such that an optical output from the optical signal passes through the optically transparent substrate of the needle array; and a processor operably associated with each of the needle array and the optical system to receive electrical data from the needle array and optical data from the optical system and process both the electrical data and the optical data.

Description

[0001] Attorney Docket No.: PURD-151 / 01WO 28593 / 707

[0002] PATENT APPLICATION

[0003] 2D AND 3D NEURAL ELECTRODES AND METHODS THEREOF

[0004] Related Application

[0005] The present application claims the benefit of and priority to U.S. provisional patent application serial number 63 / 542,491, filed October 4, 2023, the content of which is incorporated by reference herein in its entitety.

[0006] Government Support

[0007] This invention was made with government support under R21EB029740 and DP2MH136494 awarded by the National Institutes of Health, under FA9550-22-1-0078 and FA9550-23-1-0701 awarded by the Air Force Office of Scientific Research (AFOSR), and under RGY0069 / 2020 awarded by the Human Frontiers Science Program (HFSP). The government has certain rights in the invention.

[0008] Field of the Invention

[0009] The invention generally relates to 2D and 3D neural electrodes and methods thereof.

[0010] Background

[0011] The emergence of 3D transparent nanoelectrodes marks a groundbreaking advancement in brain interface technology, fundamentally transforming how we interact with and understand neural systems. Unlike conventional 2D electrode arrays, which often struggle with limited spatial resolution and can disrupt the natural architecture of brain tissue, 3D nanoelectrodes are designed to mimic the complex topography of neural environments, allowing for enhanced connectivity and reduced tissue damage. Their transparent design not only facilitates unobstructed imaging of neural processes but also enables the integration of optical stimulation techniques, paving the way for sophisticated multimodal approaches in neuroscience. This innovative architecture allows for a denser array of recording sites, capturing a broader spectrum of neural activity with unprecedented precision. As we push the boundaries of brain-computer interfaces, 3D transparent nanoelectrodes stand out as a pivotal technology, promising to unlock new pathways in understanding brain function, advancing neuroprosthetics, and developing targeted therapies for neurologicalAttorney Docket No.: PURD-151 / 01WO 28593 / 707

[0012] PATENT APPLICATION

[0013] disorders, all while minimizing the invasiveness traditionally associated with electrode implantation.

[0014] High-density electrophysiology recordings in 3 dimensions combined with 2-photon calcium imaging are a high-fidelity neuroscience technique that will enable understanding of spatial and temporal coding and computational strategies employed by the brain during complex tasks. These extracellular electrophysiological recordings have been conventionally performed using linear silicon probes (Michigan Probes) or silicon penetrating microelectrode array (Utah array). These methods are limited spatially in either 1 or 2 dimensions and are incompatible with 2-photon imaging as they are rigid and opaque. Combining electrophysiological recording with optical techniques like 2P imaging will help overcome the limitations of each other (low spatial resolution of EPhys and low temporal resolution of 2P imaging). Critically, these 3D electrodes will allow for more optimal decoding of brain activity.

[0015] Understanding the mechanisms that give rise to critical brain functions such as perceptions, actions, and memories requires monitoring the dynamic fluctuating patterns of activity within intact neural circuits at the cellular, circuit, and system levels and relating them to behavioral outcomes. However, such studies often require the animal to be head-fixed. While this allows for well-controlled stimuli and stable recordings, the animal is under severe restraint which precludes an accurate estimate of behavioral effects on circuits and circuit effects on behaviors. For example, recent studies have demonstrated prominent and widespread movement-related signals in the brain of head-fixed mice, even in primary sensory areas. Also, eye movements are large and constant in freely moving rodents but are essentially absent under head restraint. However, it is still unknown what role these signals play in sensory processing. Why are these sensory areas ‘pervaded’ by movement signals? During natural behavior, animals actively acquire sensory information as they move through the environment and use this information to guide ongoing actions. In this context, unconstrained movement-related signals could allow sensory systems to efficiently predict self-generated motion and extract additional information about the environment, thereby forming a stable internal representation of the external world. The discovery of place cells, head-direction cells and grid cells exemplify the benefits of this approach. Movement related signaling would also allow mice to deploy different behavioral strategies during sensory processing, an aspect non-determinable in head-fixedAttorney Docket No.: PURD-151 / 01WO 28593 / 707

[0016] PATENT APPLICATION

[0017] conditions. However, to probe how dynamic circuit patterns and their emergent properties shape behavioral outcomes requires mapping large populations of neurons in freely-moving animals.

[0018] The current state-of-the-art techniques for monitoring such activity patterns are either through high-density electrophysiology, which accords ultra-high temporal resolution, or via nonlinear microscopy such as two-photon imaging, albeit nascent, which offers insight into the spatial structure of activity. The flow of electricity in the form of action potentials and synaptic currents is the currency of the brain, and neural activity and synaptic changes are sensitive to millisecond timescales. High-temporal resolution is thus critical for faithfully representing millisecond precision electrical dynamics in the brain. Imaging cellular and sub-cellular patterns on the contrary helps unravel how spatial representation relates to behavior. Although high-density electrophysiology recordings can be performed using linear silicon probes in freely-moving animals, they are incapable of linking cellular scale organization to ongoing dynamics and cannot capture activity across both vertical and horizontal layers in the brain with micron scale resolution. This constraint makes it difficult to infer anything about global brain patterns and their evolution in time. On the other hand, two-photon calcium imaging lacks the temporal resolution to effectively tie electrical dynamics to the underlying spatial structure.

[0019] Summary

[0020] The invention recognizes that there is a great need for technology platforms that can combine high-resolution electrical recordings across entire volumes of brain tissue and two-photon calcium imaging in head-fixed and freely behaving animals. In that manner, the invention provides needle grids integrated onto any optical instrument, such as any microscope.

[0021] Particularly, aspects of the invention provide a transparent and flexible 3D implant capable of simultaneous electrophysiological / electrochemical recording with optical measurements such as 2-photon calcium imaging. Using novel microscale 3D metal printing, we have created customizable high-aspect-ratio needle electrodes on a flexible substrate for multi-depth recordings. While the platform can be implemented widely across the central and peripheral nervous system and used in a variety of neuroelectronic applications, the importance of this platform can be easily understood with the following description and embodiment.Attorney Docket No.: PURD-151 / 01WO 28593 / 707

[0022] PATENT APPLICATION

[0023] In certain aspects, the invention provides a system for making high-density electrophysiology in-vivo recordings, the system comprising: a needle array integrated into an optical instrument. In certain embodiments, the needle array comprising an optically transparent substrate; and a plurality of needles operably coupled to the optically transparent substrate; an optical system, wherein the optical system is arranged with respect to the needle array such that an optical output from the optical signal passes through the optically transparent substrate of the needle array; and a processor operably associated with each of the needle array and the optical system to receive electrical data from the needle array and optical data from the optical system and process both the electrical data and the optical data.

[0024] In certain embodiments, at least two of the needles of the plurality of needles are of different heights. In certain embodiments, each needle of the pluralirty of needles has a base and a tip, wherein the base is wider than the tip and each needle comprises a taper from the base to the tip. In certain embodiments, the plurality of needles are printed onto the optically transparent substrate. In certain embodiments, each needle of the pluralirty of needles is comprised of a metal that is coated with a biocompatiable polymer, wherein the tip remains unencapsulated.

[0025] In other aspects, the invention provides a method for making electrophysiology in-vivo recordings, the method comprising: providing a system that comprises: a needle array integrated into an optical instrument, and a processor operably associated with each of the needle array and the optical system to receive electrical data from the needle array and optical data from the optical system and process both the electrical data and the optical data; inserting the needle array into in-vivo tissue; directing directing an optical output from the optical system onto the in vivo tissue; and receiving to the processor electrical data from the needle array and optical data from the optical system, wherein the processor processes both the electrical data and the optical data.

[0026] In certain embodiments, at least two of the needles of the plurality of needles are of different heights. In certain embodiments, each needle of the plurality of needles has a base and a tip, wherein the base is wider than the tip and each needle comprises a taper from the base to the tip. In certain embodiments, the plurality of needles are printed onto optically transparent substrate. In certain embodiments, each needle of the pluralirty of needles is comprised of a metal that is coated with a biocompatiable polymer, wherein the tip remains unencapsulated.

[0027] In other aspects, the invention provides a needle array comprising: a flexible and optically transparent substrate; and a plurality of needles operably coupled to the flexible andAttorney Docket No.: PURD-151 / 01WO 28593 / 707

[0028] PATENT APPLICATION

[0029] optically transparent substrate, wherein each needle of the pluralirty of needles is comprised of a metal that is coated with a biocompatiable polymer, wherein the tip remains unencapsulated.

[0030] In certain embodiments, at least two of the needles of the plurality of needles are of different heights. In certain embodiments, each needle of the pluralirty of needles has a base and a tip, wherein the base is wider than the tip and each needle comprises a taper from the base to the tip. In certain embodiments, the plurality of needles are printed onto the flexible and optically transparent substrate. In certain embodiments, subject is at least one selected from the group consisting of: a silicon wafer; a printed circuit board; a glass coverslip; and a CMOS chip.

[0031] In other aspects, the invention provides a method for manufacturing a needle array, the method comprising: providing a flexible and optically transparent substrate; directly printing metal needles onto the flexible and optically transparent substrate; and encapsulating the metal needles with a biocompatible polymer, wherein the tip remains unencapsulated.

[0032] In certain embodiments, prior to printing, the flexible and optically transparent substrate is cleansed. In certain embodiments, at least one of the metal needles is printed as a single section. In certain embodiments, at least one of the metal needles is printed as multiple sections. In certain embodiments, each of the metal needles is printed with a base and a tip wherein each of the metal needles tapes from the base to the tip.

[0033] In other aspects, the invention provides a method for making high-density electrophysiology in-vivo recordings, the method comprising: providing needle array comprising a flexible and optically transparent substrate; and a plurality of needles operably coupled to the flexible and optically transparent substrate, wherein each needle of the pluralirty of needles is comprised of a metal that is coated with a biocompatiable polymer, wherein the tip remains unencapsulated; inserting needle array into in-vivo tissue; and receiving electrical signals from the needle array to a processor, wherein the processor processes the electrical signals.

[0034] In certain embodiments, at least two of the needles of the plurality of needles are of different heights. In certain embodiments, each needle of the pluralirty of needles has a base and a tip, wherein the base is wider than the tip and each needle comprises a taper from the base to the tip. In certain embodiments, the plurality of needles are printed onto the flexible and optically transparent substrate. In certain embodiments, is at least one selected from the group consisting of: a silicon wafer; a printed circuit board; a glass coverslip; and a CMOS chip.Attorney Docket No.: PURD-151 / 01WO 28593 / 707

[0035] PATENT APPLICATION

[0036] Brief Description of the Drawings

[0037] FIG. 1 compares the current prior art (left panel) to the claimed solution (right panel), and illustrates circuit dynamics during sensory processing and sleep in freely moving mice by developing the NET-2P for 3D electro-optic mapping of cortical activity.

[0038] FIG. 2 shows a proposed mechanism of traveling waves (left). By implementing NET-2P, we aim to understand how traveling waves enable precise delays to dictate synaptic plasticity (right).

[0039] FIG. 3 panels A-E show a NanoNeedle fabrication process. (Panels A-B) Schematic depicting nanoscale metal printing on flexible parylene probes. (Panel C) Needles of various aspect ratios, including multi-height designs. The multi height nature allows for 3D volumetric electrophysiology across brain tissue. (Panels D-E). Impedance spectra. (E, right) single units detected using the flexible needle probes.

[0040] FIG. 4 panels A-B show the design of the multi-channel amplifier array for Nano-Needle Array (Panel A) Circuit schematic of the CMOS IC for signal amplification and sampling. The circuit has 256 channels which are multiplexed in a 16 x 16 configuration. (Panel B) Prototype CMOS IC fabricated in TSMC’s 180 nm process.

[0041] FIG. 5 panels A-F show mapping traveling wave dynamics in head-fixed mice. (Panels A-B) Planar flexible grids with thin-gold. (Panel C) Experimental approach combining electrical and optical recordings; (Panels D-F) Preliminary data showing simultaneous mapping of TWs and cortical ensembles in mouse barrel cortex. TWs propagate with a stereotypical velocity and phase lag (Panel D,i, ii) and exhibit a sensory evoked surface potential with a late reverberatory fluctuation, which is dependent on learned associations (Panel D, iii) and motor feedback (Panel E). (Panel F) Imaging performed through the grids shows distinct ensembles (right inset) as a function of trials with and without a late spike.

[0042] FIG. 6 shows mapping traveling wave dynamics and associated ensembles during touch discrimination in freely moving mice.

[0043] FIG. 7 shows mapping traveling waves and ensembles during sleep.

[0044] FIG. 8 panels A-F show flexible nano-needle array for volumetric electrophysiological recordings. (A) Array of 64 copper needles with submicron needle tip (1&2) and array of 16 ultra-tall 1 mm needles (3). (B) Fabricated flexible 32 channel probe with alternating 100 pm and 250 pm tall needles to record from multiple layers of cortex simultaneously. (C)Attorney Docket No.: PURD-151 / 01WO 28593 / 707

[0045] PATENT APPLICATION

[0046] Summarized fabrication process for fabricating the neural probe. (D) SEM images of needle tip after RF exposure (1) and electrodeposition of Pt-NP (2). (E) The impedance of the exposed needles before and after deposition of Pt-NP. (G) Needles bend and retain shape following mechanical manipulation.

[0047] FIG. 9 shows flexible and transparent substrate for printing 3D electrodes as well as surface recording. The probe has gold electrodes sandwiched between Parylene. The SU8 layer is added at the base for ease of handling.

[0048] FIG. 10 shows nanoporous gold electrodes for high fidelity recording. The recording electrode have a nanoporous gold textured to increase the surface area which in turn reduces the electrode impedances. The impedance plots are shown in the right.

[0049] FIG. 11 shows 3D electrodes on glass canula. Metal electrodes are patterned directly on the glass canula and needle electrodes could be printed on these metal electrodes.

[0050] FIG. 12 shows direct printing of 3D electrodes on conventional CMOS ICs.

[0051] FIG. 13 shows a fabrication process for encapsulation and tip exposure of the needle electrodes.

[0052] FIG. 14 shows mechanical modelling for determining optimal geometry of the needles for damage freee insertion into the brain.

[0053] FIG. 15 shows SEMs for tip exposure using RIE etching. The figure shows the array with photo resist to protect the paryelen on the base and zoomed in exposed tip.

[0054] FIG. 16 panels A-C show chronic implant for tethered freely moving recording. (A) Chronic implant on a mouse. (B) The chronic implant with the grid and the headplate. (C) Chronic window stability with the grid.

[0055] FIG. 17 panels A-F show recording neural activity from barrel cortex of mouse using whisker touch. (A) The implantation protocol. (B) Recovery of the probes enabling reuse across multiple experiments. (C) Multiunit activity from multiple channels. (D) Clusters of distinct spikes in PCA space on a channel. (E) Touch evoked LFP across the array. (F) Touch evoked travelling waves at 250 pm depth from surface.

[0056] FIG. 18 shows simultaneous surface and translaminar neural recording.

[0057] FIG. 19 shows simultaneous surface ECoG and calcium imaging.

[0058] FIG. 20 shows simultaneous translaminar electrophysiology and calcium imaging.

[0059] FIG. 21 shows 3D electrophysiological recording with simultaneous calcium imaging.Attorney Docket No.: PURD-151 / 01WO 28593 / 707

[0060] PATENT APPLICATION FIG. 22 shows a schematic of compact 3D electrode implant with needle electrodes printed directly on conventional CMOS amplifier array for easy integration.

[0061] FIG. 23 shows a schematic of electrochemical recording of neurotransmitter levels in 3D using needle electrode array.

[0062] FIG. 24 shows integration of developed 3D needle array with open source mini-2p system for performing Ephys along with calcium imaigng in freely moving mice.

[0063] FIG. 25 shows a schematic of platform for simultaneous recording and stimulation of neurons in a cell culture or 3D organoid. The technique could also be combined with calcium imaging.

[0064] FIG. 26 shows 3D electrodes on flexible self-curling polymer substrates for creating cuff electrode array for recording and stimulation of vagus nerve or spinal cord.

[0065] Detailed Description

[0066] Local-field-potential (LFP) recordings of sensory stimulus-driven oscillations from the mammalian cortex exhibit ongoing “waves”. Animal studies have demonstrated that these waves temporally group spikes from ensembles of neurons to constrain network synchrony in fast (<35 ms) and slow (1 sec) time packets to filter inputs, modulate brain-state, and ensure efficient information coding. Traditionally, the LFP across cortical and sub-cortical regions has been thought to involve zero-lag synchrony, in which oscillations across regions are aligned perfectly in time. However, recent studies, have revealed dynamic timing differences between population responses, leading to the emergence of travelling waves (TWs) between functionally connected cortical regions. TWs can be defined as a non- stationary pattern of sensory-evoked cortical activity that spreads out from the point of maximum input in the cortex impacting neuronal excitability both within and across cortices with precise time (phase) delays. Such phase delays can entrain and destabilize local synchrony8, influence subsequent integration, shape sensory responses over a wide range of spatial and temporal scales (several ms to 100’s of ms), and aid with memory consolidation. For example, during sensory processing top-down feedback of long-range contextual and cognitive signals could be precisely timed by TWs to enable widespread synchrony without colliding with ongoing feedforward sweeps - a form of temporal conflict resolution. While studies suggest that TWs are a consequence of axonal fiber delays in superficial layers of the cortex, brain circuits are recurrent and distributed across vertical andAttorney Docket No.: PURD-151 / 01WO 28593 / 707

[0067] PATENT APPLICATION

[0068] horizontal layers suggesting travelling waves could be three-dimensional. Importantly, TWs can only be mapped when one measures with high-density electrode recordings. While the nature of TWs in the form of LFP waves is known, the underlying spiking dynamics and cellular architecture that supports such a wave are not known. Functional neural populations, termed ensembles, exhibit recurring spatial activity patterns which are semi-stable across multiple contexts (i.e. rest, sensory stimulation, and cognitive operations). Different ensembles within a given local cortical territory (0.5-lmm3) continuously remap to accommodate new stimuli, essentially outlining the “vocabulary” of the cortex. Such stable multicellular activity patterns predict perception, action, and behavior better than the sum of their constituent neurons, suggesting the importance of studying circuits at this level. Importantly, the integrity of these ensemble patterns is quantifiable and comparable across mice and manipulation with precise metrics and analytical tools, as described later (see approach sec. for preliminary data from my lab). By combining electrophysiology and imaging-based ensemble mapping we will measure how spiking and synaptic changes (LFP) across cortical layers link to functional cellular activity landscapes.

[0069] As humans, when we wake up in the morning after a good night's sleep we feel refreshed. We also think more clearly because our memories have been re-organized, a process called memory consolidation. However how the brain integrates new memories that happened during the day with old memories, without losing the older memories remains unknown. Typically, neurons in the cortex exhibit a stereotypical slow oscillation in membrane potential during sleep. These voltage dynamics are characterized by periodic shifts from a hyperpolarized (down-state) to a more depolarized (up-state) state with distinct changes in firing properties. This slow oscillation is the fundamental cellular phenomenon that organizes other sleep rhythms such as spindles (due to thalamic bursts), which are critical for memory consolidation. Emerging evidence from human studies using EEG and ECOG recordings suggest that both slow-wave oscillations as well spindles are indeed travelling waves. How these two distinct oscillations nest within one another and shape local circuit activity during sleep is not well known but critical for our understanding of consolidation, dysfunctions of which are implicated several neurological and neurodegenerative diseases. Linking cellular-scale ensemble dynamics to large-scale oscillatory patterns has never been accomplished, primarily due to limitations of performing large-scale electrophysiology and two-photon imaging simultaneously during sleep. Here, weAttorney Docket No.: PURD-151 / 01WO 28593 / 707

[0070] PATENT APPLICATION

[0071] will use our innovation which enables both large-scale 3D electrophysiology and two-photon imaging in freely moving mice to unravel the circuit features that help consolidate a learned sensory association.

[0072] We rationalize that different TW patterns are associated with unique ensemble configurations which suggest that TWs shape and recruit cellular assemblies as they traverse the region of interest (See FIG. 2We hypothesize that TWs emerge as a result of inter-region axonal fiber delays and their propagation is assisted by sparse cellular ensembles following a log-normal connectivity distribution. The built-in-time delays enable precise timing across long-distances to enable spike-timing-dependent-plasticity amidst noise. We surmise that task, movement, and context-relevant ensembles observed during the awake freely-moving state are strengthened during sleep, through dynamical shifts in excitation-inhibition balance across the somatodendritic axis which promotes plasticity and improves consolidation.

[0073] FIG. 1 compares the current prior art (left panel) to the claimed solution (right panel), and illustrates circuit dynamics during sensory processing and sleep in freely moving mice by developing the NET-2P for 3D electro-optic mapping of cortical activity. While single-unit activity, local-field-potentials (LFPs) (both stationary and travelling waves) have been measured (via ECoG, Utah arrays, voltage imaging, and EEG) across cortical regions in rodents, nonhuman-primates, and humans, the recordings suffer from poor spatial resolution to steadily map core circuit mechanisms. For example, voltage-sensitive dyes exhibit excellent temporal resolution and reflect sub-threshold signal dynamics, but lack cellular-level specificity. Further, owing to 1 -photon rather than two-photon sensitivity, these recordings suffer from scattering-induced limitations in vivo. In contrast, Utah microarrays with a pitch of -400 pm between electrodes register 0.7 units / electrode owing to their spatial scale, and are therefore not amenable for linking microcircuit spatio-temporal patterns to the mesoscale wave patterns. Here, the electrodes must be spaced by hundreds of microns to reduce tissues damage. Finally, traditional ECoG arrays due to their large electrode pitch, while sensitive to LFPs across large cortical sections, are not conducive to linking such dynamics to cellular scale patterns of activity. Two major reasons exist for this. First, is the incompatibility with modern-day imaging modalities. Second, the large electrode areas are not conducive for single-unit recordings, and render deeplayer circuit interrogation infeasible. Here, through nanofabrication and custom-built optics, weAttorney Docket No.: PURD-151 / 01WO 28593 / 707

[0074] PATENT APPLICATION

[0075] introduce a 3D Nano-Electrode Technology Integrated Two-photon (2P) Platform (NET-2P) for high-resolution spatio-temporal recording of neural dynamics in freely moving mice.

[0076] Our innovation (FIG. 1, right column) comprises of a high-density, transparent, and flexible 3D microelectrode array that conforms to the surface of the brain and facilitates simultaneous single-unit electrophysiology and LFP measurements whilst concomitantly allowing for two-photon imaging. We term this probe the NanoNeedle Net. The impact of this technology is multi-fold as it can help link both single-unit and LFP activity as well as their travelling patterns to both local and global microcircuit dynamics, can be applied across multiple species and brain regions with multi-depth capabilities, and the minimally invasive nature potentially allows for translation to human studies as the materials used are similar. The novel fabrication technique overcomes major limitations in creating high-aspect ratio nanoelectrodes and is the only platform that allows for multi-depth electrode fabrication. We further integrate this needle chip with custom-designed CMOS electronics designed in my lab. This development will fuel innovations at many levels: the design and fabrication of probes, integration with active electronics, implantable interfaces in freely-behaving animals, and development of computational and analysis infrastructure. The invention in certain aspects therefore provides transformative electrode arrays by employing readily available fabrication techniques that can be realized in any university setting. The minimally invasive nature could also allow for translation to human studies.

[0077] As part of the novel system, the NanoNeedle Net array will be integrated with a recently miniaturized two-photon laser scanning system for 3D optical sectioning in freely moving animals. The mini2P is realized using MEMs-based technology to rapidly scan across the cortex thereby allowing for fast acquisition of calcium transients with the implementation of a fixed wavelength fiber laser. To reduce pulse broadening and second-order dispersion along the fiber, we will implement prism-based compensation developed and implemented in my lab. To dissect circuit function across the columnar cortex, the addition of a micro electro-tunable lens will allow for simultaneous volumetric imaging at the sample plane. The system is coupled with a high NA water immersion objective to allow for high-resolution cellular imaging of neurons. A collimator and scan lens are selected to match the image NA of the objective and maximize transmission and excitation efficiency. The total system will weigh less than 3.5 grams. We willAttorney Docket No.: PURD-151 / 01WO 28593 / 707

[0078] PATENT APPLICATION

[0079] interleave our custom electrode array onto a cover glass to interface both the objective and penetration of the needles into the tissue.

[0080] In certain aspects we fabricate the NET-2P system in which high aspect ratio (100: 1) and multi-depth nanoneedles are integrated with CMOS amplifiers and flexible lightweight packaging and combined with a head-mounted mini two-photon microscope. In other aspects, we map the role of travelling waves, single-unit activity and cellular ensemble maps during a texture discrimination task in freely-behaving mice using the NET-2P system. We seek to unravel the spatio-temporal dynamics underlying reverberatory motor feedback in the barrel cortex. In still other aspects, we unravel the spatio-temporal dynamics underlying memory consolidation during sleep. Mice trained on texture discrimination will be measured during sleep with the NET-2P to unravel the relationship between travelling waves and sleep ensembles.

[0081] The NanoNeedle Net is realized using novel 3D nanoscale metal printing based on a CERES printer for ultra-high-aspect-ratio nanoelectrodes on a flexible transparent substrate - a process pioneered by my lab. The flexible substrate comprises of a transparent thin-film ECoG electrode (thin-film nanoporous gold) and printing is achieved directly on the electrode sites using our nanoscale electrodeposition system. As preliminary evidence, we have demonstrated 3D metal printing for our proof-of-concept needles on a parylene-based neural probe (FIG. 3 panels A-B). We were able to achieve 100pm tall needles with a tip diameter of less than 1 m (FIG. 3 panel C), 200pm-tall needles with a tip diameter of <5pm, and 1000 um-tall needles with diameters < 10pm. These gold needles are robust to physical manipulation. We can mechanically bend them and they maintain their shape, suggesting they will easily penetrate brain tissue following a durotomy (FIG. 3 panels D-E). We further show that these needle electrodes are indeed capable of reporting single unit activity as well as LFP with high SNR. The intended footprint will also allow us to use compact, easy-to-use packaging that is compatible with both off-the-shelf and custom-designed ICs

[0082] Also provided by aspects of the invention is a scalable multi-channel CMOS amplifier array for high-fidelity recording from the Nano-Needle array (FIG. 4 panel A). At present no commercial amplifier records from 256 channels at once. Our amplifier array will comprise of 256 amplifiers which are multiplexed in a 16 x 16 configuration. As a proof of concept, we designed a chopper-stabilized amplifier to improve the noise performance at very low frequencies (FIG. 4 panel B) (0.1 - 4 Hz) which is a pre-requisite for high fidelity recordingsAttorney Docket No.: PURD-151 / 01WO 28593 / 707

[0083] PATENT APPLICATION

[0084] during slow-wave sleep. The integrated input referred noise for the amplifier is 1.4 pVRMS. The amplified signal will be sampled using 16-bit ADC, sampling at 320 kHz such that each channel is sampled at 20 kHz. The sampled data is transferred out of the head-stage via a tether using a serial interface. The IC will also contain a power management and clock generation circuit for operation without any off-chip components. The IC will be flip chip bonded to the parylene probe using Anisotropic conductive film (ACF) to ensure a small form factor. The serial data from the tether will be interfaced with a PC using an FPGA.

[0085] These studies have shown that a lightweight head-mounted 2P system is indeed capable of fatigue-free exploratory behavior without impeding quality or stability of the imaging. The approach herein includes (1) a hollow-core photonic-crystal fiber to deliver 920-nm excitation, (2) fast MEMS scanner for fast point scanning, and (3) a flexible fiber bundle to collect the fluorescence. To overcome temporal dispersion in the excitation fiber, we will use on-table prechirp compensation (a feature already in place in my lab with conventional 2P) to ensure temporally precise beams at the sample plane within a weight budget of 3 grams. The design promises a FOV of over 400 x 400 pm2, and 180 pm z-scanning range. For z-scanning a micro tunable lens (pTlens) will be used weighing only 0.06 g; spanning a volume of 4.5 x 4.5 x 2.2 mm3; and a response time of less than 0.4 ms. The pTlens could be mounted near the MEMS scanner without relay optics. The laser light is fed in through a “thin” connection cable assembly comprising of a 6-core electric wire, and a 0.7-mm-diameter fiber bundle of 100 to 2000.05-mm-diameter tapered glass fibers. The reflection index and the curvature of the tapered glass rod are designed to ensure strong focusing. If for some reason this excitation assembly fails, we will use a GRIN lens with a planoconvex lens. Light collection will also be done through fiber-coupled PMTs ensuring light-weight. The Needle Net will be glued to a glass coverslip or cannula and held in place by the base plate that attaches to the head-mounted mini2P. The flex PCB with readout electronics will weigh less than 0.6 grams.

[0086] The primary somatosensory cortex (SI) receives somatotopically aligned inputs upon whisker touch. The sequence of whisker evoked activity consists of at least three parts: (i) thalamocortical excitation of SI, (ii) SI — Ml excitation and higher order motor (M2) dependent excitation, and (iii) M2 and higher-order thalamus dependent late secondary excitation of S 1. Such reverberant, recurrent excitatory activity between the frontal and sensory cortex, long outlasting the sensory stimulus, might be a fundamental mechanism underlying the activeAttorney Docket No.: PURD-151 / 01WO 28593 / 707

[0087] PATENT APPLICATION

[0088] construction of context-dependent sensory percepts - a form of top-down attention al control 19. Indeed, in preliminary experiments performed in my lab using planar ECoG recordings in head-fixed mice (FIG. 5 panels A-C), a late secondary depolarization in the form of a travelling wave (TW) in primary somatosensory barrel cortex was observed upon whisker touch (FIG. 5 panel D i, ii). We show through pharmacological controls that the late TW response reflects a motor feedback mechanism (FIG. 5 panel E). This signal correlates and causally contributes to subjective sensory percepts as evidenced by a diminished late reverberatory signal in a go-no-go whisker discrimination task (FIG. 5 panel D, iii). Using two-photon calcium imaging through the grids we observed that the late reverberatory travelling waves recruit sparse cortical ensembles in layer 2 / 3.

[0089] Head-rotation and volitional control has been shown to increase activity in M2. Also, M2 ensembles can discriminate specific motor actions associated with navigation and informs the network what is behaviorally relevant independent of the relative saliency of external events. This would suggest an increased drive to S 1 during a freely moving state as opposed to a head-fixed condition. Hence, using the NET-2P we will investigate how movement-related feedback enhances traveling wave dynamics and emergent ensembles using a two-alternative forced choice task (FIG. 4 panel A) under texture discrimination. Mice will learn to discriminate between a smooth and rough surface texture while traveling along a track using a single whisker. Upon touching a rough texture, the mouse must turn along the corridor towards the reward port. If a smooth texture is presented, the mouse must rerun the maze without receiving a reward. Mouse movement will be tracked using infrared cameras. We hypothesize that in freely moving animals, TWs can propagate across local networks in the cortex with only a few neurons spiking as the TW goes past, and this sparse code could be dependent on an enhanced motor drive. This means that the TW can drive subsets of neurons without necessarily driving high correlations in spiking activity that can impair sensory coding in SI. Using the methods outlined below, we will map texture-evoked TWs using needle electrodes across layers, including spontaneously emerging travelling waves, and link it to ensemble representation within a single barrel controlled by movement-related signaling from motor and higher-motor areas. Superficial circuit control by deep-layer neurons is not well known. Circuit mapping studies suggest that SI L2 / 3 receive strong inputs from L5A via barrel septa. Since L5 neurons receive both PoM and Ml and M2 inputs, it is conceivable that they play a critical role in modulating TW dynamics which weAttorney Docket No.: PURD-151 / 01WO 28593 / 707

[0090] PATENT APPLICATION

[0091] observe across superficial layers. Our recordings will shed light on how this circuit enables TW propagation in a freely-moving setting for the first time.

[0092] In humans, electroencephalogram recordings of early visual areas also show that late potentials correlate with consciously perceived stimuli. Accordingkly, there is an importance of late secondary potentials, resulting from reciprocal interactions between higher-order cortex and primary sensory cortex, for various forms of sensory perception in species ranging from mice to humans.

[0093] Before surgically implementing NET-2P we will map the barrel of the discrimination whisker using intrinsic imaging. In control experiments and to corroborate Needle probe recordings in a separate cohort of mice to ascertain ground-truth, we will fashion our design with through-holes for silicon probe insertion, allowing us to map the LFP, CSD, and single-units across cortical depth while recording 3D potential patterns using the Needles. Layer Specific Spike Sorting: Single unit activity and spike rates will be extracted using Kilosort23. Normalized electrode depth will be calculated based on electrode mapping and the most superficial trace of LFP. Spectrogram: Local field potentials (LFP) will be extracted using a linear regression filter. Spearman correlation will be calculated across all channels, LFP above 90% similarity will be considered for analysis. From here, behavior-triggered wavelet spectrograms will be constructed using a 2-second sliding window before and after locomotion onset. Spike-field coherence and LFP phase synchronization: Spike-field coherence of LFP will be analyzed using a combination of custom coding routines and Chronux, an open-source MATLAB package for analyzing spikefield coherence. Peristimulus LFP and single units will be extracted using the behavior-triggered window described above. Phase synchronization across trials will be derived using inter-trial phase clustering. With this method, we will assay the change in layer-specific phase synchronization. Given the possibility of a delicate E-I balance that shapes the interplay between cortical layers, we will expand our analysis to include inhibitory single units. Spontaneous travelling wave detection: the reverberating travelling wave will be detected based on its LFP spectral characteristics. Events are determined if their power is 3 SD above mean LFP power, event profile is matched to beta-theta timescales. Statistics and Power analysis: Statistical tests will be conducted using one-way ANOVA tests. Bonferroni tests will be conducted to check for Type-1 error. P-values < 0.05 will be considered significant. Graphical data will be presented as mean with SEM and SD boxplots extending to the 25th and 75th percentile. Data points outsideAttorney Docket No.: PURD-151 / 01WO 28593 / 707

[0094] PATENT APPLICATION

[0095] this range will be plotted as outliers. Power analysis of each group will be conducted using measured data population mean, effect size with 20% deviation, and a power of 0.8. We will aim for n = 10 mice per experiment, totaling a maximum of 30 mice. Mice of both sexes will be used in equal proportion.

[0096] Cell bodies and dendrites will be extracted based on constrained non-negative-matrix-factorization (CNMF). CNMF identifies spatiotemporal components based on pixels of high covariance around defined soma-dendritic centroids while accounting for background fluorescence and minimizing noise. To identify ensembles, we will first convert deconvolved calcium fluorescence into binarized raster plots of activity. We will segment the binarized raster based on periods of significant coactive activity (>95%) against a shuffled version of the raster plot. We will then measure population activity as multi-dimensional vectors corresponding to the frame-frame degree of similar temporal activity across populations of neurons, a metric defined by taking the cosine of each vector column. To ensure statistical robustness, vector columns will be circularly shuffled 1000 times with only the vector columns above 90% considered as ensemble candidates. We will search this multidimensional vector space to precisely define ensembles, using dimensional reduction-based methods such as PCA-SVD. Ensembles will be compared across brain states to ascertain their spatio-temporal structure.

[0097] Predictive coding is dependent on a comparison between higher-order feedback and feedforward thalamic sweeps. Here, higher-order inputs to SI from M2 will be enhanced due to head movement and volitional control. This will essentially alter the cortical state by ensuring a noisier cortex. A noisier environment counter-intuitively will enhance spike coding and integrative properties of neurons in the cortex.

[0098] In the case of insufficient or restricted FOV during imaging, a separate MEMs scanner with a larger optical degree angle can be introduced. Multi-plane volumetric imaging substantially reduces acquisition speed per optical plane. If this poses an issue in detecting robust calcium rise time, we will down sample the y-axis scanning direction and resample post-hoc using a linear interpolation. Although we have not encountered light-induced artifacts in our recordings, if we do observe unwanted light artifacts from the Needle Array, laser blanking and gold patteming / thinning can be introduced. Here, by imprinting sub -wavelength holes in the electrodes, I can exploit an effect known as extraordinary transmission of light (EOT) due to surface plasmons which renders metals transparent.Attorney Docket No.: PURD-151 / 01WO 28593 / 707

[0099] PATENT APPLICATION

[0100] Sleep is a brain state wherein system consolidation builds on synaptic weights and is based on a process in which reverberating activity of nascent representations are redistributed across neural circuitry for long-term storage. Sleep in mammals consists of two distinct states: slow- wave (SWS) and REM sleep. The most prominent SWS dynamics are the neocortical slow oscillations (0.1-4 Hz) and thalamocortical spindles (10-15 Hz). The second phase of REM sleep (15-25 Hz) reflects bursts of depolarized cortical activity26. Evidence of memory consolidation has been shown to occur in both SWS and REM sleep states. The mechanisms underlying memory consolidation have been poorly mapped primarily due to methodological constraints !) to induce sleep states in mice, anesthesia and sleep-deprivations have been used, resulting in several confounding results. 2) Mice must undergo extensive habituation to naturally induce sleep under head-fixation. 3) Cellular-level mapping of neural activity during sleep has been limited to either single-photon imaging in a freely moving setting, or 2-photon imaging under the head-constraint conditions. There is an unmet need to map cellular level activity concomitantly with the prominent oscillations associated with SWS and REM sleep states. Importantly, unravelling the circuit mechanisms through which mammals encode nascent experiences during sleep will yield a better understanding of dysfunctions in sleep states and memory formation. Here, we will map the oscillatory and cellular ensemble dynamics of SW S and REM states after mice have undergone the two-alternative forced choice task using the NET-2P (FIGS. 6-7). Under a freely moving condition, we will be able to unveil the underpinning of memory consolidation during a complex behavioral task.

[0101] Mice will be contained in a small arena after the two-alternative force choice task where they will undergo bouts of SWS and REM sleep monitored through fluctuations in LFP and behavior under infrared cameras. We will detect sleep spindles and slow oscillation events in a three-dimensional manner through the Nanoneedle Array. To detect discrete slow oscillation events, LFP signal will be bandpass filtered from 0.1 to 4.5 Hz and all positive-negative zero crossings of the filtered signal will be marked. Intervals between 0.4 and 2s will be extracted. Waveform power greater than 80% of average power will be considered as event of slow oscillations. To detect discrete spindles, LFP signal will be bandpass filtered between 7-15 Hz and the z-scored power will be extracted. Periods in which the power exceeds 95% of the normalized power will be considered as candidate spindle events. Spine event envelopes between 0.5 and 3 s will be considered for analysis. All spindles will be time-locked to the peak of spindleAttorney Docket No.: PURD-151 / 01WO 28593 / 707

[0102] PATENT APPLICATION

[0103] events and calcium signals will be analyzed at separate discretize windows equal to twice the duration of a single spindle event, generating pre- post-spindle calcium responses. This same method will be used for analyzing dynamics of slow oscillation events.

[0104] To map ensembles, two approaches will be considered. Emerging ensembles during SWS and REM, and ensembles during slow oscillation events and spindles. As described in Objective 2, separate ensembles will be mapped to each brain state and shared neurons across brain states will be noted and pooled as potential “joint ensemble cells”. Importantly, these cells will also be compared to the emerging ensemble during the awake state. Together, a “learning ensemble” will emerge whereby a distinct set of neurons are reactivated during a memory consolidation event. The spatio-temporal precision of this ensemble will be quantified across training sessions as the mouse begins the learn the task as well as across mice to ensure statistical robustness. Using this unique analysis coupled with 3D mapping of sleep-dependent TW events, we will ascertain how distinct oscillations across the cortical depth shape ensemble dynamics as a function of brain state. We hypothesize that excitatory neuron activity within the learning ensemble would be enhanced during sleep than awake. We posit that this is due to a shift in excitation-inhibition balance locally, which promotes plasticity as opposed to precision / rate coding during sleep. As part of this study we will compare the activity maps from Parvalbumin and Somatostatin positive interneurons (via ere lines and GCaMP6s / 8m) to elucidate this shift.

[0105] Disruptions of memory consolidation mechanisms during sleep are linked to many psychiatric and neurodegenerative disorders. Emerging evidence suggests that such disruptions are not a secondary symptom but rather the contribution of the disease state. Unravelling circuit mechanisms that emerge during awake and sleep states will allow a greater understanding and treatment of memory related disorders.

[0106] Incorporation by Reference

[0107] References and citations to other documents, such as patents, patent applications, patent publications, journals, books, papers, web contents, have been made throughout this disclosure, including to the Supplementary. The Supplementary, and all other such documents are hereby incorporated herein by reference in their entirety for all purposes.

[0108] EquivalentsAttorney Docket No.: PURD-151 / 01WO 28593 / 707

[0109] PATENT APPLICATION

[0110] The invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. The foregoing embodiments are therefore to be considered in all respects illustrative rather than limiting on the invention described herein.

[0111] EXAMPLES

[0112] High-density electrophysiology recordings in 3 dimensions combined with 2-photon calcium imaging are a high-fidelity neuroscience technique that will enable understanding of spatial and temporal coding and computational strategies employed by the brain during complex tasks. These extracellular electrophysiological recording have been conventionally performed using linear silicon probes (Michigan Probes) or silicon penetrating microelectrode array (Utah array). These methods are limited spatially in either 1 or 2 dimensions and are incompatible with 2-photon imaging as they are rigid and opaque. Combining electrophysiological recording with optical techniques like 2P imaging will help overcome the limitations of each other (low spatial resolution of EPhys and low temporal resolution of 2P imaging).

[0113] Here, we have developed a transparent and flexible 3D implant capable of simultaneous electrophysiological / electrochemical recording with optical measurements such as 2-photon calcium imaging. Using novel microscale 3D metal printing, we have created customizable high-aspect-ratio needle electrodes on a flexible substrate for multi -depth recordings.

[0114] Example 1 : Device description

[0115] We have developed a robust fabrication process for ultra-high-aspect-ratio needles on a soft and transparent neural probe. Our proprietary fabrication involves the innovative process of directly printing metal nano-needles onto the pads of the planar neural probe using the Exaddon CERES Printer. CERES is a 3D printer that can print pure metal structures at nanoscale with submicron pitch accuracy. Additionally, the CERES system can make patterns using liquids and nanoparticles of different materials (copper, gold and platinum). The 3D printing is based on electrochemical precise deposition of metal ions via reduction in a three-electrode electrochemical cell. The fundamental building block of the 3D structure is called a “voxel” which can have volume from 0.6um3to 80um3. The voxel is created by precise dispensing of liquid ink (in femtoliters) containing metal ions through a custom AFM tip with a microchannel. The dispensed metals ions when in contact with the working electrode get reduced to solid metal atoms thus forming a voxel. The voxel size is determined by the amount of liquid ink that isAttorney Docket No.: PURD-151 / 01WO 28593 / 707

[0116] PATENT APPLICATION

[0117] dispensed which is decided by the pressure in the microchannel, which can be varied dynamically throughout the printing process.

[0118] The printer is capable of printing complex structures like 3D spirals as well as cantilevers. Here a stack of voxels is printed on a conducting substrate with varying pressures, to create a high aspect ratio tapering needle electrode with tip dimensions less than lum. This process is repeated to create an array of these electrodes. We have demonstrated fabrication of an array of 32 and 64 high aspect ratio needle like structures with the tip diameter less than 1 pm (FIG. 8 panel A reference identifiers 1-2). We can print needles with varying heights from 250 pm to ultra-tall 1 mm needles (FIG. 8 panel A reference identifier 3). FIG. 8 panel B shows a finished probe with alternating 100 pm and 250 pm needles.

[0119] Example 2: Fabrication Process

[0120] Fabrication of printing substrates (Grids)

[0121] The needle electrodes could be fabricated on various substrates as long as the printing area is conducting (Gold, Platinum, ITO and PEDOT) and has low resistance electrical contact to the working electrode of the printing chamber. The printing substrate is submerged in a printing chamber solution (either acidic or basic solution depending on the printing ink).

[0122] Flexible and transparent substrates (Grids)

[0123] The flexible and transparent probes are built on a silicon wafer using conventional photolithography approach. The flexible probes have metal traces and electrodes that are sandwiched between two layers of transparent and flexible polymer (like parylene C, Polyamide, etc). The electrode sites are then opened for printing, by selective RF plasma etching on the electrode sites using a photoresist mask (AZ10XT). See FIG. 9.

[0124] To increase the surface area and decrease the electrode impedance, the electrode surface is textured with nanoporous gold. The electrode surface could also be electroplated with Platinum nano particles (PtNP) or PEDOT:PSS for decreasing the impedance. FIG. 10 shows nanoporous gold electrodes for high fidelity recording. The recording electrode have a nanoporous gold textured to increase the surface area which in turn reduces the electrode impedances. The impedance plots are shown in the right.Attorney Docket No.: PURD-151 / 01WO 28593 / 707

[0125] PATENT APPLICATION

[0126] Printed circuit boards (flex and rigid)

[0127] The needle electrode could be fabricated directly on a printed circuit board (either rigid or flex) for easy and rapid manufacturing of electrode arrays for both in-vivo and in-vitro applications.

[0128] Rigid substrates (glass cannula with metal electrodes)

[0129] The needles could also be printed directly on glass coverslips with patterned electrodes for use in optical recordings. The needle electrode array on glass could be used in recording electrical signals as well as performing optical measurements on a cell culture or 3D organoids.

[0130] FIG. 11 shows 3D electrodes on glass canula. Metal electrodes are patterned directly on the glass canula and needle electrodes could be printed on these metal electrodes.

[0131] CMOS chips

[0132] The needle electrodes can be directly fabricated on the bonding pads on a CMOS IC with amplifiers. Direct printing on the CMOS IC will ensure low parasitic capacitances for high SNR recording as well as compact packaging for easy use in chronic freely moving applications. The CMOS IC can have an amplifier array, stimulation circuit as well as digital processor for online spike analysis. The entire package can be implanted subdurally and can communicate with other devices wirelessly. See FIG. 12.

[0133] Example 3: Fabrication of Needles on Grids

[0134] FIG. 13 shows a fabrication process for encapsulation and tip exposure of the needle electrodes. The printing substrate is cleaned using O2 plasma (O2 - 40 seem, Ar - 2, Power -100W for 30 sec) for better adhesion of the needles and getting rid of particles. The printing target on the substrate should be conducting (metals like gold, Pt, ITO and polymers like PEDOT) and shorted to the working electrode of the printing chamber. The substrate is submerged in an acidic chamber solution. The working electrode voltage is set to the reduction potential of the metal ion in the printing ink and the needles are printed one by one. For taller needles, the needles are printed in sections to avoid collision of the printing tip with the printed structures. The needles have a wider base with diameter of 7-10um, with tapering starting fromAttorney Docket No.: PURD-151 / 01WO 28593 / 707

[0135] PATENT APPLICATION

[0136] 50um above the base. Using COMSOL mechanical modelling, the geometry of the needle is optimized to have critical load factor greater than 3, to prevent buckling during insertion into the brain tissue. The geometry of the needle is also optimized to prevent bending due to lateral forces that might be experienced during handling and next processing steps. After printing the substate is rinsed with DI water to wash away all the printing solution and dried carefully with N2.

[0137] FIG. 14 shows mechanical modelling for determining optimal geometry of the needles for damage freee insertion into the brain.

[0138] Example 4: Encapsulation and tip exposure

[0139] FIG. 15 shows SEMs for tip exposure using RTE etching. The figure shows the array with photo resist to protect the paryelen on the base and zoomed in exposed tip. The needle is then encapsulated with 800nm-1000nm of biocompatible Parylene-C polymer. For localizing the recording area of individual needle, only the tip of the needle is exposed. The tip can be selectively exposed using 3 different methods:

[0140] a. Focused Ion Beam Etching - Using an Focused ion beam of Ar ions, the tip of each individual needle is etched and then capped with platinum.

[0141] b. Femtosecond laser ablation - The substrate is submerged in water, and a femtosecond laser is focused on the tip of each individual needle to etch away the parylene. The power of the laser is set at 340 mW and the beam is focused on the tip for 30 seconds.

[0142] c. Anisotropic O2 RIE - A photoresist (AZ 1518) is spin coated on the substrate to protect the probe and the parylene coating at the base of the needles. Using O2 RIE (O2 - 30 seem, Ar - 2, Power - 150W for 8 min), the parylene encapsulation at the tip is selectively etched.

[0143] Example 5: Functionalization of the needle tip

[0144] The needles with the encapsulation and exposed tip have impedances in range of 2-5 MQAttorney Docket No.: PURD-151 / 01WO 28593 / 707

[0145] PATENT APPLICATION

[0146] due to low surface area that is exposed. This range of impedance is not suitable for high-fidelity recording of biopotentials (LFP and spiking activity). To decrease the impedance, we electroplate either platinum nanoparticles or PEDOT:PSS. After electroplating the impedance of the needles are in the range of 300-500 k£l which allows for high SNR recordings of biopotentials. For electrochemical measurements, the tip can be functionalized with polymer of PEDOT:CNT (carbon nanotubes).

[0147] Example 6: Packaging of the probe

[0148] For needles printed on a flexible substrate, the flexible probe with the needles is released from the silicon wafer while submerged under water. The base of the flexible probe is then attached to a PEEK base for ease of handling and integration with the interface PCB. The interface PCB connects the probe to the amplifier array. The interface between the PCB and the probe is done using a zero-insertion force (ZIF) connector. The probe can be directly wire bonded to a rigid or flex PCB for more compact packaging.

[0149] Example 7: Combining electrophysiological recording with optical measurements

[0150] For experiments combining electrophysiological recording with optical measurements, the probe head is glued to a glass coverslip using an UV curable optical glue. For chronic experiments, the interface PCB is attached to a titanium headplate using a custom 3D printed platform and implanted on the skull of the animal (See FIG. 16 panels A-C). For reducing the footprint, size and the weight the probe can be directly wire bonded to a flex PCB with amplifier and a wireless module.

[0151] Example 8: In-vivo electrophysiological recordings

[0152] For in-vivo recording, the recording end of the probe with the needles is attached to a circular cover glass using a UV curable optical glue. A cranial window roughly the shape of the coverslip is made on the target region. Duratomy is performed, as the tough dura mater can cause damage to the needles as well as the brain tissue during insertion. Using vacuum, a pipette tip is attached to the coverslip to guide the needles on top of the cranial window. Then using a manipulator, the needles are inserted into the brain at speed of lum per second (FIG. 17 panel A). Because of the scale of the needles, there is no buckling and bleeding. After the insertion, theAttorney Docket No.: PURD-151 / 01WO 28593 / 707

[0153] PATENT APPLICATION

[0154] probe is allowed to stabilize for 30 mins and then the recording is started. Using a 32-channel probe with 250 pm needles we were able to successfully record touch-evoked LFP and multiunit spiking activity (MU A) in the barrel cortex in an awake mouse. FIG. 17 panel C shows the MUA on the channels. Distinct spikes on channels were separated by clustering in the PCA space (FIG. 17 panel D). FIG. 17 panel E shows the trial average LFP response following whisker stimulation. FIG. 17 panel F shows one of the wide band travelling waves that was detected simultaneously from the probe. The needle array can be recovered after acute implantation, facilitating reuse across multiple experiments (FIG. 17 panel B).

[0155] Example 9: Multimodal Recording

[0156] The developed 2D and 3D platforms could be used simultaneously with commercially available neural recording technologies like silicon probe, 2-P calcium imaging and wide field imaging.

[0157] FIG. 18 shows simultaneous surface ECoG and translaminar electrophysiology.

[0158] FIG. 19 shows simultaneous surface ECoG and calcium imaging.

[0159] FIG. 20 shows simultaneous translaminar electrophysiology and calcium imaging.

[0160] FIG. 21 shows 3D electrophysiological recording with simultaneous calcium imaging. FIG. 22 shows highly compact and integrated neural implant with direct printing of electrodes on conventional CMOS amplifier arrays.

[0161] FIG. 23 shows electrochemical recording for detecting electrochemically active molecules like neurotransmitters.

[0162] Example 10: Freely moving simultaneous EPhys and 2-photon imaging

[0163] FIG. 24 shows integration of developed 3D needle array with open source mini-2p system for performing Ephys along with calcium imaigng in freely moving mice. The flexible and transparent nature of the developed 3D electrodes allows for integration with open-source microscopy platform like mini-2p. This allows for performing simultaneous EPhys and imaging of neural activity in freely moving rodents.

[0164] Example 11 : Recording and stimulation of 3D organoids and cell cultures

[0165] FIG. 25 shows a schematic of platform for simultaneous recording and stimulation ofAttorney Docket No.: PURD-151 / 01WO 28593 / 707

[0166] PATENT APPLICATION

[0167] neurons in a cell culture or 3D organoid. The technique could also be combined with calcium imaging.

[0168] Example 12: Recording and stimulation of spinal cord and vagus nerve

[0169] FIG. 26 shows 3D electrodes on flexible self-curling polymer substrates for creating cuff electrode array for recording and stimulation of vagus nerve or spinal cord. The needle electrodes could be printed on flexible self-curling polymer substrate to create cuff electrode to record from vagus nerve or spinal cord.

Claims

Attorney Docket No.: PURD-151 / 01WO 28593 / 707PATENT APPLICATIONWhat is claimed is:

1. A system for making high-density electrophysiology in-vivo recordings, the system comprising: a needle array integrated into an optical instrument.

2. The system of claim 1, wherein the needle array comprising an optically transparent substrate; and a plurality of needles operably coupled to the optically transparent substrate;an optical system, wherein the optical system is arranged with respect to the needle array such that an optical output from the optical signal passes through the optically transparent substrate of the needle array; anda processor operably associated with each of the needle array and the optical system to receive electrical data from the needle array and optical data from the optical system and process both the electrical data and the optical data.

3. The system of claim 2, wherein at least two of the needles of the plurality of needles are of different heights.

4. The system of claim 2, wherein each needle of the pluralirty of needles has a base and a tip, wherein the base is wider than the tip and each needle comprises a taper from the base to the tip.

5. The system of claim 2, wherein the plurality of needles are printed onto the optically transparent substrate.6 The system of claim 2, wherein each needle of the pluralirty of needles is comprised of a metal that is coated with a biocompatiable polymer, wherein the tip remains unencapsulated.

7. A method for making electrophysiology in-vivo recordings, the method comprising:providing a system that comprises: a needle array integrated into an optical instrument, and a processor operably associated with each of the needle array and the optical system to receive electrical data from the needle array and optical data from the optical system and process both the electrical data and the optical data;Attorney Docket No.: PURD-151 / 01WO 28593 / 707PATENT APPLICATIONinserting the needle array into in-vivo tissue;directing directing an optical output from the optical system onto the in vivo tissue; and receiving to the processor electrical data from the needle array and optical data from the optical system, wherein the processor processes both the electrical data and the optical data.

8. The method claim 7, wherein at least two of the needles of the plurality of needles are of different heights.

9. The method claim 7, wherein each needle of the plurality of needles has a base and a tip, wherein the base is wider than the tip and each needle comprises a taper from the base to the tip.

10. The method claim 7, wherein the plurality of needles are printed onto optically transparent substrate.

11. The method claim 7, wherein each needle of the pluralirty of needles is comprised of a metal that is coated with a biocompatiable polymer, wherein the tip remains unencapsulated.

12. A needle array comprising:a flexible and optically transparent substrate; anda plurality of needles operably coupled to the flexible and optically transparent substrate, wherein each needle of the pluralirty of needles is comprised of a metal that is coated with a biocompatiable polymer, wherein the tip remains unencapsulated.

13. The needle array of claim 12, wherein at least two of the needles of the plurality of needles are of different heights.

14. The needle array of claim 12, wherein each needle of the pluralirty of needles has a base and a tip, wherein the base is wider than the tip and each needle comprises a taper from the base to the tip.Attorney Docket No.: PURD-151 / 01WO 28593 / 707PATENT APPLICATION15. The needle array of claim 12, wherein the plurality of needles are printed onto the flexible and optically transparent substrate.

16. The needle array of claim 12, wherein the subject is at least one selected from the group consisting of: a silicon wafer; a printed circuit board; a glass coverslip; and a CMOS chip.

17. A method for manufacturing a needle array, the method comprising:providing a flexible and optically transparent substrate;directly printing metal needles onto the flexible and optically transparent substrate; and encapsulating the metal needles with a biocompatible polymer, wherein the tip remains unencapsulated.

18. The method of claim 17, wherein prior to printing, the flexible and optically transparent substrate is cleansed.

19. The method of claim 17, wherein at least one of the metal needles is printed as a single section.

20. The method of claim 17, wherein at least one of the metal needles is printed as multiple sections.

21. The method of claim 17, wherein each of the metal needles is printed with a base and a tip wherein each of the metal needles tapes from the base to the tip.

22. A method for making high-density electrophysiology in-vivo recordings, the method comprising:providing needle array comprising a flexible and optically transparent substrate; and a plurality of needles operably coupled to the flexible and optically transparent substrate, wherein each needle of the pluralirty of needles is comprised of a metal that is coated with a biocompatiable polymer, wherein the tip remains unencapsulated;inserting needle array into in-vivo tissue; andAttorney Docket No.: PURD-151 / 01WO 28593 / 707PATENT APPLICATIONreceiving electrical signals from the needle array to a processor, wherein the processor processes the electrical signals.

23. The method claim 22, wherein at least two of the needles of the plurality of needles are of different heights.

24. The method claim 22, wherein each needle of the pluralirty of needles has a base and a tip, wherein the base is wider than the tip and each needle comprises a taper from the base to the tip.

25. The method claim 22, wherein the plurality of needles are printed onto the flexible and optically transparent substrate.

26. The method claim 22, wherein the substrate is at least one selected from the group consisting of: a silicon wafer; a printed circuit board; a glass coverslip; and a CMOS chip.

Images