Abstract
Understanding human brain function requires tools capable of capturing the interplay between electrical activity and neurochemical signaling in real time. Current approaches measure electrophysiology or neurotransmitter dynamics in isolation and lack the spatiotemporal resolution needed for intraoperative human studies. We developed the Multifunctional Apparatus for Voltammetry, Electrophysiology, and Neuromodulation (MAVEN), a compact, battery-powered platform, to overcome these limitations and enable concurrent electrophysiological recordings, phasic and tonic neurochemical sensing, and programmable neurostimulation delivery. Herein, we present the preclinical validation of MAVEN in small- and large-animal models using addictive substances to elicit robust changes in dopamine and serotonin dynamics while also delivering programmable neurostimulation. Across models, MAVEN minimized stimulation artifacts and provided stable multimodal readouts, establishing feasibility and safety for real-time monitoring. Designed for both basic and clinical neuroscience, MAVEN enables investigations into DBS mechanisms, the neurochemical basis of neuropsychiatric disorders, and the potential role of neurotransmitters as actionable biomarkers for closed-loop neuromodulation. This has immediate relevance in the era of personalized neuromodulation and overcomes prior technical limitations that precluded direct investigation of human disease.
Keywords: neurochemical sensing, electrophysiology, deep brain stimulation, voltammetry, electrical neuromodulation, pharmacological neuromodulation, intraoperative monitoring
1. Introduction
Deep brain stimulation (DBS) is an established neuromodulation technique that delivers targeted electrical stimulation to precise subcortical brain regions, restoring function in otherwise dysfunctional neural circuits implicated in various neurologic and psychiatric disorders (Oesterle et al., 2025; Da Cunha et al., 2015). Since its initial clinical introduction in the late 1980s, DBS has been approved by the United States Food and Drug Administration for the treatment of essential tremor, Parkinson’s disease, dystonia, and epilepsy, with additional humanitarian device exemptions granted for obsessive-compulsive disorder and Tourette syndrome. Despite over three decades of clinical use and implantation in more than 244,000 patients worldwide (Sandoval-Pistorius, 2023), the exact mechanisms underlying DBS efficacy remain incompletely understood. Moreover, the pathogenic neural circuit dynamics underlying investigational indications for DBS, including treatment-resistant depression and substance use disorders, remain poorly characterized (Chan et al., 2024; Castrioto et al., 2014; Lind et al., 2007). Both normal and pathological brain functions are shaped by the coordinated interplay between electrophysiological activity and neurotransmitter (e.g., dopamine or serotonin) dynamics (Koob and Volkow, 2016; Wise, 2009; McCormick and Prince, 1987; Marder, 1984). Existing research methodologies, however, offer limited capacity to perform integrated, real-time analysis combining electrophysiological recordings and neurochemical sensing. Traditional electrophysiological techniques, such as single-unit, multi-unit, and local field potential (LFP) recordings, enable high temporal resolution critical during DBS electrode implantation but cannot provide essential neurochemical context (Blumenfeld and Brontë-Stewart, 2015; Little and Brown, 2014). Conversely, traditional neurochemical measurement techniques like microdialysis, while informative about neurotransmitter dynamics, suffer from poor temporal resolution, yielding only static snapshots rather than continuous, real-time data. Similarly, neuroimaging methods, including functional magnetic resonance imaging (fMRI) and positron emission tomography (PET), infer neural activity indirectly, lacking the temporal and spatial resolution required for intraoperative decision-making and understanding DBS mechanisms. Consequently, no existing methodology fully captures the complexity of a neural circuit, highlighting a critical need for integrated platforms capable of simultaneously measuring electrophysiological and neurochemical signals with high spatiotemporal fidelity in vivo (Li et al., 2024; Bang et al., 2020; Paek et al., 2013; Hatsopoulos and Donoghue, 2009; Lee et al., 2004).
In neuroscience, voltammetry provides a powerful approach for measuring neurochemical signals with high temporal resolution by applying varying electrical potentials to an electrode in brain tissue and recording the resulting current. Techniques such as fast-scan cyclic voltammetry (FSCV) facilitate the detection of rapid, action-potential-induced (phasic) fluctuations in neurotransmitter concentrations at millisecond resolution. Alternatively, multiple cyclic square wave voltammetry (MCSWV) enables the measurement of basal (tonic) neurotransmitter levels, which reflect slower, sustained changes crucial for maintaining the extracellular homeostatic environment (Shin et al., 2021; Oh et al., 2018; Grace, 1995). While significant strides have been made in leveraging FSCV methods intraoperatively during human DBS surgery (Batten et al., 2024; Sands et al., 2023; Bang et al., 2020; Kishida et al., 2016; Chang et al., 2012; Kishida et al., 2011), these applications have largely remained restricted to research settings and address phasic neurotransmitter events. Tonic neurotransmitter concentrations, though currently not measured during clinical DBS procedures, are essential for fully understanding neural circuit dysfunction, as they more accurately mirror the chronicity and severity of neuropsychiatric disorders (Da Cunha et al., 2012; Li et al., 2021; Lipton et al., 2019). Thus, expanding voltammetric approaches to encompass both phasic and tonic measurements within clinical workflows would represent a critical advancement toward comprehensive neural circuit characterization and potentially therapeutic optimization.
Thus, the ideal instrument for in vivo human neural circuit research during DBS surgery should: (1) enable near-simultaneous, real-time electrophysiological signal recordings alongside both phasic and tonic neurochemical measurements, (2) facilitate recordings both prior to stimulation (to characterize disease pathogenesis) and during concurrent stimulation (to elucidate DBS mechanisms), and (3) integrate seamlessly into standard surgical workflows. Previously, development of such a transformative platform has faced technical challenges stemming from diverse signal processing requirements across different modalities, electrode invasiveness and biocompatibility consideration, and ethical concerns regarding deployment in the human setting. Motivated by these challenges, we developed the Multifunctional Apparatus for Voltammetry, Electrophysiology, and Neuromodulation (MAVEN), a portable, multimodal platform engineered explicitly for integration into standard DBS surgical workflows. MAVEN enables near-simultaneous, real-time acquisition of electrophysiological and comprehensive neurochemical (both tonic and phasic) signals in vivo, alongside delivery of programmable electrical stimulation.
MAVEN was designed with human safety features with the eventual goal of translating this technology to human clinical studies intraoperatively. The overall field of neuromodulation is trending toward personalized, closed-loop implantable systems that, by definition, are adaptive to in situ biomarker feedback via real-time sensing mechanisms. As MAVEN was also designed as a versatile instrument capable of measuring electrophysiological signals concomitantly with tonic and phasic neurochemical levels in the brain, it offers the unique possibility of using neurotransmitters or electrophysiological proxies of dynamic neurotransmitter patterns as biomarkers. Consequently, such studies could clarify DBS mechanisms, enable fully personalized stimulation programs, and ultimately broaden therapeutic indications.
Herein, we present the preclinical validation of MAVEN in small (rodent) and large (swine) animal models, demonstrating its capability to measure neurochemical dynamics in response to both electrical neuromodulation and pharmacological interventions. The translational potential of MAVEN lies in its ability to facilitate unprecedented insights into neuropsychiatric disorder pathophysiology, refine DBS targeting and efficacy through neurochemical guidance, and provide a foundation for biomarker-driven, closed-loop neuromodulation therapies in the era of personalized neuromodulation.
2. Materials and Methods
Note that the in vitro characterization and validation protocols for MAVEN testing can be found in the Supplementary file.
2.1. Electrode fabrication
MAVEN is compatible with multiple sensors and electrode types, including conventional carbon-fiber microelectrodes (CFMs) and flexible laser-induced graphene (LIG) sensors. CFMs were fabricated as described previously (Oh et al., 2018; Chang et al., 2013). A single AS4 carbon fiber (diameter = 7 μm, Hexel, Stamford, CT) was embedded into polyimide-coated silica tubing (20μm inner diameter, ID; 90μm outer diameter, OD; Polymicro Technologies, Arizona, USA) and attached to a nitinol extension wire (Nitinol #1; Fort Wayne Metals, IN) using silver-based conductive paste. This was then insulated with polyimide tubing (0.0089” ID, 0.0134” OD; Vention Medical, New Hampshire, USA). The exposed carbon fiber was trimmed microscopically to a length of 50–100 μm and electropolymerized with a low-density PEDOT:Nafion coating for sensing (Vreeland et al., 2015). To prepare an Ag/AgCl reference electrode, an exposed tip of Teflon-coated silver wire (A-M systems, Inc., Sequim, WA) was chlorinated using a 9-V dry cell battery in normal saline. A soft LIG electrode sensor specifically designed for implantation in mice was fabricated as previously described (Li et al., 2022).
2.2. Data analysis
FSCV and N-shaped FSCV (N-FSCV) data were analyzed and extracted using WincsWare. For MCSWV and N-shaped MCSWV (N-MCSWV), the outputs of the CFMs were analyzed using MATLAB (MathWorks Inc., Natick, MA, USA) to generate voltammograms and color plots, as described previously (Yuen et al., 2023a; Goyal et al., 2022). For FPA and electrophysiology, all data were analyzed in MATLAB.
3. Results and Discussion
3.1. MAVEN and WincsWare software
There exists commercially available electrochemical and electrophysiological recording equipment, which we have previously reviewed (Rojas-Cabrera et al., 2020). However, the aforementioned equipment has one or more of the following restrictions: microdialysis-based approaches suffer from poor temporal resolution and tissue damage; voltammetry systems are typically limited to either phasic (FSCV) or tonic (MCSWV) neurotransmitter measurements, but not both; electrophysiological systems, such as electroencephalography (EEG) and electrocorticography (ECoG) provide broad coverage but limited spatial resolution or require invasive implants; and integrated devices like UEI (Universal Electrochemistry Instrument, Chapel Hill, NC) or Pinnacle (Pinnacle Technology Inc., Lawrence, KS) remain research tools not designed for human intraoperative use and often lacking multimodal capability, artifact mitigation, or clinical workflow integration. For this reason, MAVEN was designed to overcome the challenge of integrating multiple independent recording and stimulation platforms that currently complicate multimodal brain research. The lineage of MAVEN and its related systems is summarized in Supplementary Table 1.
MAVEN’s capabilities consist of neurochemical sensing, electrophysiology recording, and stimulation delivery, as depicted in Figure 1a. Neurochemical sensing can be performed using voltammetry and fixed potential amperometry (FPA). Voltammetry is capable of distinguishing phasic and tonic concentrations of specific neurotransmitters in near-real time: phasic measurements include FSCV and N-FSCV for DA and serotonin (5-HT), respectively, while tonic measurements include MCSWV and N-MCSWV for DA and 5-HT, respectively (Shin et al., 2021; Oh et al., 2018; Heien et al., 2004; Jackson et al., 1995).
Fig. 1. Device concept and internal architecture of the Multifunctional Apparatus for Voltammetry, Electrophysiology, and Neuromodulation (MAVEN).

(a) Schematic of MAVEN in an experimental setting of human DBS surgery, illustrating its capacity for simultaneous sensing, recording, and modulation. (b) Block diagram of the system architecture, highlighting separate acquisition and stimulator circuitry, Bluetooth transceivers, and the HOIST fiber-optic interface. (c) Mechanical rendering of the MAVEN module with its battery pack (left), alongside an exploded view of the component assemblies (right). FSCV, fast-scan cyclic voltammetry; MCSWV, multiple cyclic square-wave voltammetry; LFP, local field potential; DBS, deep brain stimulation; DAC, digital-to-analog converter; ADC, analog-to-digital converter; MUX, multiplexer; TIA, transimpedance amplifier; LED, light-emitting diode.
The MAVEN system comprises three coordinated key elements (Fig. 1b): (1) Separate acquisition and stimulation circuitry boards housed within the handheld module; (2) WincsWare software, a graphical user interface (GUI) installed on a host computer for control and data visualization; and (3) Harmoni Optical Interface and Synchronization Transceiver (HOIST), which provides high-bandwidth, fiber-optic communication and hardware synchronization with external instrumentation. The MAVEN module is divided between two splash-resistant, polymer enclosures that dock onto a rechargeable battery pack (Fig. 1c). The internal architecture of MAVEN includes several printed-circuit-board assemblies (PCBAs) dedicated to signal acquisition or electrical stimulation. The acquisition board digitalizes incoming electrochemical and electrophysiological signals using transimpedance amplifiers, analog-to-digital (ADC) and digital-to-analog (DAC) converters, and an Intan RHD2132 integrated circuit; and an on-board field-programmable gate array (FPGA) governs ramp generation, real-time control, and data handling. The system incorporates one high-speed ramp circuit for MCSWV, while four low-speed ramps handle FSCV. Additional ramps can be added modularly, allowing channel counts to scale without enlarging the device footprint. The companion stimulator board delivers fully isolated voltage- or current-controlled waveforms (50 mV-10 V or 10 μA-10 mA) and incorporates redundant fault protection to satisfy safety requirements for human use.
WincsWare connects to MAVEN either wirelessly through a Bluetooth link (Bluegiga WT41-A) or via the HOIST interface. HOIST attaches to the host computer over USB, drawing its power from the same port, and relays data to MAVEN through 1-mm core plastic optical fibers (Industrial Fiber Optics, Inc., Tempe, AZ) coupled to Broadcom/Avago transmitter and receiver modules. This configuration ensures that the MAVEN device remains electrically decoupled from any main power sources, maintaining subject safety during electrode-based data acquisition and stimulation, as shown in Fig. 1a. WincsWare includes an application programming interface (API) for MATLAB (The MathWorks Inc., Natick, MA), which allows users to implement customized closed-loop neuromodulation paradigms in which stimulation parameters can be dynamically adjusted based on neurotransmitter levels or electrophysiologic signals.
The HOIST provides three additional advantages. First, it can replace Bluetooth to provide a higher-bandwidth communication link (approximately 500 kbps) between MAVEN and WincsWare, supporting high-rate acquisition of electrochemical and electrophysiological data. Second, by receiving a transistor-transistor logic (TTL) signal from the MRI system, HOIST can pause data collection during radio-frequency pulses, allowing artifact-free recording alongside structural or functional MRI. Third, its TTL in/out lines synchronize MAVEN with other instruments (e.g., electromyography systems, alternate stimulators, hand accelerometers for tremor measurement), which streamlines multimodal data acquisition in both laboratory and operating-room settings.
The WincsWare software is a lightweight C#/.NET application developed in Microsoft Visual Studio that orchestrates all data acquisition and stimulation functions of the MAVEN system with sub-second latency. It runs efficiently on a standard Windows 10 or 11 PC. Within the GUI, users can fine-tune standard MCSWV and FSCV waveforms and charge-balanced stimulus sequences by adjusting a variety of waveform parameters, including baseline voltage, peak voltage, and scan rate, or they can import arbitrary electrochemical or stimulation waveforms from comma-separated value (CSV) files. Complex, charge-balanced stimulus trains are configured in the same manner. Time-stamped text annotations can be inserted at any point during live acquisition or playback, and data can be exported as high-resolution images or Excel-compatible files for downstream analyses.
3.2. MAVEN enables both phasic and tonic voltammetric recordings in vitro
Like its predecessors, MAVEN uses FSCV to capture transient (phasic) changes in DA concentration. During FSCV, a triangular voltage waveform is applied to a CFM, driving rapid oxidation-reduction reactions of nearby electro-active species; the resulting current is transmitted back to WincsWare for real-time visualization and post-hoc identification of analytes. Phasic 5-HT signals are obtained by substituting the conventional triangular sweep with an N-shaped waveform (N-FSCV). In a series of in vitro experiments (Supp. Fig. 1), both FSCV and N-FSCV reliably detected sub-micromolar bolus additions of DA and 5-HT, respectively. The flow-injection system (FIAlab 3200, FIAlab Instruments, Seattle, WA) consists of a switching valve with a six-port loop injector and syringe pump (Supp. Fig. 1a). A representative image of the WincsWare interface when selecting parameters for the FSCV waveform is shown in Supp. Fig. 1b–I and depicts a pyramidal waveform ranging from −0.4 to 1.3V, swept at 400 V/s with a 10 Hz scan rate. After parameters have been set, FSCV runs continuously, and WincsWare displays data in real-time as shown in Supp. Fig. 1b–II, which depicts the background-current cyclic voltammogram in the top left panel and the background-subtracted cyclic voltammogram in the bottom left panel. Each CFM was placed in a flowing stream of Tris buffer, while multiple five-second boluses of a DA (e.g., 0.5, 1.0, 1.5, 2.0, and 2.5 μM) or 5-HT (e.g., 0.1, 0.2, 0.3, 0.5, and 1.0 μM) were injected via an electronic loop injector at 2 mL/min. The current detected by the CFM was recorded for each concentration and repeated for five CFMs. Supp. Fig. 1c shows a representative pseudo-color plot following injection of 2.5 μM of DA. Supp. Fig. 1d shows the voltammograms for each DA concentration (0.5–2.5 μM) tested from a single CFM electrode, while Supp. Fig. 1e plots the averaged evoked currents across the 5 CFMs against DA concentration, with a quadratic regression applied to the data (0.5μM= 8.23 ± 5.37 nA, 1.0μM = 15.62 ± 5.53 nA, 1.5μM = 22.87 ± 5.72 nA, 2.0μM = 30.06 ± 5.98 nA, 2.5μM= 38.74 ± 9.05 nA; R2 = 0.9994, error bars represent standard deviation). The same experimental set up was used to test N-FSCV for simulated detection of evoked 5-HT release, and the corresponding pseudo-color plot, voltammograms, and calibration curve (0.1μM = 7.06 ± 3.37 nA, 0.2μM = 8.82 ± 1.48 nA, 0.3μM = 11.11 ± 2.51 nA, 0.5μM = 15.65 ± 2.34 nA, 1.0μM = 25.78 ± 2.60 nA; R2 = 0.9995, error bars represent standard deviation) are depicted in Supp. Fig. 1f–h. The N-FSCV waveform sweeps at 400 V/s from 0.2 to 1.0 V, down to −0.1 V, then back up to 0.2 V, with a 10-Hz scan rate.
The ability to record tonic extracellular levels of neurotransmitters is particularly important because, whereas phasic fluctuations mirror moment-to-moment firing, tonic levels represent longer-term neurochemical homeostasis and often correlate with disease severity. Originally, MCSWV was a breakthrough for tonic sensing; however, it relied on bulky data-acquisition hardware and provided no electrophysiological context. By integrating MCSWV into a compact platform, MAVEN resolves both shortcomings and, crucially, can deliver DBS-like stimulation that is temporally interleaved with either phasic or tonic measurements, thereby eliminating the stimulation artifacts that plagued earlier systems.
To validate tonic sensing, we incrementally added five DA concentrations (0.1–0.5 μM) to a Tris buffer bath while applying MCSWV to a submerged CFM (Supp. Fig. 2a). Each concentration was recorded for 10 min to allow signal stabilization (Supp. Fig. 2b), and the experiment was repeated with four separate electrodes. As portrayed in the WincsWare interface (Supp. Fig. 2c), the resulting data are visualized as a pseudo-color voltammogram (Supp. Fig. 2d), where the x-axis shows the staircase center potentials sweeping from −0.2 V to +0.9 V, and the y-axis displays the square-wave amplitude (ESW, ±0.4 V). A representative charge vs. time graph illustrates monotonic increases as DA concentration rises (Supp. Fig. 2e), and the pooled calibration curve exhibits a quadratic fit (R2 = 0.9775; n = 4 CFMs) across the tested range (100 nM=8.54 ± 0.35 pC, 200 nM=13.25 ± 0.37 pC, 300 nM=16.98 ± 0.26 pC, 400 nM=19.28 ± 0.37 pC, 500 nM=22.24 ± 0.53 pC; R2 = 0.9775) (Supp. Fig. 2f). An identical protocol using N-MCSWV confirmed analogous performance for tonic 5-HT detection, with the pooled calibration curve exhibiting a quadratic fit (R2 = 0.9332; n = 4 CFMs) across the tested range (10 nM=2.10 ± 0.28 pC, 20 nM=4.48 ± 0.49 pC, 50 nM=10.69 ± 1.09 pC, 100 nM=19.37 ± 1.49 pC, 200 nM=33.31 ± 3.41 pC) (Supp. Fig. 2g–i). The quadratic fit for tonic DA and 5-HT calibration curves has been well-described in the literature (Shin et al., 2021; Oh et al., 2018; Atcherley et al., 2013).
3.3. MAVEN enables artifact-free neuromodulation studies by interleaving stimulation with phasic or tonic voltammetry in vivo
In WincsWare, users can define their stimulation parameters of interest (Fig. 2a, panel I). Two stimulation modes are available: burst mode, which repeats a uniform waveform at a specified frequency; and arbitrary mode, which enables customized pulse sequences. Users can further configure monophasic or biphasic pulses by specifying frequency, train duration, and amplitude, thereby enabling simulation of human stimulation protocols. Prior to delivery of stimulation, WincsWare graphically renders the resulting waveform for preview and verification (Fig. 2a, panel II). Crucially, WincsWare enables electrical stimulation delivery and continuous phasic measurements (using FSCV) simultaneously without introducing interference. When electrical stimulation is delivered during ongoing FSCV acquisition, the software overlays the pulse train and FSCV sync signals in a shared display (Fig. 2a, panel II), helping to preserve signal integrity without requiring post hoc artifact correction. This synchronization has an assistance feature, wherein the program will alert users when the programmed stimulation parameters overlap inappropriately with FSCV waveforms and will recommend compatible parameter adjustments.
Fig. 2. Interleaved electrical stimulation with phasic or tonic voltammetry.

(a) The WincsWare stimulation interface displaying electrical stimulation parameter input options (panel I) and the stimulation waveform (red line) alongside the FSCV sync pulse (blue line) (panel II). (b) Schematic of implantation of CFM into the NAc and stimulating electrode into the VTA of an anesthetized rat. (c) Phasic study with interleaving stimulation function off: FSCV waveform and electrical train superimposed (left), the resulting FSCV pseudo-color plot (middle), and DA voltammogram (right) featuring stimulation artifacts (arrows). (d) Same experiment with interleaving on; artifacts are absent. (e) Tonic study with interleaving off: 60-min N-MCSWV trace of 5-HT (left) showing artifactual spikes during the 30-min DBS epoch; corresponding voltammograms pre-DBS (middle) and during DBS (right) reveal checkered artifacts. (f) Interleaving on: clean 5-HT trace (left) and artifact-free voltammograms pre- and during DBS (middle, right). Panel b was created using Biorender.com. CFM, carbon-fiber microelectrode; NAc, nucleus accumbens; VTA, ventral tegmental area; FSCV, fast-scan cyclic voltammetry; DA, dopamine; 5-HT, serotonin; N-MCSWV, N-shaped multiple cyclic square-wave voltammetry; DBS, deep brain stimulation.
We validated this capability in anesthetized rats fitted with a CFM in the NAc and a stimulating electrode in the VTA (Fig. 2b). For phasic measurements, a 2-second VTA pulse train evoked DA release that was monitored in the NAc with FSCV. With interleaving disabled, stimulation artifacts distorted both the pseudo-color plot and background-subtracted voltammogram (Fig. 2c). However, activating interleaving eliminated these artifacts, yielding clean redox activity and DA transients (Fig. 2d).
The same surgical preparation was then used for tonic studies. Continuous DBS-like stimulation was applied for 30 minutes while N-MCSWV tracked extracellular 5-HT. Without interleaving, large spikes contaminated the 5-HT trace and produced a checkered artifact pattern in the voltammogram (Fig. 2e). When interleaving was enabled, the artifacts disappeared and a gradual decline in tonic 5-HT, consistent with prior observations of VTA stimulation, was resolved clearly (Fig. 2f). We previously observed a comparable decline in tonic DA levels in the NAc during a similar duration of DBS applied to the VTA (Yuen, et al., 2023a). Together, these results demonstrate that MAVEN can deliver electrical neuromodulation and, in the same experiment, record artifact-free phasic or tonic neurochemical signals from a single CFM.
3.3. MAVEN performs interleaved voltammetric and electrophysiological recordings from a single electrode in vivo – nicotine in rats
MAVEN integrates multiple recording modalities (voltammetry [FSCV or MCSWV], LFP, single-unit activity) within a single channel, thereby enabling true multimodal data collection. Figure 3 demonstrates this capability using MCSWV and LFP to measure tonic DA and electrophysiological activity, respectively, in the rat nucleus accumbens (NAc) during systemic nicotine administration. Changes in tonic DA levels in the NAc in response to various pharmacological and electrical stimuli have been reported previously (Goyal et al., 2024; Yuen et al., 2023b); and nicotine, acting through presynaptic nicotinic acetylcholine receptors on dopaminergic neurons, increases neuron firing (Picciotto et al., 1998; Pidoplichko, et al., 1997; Grenhoff et al., 1986) and extracellular DA concentrations (Lucente et al., 2022; Di Chiara, 2002). In our experiment (Fig. 3a), the CFM was stereotaxically positioned in the NAc, and WincsWare was set to alternate MCSWV sweeps with LFP acquisition (Fig. 3b). Nicotine (800 μg/kg, s.c.) elevated tonic DA from 325.1 nM (Fig. 3c) to 460.0 nM (Fig. 3f). Concomitantly, LFP recordings displayed a marked increase in beta-band power after nicotine administration (Fig. 3g, h) relative to pre-nicotine traces (Fig. 3d, e), echoing prior reports (Sutton et al., 2016; Marsálek et al., 2006). These results validate MAVEN’s ability to acquire synchronized neurochemical and electrophysiological readouts from the same electrode in vivo.
Fig. 3. Simultaneous tonic dopamine sensing and electrophysiological recording with MAVEN during nicotine stimuli in vivo.

(a) Schematic of a CFM implantation into the NAc of an anesthetized rat. (b) WincsWare interface showing interleaved MCSWV and electrophysiological recordings. (c) Pseudocolor MCSWV voltammogram illustrating baseline tonic DA (325 nM) before nicotine (800 μg/kg s.c.) injection. (d) Representative raw LFP trace before nicotine; red arrows denote periodic noise artifacts. (e) Power spectrum of trace in (d). (f) Voltammogram 10 min after nicotine, revealing elevated tonic DA (460 nM). (g) Raw LFP trace post-nicotine. (h) Corresponding power spectrum showing increased beta-band power. Panel a was created using BioRender.com. CFM, carbon-fiber microelectrode; LFP, local field potential; MCSWV, multiple cyclic square-wave voltammetry; NAc, nucleus accumbens; DA, dopamine.
3.4. MAVEN characterizes tonic neurochemical changes during opioid exposure in rat and swine animal models
3.4.1. Rat model
Over 40 million people in the United States alone suffer from substance use disorders (SUD). Opioids are the leading cause of overdose and were responsible for over 100,000 deaths and 800,000 nonfatal overdoses in 2022 (Spencer et al., 2022). Converging evidence suggests that addiction’s hallmark, habitual drug seeking, is driven by sustained shifts in DA and 5-HT levels within the NAc rather than by brief phasic bursts (Lipton et al., 2019; Everitt and Robbins, 2016). Voltammetric studies, including our own, have begun to track these slow neurochemical dynamics in vivo during exposure to opioids and other drugs of abuse (Goyal et al., 2024; Yuen et al., 2023a and 2023b; Yuen et al., 2021). Repeated intake remodels reward circuitry through neuroplastic adaptations, and the resulting behavioral phenotype appears to be reinforced by the altered extracellular milieu (Vander Weele et al., 2014; Doyon et al., 2013; Wise, 2009; Di Chiara and Imperato, 1988). Although 5-HT’s role is less defined than that of DA, animal data indicate that it can modulate DA tone directly, further shaping reward signaling (Pomrenze et al., 2022; Y. Li et al., 2021; Haleem et al., 2018; Rothman and Baumann, 2006). Leveraging MAVEN’s ability to record both transmitters simultaneously, we set out to clarify this interplay during acute opioid administration.
To evaluate MAVEN’s utility for investigating molecular mechanisms underlying SUD, we conducted two acute fentanyl challenges in urethane-anesthetized rats (n = 2 per neurotransmitter). A CFM was implanted in the NAc core and a concentric stimulating electrode in the VTA as previously described. Following a ≥ 30-minute baseline, either normal saline (1 mL/kg) or fentanyl was delivered intravenously, and tonic neurotransmitter levels were tracked with multiple cyclic square-wave voltammetry (MCSWV for DA; and N-MCSWV for 5-HT).
In the MCSWV group, tonic DA levels in the NAc remained stable following saline administration (Fig. 4a), while fentanyl (30 μg/kg i.v.) increased DA after injection (Fig. 4b). Tonic DA levels peaked within minutes before both rats expired due to hypoxia as captured on the trace (Fig. 4b), similar to that observed in prior studies (Barath et al., 2020) (baseline 573.4 ± 52.1 nM; saline 585.2 ± 60.2 nM; peak post-fentanyl 830.4 ± 101.9 nM) (Fig. 4c). These values are similar to the existing literature (Oh, et al., 2018).
Figure 4. MAVEN quantifies fentanyl-evoked changes in tonic dopamine and serotonin in the rat nucleus accumbens.

(a) Systemic saline administration yielded no change in accumbal DA over time, while (b) fentanyl resulted in an increase in tonic DA within five minutes of administration. (c) Average tonic DA levels and (d) representative MCSWV pseudo-color plots at baseline, following saline, and following fentanyl administration. (e) Saline yielded no change in accumbal 5-HT over time, while (b) fentanyl resulted in an increase in tonic 5-HT. (g) Average tonic 5-HT levels and (h) representative N-MCSWV pseudo-color plots at baseline, following saline, and following fentanyl administration.
Similarly, tonic 5-HT levels remained unchanged following saline (Fig. 4e) but increased following fentanyl administration (20 μg/kg i.v.) (baseline 99.8 ± 28.3 nM; saline = 100.1 ± 17.4 nM; peak post-fentanyl = 260 ± 56.6 nM) (Fig. 4f–g). Pseudo-color plots confirmed clean oxidation-reduction signatures at pre-drug, saline-only, and peak post-drug administration time points for MCSWV (Fig. 4d) and N-MCSWV (Fig. 4h), similar to prior studies (Shin et al., 2021). Collectively, these results demonstrate MAVEN’s capability to capture pharmacologic responses in tonic neurotransmitter concentrations in near real-time, underscoring its utility as a mechanistic research tool for in vivo pharmacologic studies.
3.4.2. Swine model
The clinical relevance of tonic DA dynamics motivated us to test MCSWV in a large-animal model that more closely approximates human neuroanatomy than rodents (Shon et al., 2010a). The swine brain, owing to its large volume (approximately 160 g) and gyrencephalic cortical surface, resembles the human brain better than rodent models, and is comparable to that of non-human primates (Lind et al., 2007). Utilizing a frame-based stereotactic system clinically approved for human DBS surgery (Scheitler et al., 2025; Scheitler et al., 2024), we used standard surgical instruments and commercial DBS planning software (BrainLab, Munich, Germany) to implant a CFM into the swine NAc and a stimulating electrode into the swine VTA, as shown in Fig. 5a. In response to fentanyl administration, tonic accumbal DA levels demonstrated a steep, monotonic rise (Fig. 5b), an increase corroborated by the corresponding pre-fentanyl and peak-fentanyl pseudocolor plots and voltammograms (Fig. 5c and 5d).
Fig. 5. Real-time measurement of tonic dopamine in a swine model of fentanyl exposure during stereotactic deep brain stimulation surgery.

(a) Intra-operative photograph showing a swine model with the stereotactic head frame, recording apparatus (microdrive, CFM, Ag/AgCl reference electrode, and stimulating electrode), and the MAVEN unit. (b) Tonic DA concentration (nM) in the NAc vs. time. Dashed vertical line marks the time point of fentanyl injection (30 μg/kg i.v., n = 2). Black arrow corresponds to time of fentanyl administration (50 μg/kg i.v.). (c) MCSWV pseudocolor plot and voltammogram prior to fentanyl administration. Pre-fentanyl tonic DA concentration was 272.52 nM. (d) Pseudocolor plot and voltammogram for peak-fentanyl. Post-fentanyl tonic DA concentration was 332.08 nM. (e) Left: stereotactic plan for the NAc target; right: post-mortem Nissl stain confirming accurate CFM placement (dashed circle). (f) Left: stereotactic plan for the VTA target; right: intra-operative radiograph verifying correct stimulating lead position. (g) Representative tractography of the mesolimbic pathway (green) with planned trajectories to the NAc (yellow) and VTA (blue) targets. CFM, carbon-fiber microelectrode; DA, dopamine; NAc, nucleus accumbens; VTA, ventral tegmental area; CN, caudate nucleus; IC, internal capsule; LV, lateral ventricle; Put, putamen.
This swine study advances translational DBS research in three ways. First, it demonstrates that MAVEN can be integrated seamlessly into a clinical stereotactic workflow, supporting real-time neurochemical monitoring during large-animal surgery. Second, the use of FDA-cleared planning software establishes a direct path to first-in-human studies. Third, because the NAc and VTA are small structures that are not easily visualized on structural MRI, we used the software’s tractography module to map the mesolimbic pathway and refine our targets directly along the dopaminergic projection fibers. This is a novel application, as tractography-guided DBS is an emerging tool for patient-specific neuromodulation (Chan et al., 2024). Figure 5 presents the direct targeting and trajectory plans for the swine NAc (Fig. 5e) and VTA (Fig. 5f), along with their respective confirmations (postmortem histology for the NAc and intraoperative radiograph for the VTA) and the corresponding mesolimbic tractography (Fig. 5g). Taken together, these data confirm that MAVEN can safely record tonic neurochemicals during image-guided DBS surgery in a large animal model and that it can be readily adaptable to existing human surgical workflows.
3.4.3. MAVEN interfaces with soft laser-induced graphene electrodes for chronic recordings in mice
Chronic neurochemical monitoring in small animals is often hampered by the rigidity and fragility of conventional CFMs. To address this limitation, we paired MAVEN with a soft, laser-induced graphene (LIG) electrode sensor (Li et al., 2022) specifically designed for long-term implantation in mice. Scanning-electron microscopy confirmed the porous microstructure of the LIG surface (Fig. 6a). In vitro, the sensor produced a well-resolved oxidation-reduction signature for 500 nM DA with MAVEN using FSCV (Fig. 6b, c). Four days after implantation in the dorsal striatum of a head-fixed, awake mouse, the same sensor captured spontaneous DA transients in vivo (Fig. 6d, e). These results demonstrate that MAVEN is capable of producing high-fidelity electrochemical signals from not only CFM but also from other carbon-based fabricated microelectrodes.
Fig. 6. Compatibility of MAVEN with laser-induced graphene (LIG) sensors for dopamine detection.

(a) Scanning-electron micrograph of the porous LIG electrode surface (scale bar, 5 μm). (b, c) In vitro FSCV showing the oxidation-reduction pseudocolor plot (b) and background-subtracted voltammogram (c) for 500 nM dopamine. (d) Schematic illustration of the chronic recording setup. A head-fixed mouse is connected via a lightweight tether to the MAVEN module on a mobile platform. (e) Representative FSCV pseudocolor plot acquired on post-operative day 4, revealing spontaneous striatal dopamine transients. Panel d was created using Biorender.com.
4. Conclusions
Designed for both basic and clinical neuroscience, here we have demonstrated the utility of the compact, battery-powered MAVEN platform to acquire near real-time phasic and tonic DA and 5-HT sensing in vivo, record near-simultaneous electrophysiological data, and interleave programmable electrical stimulation, all within the same procedure. By integrating these modalities, together with artifact-free interleaving, a Bluetooth or fiber-optic data link, and an intuitive WincsWare software interface, MAVEN lowers the technical barrier for laboratories that lack custom engineering expertise. By integrating hardware and software in a platform that is ready for commercialization, MAVEN confers at least six near-term opportunities: (1) investigating how neurochemical and electrophysiological signatures are related to human behavior and cognition during awake procedures, (2) elucidating the mechanisms of DBS after decades of empirical application, (3) probing in vivo neurochemical dynamics that underpin neuropsychiatric disorders, accelerating drug discovery and personalized therapeutic strategies, (4) mapping neural circuit function by stimulating one node and measuring downstream neurochemical and concomitant electrophysiological responses from a single CFM or alternative sensor, (5) establishing neurotransmitters as actionable biomarkers for future closed-loop neuromodulation, and (6) disentangling the distinct contributions of phasic extracellular neurotransmitter release vs. tonic neurotransmitter levels to electrophysiology, pharmacotherapy, and neuromodulation. MAVEN also has immediate relevance for human DBS surgery, in which the targeting of subcortical neuroanatomic regions, such as at the border of the internal capsule and the NAc (e.g., during DBS surgery for OCD), could be refined through neurochemical guidance, as electrophysiological signals alone are insufficient to guide targeting (Li et al., 2024). Accordingly, MAVEN, a platform that is ready for commercialization, is positioned to accelerate the transition toward personalized neuromodulation wherein neurochemical signatures guide both timing and target of intervention.
Beyond the central nervous system, MAVEN’s versatility is equally valuable. Indeed, the same voltammetric and amperometric capabilities can quantify analytes in blood, urine, sweat, and cerebrospinal fluid. Prior work has already used voltammetry to measure uric acid, bilirubin, and albumin in serum (Kokoskarova et al., 2023); cortisol and drug metabolites in urine (Hadad et al., 2023; Kassa et al., 2023); atherosclerosis biomarkers in sweat (Wei et al., 2023); antibiotics in blood (Feier et al., 2019); and metabolic components in cerebrospinal fluid (Schirinzi et al., 2021). Altogether, MAVEN is poised to serve as a general-purpose diagnostic and pre-clinical and clinical Neuroscience research tool across multiple central and peripheral systems.
Supplementary Material
Acknowledgments
We gratefully acknowledge the late Ken Kressin (1961–2019) for his pivotal contributions as a software architect and firmware designer, as well as the support of the Mayo Clinic X-Ray Imaging Core. This work was funded by the National Institute of Neurological Disorders and Stroke (R01 NS112176, F31 NS11520, R42 NS125895, R01 NS129549); the Agency for Healthcare Research and Quality (R18 HS029774); the National Institute of General Medical Sciences (T32 GM065841); the Grainger Foundation; the Terrance and Bette Noble Foundation; the Society of Neurological Surgeons Neurosurgeon-Scientist Training Program (to K.M.S.); the Mayo Clinic Kern Center for the Science of Health Care Delivery; and the National Research Foundation of Korea (NRF-2021R1A2B5B02002437).
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://
Footnotes
Declaration of competing interest
The investigators and Mayo Clinic hold intellectual-property interests in the technology described and could benefit financially from its successful development and commercialization.
Data availability
Data will be made available on request.
References
- Agnesi F, Tye SJ, Bledsoe JM, Griessenauer CJ, Kimble CJ, Sieck GC, Bennet KE, Garris PA, Blaha CD, Lee KH. J Neurosurg. 2009. Oct;111(4):701–11. doi: 10.3171/2009.3.JNS0990. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Atcherley CW, Laude ND, Parent KL, Heien ML. Langmuir. 2013. Dec 3;29(48):14885–92. doi: 10.1021/la402686s. [DOI] [PubMed] [Google Scholar]
- Bang D, Kishida KT, Lohrenz T, White JP, Laxton AW, Tatter SB, Fleming SM, Montague PR. Neuron 108, 999–1010.e1016 (2020). doi: 10.1016/j.neuron.2020.09.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Barath AS, Rusheen AE, Rojas Cabrera JM, Price JB, Owen RL, Shin H, Jang DP, Blaha CD, Lee KH, Oh Y. Front Neurosci. (2020) Aug 18;14:869. doi: 10.3389/fnins.2020.00869. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Batten SR, Bang D, Kopell BH, Davis AN, Heflin M, Fu Q, Perl O, Ziafat K, Hashemi A, Saez I, Barbosa LS, Twomey T, Lohrenz T, White JP, Dayan P, Charney AW, Figee M, Mayberg HS, Kishida KT, Gu X, Montague PR. Nat Hum Behav 8, 718–728 (2024). doi: 10.1038/s41562-024-01831-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Bennet KE, Tomshine JR, Min HK, Manciu FS, Marsh MP, Paek SB, Settell ML, Nicolai EN, Blaha CD, Kouzani AZ, Chang SY, Lee KH. Front Hum Neurosci. (2016) Mar 15;10:102. doi: 10.3389/fnhum.2016.00102. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Blumenfeld Z, Brontë-Stewart H Neuropsychol Rev 25, 384–397 (2015). doi: 10.1007/s11065-015-9308-7. [DOI] [PubMed] [Google Scholar]
- Castrioto A, Lhommée E, Moro E, Krack P. Lancet Neurol 13, 287–305 (2014). doi: 10.1016/S1474-4422(13)70294-1. [DOI] [PubMed] [Google Scholar]
- Chan JL, Carpentier AV, Middlebrooks EH, Okun MS, Wong JK. Expert Rev Neurother 24, 11–24 (2024). doi: 10.1080/14737175.2023.2289573. [DOI] [PubMed] [Google Scholar]
- Chang SY, Shon YM, Agnesi F, Lee KH. Annu Int Conf IEEE Eng Med Biol Soc. (2009);2009:3294–7. doi: 10.1109/IEMBS.2009.5333735. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Chang SY, Kim I, Marsh MP, Jang DP, Hwang SC, Van Gompel JJ, Goerss SJ, Kimble CJ, Bennet KE, Garris PA, Blaha CD, Lee KH. Mayo Clin Proc 87, 760–765 (2012). doi: 10.1016/j.mayocp.2012.05.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Chang SY, Kimble CJ, Kim I, Paek SB, Kressin KR, Boesche JB, Whitlock SV, Eaker DR, Kasasbeh A, Horne AE, Blaha CD, Bennet KE, Lee KH. J Neurosurg 119(6), 1556–1565 (2013). doi: 10.3171/2013.8.JNS122142. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Da Cunha C, Gomez-A A, Blaha CD. Reviews in Neuroscience (Special Issue on The emotional brain and its relation to psychopathology (Editor-in-Chief: Huston Joseph; Guest Editors: Campolongo Patrizia, Gasbarri Antonella and Tomaz Carlos). 23(5–6):747–67 (2012). [DOI] [PubMed] [Google Scholar]
- Da Cunha C, Boschen SL, Gómez-A A, Ross EK, Gibson WS, Min HK, Lee KH, Blaha CD. Neurosci Biobehav Rev. 2015. Nov;58:186–210. doi: 10.1016/j.neubiorev.2015.02.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Di Chiara G Eur J Pharmacol 393, 295–314 (2000). doi: 10.1016/s0014-2999(00)00122-9. [DOI] [PubMed] [Google Scholar]
- Di Chiara G, Imperato A. Proc Natl Acad Sci USA 85, 5274–5278 (1988). doi: 10.1073/pnas.85.14.5274. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Doyon WM, Thomas AM, Ostroumov A, Dong Y, Dani JA. Biochem Pharmacol 86, 1181–1193 (2013). doi: 10.1016/j.bcp.2013.07.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Everitt BJ, Robbins TW. Annu Rev Psychol 67, 23–50 (2016). doi: 10.1146/annurev-psych-122414-033457. [DOI] [PubMed] [Google Scholar]
- Feier B, Blidar A, Vlase L, Cristea C. Electrochemistry Communications.;104:106474 (2019). [Google Scholar]
- Goyal A, Cabrera JR, Blaha CD, Lee KH, Shin H, Oh Y. Biomed Eng Lett 14, 1347–1354 (2024). doi: 10.1007/s13534-024-00408-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Goyal A, Hwang S, Rusheen AE, Blaha CD, Bennet KE, Lee KH, Jang DP, Oh Y, Shin H. Front Neurosci 16, (2022). doi: 10.3389/fnins.2022.899436. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Grace AA. Drug Alcohol Depend 37, 111–129 (1995). doi: 10.1016/0376-8716(94)01066-t. [DOI] [PubMed] [Google Scholar]
- Grenhoff J, Aston-Jones G, Svensson TH. Acta Physiol Scand 128, 351–358 (1986). doi: 10.1111/j.1748-1716.1986.tb07988.x. [DOI] [PubMed] [Google Scholar]
- Hadad M, Hadad N, Zestos AG. Biosensors (Basel) 13, (2023). doi: 10.3390/bios13060626. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Haleem DJ, Nawaz S, Salman T. Pharmacol Biochem Behav 170, 71–78 (2018). doi: 10.1016/j.pbb.2018.05.010. [DOI] [PubMed] [Google Scholar]
- Hatsopoulos NG, Donoghue JP. Annu Rev Neurosci 32, 249–266 (2009). doi: 10.1146/annurev.neuro.051508.135241. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Heien ML, Johnson MA, Wightman RM. Analytical chemistry. 2004. Oct 1;76(19):5697–704. [DOI] [PubMed] [Google Scholar]
- Jackson BP, Dietz SM, Wightman RM. Anal Chem. (1995) Mar 15;67(6):1115–20. doi: 10.1021/ac00102a015. [DOI] [PubMed] [Google Scholar]
- Kassa A, Bitew Z, Abebe A. Sens Bio-Sens Res 39, 100554 (2023). [Google Scholar]
- Kishida KT, Sandberg SG, Lohrenz T, Comair YG, Sáez I, Phillips PE, Montague PR. PLoS One 6, e23291 (2011). doi: 10.1371/journal.pone.0023291. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kishida KT, Saez I, Lohrenz T, Witcher MR, Laxton AW, Tatter SB, White JP, Ellis TL, Phillips PE, Montague PR. Proc Natl Acad Sci U S A 113, 200–205 (2016). doi: 10.1073/pnas.1513619112. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kokoskarova P, Stojanov L, Najkov K, Ristovska N, Ruskovska T, Skrzypek S, Mirceski V. Sci Rep 13, 8485 (2023). doi: 10.1038/s41598-023-34350-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Koob GF, Volkow ND Lancet Psychiatry 3, 760–773 (2016). doi: 10.1016/S2215-0366(16)00104-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lee KH, Chang SY, Roberts DW, Kim U. J Neurosurg 101, 511–517 (2004). doi: 10.3171/jns.2004.101.3.0511. [DOI] [PubMed] [Google Scholar]
- Lee KH, Lujan JL, Trevathan JK, Ross EK, Bartoletta JJ, Park HO, Paek SB, Nicolai EN, Lee JH, Min HK, Kimble CJ, Blaha CD, Bennet KE. Sci Rep. (2017) Apr 28;7:46675. doi: 10.1038/srep46675 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Li J, Liu Y, Yuan L, Zhang B, Bishop ES, Wang K, Tang J, Zheng YQ, Xu W, Niu S, Beker L, Li TL, Chen G, Diyaolu M, Thomas AL, Mottini V, Tok JB, Dunn JCY, Cui B, Pașca SP, Cui Y, Habtezion A, Chen X, Bao Z. Nature 606, 94–101 (2022). doi: 10.1038/s41586-022-04615-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Li W, Li N, Wang X, Chen L, Su M, Zheng Z, Li J, Wang X, Jing D, Wang X, Ge S. Neurosci Lett 836, 137884 (2024). doi: 10.1016/j.neulet.2024.137884. [DOI] [PubMed] [Google Scholar]
- Li W, Li N, Wang X, Chen L, Su M, Zheng Z, Li J, Wang X, Jing D, Wang X, Ge S. Science 373, 1252–1256 (2021). doi: 10.1126/science.abi9086. [DOI] [Google Scholar]
- Lind NM, Moustgaard A, Jelsing J, Vajta G, Cumming P, Hansen AK. Neurosci Biobehav Rev 31, 728–751 (2007). doi: 10.1016/j.neubiorev.2007.02.003. [DOI] [PubMed] [Google Scholar]
- Lipton DM, Gonzales BJ, Citri A. Front Syst Neurosci 13, 28 (2019). doi: 10.3389/fnsys.2019.00028. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Little S, Brown P. Parkinsonism Relat Disord 20 Suppl 1, S44–48 (2014). doi: 10.1016/S1353-8020(13)70013-0. [DOI] [PubMed] [Google Scholar]
- Lucente E, Soderpalm B, Ericson M, Adermark L. Front Mol Neurosci 15, 1104648 (2022). doi: 10.3389/fnmol.2022.1104648. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Marder E Trends Neurosci 7, 48–53 (1984). [Google Scholar]
- Marsálek T, Matousek V, Mautner P, Merta M, Moucek R. Neural Netw World 16, 41–60 (2006). [Google Scholar]
- McCormick DA, Prince DA. J Mind Behav 8, 573–590 (1987). [Google Scholar]
- Min HK, Ross EK, Jo HJ, Cho S, Settell ML, Jeong JH, Duffy PS, Chang SY, Bennet KE, Blaha CD, Lee KH. J Neurosci. (2016) Jun 1;36(22):6022–9. doi: 10.1523/JNEUROSCI.0403-16.2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Oesterle TS, Bormann NL, Al-Soleiti M, Kung S, Singh B, McGinnis MT, Correa da Costa S, Rummans T, Chauhan M, Rojas Cabrera JM, Vettleson-Trutza SA, Scheitler KM, Shin H, Lee KH, Gold MS. Brain Sci. 2025. Jul 6;15(7):723. doi: 10.3390/brainsci15070723. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Oh Y, Heien ML, Park C, Kang YM, Kim J, Boschen SL, Shin H, Cho HU, Blaha CD, Bennet KE, Lee HK, Jung SJ, Kim IY, Lee KH, Jang DP. Biosens Bioelectron 121, 174–182 (2018). doi: 10.1016/j.bios.2018.08.034. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Paxinos G, Watson C. The rat brain in stereotaxic coordinates. 3rd ed. San Diego: Academic Press; 1997. [Google Scholar]
- Paek SB, Knight EJ, Chang SY, Lujan JL, Jang DP, Bennet KE, Lee KH. Biomed Eng Lett 3, 22–31 (2013). doi: 10.1007/s13534-013-0086-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Picciotto MR, Zoli M, Rimondini R, Léna C, Marubio LM, Pich EM, Fuxe K, Changeux JP. Nature 391, 173–177 (1998). doi: 10.1038/34413. [DOI] [PubMed] [Google Scholar]
- Pidoplichko VI, DeBiasi M, Williams JT, Dani JA. Nature 390, 401–404 (1997). doi: 10.1038/37120. [DOI] [PubMed] [Google Scholar]
- Pomrenze MB, Pinto DF, Neumann PA, Llorach P, Tucciarone JM, Morishita W, Eshel N, Heifets BD, Malenka RC. Neuron 110, 4125–4143.e4126 (2022). doi: 10.1016/j.neuron.2022.09.024. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Rojas Cabrera JM, Price JB, Rusheen AE, Goyal A, Jondal D, Barath AS, Shin H, Chang SY, Bennet KE, Blaha CD, Lee KH, Oh Y. Rev Anal Chem. (2020);39(1):188–199. doi: 10.1515/revac-2020-0117. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Rothman RB, Baumann MH. Ann N Y Acad Sci 1074, 245–260 (2006). doi: 10.1196/annals.1369.064. [DOI] [PubMed] [Google Scholar]
- Rusheen AE, Gee TA, Jang DP, Blaha CD, Bennet KE, Lee KH, Heien ML, Oh Y. Trends Analyt Chem. (2020) Nov;132:116049. doi: 10.1016/j.trac.2020.116049. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Sands LP, Jiang A, Liebenow B, DiMarco E, Laxton AW, Tatter SB, Montague PR, Kishida KT. Sci Adv. 9(48). doi: 10.1126/sciadv.adi4927. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Sandoval-Pistorius SS, Hacker ML, Waters AC, Wang J, Provenza NR, de Hemptinne C, Johnson KA, Morrison MA, Cernera S. J Neurosci. 2023. Nov 8;43(45):7575–7586. doi: 10.1523/JNEUROSCI.1427-23.2023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Scheitler KM, Rusheen AE, Yuen J, Goyal A, Hong S, Osman GM, Sharaf B, Klassen BT, Grewal SS, Miller KJ, Shin H, Oh Y, Lee KH. J Neurosurg 141, 406–411 (2024). doi: 10.3171/2024.1.JNS232563. [DOI] [PubMed] [Google Scholar]
- Scheitler KM, Rojas-Cabrera JM, Vettleson-Trutza SA, Tsai ST, Pons-Monnier GK, El-Gohary MM, Scheer R, Kwak Y, Barone DG, Blaha CD, Oesterle TS, Shin H, Lee KH, Oh Y. Brain Stimul. 18(5), 1441–1443 (2025). doi: 10.1016/j.brs.2025.07.007. [DOI] [PubMed] [Google Scholar]
- Schirinzi T, Cordella A, Mercuri NB, D'Amico A, Palombi A, Zompanti A, Grasso S, Pennazza G, Santonico M. Sensors (Basel) 21, (2021). doi: 10.3390/s21113767. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Shin H, Goyal A, Barnett JH, Rusheen AE, Yuen J, Jha R, Hwang SM, Kang Y, Park C, Cho HU, Blaha CD, Bennet KE, Oh Y, Heien ML, Jang DP, Lee KH. Anal Chem 93(51), 16987–16994 (2021). doi: 10.1021/acs.analchem.1c02131. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Shon YM, Lee KH, Goerss SJ, Kim IY, Kimble C, Van Gompel JJ, Bennet K, Blaha CD, Chang SY. Neurosci Lett. (2010a) May 21;475(3):136–40. doi: 10.1016/j.neulet.2010.03.060. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Shon YM, Chang SY, Tye SJ, Kimble CJ, Bennet KE, Blaha CD, Lee KH. J Neurosurg. (2010b);112(3):539–48. doi: 10.3171/2009.7.JNS09787. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Spencer MR, Minino AM, Warner M, Drug Overdose Deaths in the United States, 2001–2021. NCHS Data Brief, 1–8 (2022). [PubMed] [Google Scholar]
- Sutton SK, Van Rensburg KJ, Jentink KG, Drobes DJ, Evans DE. Psychopharmacology 233, 2301–2308 (2016). doi: 10.1007/s00213-016-4276-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Van Gompel JJ, Bower MR, Worrell GA, Stead M, Chang SY, Goerss SJ, Kim I, Bennet KE, Meyer FB, Marsh WR, Blaha CD, Lee KH. Epilepsia. (2014) Feb;55(2):233–44. doi: 10.1111/epi.12511. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Vander Weele CM et al. , Eur J Neurosci 40, 3041–3054 (2014). doi: 10.1111/ejn.12709. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Vreeland RF, Atcherley CW, Russell WS, Xie JY, Lu D, Laude ND, Porreca F, Heien ML. Anal Chem 87, 2600–2607 (2015). doi: 10.1021/ac502165f. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wei J, Zhang X, Chang Q, Mugo SM, Zhang Q. Anal Chem 95, 15786–15794 (2023). doi: 10.1021/acs.analchem.3c03310. [DOI] [PubMed] [Google Scholar]
- Wise RA. Trends Neurosci 32, 517–524 (2009). doi: 10.1016/j.tins.2009.06.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Yuen J, Goyal A, Rusheen AE, Kouzani AZ, Berk M, Kim JH, Tye SJ, Blaha CD, Bennet KE, Jang DP, Lee KH, Shin H, Oh Y. Front Pharmacol 12, 705254 (2021). doi: 10.3389/fphar.2021.705254. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Yuen J, Goyal A, Rusheen AE, Kouzani AZ, Berk M, Kim JH, Tye SJ, Abulseoud OA, Oesterle TS, Blaha CD, Bennet KE, Lee KH, Oh Y, Shin H. Front Pharmacol 14, 1199655 (2023a). doi: 10.3389/fphar.2023.1199655. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Yuen J, Goyal A, Rusheen AE, Kouzani AZ, Berk M, Kim JH, Tye SJ, Blaha CD, Bennet KE, Lee KH, Shin H, Oh Y. Front Neurosci 17, 1061578 (2023b). doi: 10.3389/fnins.2023.1061578. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data will be made available on request.
