Biology subjects

Hernandez-Reynoso, A. G.

Publications and source records attributed to Hernandez-Reynoso, A. G..

4 recordsLinked to original sources

bioRxiv2023

Planar Amorphous Silicon Carbide Microelectrode Arrays for Chronic Recording in Rat Motor Cortex

Chronic implantation of intracortical microelectrode arrays (MEAs) capable of recording from individual neurons can be used for the development of brain-machine interfaces. However, these devices show reduced recording capabilities under chronic conditions due, at least in part, to the brains foreign body response. This creates a need for MEAs that can minimize the foreign body response to enable long-term recording. A potential approach to reduce the foreign body response is the use of ultrathin MEAs. Here, we fabricated ultrathin (cross-sectional area: 160 {micro}m2) amorphous silicon carbide (a-SiC) MEAs with sixteen electrode channels and implanted them into the motor cortex of seven female Sprague-Dawley rats. A-SiC was chosen as the fabrication base for its high chemical stability, good electrical insulation properties, and amenability to thin film fabrication techniques. Electrochemical analysis and neural recordings were performed weekly for 4 months. MEAs were characterized in vitro pre-implantation and in vivo using electrochemical impedance spectroscopy and cyclic voltammetry at 50 mV/s and 50,000 mV/s. Neural recordings were analyzed for single unit activity. At the end of the study, animals were sacrificed for immunohistochemistry analysis. We observed statistically significant, but small, increases in 1 and 30 kHz impedance values and 50,000 mV/s charge storage capacity over the 16-week implantation period. Slow sweep 50 mV/s CV and 1 Hz impedance did not significantly change over time. Impedance values increased from 11.6 M{Omega} to 13.5 M{Omega} at 1 Hz, 1.2 M{Omega} to 2.9 M{Omega} at 1 kHz, and 0.11 M{Omega} to 0.13 M{Omega} at 30 kHz over 16 weeks. The median charge storage capacity of the implanted electrodes at 50 mV/s were 58.1 mC/cm2 on week 1 and 55.9 mC/cm2 on week 16, and at 50,000 mV/s were 4.27 mC/cm2 on week 1 and 5.93 mC/cm2 on week 16. Devices were able to record neural activity from 92% of all active channels at the beginning of the study, At the study endpoint, a-SiC devices were still recording single-unit activity on 51% of electrochemically active electrode channels. In addition, we observed that the signal-to-noise ratio experienced a small decline of only -0.19 per week. We also classified the units as fast and slow spiking based on the trough-to-peak time. Although the overall presence of single units declined, fast and slow spiking units declined at a similar rate. Furthermore, immunohistochemistry showed minimal foreign body response to the a-SiC devices, as highlighted by statistically insignificant differences in activated glial cells between implanted brains slices and contralateral sham slices, as evidenced by GFAP staining. NeuN staining revealed the presence of neural cell bodies close to the implantation site, again statistically not different from a contralateral sham slice. These results support the use of ultrathin a-SiC MEAs for long-term implantation and use in brain-machine interfaces.

bioengineering↗

bioRxiv2023

NeurostimML: A machine learning model for predicting neurostimulation-induced tissue damage

ObjectiveThe safe delivery of electrical current to neural tissue depends on many factors, yet previous methods for predicting tissue damage rely on only a few stimulation parameters. Here, we report the development of a machine learning approach that could lead to a more reliable method for predicting electrical stimulation-induced tissue damage by incorporating additional stimulation parameters. ApproachA literature search was conducted to build an initial database of tissue response information after electrical stimulation, categorized as either damaging or non-damaging. Subsequently, we used ordinal encoding and random forest for feature selection, and investigated four machine learning models for classification: Logistic Regression, K-nearest Neighbor, Random Forest, and Multilayer Perceptron. Finally, we compared the results of these models against the accuracy of the Shannon equation. Main ResultsWe compiled a database with 387 unique stimulation parameter combinations collected from 58 independent studies conducted over a period of 47 years, with 195 (51%) categorized as non-damaging and 190 (49%) categorized as damaging. The features selected for building our model with a Random Forest algorithm were: waveform shape, geometric surface area, pulse width, frequency, pulse amplitude, charge per phase, charge density, current density, duty cycle, daily stimulation duration, daily number of pulses delivered, and daily accumulated charge. The Shannon equation yielded an accuracy of 63.9% using a k value of 1.79. In contrast, the Random Forest algorithm was able to robustly predict whether a set of stimulation parameters was classified as damaging or non-damaging with an accuracy of 88.3%. SignificanceThis novel Random Forest model can facilitate more informed decision making in the selection of neuromodulation parameters for both research studies and clinical practice. This study represents the first approach to use machine learning in the prediction of stimulation-induced neural tissue damage, and lays the groundwork for neurostimulation driven by machine learning models.

neuroscience↗

bioRxiv2023

Isolation of the murine Glut1 deficient thalamocortical circuit: wavelet characterization and reverse glucose dependence of low and gamma frequency oscillations

Glucose represents the principal brain energy source. Thus, not unexpectedly, genetic glucose transporter 1 (Glut1) deficiency (G1D) manifests with encephalopathy. G1D seizures, which constitute a prominent disease manifestation, often prove refractory to medications but may respond to therapeutic diets. These seizures are associated with aberrant thalamocortical oscillations as inferred from human electroencephalography and functional imaging. Mouse electrophysiological recordings indicate that inhibitory neuron failure in thalamus and cortex underlies these abnormalities. This provides the motivation to develop a neural circuit testbed to characterize the mechanisms of thalamocortical synchronization and the effects of known or novel interventions. To this end, we used mouse thalamocortical slices on multielectrode arrays and characterized spontaneous low frequency oscillations and less frequent 30-50 Hz or gamma oscillations under near-physiological bath glucose concentration. Using the cortical recordings from layer IV, we quantified oscillation epochs via an automated wavelet-based algorithm. This method proved analytically superior to power spectral density, short-time Fourier transform or amplitude-threshold detection. As expected from human observations, increased bath glucose reduced the lower frequency oscillations while augmenting the gamma oscillations, likely reflecting strengthened inhibitory neuron activity. This approach provides an ex vivo method for the evaluation of mechanisms, fuels, and pharmacological agents in a crucial G1D epileptogenic circuit.

neuroscience↗

bioRxiv2023

Behavioral Paradigm for the Evaluation of Stimulation-Evoked Somatosensory Perception Thresholds in Rats

Intracortical microstimulation (ICMS) of the somatosensory cortex via penetrating microelectrode arrays (MEAs) can evoke cutaneous and proprioceptive sensations for restoration of perception in individuals with spinal cord injuries. However, ICMS current amplitudes needed to evoke these sensory percepts tend to change over time following implantation. Animal models have been used to investigate the mechanisms by which these changes occur and aid in the development of new engineering strategies to mitigate such changes. Non-human primates are commonly the animal of choice for investigating ICMS, but ethical concerns exist regarding their use. Rodents are a preferred animal model due to their availability, affordability, and ease of handling, but there are limited choices of behavioral tasks for investigating ICMS. In this study, we investigated the application of an innovative behavioral go/no-go paradigm capable of estimating ICMS-evoked sensory perception thresholds in freely moving rats. We divided animals into two groups, one receiving ICMS and a control group receiving auditory tones. Then, we trained the animals to nose-poke - a well-established behavioral task for rats - following either a suprathreshold ICMS current-controlled pulse train or frequency-controlled auditory tone. Animals received a sugar pellet reward when nose-poking correctly. When nose-poking incorrectly, animals received a mild air puff. After animals became proficient in this task, as defined by accuracy, precision, and other performance metrics, they continued to the next phase for perception threshold detection, where we varied the ICMS amplitude using a modified staircase method. Finally, we used non-linear regression to estimate perception thresholds. Results indicated that our behavioral protocol could estimate ICMS perception thresholds based on [~]95% accuracy of rat nose-poke responses to the conditioned stimulus. This behavioral paradigm provides a robust methodology for evaluating stimulation-evoked somatosensory percepts in rats comparable to the evaluation of auditory percepts. In future studies, this validated methodology can be used to study the performance of novel MEA device technologies on ICMS-evoked perception threshold stability using freely moving rats or to investigate information processing principles in neural circuits related to sensory perception discrimination.

neuroscience↗