Journal: Nature Communications

Article Title: Biomimetic computer-to-brain communication enhancing naturalistic touch sensations via peripheral nerve stimulation

doi: 10.1038/s41467-024-45190-6

Figure Lengend Snippet: The successful development of a somatosensory neuroprosthesis is based on three main pillars: (1) In-silico models of the biological sensory processing have to be exploited for emulating the natural neural activation of the nervous system to external tactile stimuli (blue segment); (2) animal proof of concept allows for experimental validation of the mechanisms behind the use of specific neurostimulation strategies defined with the use of modeling (orange segment); DRG—dorsal root ganglion. (3) A rigorous clinical validation of the biomimetic technology with implanted humans has to be performed in order to assess the functional outcomes in real-life scenarios (green segment). The results from the clinical trials will then allow us to collect relevant data exploitable for improving computational modeling. (Illustration credit: Giacomo Valle, ETH Zurich).

Article Snippet: Examples of spiking activity over time during each condition are presented using a peri-stimulus time histogram (PSTH) (right) with a time bin of 50 ms. Brown lines represent the envelope of neural activity. b Comparing local field potential (LFP) recorded in DRG resulting from natural touch in biomimetic and tonic stimulation.

Techniques: In Silico, Activation Assay, Biomarker Discovery, Functional Assay, Clinical Proteomics

Journal: Nature Communications

Article Title: Biomimetic computer-to-brain communication enhancing naturalistic touch sensations via peripheral nerve stimulation

doi: 10.1038/s41467-024-45190-6

Figure Lengend Snippet: a Decerebrated cat experimental setup. We stimulated the tibial nerve with a cuff electrode and recorded neural response on the spinal level; upper part: exposed L6 vertebrae and dorsal root ganglion (DRG) with examples of recorded neural spikes from spinal linear electrode probe and DRG UTAH array. b Obtaining multiunit neural activity. We filtered the signal to extract the spiking component and detect the neural action potentials using the thresholding algorithm. c left: Example of biomimetic stimulation paradigm and recorded response signal in one channel of the spinal cord and DRG electrodes. Neural activity is presented and quantified with a raster plot (black dots) and peri-stimulus time histogram (PSTH, yellow, bin length 10 ms). Each row of the raster plot represents the response to a single biomimetic pattern (2 s), while each dot corresponds to an action potential. Right: We compared the PSTHs from different conditions and presented them using cross-correlation (blue: L6 neural response and stimulation pattern; red: DRG neural response and stimulation pattern; black: L6 and DRG neural responses). Source data are provided as a Source Data file.

Article Snippet: Examples of spiking activity over time during each condition are presented using a peri-stimulus time histogram (PSTH) (right) with a time bin of 50 ms. Brown lines represent the envelope of neural activity. b Comparing local field potential (LFP) recorded in DRG resulting from natural touch in biomimetic and tonic stimulation.

Techniques: Activity Assay

Journal: Nature Communications

Article Title: Biomimetic computer-to-brain communication enhancing naturalistic touch sensations via peripheral nerve stimulation

doi: 10.1038/s41467-024-45190-6

Figure Lengend Snippet: a Comparing multiunit neural activity as a response to natural touch, biomimetic, or tonic (50 Hz) stimulation. We compared the signal recorded in DRG (dorsal root ganglion) and spinal cord (level L6). The overall amount of neural activity during each condition is summed, normalized, and presented with bars for comparison (left). Examples of spiking activity over time during each condition are presented using a peri-stimulus time histogram (PSTH) (right) with a time bin of 50 ms. Brown lines represent the envelope of neural activity. b Comparing local field potential (LFP) recorded in DRG resulting from natural touch in biomimetic and tonic stimulation. Another natural touch response, recorded in cat 2, was added to the analysis. We compared the distribution of LFP amplitude values using the Kullback–Leibler divergence metric. c Comparing current source density (CSD) calculated from LFP recorded in the spinal cord resulting from natural touch, biomimetic, or tonic stimulation. CSD is normalized for each condition and presented along the length of the electrode with 100 ms bin for electrical stimulation and 40 ms for natural touch condition. d Left: correlation of CSD between biomimetic/tonic and natural touch condition and biomimetic and tonic stimulation, channel by channel, color-coded. Right: Histogram and cumulative distribution function (cdf) of the correlation coefficient values resulting from comparing biomimetic/tonic stimulation and natural touch condition (top/middle) and biomimetic and tonic stimulation (bottom). The blue line represents the cdf when the recording channels are matched and compared. The red line corresponds to cdf when channels are randomly shuffled and compared. Source data are provided as a Source Data file.

Article Snippet: Examples of spiking activity over time during each condition are presented using a peri-stimulus time histogram (PSTH) (right) with a time bin of 50 ms. Brown lines represent the envelope of neural activity. b Comparing local field potential (LFP) recorded in DRG resulting from natural touch in biomimetic and tonic stimulation.

Techniques: Activity Assay, Comparison