The goal of the UCSF RC+S benchtop is to create a physical environment to gain experience using the different functionalities of the RC+S device. A first version of benchtop setup for RC+S was developed by our colleague Maria Olaru UCSF benchtop system. Here we extend that original version of the setup. We add work on a circuit model to recreate electrode-tissue load characteristics.
The RC+S benchtop can be used to learn (train) how to use RC+S software developed at our lab (see Researcher Facing Application and Patient Facing Application). For that purpose (software navigation, changing settings, configuration files, etc), there is no need to use benchtop hardware configurations, so you may want to skip this tutorial.
In case you need benchtop modalities with 'some-how' representative electrode-tissue electrical properties, this tutorial may be of value to get you started. A few reasons why you may be interested in this tutorial:
- you want to understand how different stimulation settings may interfere with different sense settings
- you want to play around with different electrode-tissue impedance configurations (e.g. perfect resistor match load, mismatch resistor load, mismatch capacitive load)
- you want to study how embedded adaptive DBS perfors given different adaptive stimulation settings
- you want to study a 'some-how' realistic settings where 'fast' adaptive stimulation (stimulation response from low to high leved in ~100-200 ms) given different load (perfect match resistor load (it will work :) and then capacitive mismtach (it will not work, there will be self-triggering, as we see in patients)
- you want to know what exactly means stimulation pulse with Active and Passive recharge
- you want to assess 'how good' :) or 'how bad' :( we are able to quantify 'off-device' power from the streamed time-domain signal
- ...
This is work-in-progress in the lab and there is a moving effort with Oxford ([email protected]) and Brown labs ([email protected]) to further develop and build up a standarized (universal) test-bench architecture accross sites. An example of a first step towards this is the neuroDAC, an audio-DAC signal generator to generate microvolt level signals (see neuroDAC publication). BTW, we have a neuroDAC board (v1.0) in our lab! (see the first drawer below the bench desk). As you will read below, there is a few options to choose a signal generator or another, or even not to use any (just let the sense sginal to be background amplifier noise) and that will really depend on what the goal of your bench test is!
An overview of the different hardware and software elements that can be used with the benchtop are listed below:
Hardware
- RC+S: INS and lead extensor cable
- Electrode tissue interface ('star Load', initial prototype implemented on a bread-board)
- Resistor unbalanced (1Kohm vs 4K7)
- Resistor balanced (1Kohm)
- Resistor and capacitor unbalanced
- Signal input/output:
- Signal generator (config explained in this README)
- neuroDAC (audio DAQ) (not exaplined here - neuroDAC paper)
- preSonus (audio DAQ) (not explained here - preSonus benchtop)
- Signal acquistion: NI myDAQ (not explained here - labview scripts for datalogging and RC+S example dataset recorded with the benchtop configuraiton explained below
Example Benchtop Configuraiton 1)
Software
- UCSF software to interface RCS:
- Signal input/output
- Signal generator PiscoScope 6 (only tested in Windows version)
- neuroDAC
- preSonuns
- Signal acquisition NI LabView 2019
In this configuration the RC+S benchtop is interfaced with the signal generator.
The bread board can be used for 3 different conditions of the 'star load':
- Resistive unbalanced (mismatch)
- Resistive balanced (match)
- Resisitive and capacitive unbalanced (closer to real conditions; results in self-triggering when testing 'fast aDBS')
