Modular Data Acquisition System for Recording Activity and Electrical Stimulation of Brain Tissue Using Dedicated Electronics

• Published in: Sensors (Basel, Switzerland) Switzerland), 2021-06, Vol.21 (13), p.4423
• Main Authors: Jurgielewicz, Paweł, Fiutowski, Tomasz, Kublik, Ewa, Skoczeń, Andrzej, Szypulska, Małgorzata, Wiącek, Piotr, Hottowy, Paweł, Mindur, Bartosz
• Format: Article
• Language: English
• Subjects: data acquisition; electrical stimulation; microelectrodes; neural interfaces; neurophysiology; software engineering
• ISSN: 1424-8220; 1424-8220
• DOI: 10.3390/s21134423

In this paper, we present a modular Data Acquisition (DAQ) system for simultaneous electrical stimulation and recording of brain activity. The DAQ system is designed to work with custom-designed Application Specific Integrated Circuit (ASIC) called Neurostim-3 and a variety of commercially available Multi-Electrode Arrays (MEAs). The system can control simultaneously up to 512 independent bidirectional i.e., input-output channels. We present in-depth insight into both hardware and software architectures and discuss relationships between cooperating parts of that system. The particular focus of this study was the exploration of efficient software design so that it could perform all its tasks in real-time using a standard Personal Computer (PC) without the need for data precomputation even for the most demanding experiment scenarios. Not only do we show bare performance metrics, but we also used this software to characterise signal processing capabilities of Neurostim-3 (e.g., gain linearity, transmission band) so that to obtain information on how well it can handle neural signals in real-world applications. The results indicate that each Neurostim-3 channel exhibits signal gain linearity in a wide range of input signal amplitudes. Moreover, their high-pass cut-off frequency gets close to 0.6Hz making it suitable for recording both Local Field Potential (LFP) and spiking brain activity signals. Additionally, the current stimulation circuitry was checked in terms of the ability to reproduce complex patterns. Finally, we present data acquired using our system from the experiments on a living rat’s brain, which proved we obtained physiological data from non-stimulated and stimulated tissue. The presented results lead us to conclude that our hardware and software can work efficiently and effectively in tandem giving valuable insights into how information is being processed by the brain.