The intricate dance of neurons orchestrates vital bodily functions and complex thoughts within a vast network. Traditional neuroscience, relying on in vivo electrophysiology within living organisms, grapples with the brain’s overarching complexity. An alternative method involves extracting cells and studying them in vitro, on a culture dish, offering researchers better control and precision in probing neural processes.
A recent study featured in Advanced Science introduces a cost-effective, open-source in vitro system for interfacing with neurons, providing a more accessible avenue for researchers keen on delving into neural interactions. This study is part of the Mind in Vitro (MiV) project, aiming to understand how neurons interact for a deeper grasp of complex systems like the brain. Beyond comprehension, the MiV project envisions utilizing in vitro neural networks for computation, involving an interdisciplinary team from computer science, engineering, neurobiology, physiology, and more.
Zhi (Andrew) Dou, a graduate student in the Gazzola lab leading the project, explains, “The goal of the MiV project is to ultimately use neurons for computation,” envisioning a dynamic and energy-efficient system distinct from traditional computing.
The MiV study, led by Mattia Gazzola and Xiaotian Zhang, describes an innovative approach to measuring neuron activity using micro-electrode array (MEA) technology. While similar commercial systems exist, they often come with a hefty price tag and cater to specific experimental approaches.
Zhang notes, “The problem with the current technology for interfacing with neurons is that it’s mainly commercial systems, standardized for specific testing conditions within biochemistry or traditional neuroscience.” The MiV project seeks to move away from this standardization, aiming for full control and customization.
In the MiV apparatus, cells are placed on a plate with MEAs, allowing technology to interface with neural substrates. Electrodes detect neuron voltage, amplified and sent to a computer for data processing. The MiV system boasts over 500 electrodes compared to the standard 60-electrode commercial system, enabling more extensive data collection. Notably, it incorporates portability, bi-directional communication with neurons, imaging capabilities during recordings, and the ability to test various input types.
Despite these enhancements, the MiV system’s cost is remarkably 10 times cheaper than commercial alternatives, emphasizing affordability and complete customization. To promote wider access, the researchers have made their hardware and software models for the MiV system open source and freely available online.