AN INTEGRATED PLATFORM FOR REAL-TIME VOLTAGE IMAGING AND FEEDBACK CONTROL

Understanding how brain activity controls our thoughts, sensations and movements is a central goal in neuroscience. However, causally inferring how specific population dynamics dictate behavior remains challenging, in particular because technologies for real-time monitoring and manipulating neuronal activity are currently underdeveloped. Among available techniques for monitoring neuronal electrical activity, voltage imaging is uniquely powerful, enabling direct visualization of membrane potential dynamics from subcellular compartments to large populations. When combined with optogenetics, it enables simultaneous optical recording and manipulation of targeted neurons or subcellular regions — an approach known as all-optical electrophysiology. Yet, closing the loop between activity and manipulation requires analysis at sub-millisecond timescales, which has remained a major technical bottleneck. Here we present an integrated software platform that unifies hardware control, high-speed data streaming, and GPU-accelerated analysis to enable real-time voltage imaging and activity-dependent optogenetic manipulation. The system provides plug-and-play control of cameras, light sources, digital micromirror devices, and TTL-triggered hardware. Our streaming architecture supports real-time processing at up to 500 × 500 pixels at 1000 fps, enabling rapid extraction of neuronal electrical signals even in fast-spiking cells. The platform is designed to work with diverse hardware setups and includes a streamlined setup procedure to facilitate user-specific configurations of imaging and feedback manipulation. We believe this development enables researchers to manipulate neuronal populations based on their ongoing activity, allowing causal tests of how real-time neuronal dynamics shape behavior.