StentorCam: A Modular High-Throughput System for Behavioral Imaging in Multiwell Plates
Diana Ceron, Leonard Chau, Kimi Lee, Addy Brien, Keith Curry, Jacob Vasquez
Department of Biology
Faculty Supervisor: Raymond Esquerra
StentorCam is a low-cost, modular imaging system built on the RoboCam architecture that enables automated, repeatable video recording across multiwell formats, including 96-well plates, to study the behavior of motile microorganisms. By repurposing a consumer 3D printer for accurate, programmable positioning, the system combines a Raspberry Pi with Python and OpenCV to coordinate instrument control, imaging, and data collection. To reduce disturbance during experiments and enable long-term observation, StentorCam uses infrared (850 nm) illumination for “dark” imaging while continuously tracking organism movement. This version introduces important design improvements over the original RoboCam, enhancing imaging stability, ease of use, and overall performance.
The platform uniquely supports synchronized, programmable delivery of external stimuli to induce and measure behavioral changes, demonstrated here with controlled optical stimulation (e.g., laser excitation). Recorded videos are easily processed with open-source tracking tools such as Fiji/TrackMate, followed by Python-based analysis to extract trajectories and kinematic data, such as swimming speed. We validate the system using the light-sensitive ciliate Stentor coeruleus to study photophobic and phototactic responses. Ultimately, StentorCam functions as a versatile, scalable platform for stimulus-related behavioral imaging, enabling accessible, high-throughput experiments across various small, motile biological systems.