Researchers at the Weizmann Institute of Science have introduced BAMBI, a calcium imaging-based brain-computer interface designed for long-term tracking of neuronal activity in freely behaving mice. The system, detailed in a new study published in the Journal of Neuroscience Methods, combines miniaturized fluorescence microscopy with real-time decoding and position tracking to monitor the same neurons over extended periods without restricting animal movement.

BAMBI stands for a Ca²⁺ Imaging–Based Brain–Computer Interface for Longitudinal Neuronal Tracking in Freely Behaving Mice. It builds on established techniques in one-photon calcium imaging using head-mounted miniscopes, allowing scientists to record activity from thousands of neurons in regions such as the hippocampus while mice explore their environment naturally.

Core Technology and Design Features

The platform integrates several key components. A head-mounted miniscope captures fluorescence signals from calcium indicators expressed in neurons. Custom software performs online motion correction, neuron identification, and activity extraction. A dedicated position tracking module records the mouse’s location in real time, synchronizing behavioral data with neural signals.

This setup supports closed-loop experiments where decoded neural activity can trigger feedback or stimulation. Unlike traditional tethered systems, BAMBI maintains high stability across multiple days or weeks, addressing a major challenge in longitudinal studies: reliably identifying the same cells over time despite brain movement and tissue changes.

Authors Linor Balilti-Turgeman, Nitzan Shalvi, Or Pinchasof, Nitzan Geva, Daniel Deitch, Alon Rubin, and Yaniv Ziv developed the system with an emphasis on open-source compatibility and ease of integration with existing miniscope hardware.

Applications in Neuroscience Research

Longitudinal neuronal tracking is essential for understanding how memories form, how neural codes evolve, and how circuits adapt during learning or disease. Previous work from the Ziv lab has used calcium imaging to follow hippocampal place cells over weeks, revealing both stability and representational drift in spatial maps.

BAMBI extends these capabilities by adding real-time brain-computer interface functionality. Researchers can now decode intended behaviors or cognitive states from population activity and deliver immediate feedback, opening avenues for studying causal relationships between neural patterns and actions in awake, unrestrained animals.

Potential uses include investigating memory consolidation during sleep, testing neurofeedback paradigms, and modeling neurological conditions in genetic mouse lines. The system’s position tracking also enables precise correlation of neural activity with spatial behavior without additional hardware.

Photo by Bioscience Image Library by Fayette Reynolds on Unsplash

Technical Advantages Over Existing Methods

Traditional brain-computer interfaces in rodents often rely on electrode arrays or require head fixation, limiting natural behavior. Optical approaches using calcium indicators provide cellular resolution and can track larger populations, but real-time processing has been computationally demanding.

BAMBI incorporates efficient algorithms for motion stabilization and trace extraction that run with low latency. The inclusion of simultaneous position tracking creates a multimodal dataset valuable for both basic research and the development of more advanced interfaces.

Because the system works with freely behaving animals, it preserves ethologically relevant behaviors that are difficult to study under restraint. This is particularly important for hippocampal and cortical circuits involved in navigation, decision-making, and episodic memory.

Context Within Broader Brain-Computer Interface Field

Brain-computer interfaces have advanced rapidly in both human clinical applications and basic neuroscience. In rodents, optical BCIs complement electrical methods by offering genetic targeting and visualization of specific cell types.

Related developments include real-time decoding systems for miniscopes and neurofeedback protocols that modulate sharp-wave ripples or replay events. BAMBI’s emphasis on longitudinal stability positions it as a tool for studying slow processes such as memory stabilization or circuit reorganization after injury.

Institutions like the Weizmann Institute of Science continue to push boundaries in systems neuroscience through such innovations. The work aligns with global efforts to create more naturalistic experimental preparations that better model human brain function.

Implications for Academic Research and Training

Tools like BAMBI lower barriers for labs studying long-term neural dynamics. Graduate students and postdoctoral researchers can now design experiments that track individual neurons across behavioral training sessions lasting weeks, generating richer datasets for thesis work or publications.

University neuroscience programs may incorporate similar open-source platforms into curricula, giving trainees hands-on experience with modern imaging and decoding techniques. This prepares the next generation of researchers for careers in both academia and neurotechnology industries.

Funding agencies increasingly prioritize methods that enable reproducible, longitudinal studies. Systems that combine imaging, behavior, and real-time analysis support these priorities by producing multimodal data suitable for sharing and meta-analysis.

Photo by Google DeepMind on Unsplash

Future Directions and Outlook

Further refinements could expand BAMBI to additional brain regions, integrate optogenetic stimulation for closed-loop control, or combine it with other modalities such as fiber photometry or electrophysiology. Improvements in indicator brightness and miniscope miniaturization will likely enhance signal quality and reduce implant size.

As the technology matures, it may contribute to translational research on disorders involving circuit dysfunction, such as epilepsy, Alzheimer’s disease, or psychiatric conditions. Mouse models remain central to preclinical validation, and longitudinal cellular-resolution data can bridge gaps between molecular mechanisms and behavioral outcomes.

Collaborations between engineering, computer science, and biology departments will be essential for scaling these interfaces and developing user-friendly analysis pipelines.

Accessing the Original Research

The full study appears in the Journal of Neuroscience Methods. Readers can access the publication at https://www.sciencedirect.com/science/article/pii/S0165027026001627. The authors are Linor Balilti-Turgeman, Nitzan Shalvi, Or Pinchasof, Nitzan Geva, Daniel Deitch, Alon Rubin, and Yaniv Ziv, affiliated with the Department of Brain Sciences at the Weizmann Institute of Science.

Additional information about ongoing work in the Ziv laboratory is available on the lab website.