Analysis of an open source, closed-loop, realtime system for hippocampal sharp-wave ripple disruption

Area of Science:

• Neuroscience
• Computational Neuroscience
• Systems Neuroscience

Background:

• Closed-loop neural modulation allows causal inference between neural circuits and behavior.
• Sharp-wave ripples (SWRs) in the hippocampus are critical for memory and require precise, low-latency perturbation.
• Existing SWR disruption systems lack detailed performance analysis, hindering experimental design.

Purpose of the Study:

• To develop and evaluate a low-latency, closed-loop system for detecting hippocampal sharp-wave ripples (SWRs).
• To characterize the performance trade-offs of SWR detection latency across different hardware and algorithmic parameters.
• To provide a modular, open-source framework for real-time neural interfacing.

Main Methods:

• Developed a real-time SWR detection algorithm integrated into an open-source neural data acquisition suite.
• Utilized synthetic data to explore the detection algorithm's parameter space.
• Quantified in vivo system performance and latency using both USB and Ethernet data acquisition hardware.

Main Results:

• Signal detection latency comprises data acquisition (7.5–13.8 ms for USB, 1.35–2.6 ms for Ethernet) and algorithmic components.
• Ethernet hardware achieved algorithmic latencies of ~20–66 ms with <10 false SWR detections per minute.
• Performance is highly dependent on algorithmic parameter choices and hardware selection.

Conclusions:

• Characterized the performance of a low-latency closed-loop SWR detection system.
• Established a framework for analyzing closed-loop neural interfacing systems.
• The open-source, modular system facilitates precise causal closed-loop experiments in neuroscience.