A Low-Cost Modular Multi-Region Electrode for Distributed Network Recording and Brain State Decoding

Background: Precise decoding of brain states is essential for closed-loop neuromodulation, but current electrodes and recording strategies are inadequate. Multi-region recording offers network-level advantages over single-region approaches, yet remains underdeveloped due to the lack of low-cost, flexible multi-region electrodes and standardized workflows. Methods: We developed a low-cost, modular, silica capillary tube-based 16-channel electrode for multi-region local field potential (LFP) recording in small animals, along with an integrated workflow for spectral analysis, functional connectivity assessment, and machine learning-based brain state decoding. Results: Our electrode design enables flexible customization of target regions and low-cost (~19.43 USD/unit) assembly without specialized equipment. In vivo validation in rats targeting eight emotional network nuclei achieved 79.2% implantation accuracy, with stable LFP recordings maintained for over 3 months. In a reserpine-induced depression model, spectral analysis revealed state-specific oscillatory changes including reduced alpha power in the infralimbic cortex. Inter-regional functional connectivity analysis further captured drug-induced network-level synchronization changes. We also developed a machine learning pipeline with random forest classifier that achieved ~96.4% accuracy in decoding brain states from the multi-region signals. Conclusions: This modular multi-region electrode provides a practical, adaptable, and low-cost platform for long-term distributed network recording, supporting quantitative LFP analysis, functional connectivity assessment, and high-accuracy brain state decoding, laying a technical foundation for preclinical closed-loop neuromodulation research.