Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
February 12, 2026IEEE Transactions on Biomedical Circuits and Systems

Memristive Reservoir Computing Circuit for Real-Time Prediction of Epilepsy

View Full Paper
Ask AI
Bookmark
Share

Authors

LLLun LuNational University of Defense TechnologyAXAo XuNational University of Defense TechnologyYXY. C. XuChangchun University of Science and Technology

Discussion

Loading...

Member takes

Overview

Novel memristor-based system predicts seizures in real-time, indicating a step forward in low-power healthcare solutions.

Key Points

  • The aim is to develop a memristor-based system that enables real-time prediction of epileptic seizures while overcoming hardware limitations.
  • Proposed architecture utilizes memristor-based reservoir computing.
  • Implemented nonlinear temporal feature extraction using volatile memristors.
  • Integrated non-volatile memristor crossbar arrays for efficient data processing.
  • Conducted hardware-aware evaluations to assess system robustness.
  • Achieved over 97% accuracy in simulation and exceeded 95% accuracy in hardware experiments.
  • Maintained stable performance under substantial noise conditions.
  • Utilized only 1,700 trainable parameters, showcasing efficiency.

Cite This Study

Lu et al. (2026) studied this question.

synapsesocial.com/papers/698d6e1a5be6419ac0d5386fhttps://doi.org/10.1109/tbcas.2026.3662427
View Full Paper
Ask AI
Bookmark
Share