Physically Controlled Synaptic Plasticity and Reconfigurable Analog Computing Systems for Healthcare Data Processing
발표자
이윤찬 (연세대학교)
연구책임자
이윤찬 (연세대학교)
조정호 (연세대)
초록
내용
Efficiently processing the massive influx of unstructured and dynamic biometric data is a significant hurdle for conventional computing resources. To address this, we propose a neuromorphic solution inspired by biological neural networks. In this study, we present an electrolyte-gated organic transistor that physically realizes the transition between short-term and long-term plasticity (STP and LTP). By utilizing a photochemical cross-linking reaction, we precisely restricted ion penetration within the organic channel, thereby tuning the device's memory retention characteristics. This physical tunability enabled the creation of reconfigurable synaptic logic gates capable of executing medical diagnostic algorithms directly, without the need for additional weight-update processes. Ultimately, the system demonstrated the practical capability to process biometric information across varying time scales for effective healthcare monitoring.