Multimodal In‐Sensor Computing with Dual‐Phase Organic Synapses for Wearable Fitness Monitoring
Yanran Mao, Yongsuk Choi, Chuan Qian, Dong Gue Roe, Seonkwon Kim, Yuehong Liu, Diandian Chen, Dongsheng Tang, Jia Sun, Jeong Ho Cho
Abstract
With the advancement of wearable and mobile devices, demand for the real-time, low-power processing of physiological and environmental signals is growing rapidly. To achieve this, neuromorphic systems that employ artificial synapses for analog signal processing and parallel computing represent a promising strategy. In this study, a synaptic sensor is developed that simultaneously responds to human respiration and ambient ultraviolet (UV) light, enabling multimodal analog data processing. The proposed device is fabricated using the organic semiconductor 5,5'-Di(4-biphenylyl)-2,2'-bithiophene, which has distinct bulk and channel phases. Human respiration-induced airflow is converted into a synaptic current via charge trapping triggered by the interaction between molecules of water and the bulk phase, leading to real-time detection of the respiratory rate. The inherent photosensitivity of the device also allows for simultaneous UV detection, thus capturing the environmental exposure conditions. Using these multimodal sensing and processing capabilities, a real-time feedback system is implemented that supports exercise monitoring by integrating physiological and environmental information. This work demonstrates the potential use of synaptic sensors as front-end components in wearable neuromorphic platforms, offering a compact, energy-efficient, and intelligent interface for healthcare and personalized information services.