Paper-Based Sensor Paves Way for Wearable Sensors in Health Monitoring

Wearable artificial intelligence (AI) sensors that possess sensing and cognitive abilities are attracting significant interest in health monitoring. The development of self-powered AI sensors that function efficiently and have similar low energy consumption as the human brain is essential. Physical reservoir computing (PRC), which employs physical phenomena to mimic brain functions, provides a solution for energy-efficient architecture. However, developing flexible, disposable sensors using PRC that can process optical signals with sub-second response times for biological applications has remained a challenge. Researchers have now developed disposable, flexible paper-based optoelectronic devices using nanocellulose and zinc oxide (ZnO) nanoparticles for PRC applications.

This flexible paper-based sensor developed by researchers from Tokyo University of Science (TUS, Tokyo, Japan) functions similarly to the human brain, where information is transferred across a network of neurons via synapses. Each neuron's ability to independently process information allows for simultaneous multiple task handling, making the brain more efficient than conventional computing systems. To mimic this capability, the researchers designed a photo-electronic artificial synapse device composed of gold electrodes on top of a 10 µm transparent film consisting of ZnO nanoparticles and cellulose nanofibers (CNFs). The transparent film allows light to pass through, enabling it to handle optical input signals representing various biological information. Moreover, the cellulose nanofibers impart flexibility to the film and can be easily disposed of by incineration. Additionally, the ZnO nanoparticles are photoresponsive and generate a photocurrent upon exposure to pulsed UV light and a constant voltage. This photocurrent mimics the responses transmitted by synapsis in the human brain, allowing the device to interpret and process biological information received from optical sensors

Remarkably, the device can differentiate between 4-bit input optical pulses and produce unique currents in response to time-series optical inputs, with sub-second response times essential for monitoring health signals. Furthermore, when exposed to two successive light pulses, the electrical current response was stronger for the second pulse. This feature, known as post-potentiation facilitation, aids in short-term memory processes in the brain and improves pattern recognition. The researchers tested this by converting MNIST images, a dataset of handwritten digits, into 4-bit optical pulses and then irradiating the film with these pulses and measuring the current response. Using this data as input, a neural network recognized handwritten numbers with 88% accuracy. The researchers also confirmed the device's durability by performing repeated bending and stretching tests, and found no loss in recognition capability even after 1,000 cycles, thus demonstrating its resilience and potential for repeated use in health monitoring applications.

“A paper-based optoelectronic synaptic device composed of nanocellulose and ZnO was developed for realizing physical reservoir computing,” said TUS Associate Professor Takashi Ikuno who led the study. “This device exhibits synaptic behavior and cognitive tasks at a suitable timescale for health monitoring.”

Related Links:
TUS