Flexible artificial intelligence optoelectronic sensors

Artificial intelligence (AI) is known for its high energy consumption, especially in data-intensive tasks such as health monitoring. To address this, researchers at Tokyo University of Science (TUS) have developed a flexible paper-based sensor composed of nanocellulose and zinc oxide (ZnO) nanoparticles operating like the human eyes and brain. The sensor is energy-efficient, responds to optical input in real-time, and is flexible and easy to dispose of.

AI’s capability comes at a very high energy cost, with estimates indicating that training OPEN AI’s GPT-3 model consumed more than 1,287MWh, enough to supply an average U.S. household for 120 years. This energy cost poses a substantial roadblock, particularly for using AI in large-scale applications such as health monitoring where large amounts of critical health information are sent to centralized data centers for processing.

Achieving AI-based health monitoring and biological diagnosis requires a standalone sensor operating without constant connection to a central server. Additionally, the sensor must have a low power consumption for prolonged use, should be capable of handling rapidly changing biological signals for real-time monitoring, be flexible enough to attach comfortably to the human body, and be easy to make and dispose of for hygiene reasons.

Considering these criteria, TUS researchers led by Associate Professor Takashi Ikuno developed a flexible paper-based sensor operating like the human brain. Their findings were published in the journal Advanced Electronic Materials.

“A paper-based optoelectronic synaptic device composed of nanocellulose and ZnO was developed for realizing physical reservoir computing. This device exhibits synaptic behavior and cognitive tasks at a suitable timescale for health monitoring,” Ikuno says.

In the human brain, information travels between networks of neurons through synapses. Each neuron can process information on its own, enabling the brain to handle multiple tasks at the same time. To mimic this capability, the researchers fabricated a photo-electronic artificial synapse device composed of gold electrodes on top of a 10µm transparent film consisting of ZnO nanoparticles and cellulose nanofibers (CNFs).

The transparent film allows light to pass through and handle optical input signals representing various biological information. The cellulose nanofibers impart flexibility and can be easily disposed of by incineration, and the ZnO nanoparticles are photoresponsive and generate a photocurrent when exposed to pulsed UV light and a constant voltage. This photocurrent mimics the responses transmitted by synapsis in the human brain, enabling the device to interpret and process biological information received from optical sensors.

The film could distinguish 4-bit input optical pulses and generate distinct currents in response to time-series optical input, with a rapid response time on the order of sub-seconds. This quick response is crucial for detecting sudden changes or abnormalities in health-related signals. Furthermore, when exposed to two successive light pulses, the electrical current response was stronger for the second pulse. This behavior termed post-potentiation facilitation contributes to short-term memory processes in the brain and enhances the synapses’ ability to detect and respond to familiar patterns.

To test this, the researchers converted a dataset of handwritten digits into 4-bit optical pulses. They then irradiated the film with these pulses and measured the current response. Using this data as input, a neural network recognized handwritten numbers with an accuracy of 88%.

Remarkably, this handwritten-digit recognition capability remained unaffected even when the device was repeatedly bent and stretched up to 1,000x.