From creating images, generating text, and enabling self-driving cars, the potential uses of artificial intelligence (AI) are vast and transformative. Nonetheless, all this capability comes at a really high energy cost. As an illustration, estimates indicate that training OPEN AI’s popular GPT-3 model consumed over 1,287 MWh, enough to produce a median U.S. household for 120 years. This energy cost poses a considerable roadblock, particularly for using AI in large-scale applications like health monitoring where large amounts of critical health information are sent to centralized data centers for processing. This not only consumes a number of energy but additionally raises concerns about sustainability, bandwidth overload, and communication delays.
Achieving AI-based health monitoring and biological diagnosis requires a standalone sensor that operates independently without the necessity for constant connection to a central server. At the identical time, the sensor should have a low power consumption for prolonged use, ought to be able to handling the rapidly changing biological signals for real-time monitoring, be flexible enough to connect comfortably to the human body, and be easy to make and eliminate as a consequence of the necessity for frequent replacements for hygiene reasons.
Considering these criteria, researchers from Tokyo University of Science (TUS) led by Associate Professor Takashi Ikuno have developed a versatile paper-based sensor that operates just like the human brain. Their findings were published online within the journal Advanced Electronic Materialson 22 February 2024.
“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 an appropriate timescale for health monitoring,” says Dr. Ikuno.
Within the human brain, information travels between networks of neurons through synapses. Each neuron can process information by itself, enabling the brain to handle multiple tasks at the identical time. This ability for parallel processing makes the brain rather more efficient in comparison with traditional computing systems. To mimic this capability, the researchers fabricated a photo-electronic artificial synapse device composed of gold electrodes on top of a ten µm transparent film consisting of zinc oxide (ZnO) nanoparticles and cellulose nanofibers (CNFs).
The transparent film serves three foremost purposes. Firstly, it allows light to go through, enabling it to handle optical input signals representing various biological information. Secondly, the cellulose nanofibers impart flexibility and may be easily disposed of by incineration. Thirdly, the ZnO nanoparticles are photoresponsive and generate a photocurrent when exposed to pulsed UV light and a continuing voltage. This photocurrent mimics the responses transmitted by synapsis within the human brain, enabling the device to interpret and process biological information received from optical sensors.
Notably, the film was able to tell apart 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 subseconds. This quick response is crucial for detecting sudden changes or abnormalities in health-related signals. Moreover, when exposed to 2 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 within the brain and enhances the power of synapses to detect and reply to familiar patterns.
To check this, the researchers converted MNIST images, a dataset of handwritten digits, into 4-bit optical pulses. They then irradiated the film with these pulses and measured the present response. Using this data as input, a neural network was capable of recognize handwritten numbers with an accuracy of 88%.
Remarkably, this handwritten-digit recognition capability remained unaffected even when the device was repeatedly bent and stretched as much as 1,000 times, demonstrating its ruggedness and feasibility for repeated use. “This study highlights the potential of embedding semiconductor nanoparticles in flexible CNF movies to be used as flexible synaptic devices for PRC,” concludes Dr. Ikuno.
Allow us to hope that these advancements pave the way in which for wearable sensors in health monitoring applications!
