Energy-efficient AI health monitoring Flexible sensor on paper base works like a human brain

Related Vendors

FAULHABER Drive Systems

Wuxi InfiMotion Technology Co., Ltd.

Researchers at the Tokyo University of Science have designed a paper-based sensor that not only works like a human brain but also addresses issues such as energy efficiency, sustainability, bandwidth congestion, and communication delays. The sensor has potential for use in health monitoring.

AI, used in the most diverse applications, opens up possibilities for diagnosis and identification that researchers and developers could not have even dreamed of a few years or decades ago. In terms of medicine and healthcare, AI can help diagnose diseases faster, create personalized treatment plans, and analyze medical images. Advances in AI enable the discovery of new drugs and more precise predictions of disease progression. However, this requires an appropriate data situation that is not only extensive but also up-to-date.

However, artificial intelligence does not come with advantages only. The use of AI in and for all conceivable situations raises questions regarding energy efficiency, sustainability, bandwidth usage, and communication reliability when deployed in large-scale health monitoring applications. After all, such use continually sends large amounts of critical health data for processing to central data centers.

AI-supported health monitoring with obstacles

For AI-supported health monitoring and biological diagnosis, a stand-alone sensor is required that operates independently and does not need to be constantly connected to a central server, analyzed members of the Tokyo University of Science. Additionally, the sensor must have low power consumption for extended use, process quickly changing biological signals in real-time, be flexibly attached to the human body, and, from a hygiene perspective, be easily manufactured and disposed of.

The researchers Komatsu, Nosoda, Tokiwa, and Ikuno took on this task and developed a flexible sensor on a paper basis that works like the human brain. The results of their work can be viewed online in the journal Advanced Electronic Materials.

"A paper-based optoelectronic synaptic device, made of nanocellulose and ZnO, was developed to enable Physical Reservoir Computing. This device demonstrates synaptic behavior and cognitive tasks on a timescale suitable for health monitoring," said Dr. Ikuno.

Efficiency like in the human brain

In the human brain, information is transmitted between networks of neurons via synapses. Each neuron can process information on its own, allowing the brain to perform multiple tasks simultaneously. This capability for parallel processing makes the brain much more efficient compared to conventional computer systems, especially in terms of energy consumption.

To mimic this capability, the researchers manufactured a photoelectronic artificial synapse consisting of gold electrodes on a 10 µm thick transparent film made of zinc oxide (ZnO) nanoparticles and cellulose nanofibers (CNFs). The transparent film serves three purposes, the researchers explain. First, it processes optical input signals that represent various biological information. Second, cellulose nanofibers provide flexibility and facilitate disposal by burning. Third, ZnO nanoparticles generate a photocurrent that allows similar reactions to synapses in the human brain, enabling the interpretation and processing of optical signals.

Flexible, synaptic devices

The flexible film demonstrated the ability to distinguish 4-bit optical input pulses and respond quickly. This rapid response time is crucial for detecting sudden changes in health-related signals. Moreover, the reaction to the electrical current increased with repeated light pulses, contributing to improved short-term memory behavior similar to that in the brain. The device could also reliably recognize handwritten digits even after repeated bending and stretching, highlighting its robustness and deployability. "This study underscores the potential of embedding semiconductor nanoparticles in flexible CNF films for use as flexible synaptic devices for PRC," said Dr. Ikuno. (sb)