Japanese scientists have successfully developed a multi-mode smart sensor patch that is able to monitor users' health symptoms, such as arrhythmias, coughs, and fall risks, through edge computing mode. The smart sensor patch incorporates a Bluetooth module to connect with a smartphone and perform real-time data processing, eliminating the need for on-grid or cloud-based data analysis. Increasingly, edge computing is being used on smartphones to analyze data collected by multimodal flexible wearable sensor patches capable of detecting arrhythmias, coughs, and falls.
Wearable sensors are devices that can be worn on the body to measure physical status. They fall under the umbrella of the Internet of Things (IoT) and show great potential for health monitoring. These sensors generate a lot of data that needs to be processed before it can be understood. The field of computing that processes this data, i.e., processing on the sensor itself or on the device connected to it, rather than on a remote cloud server, is known as edge computing. Edge computing is a key component of wearable sensor technology.
"The goal of our research is to design a multimodal sensor patch that can process and interpret data using edge computing to detect early disease symptoms in everyday life," explains Professor Takei.
Figure: The Future of Healthcare: The Combination of Smart Sensor Patches and Edge Computing
The research team made sensors that can monitor heart activity (via ECG), breathing, skin temperature, and humidity due to sweating. After confirming that these sensors are suitable for long-term use, they are integrated into a flexible film (sensor patch) that adheres to the skin. The sensor patch also contains a Bluetooth module for easy connection to a smartphone.
The research team first tested the ability of the sensor patch to detect physiological changes in three volunteers, who wore the patch on their chest. The sensor patch monitors the volunteers' vital signs at wet-bulb temperatures (used to determine the likelihood of heat stress) above 22°C and 29°C. "Although we had a smaller test group, we observed changes in their vital signs at high temperatures. This observation may ultimately help identify the symptoms of early heat stress," says Professor Takei.
The team also developed a machine learning program that processes the recorded data to detect other symptoms such as arrhythmias, coughs, and falls. "In addition to the analysis on the computer," adds Associate Professor Nakajima, "we have also designed an edge computing application that can perform the same analysis on a smartphone." We achieved more than 80% accuracy in our predictions. ”
"Important advances in this research lie in the integration of multimodal flexible sensors, real-time machine learning data analysis, and remote vital monitoring via smartphones," Professor Takei concluded. "One of the drawbacks of our system is that model training cannot be done on a smartphone, but must be done on a computer; However, this problem can be solved by simplifying data processing. "This research is driving the development of patch-based edge computing systems in the field of telemedicine or remote diagnosis.
