A Featherweight Finger-Inspired Wearable Pressure Sensor Could Revolutionize Breath Tracking
Monitoring and diagnosis of respiratory conditions could get a lot easier, thanks to a team of researchers in China.
Researchers from the North University of China and Xiamen University have created a lightweight wearable designed to turn pressure readings from the wrist into accurate breath monitoring — taking inspiration from the sensitivity of the human finger.
"Our mission was to bridge the gap between high-precision monitoring and wearable comfort," explains senior and co-corresponding author Libo Gao of the team's work. "We've shown that you can track respiration with clinical accuracy — without putting anything on your chest or face. This could be a game-changer in how we approach remote health monitoring, especially for patients who need round-the-clock care."
Traditional approaches to monitoring a patient's respiration patterns over time rely on devices strapped to the chest or secured near the mouth or nose — neither of which is conductive to long-term monitoring or even short-term patient comfort. The researchers' proposed solution: a high-sensitivity yet robust pressure sensor that can be worn on the wrist, weighing just 9g (around 0.32oz).
The pressure sensor's design was inspired by that of the human fingertip, and is made with a flexble substrate onto which the patterned sensor is directly printed. It's designed not to measure respiration directly, which doesn't take place anywhere near the wrist, but its effects on pulse waves known as respiration-induced amplitude variation (RIAV), respiration-induced fluctuation in ventricular filling (RIFV), and respiration-induced variation in baseline (RIIV).
Data gathered by the sensor are transmitted to a nearby smartphone over Bluetooth by an Espressif ESP32 microcontroller and processed using a hybrid Residual Network - Bidirectional Long Short-Term Memory (ResNet-BiLSTM) neural network. This, the team says, can track breathing and classify it a range of states with a claimed 99.5 percent accuracy.
The researchers are positioning the sensor and machine learning model as ideal for diagnosing and monitoring patients with respiratory conditions, but also suggest it could go still further — proposing its use by athletes, high-altitude workers, and even astronauts for continuous respiration tacking even while working.
The team's work has been published in the journal Microsystems & Engineering under open-access terms.