A flexible wristband with a fingertip-inspired sensor and AI decoding can track breathing patterns with over 99.5% accuracy.

A lightweight, flexible wristband can now accurately decode breathing patterns by analyzing pulse signals, achieving over 99.5% accuracy with the help of a fingertip-inspired pressure sensor and deep learning. Developed by a team including researchers from North University of China and Xiamen University, this system offers an effortless and unobtrusive alternative to bulky chest straps or nasal devices, enabling continuous, real-time respiratory monitoring. This breakthrough could transform respiratory care, especially for individuals managing chronic conditions or recovering from illness.

Chronic Respiratory Diseases and Monitoring Challenges

Chronic respiratory diseases affect millions globally and require constant monitoring to manage symptoms and prevent complications. However, existing technologies – such as chest belts or nasal prongs – can be intrusive and unsuitable for long-term daily use. Recent approaches have explored indirect signals like heart rate or pulse waveforms, offering better comfort but often sacrificing accuracy.

Traditional signal classification methods fall short when faced with the complexity of human respiration. Due to these challenges, there is a growing demand for devices that are not only comfortable to wear but also capable of extracting high-fidelity respiratory data over extended periods. Based on these challenges, there is a pressing need to develop advanced systems for long-term, precise respiratory monitoring.

Innovative Wearable System With Artificial Intelligence

Researchers from North University of China and Xiamen University have unveiled a novel wearable system that reads respiratory patterns directly from wrist pulse signals. Published in Microsystems & Nanoengineering, the study details a miniaturized, AI-powered device that integrates a flexible pressure sensor with a deep neural network. The sensor system enables continuous respiratory tracking with high comfort and clinical-grade precision, offering a promising alternative to traditional respiratory monitors.

Fingertip-Inspired Sensor and Deep Neural Network Design

At the heart of the system lies a 300-μm-thick flexible pressure sensor, inspired by the structure of a human fingertip. Printed on a thermoplastic polyurethane (TPU) substrate, the sensor is capable of detecting subtle fluctuations in pulse waves caused by respiration.

These signals-categorized as respiration-induced amplitude variation (RIAV), respiration-induced fluctuation in ventricular filling (RIFV), and respiration-induced variation in baseline (RIIV)-are transmitted to a mobile application via Bluetooth and processed using a hybrid Residual Network – Bidirectional Long Short-Term Memory (ResNet-BiLSTM) neural network. This deep learning model captures the temporal and spatial dynamics of respiratory patterns, classifying slow, normal, fast, and simulated breathing states with a remarkable 99.5% accuracy.

Lightweight Design and Real-World Testing

The device’s ultra-lightweight construction (just 9 grams), skin-conforming design, and long-term mechanical stability allow it to be worn for hours without discomfort.

Testing involved 13 human volunteers and machine-simulated breathing, demonstrating the sensor’s robustness across real and artificial respiratory scenarios. By avoiding reliance on chest placement or airflow proximity, the device simplifies setup and enhances usability-making it ideal for users in everyday environments.

Potential Impact on Chronic Disease Management

“Our mission was to bridge the gap between high-precision monitoring and wearable comfort,” said Dr. Libo Gao, senior author of the study. “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.”

The device’s ability to deliver accurate respiratory insights in a comfortable, wearable format opens new doors for chronic disease management, eldercare, and telemedicine. Its seamless integration with mobile platforms allows for real-time alerts and long-term data logging-tools that are essential for early intervention in conditions like COPD or sleep apnea. Beyond healthcare, this innovation could also benefit athletes, astronauts, or high-altitude workers by offering continuous respiration tracking in dynamic environments. As wearable tech evolves, this wrist-worn system could become a cornerstone in the future of personalized respiratory health monitoring.