Fully Integrated Patch Based on Lamellar Porous Film Assisted GaN Optopairs for Wireless Intelligent Respiratory Monitoring.
Zecong LiuJunjie SuKemeng ZhouBinlu YuYuanjing LinKwai Hei LiPublished in: Nano letters (2023)
Respiratory pattern is one of the most crucial indicators for accessing human health, but there has been limited success in implementing fast-responsive, affordable, and miniaturized platforms with the capability for smart recognition. Herein, a fully integrated and flexible patch for wireless intelligent respiratory monitoring based on a lamellar porous film functionalized GaN optoelectronic chip with a desirable response to relative humidity (RH) variation is reported. The submillimeter-sized GaN device exhibits a high sensitivity of 13.2 nA/%RH at 2-70%RH and 61.5 nA/%RH at 70-90%RH, and a fast response/recovery time of 12.5 s/6 s. With the integration of a wireless data transmission module and the assistance of machine learning based on 1-D convolutional neural networks, seven breathing patterns are identified with an overall classification accuracy of >96%. This integrated and flexible on-mask sensing platform successfully demonstrates real-time and intelligent respiratory monitoring capability, showing great promise for practical healthcare applications.
Keyphrases
- machine learning
- human health
- convolutional neural network
- healthcare
- deep learning
- big data
- risk assessment
- respiratory tract
- high throughput
- artificial intelligence
- climate change
- room temperature
- low cost
- electronic health record
- highly efficient
- mass spectrometry
- circulating tumor cells
- social media
- cancer therapy
- quality improvement
- obstructive sleep apnea
- data analysis
- tandem mass spectrometry
- health insurance