Advantages:
- An AI-assisted contactless device that measures a patient’s breathing pattern and provides early and rapid detection of COVID-19 respiratory symptoms, in under 5 minutes.
- The ultrahigh sensitivity enables real-time monitoring of abnormal variations in a patient’s breath across multiple stages of disease.
- The portable device that facilitates effective screening and tracking in both clinical and remote environments. It can be utilized for monitoring apnea during sleep.
Summary:
The COVID-19 pandemic presents a significant public health challenge. Accurate, rapid detection and real-time monitoring of symptoms like shortness of breath, cough, and fever are crucial. However, current respiratory diagnostic devices often deliver inaccurate or limited results, have lengthy processing times, or need physical contact with patients. Thus, there's an urgent need for contactless devices that enable early and fast detection and real-time monitoring of the disease's progression.
To address the need for rapid and accurate detection of COVID-19 and its variants, our researchers developed the contactless USF-MSML (Ultra-sensitive Magnetic Sensor with Machine Learning) detector. This innovative device uses an ultra-sensitive magnetic sensor and machine learning (ML) to monitor breathing symptoms, sleep patterns, and disease progression in real-time. Converting magnetic oscillations from breathing patterns into voltage changes enables ML-assisted analysis of one’s breath variation to identify respiratory diseases like COVID-19. Enhanced with predictive analytics, the device provides early, accurate screening within five minutes and real-time disease tracking, useful in both clinical and remote settings. The device successfully tested COVID-19 patients in a hospital in Vietnam. Our newly developed technology has shown the effectiveness of respiratory signal features in distinguishing COVID-19 and other respiratory patients from healthy individuals, making a significant contribution to global efforts in combating the pandemic.1
The image above provides an overview of the research framework and the COVID-19 diagnosis and monitoring system: (a) Schematic representation of the magnetic respiratory monitoring system capturing respiratory patterns using the Hall effect sensor in tandem with a small permanent magnet; (b) Breath test protocol utilized during data collection, incorporating three distinct breathing styles: normal breath, holding breath, and deep breath; (c) Signal processing methods and feature extraction used to analyze the acquired respiratory signals. (d) Illustration of ML model for COVID-19 diagnosis based on extracted respiratory features.1
Desired Partnerships:
- License
- Sponsored Research
- Co-Development
Reference:
D. Nguyen, P.K. Huynh, V.D.A. Bui, K.Y. Hwang, N. Jain, C. Nguyen, L.H.N. Minh, L.V. Truong, X.T. Nguyen, D.H. Nguyen, L.T. Dung, T.Q. Le, and M.H. Phan,
“Real-Time Magnetic Tracking and Diagnosis of COVID-19 via Machine Learning,”
npj Digital Medicine 2024 (under review); https://arxiv.org/abs/2311.00737.