Noncontact Monitoring of Dehydration Using RF Data Collected Off the Chest and the Hand

• Published in: IEEE sensors journal Vol. 24; no. 3; pp. 3574 - 3582
• Main Authors: Buttar, Hasan Mujtaba, Pervez, Kawish, Rahman, Muhammad Mahboob Ur, Mian, Adnan Noor, Riaz, Kashif, Abbasi, Qammer H.
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.02.2024; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Accuracy; Biomedical monitoring; Blood flow; Classifiers; COVID-19; Data collection; Decision trees; Dehydration; Frequency response; Keywords COVID-19; Machine learning; machine learning (ML); Monitoring; Neural networks; noncontact methods; OFDM; Orthogonal Frequency Division Multiplexing; Radio frequency; Radio signals; radio-frequency (RF)-based methods; RF signals; Sensors; Software radio; software-defined radio (SDR); Support vector machines
• ISSN: 1530-437X; 1558-1748
• DOI: 10.1109/JSEN.2023.3334590

We report a novel noncontact method for dehydration monitoring. We utilize a transmit software-defined radio (SDR) that impinges a wideband radio-frequency (RF) signal (of frequency 5.23 GHz) onto either the chest or the hand of a subject who sits nearby. Furthermore, another SDR in the closed vicinity collects the reflected RF signals. The two SDRs exchange orthogonal frequency division multiplexing (OFDM) signal, whose individual subcarriers get modulated once it reflects off (passes through) the chest (the hand) of the subject. This way, the signal collected by the receive SDR consists of channel frequency response (CFR) that captures the variation in the blood osmolality due to dehydration. The received raw CFR data is then passed through a handful of machine-learning (ML) classifiers that classify each subject as either hydrated or dehydrated. To train our ML classifiers, we have constructed our custom dataset by collecting data from five Muslim subjects who were fasting during the month of Ramadan. Specifically, we have implemented and tested the following ML classifiers: <inline-formula> <tex-math notation="LaTeX">{k} </tex-math></inline-formula>-nearest neighbor (KNN), support vector machines (SVMs), decision tree (DT), ensemble classifier, and a neural network (NN) classifier. Among all the classifiers, the neural network classifier achieved the best classification accuracy, that is, an accuracy of 93.8% (96.15%) for the proposed chest-based (hand-based) method. Compared to the prior contact-based method where the reported accuracy is 97.83%, our proposed noncontact method provides slightly less accuracy than that reported in the literature for the contact-based method; nevertheless, the advantages of our noncontact dehydration method speak for themselves.