Snoring and breath-stopping are key symptoms of sleep apnea. Most existing studies primarily focus on wearable devices or smartphone-based systems. Wearable devices can be uncomfortable, while smartphone-based systems often require specific angles, distances, or positions, making them sensitive to environmental changes. This paper proposes a contactless and engagement-free system for snoring and breath-stopping detection using a seismic sensor. Distinguishing between snoring, breath-stopping, and normal breathing from raw data alone is challenging. Snoring features typically reside in a higher frequency range than breath-stopping and normal breathing, with breath-stopping features appearing in a lower frequency range. Calculating the differential and integral of raw data can enhance features in low and high frequencies, respectively. We introduce AFFCL, an attention feature fusion and contrastive learning framework to leverage information from differential and integral signals. AFFCL generates both shared and exclusive features from the differential and integral signals and employs an attention mechanism for feature fusion. Additionally, cluster-level supervised contrastive learning in AFFCL further enhances system performance. Our system has been performed 5-fold cross-validation on 44 people, which achieves an average accuracy of 93.40% and an F1 score of 92.42%. The accuracy for detecting breath-stopping, snoring, and normal breathing are 89.54%, 94.60%, and 96.06%, respectively. Evaluation results demonstrate that our system effectively identifies breath-stopping and snoring.

Monitoring vital signs, such as heart rate (HR) and respiratory rate (RR) is critical for veterinary medicine. The existing contact based systems are difficult to use on animals for longer periods as they may cause movement restriction. Contactless solutions have recently gained more popularity due to their ease-of-use. However, there are very few validated systems for animals. In this study, we propose a contactless vital signs monitoring system, CageDot for animals. Our system provides continuous real-time monitoring of HR and RR during hospitalization. The CageDot is placed under the animal cage and detects the heart vibrations using a geophone sensor. The CageDot also has a signal quality control algorithm to address the problem of obtaining high-quality cardiac data in real-life noisy environments, such as hospitals. Compared to the existing works that use controlled environments, this is especially significant. The algorithm includes several steps that include background noise/movement removal, subject movement detection, heartbeat extraction, and vital signs estimation. The experimental results on 16 hospitalized dogs and cats show that this system can achieve high accuracy for vital signs monitoring with a mean absolute error (MAE) of 4.71 for HR (3.8% error rate) and 1.28 (8% error rate) for RR.

The evolution of technology has paved way for smart and improved healthcare. Invasive and contact-based methods, that have been the gold-standard for healthcare monitoring, are being replaced by contact-less methods to enable engagement-free and continuous remote patient monitoring. These methods are getting more and more popular, not just for humans, but for animals too. In our study, we develop an innovative contact-less monitoring system, CageDot, that uses geophone sensors for tracking animal activities and following infectious disease progress. The activity levels of the animals can act as a surrogate for animal health and help to document illness and recovery with treatment. Our proposed system was extensively tested by performing three multi-week experiments over a period of 1.5 years. The results show that the CageDot system can accurately and continuously track the animal activities and follow infectious disease progress in a non-invasive and contact-free way. This has a significant application and impact in the future drug and vaccine development, because the existing approach relies on daily health check by humans, which is time consuming and not continuous that may potentially skip important events.

Health monitoring is essential for both humans and animals in daily life. While numerous health monitoring systems have been developed, the majority are designed exclusively for either humans or animals, and most require direct physical contact. We have developed BedDot, a contactless health monitoring system for both humans and animals using a seismic sensor. BedDot can be deployed in various environments, such as bed and seat settings for humans, as well as in cages for animals, to monitor occupancy, heart rate (HR), respiratory rate (RR), and blood pressure (BP). Our system demonstrates high accuracy in the clinical experiments of 150 patients and 16 dogs and cats.

COVID-19 has devastated the entire world for the past couple of years. Timely and efficient detection and identification of a virus are crucial in preventing the wider virus spread. By using intelligent sensors based on Surface-Enhanced Raman Scattering (SERS), it is possible to detect and identify virus automatically. In this study, we successfully applied the XGBoost Algorithm (Supervised Machine Learning) to classify the type of the virus using the SERS sensor data. The supervised approach has a limitation when a new type of virus arises, whose shape is different from the previously known samples. To tackle this problem, we investigated the unsupervised learning approaches that can cluster the virus data into different groups without labeled data. The unsupervised approach presented in this paper is called k-Shape Clustering. This technique compares the cross-correlation between different samples and then clusters them into similar or different groups. If a subvariant of a virus emerges, it would be clustered into the existing virus groups; if a new type of virus is found, it would be clustered into a new group. Both of the approaches have shown very promising results based on extensive evaluations.