Abstract
The Internet of Things (IoT) is a trending model in the wake of recent advancements in ubiquitous sensors and smart devices, and is rapidly being deployed in communications, infrastructure, transportation and healthcare services. The Internet of Medical Things (IoMT) is a subset of the IoT and provides a layered architecture for connecting individuals with mobile devices and wearables, such that their vital physiological data can be captured and analyzed non-invasively using smart sensors embedded within these devices. Currently available wearables have embedded sensing modules for measuring movement, direction, light and pressure. Actigraphs are one such type of wearables which exclusively employ the use of accelerometers for capturing human movement-based vibration data. The main intention of this research work is the analysis of unstructured, non-stationary actigraphy signals. This study aims to address three key objectives: (i) enabling compression and denoising of actigraphy data during acquisition; (ii) extracting regions of interest from the actigraphy data, and; (iii) deriving actigraphy specific features for improving the activity classification accuracy. These have been achieved through three key contributions, namely: (A) a signal encoding framework for data compression and denoising, (B) two novel adaptive segmentation schemes which help in extracting specific movement information from actigraphy data, and (C) two key actigraphy specific features, which quantify limb movements, and hence provide a better classification accuracy using machine learning algorithms. The outcome of this research work is a device-independent actigraphy analysis application for estimating the severity of neuromuscular diseases, identifying types of daily activity, and classify joint degeneration severity, which has been tested on five different actigraphy datasets from wake and sleep states. Compared to traditional signal processing methods, the proposed algorithms ensured a 20-90% increase in SNR (signal-to-noise ratio), 50-80% reduction in bit rate, 50-90% data compression, and an increase in activity recognition accuracy by over 5-20%. Results from this systematic investigation indicate that data analysis conducted at the acquisition source, optimizes signal denoising, memory and power usage, and activity recognition, thereby promoting an edge computing approach to physiological signal analysis using wearables in a low resource environment.
Highlights
The idea of connecting human civilization on a global scale gave birth to telephony in the 1920s
Considering that this research is based on actigraphy-based wearables, we first explore the basic concepts of Internet of Things (IoT), examine its definition, structure and describe various layers of its architecture
This chapter begins by applying an initial set of signal analysis algorithms to short-duration, single-axial actigraphy data to test the efficacy of assessing actigraphy signals for estimating the severity of periodic limb movements (PLMs) in sleep
Summary
The idea of connecting human civilization on a global scale gave birth to telephony in the 1920s. Following the development of commercial websites and systems in early 90s, we have witnessed rapid growth in the development of mobile phones, music players, laptops, tablets, smart phones and their related data encoding and communication protocols This has started a trend to standardize communication pathways between devices and users, and sensor design, depending on specific applications. The algorithms installed on these devices, analyze the input signals for generating user-friendly feedback such as step count, calories burnt, miles ran and pulse rate variability, which is useful in maintaining individual fitness levels [35] Though these devices do not provide vital health information from sensor data, a systematic investigation indicates that motion data could be a primal contributing factor in assessing movement-related disorders or disabilities, and could be used for developing surrogate markers for detecting or predicting the onset of various diseases such as Parkinson’s disease [57]
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.