Internet of Things in Biomedical Engineering

Valentina Emilia Balas, Le Hoang Son, Sudan Jha, Manju Khari, Raghvendra Kumar

• Publisher: Academic Press
• Year: 2019
• ISBN:
  9780128173572
• Topic: Sciences biologiques
• Subtopic: Biotechnologie

Internet of Things in Biomedical Engineering presents the most current research in Internet of Things (IoT) applications for clinical patient monitoring and treatment. The book takes a systems-level approach for both human-factors and the technical aspects of networking, databases and privacy. Sections delve into the latest advances and cutting-edge technologies, starting with an overview of the Internet of Things and biomedical engineering, as well as a focus on 'daily life.' Contributors from various experts then discuss 'computer assisted anthropology, ' CLOUDFALL, and image guided surgery, as well as bio-informatics and data mining.

This comprehensive coverage of the industry and technology is a perfect resource for students and researchers interested in the topic.

• Presents recent advances in IoT for biomedical engineering, covering biometrics, bioinformatics, artificial intelligence, computer vision and various network applications
• Discusses big data and data mining in healthcare and other IoT based biomedical data analysis
• Includes discussions on a variety of IoT applications and medical information systems
• Includes case studies and applications, as well as examples on how to automate data analysis with Perl R in IoT

Chapter 1

CLOUDFALL 1.0: A Smart, Low-Cost, IoT-Based Fall-Detection Sensor Node

Vikram Puri^⁎; Kalpna Gautam^†; Jolanda G. Tromp^⁎; Chung Van Le^⁎; Nidhi Sachdeva^‡; Tri Huu Minh Tran^{⁎ ⁎} DuyTan University, Da Nang, Vietnam
^† G.N.D.U Regional Campus, Jalandhar, India
^‡ Fairleigh Dickinson University, Vancouver, BC, Canada

Abstract

A fall can cause major problems if the injuries are left untreated too long and can possibly lead to serious brain or bone injuries, especially for the elderly. Injuries from a fall reduce the patient’s ability to enjoy social interaction and perform day-to-day activities; fear of future falls, isolation, and reduced independence can lead to depression. To reduce after-effects and injuries from falls, getting medical care as quickly as possible is vital. Smart IoT-based wearable sensors can provide a means to overcome the problems associated with a fall. Wearable sensor nodes make it possible to deliver a high degree of monitoring and reliability. This chapter focuses on the requirements and specifications of a low-cost, high-reliability, and energy-efficient sensor node, proposing a device to meet these requirements and exploring different parameters such as communication interface and processing speed. It also provides both hardware and software designs for a successful implementation of the wearable device in terms of quality of service (QoS). The design description provided clearly demonstrates that the proposed sensor node is energy efficient and better at detecting falls than other fall-detection systems.

Keywords

Internet of things (IoT); Fall detection; ThingSpeak; Sensor node; Wearable sensor

1.1 Introduction

Falls have become a major health issue for the elderly worldwide. According to the World Health Organization (WHO), the elderly segment of the population currently is 8.5% and it will be increased by 16.7% of the total population by 2050 [1]. With reference to these tendencies, many developing countries are putting health policies in place in an effort to make older people more active and independent [2]. In the same context, the emerging concept of active and healthy aging (AHA) has faced numerous challenges, but it also holds great possibilities for societal development over the next few decades. The idea behind AHA is broad, and involves improving the quality of life as well as increasing security for the elderly. A 30% increase has been reported in people suffering one or more falls per year from the total population. For people over the age of 80, this number reaches 50%, according to WHO. These falls are responsible for many hip and wrist fractures and sometimes lead to more serious problems like dehydration, head injuries, and even death [3].

Currently, many companies are working on solutions to the fall detection problem. Fall detection systems for seniors can be categorized into three types: (1) nonwearable-based system (NWS), (2) wearable-based system (WBS), and (3) hybrid-based system (HBS).

A nonwearable fall system is embedded in the environment with the support of a camera, depth-detection camera, and heat-sensing camera for the detection of falls [4]. The main problems in nonwearable systems are the high cost and lack of privacy for detecting falls. To eliminate these problems, a wearable-based system has been introduced. Usually, a wearable system makes use of an inertial accelerometer and a gyroscope directly attached to the body of the individual. In addition, digital accelerometers are being widely used because they provides many benefits, such as low power consumption, light weight, miniature size, and—most crucial—ease of operation. Some current studies deal with the implementation of machine learning (ML) in fall detection systems [5–8]. These works proposed a threshold-based algorithm to deal with the challenges such as computational cost and complexity and also to increase the accuracy of the system. ML is a computer science technique that involves using statistical models to produce an automated prediction system (Fig. 1.1).

Fig. 1.1 General architecture for wearable sensor.

In the general architecture of wearable sensors [9], the system is categorized into three parts: (1) the wearable sensor itself; (2) the smart IoT gateway, and (3) the web-server/user terminal. In the general architecture, a wearable sensor node is connected to fall detection sensors through a Wi-Fi module. The Wi-Fi module is further connected to the smart gateway for better communication between the sensors and the web server. The main role of the web server is to store, process, and analyze data and send the encoded data to user terminals for monitoring.

The main concept of this chapter is to design an internet-connected fall detection system. In addition, the chapter includes studies and comparisons of our proposed system in terms of energy efficiency and accuracy.

1.2 Related Studies

Shahzad and Kim [10] presented an android application called FallDroid, which is used to detect a fall and also sends an emergency alert. This application uses a two-step algorithm that has achieved excellent classification results on fall-like events. This application is useful for sending and storing data on cloud servers. Yacchirema et al. [11] ...

Table of contents

• Cover image
• Title page
• Table of Contents
• Copyright
• About the editors
• About the contributors
• Preface
• About the book
• Chapter 1: CLOUDFALL 1.0: A Smart, Low-Cost, IoT-Based Fall-Detection Sensor Node
• Chapter 2: Computer-Assisted Anthropology
• Chapter 3: Modeling a Fuzzy System for Diagnosis of Disease Syndromes of Traditional Vietnamese Medicine Combining Positive and Negative Rules
• Chapter 4: Image-Guided Surgery Through Internet of Things
• Chapter 5: IoT in Agriculture Investigation on Plant Diseases and Nutrient Level Using Image Analysis Techniques
• Chapter 6: Internet of Things in Healthcare: A Brief Overview
• Chapter 7: Artificial Intelligence Based Diagnostics, Therapeutics and Applications in Biomedical Engineering and Bioinformatics
• Chapter 8: Why Big Data and What Is It? Basic to Advanced Big Data Journey for the Medical Industry
• Chapter 9: Internet of Things Application in Life Sciences
• Chapter 10: Emerging Trends of IoT-Based Applications in Day-to-Day Life
• Chapter 11: Combining Predictive Analytics and Artificial Intelligence With Human Intelligence in IoT-Based Image-Guided Surgery
• Chapter 12: Internet of Things Technologies
• Chapter 13: Medical Big Data Mining and Processing in e-Healthcare
• Index