Knowledge Modelling and Big Data Analytics in Healthcare

Advances and Applications

Mayuri Mehta, Kalpdrum Passi, Indranath Chatterjee, Rajan Patel, Mayuri Mehta, Kalpdrum Passi, Indranath Chatterjee, Rajan Patel

• Publisher: CRC Press
• Year: 2021
• Print ISBN:
  9780367696610
• eBook ISBN:
  9781000478068
• Topic: Medicine
• Subtopic: Data Mining

Knowledge Modelling and Big Data Analytics in Healthcare: Advances and Applications focuses on automated analytical techniques for healthcare applications used to extract knowledge from a vast amount of data. It brings together a variety of different aspects of the healthcare system and aids in the decision-making processes for healthcare professionals.

The editors connect four contemporary areas of research rarely brought together in one book: artificial intelligence, big data analytics, knowledge modelling, and healthcare. They present state-of-the-art research from the healthcare sector, including research on medical imaging, healthcare analysis, and the applications of artificial intelligence in drug discovery.

This book is intended for data scientists, academicians, and industry professionals in the healthcare sector.

Section I

Big Data in Healthcare

1

Intelligent Healthcare Systems

A Design Overview of Existing Systems and Technologies

S. Saravanan, R. Saminathan, and P. Anbalagan

DOI: 10.1201/9781003142751-2

CONTENTS

1.1 Introduction

1.2 Contributions in This Chapter

1.3 Healthcare Monitoring in IoT Using Big Data Analytics

1.4 Industrial IoT-Enabled Framework for Health Using the Cloud

1.5 Patient Health Monitoring System Based on the IoT

1.6 IoT Smart Health Monitoring System

1.7 IoT-Based Healthcare Monitoring System

1.8 Monitoring of Human Body Signal Based on LoRa Wireless Network System

1.9 A Framework for Healthcare System Using Cloud Computing

1.10 Wearable Devices

1.11 Precision Medicine and Health State Monitoring through Wearable Devices

1.12 Conclusion

References

1.1 INTRODUCTION

As the world population increases, chronic diseases and critical health issues also increase day by day. To seek out health problems, a smart health monitoring system should be developed throughout the world. In today’s world, people are busy with their work schedules; they do not have time to visit hospitals or healthcare providers. This affects people’s health status and creates greater problems. In order to improve the health status, a smart healthcare monitoring system should be installed for periodic health checkups (Al Brashdi et al. 2018). A smart system sends an alert to the concerned healthcare provider or hospital or to the concerned patient about the patient’s health status. The smart healthcare system (Figure 1.1) monitors the patient’s temperature, blood pressure, heart rate, respiration rate, oxygen rate through SpO2 and the like. If there is a change in these parameters, the alert short message service (SMS) will be transferred to the patient’s caretaker or doctors or healthcare provider. For an efficient, smart healthcare system, the Internet of Things (IoT) is implemented. Generally, IoT is an efficient and promising technology in this modern world. The IoT consists of a collection of physical devices, such as sensors, computing devices, and storage devices, that work based on the data collected from the sensors, and the data are stored in the form of a database (Dineshkumar et al. 2016; Hossain et al. 2016; Medvediev et al. 2018). The stored database in the IoT can be accessed and controlled anywhere anytime by Internet technologies. Table 1.1 depicts the pros and cons of the present healthcare system.

FIGURE 1.1 Intelligent Healthcare.

TABLE 1.1 Pros and Cons of Present–Day Healthcare System

S. No	Pros	Cons
1.	Downsizing	Hospital downsizing
2.	Cost reduction	Staff workload
3.	Reduce the cost of labor	No foresight in the management
4.	Reduced budget	No adequate care for patient

The IoT framework consists of three architectural layers, namely, (1) the IoT device layer, which is on the client side; (2) the IoT gateway layer, which is on the server side; and (3) the IoT platform layer, which works as a pathway between the clients and the operator. The building blocks of the IoT are sensors, actuators, data acquisition devices, preprocessing devices and cloud platform. The sensors in the system can be used either as an embedded device or as self-supporting devices that are used to collect the tele-metric data. There is another device called an actuator. Actuators are the devices that are used to convert physical movement of data into another form. When actuators are connected with sensors, the data collected are converted into physical data. Based on the collected data, the IoT device analyzes the data and orders the actuators to perform a certain task or activity. The data acquired from the sensors are data and resource constrained, and the data consume a large amount of power, which leads to device failure. Even though the sensors and the actuators are acting together (Ananth et al. 2019), the data acquisition process is the most important stage in the IoT, where the collection of data, filtration and storage-based platforms like cloud computing are used. Gateways in the IoT are used to convert the sensor data into some format that can be transferrable to some other format. Gateways are used to control, transfer, filter and select data to minimize the storage level in the cloud-based platform. Gateways act as local preprocessors of sensor data that are ready for further processing. The gateway acts as a security. The gateway is accountable for monitoring the data stream on both the client and server sides with the proper encryption and authentication. In the IoT, the data transfer speed can be enhanced by using edge computing (Shahidul Islam et al. 2019). Edge computing is used to speed up the data analysis in the IoT platform. In edge computing, the base station is close to the server, and it is easy to collect and process the data. This, in turn, reduces the power consumption of the network infrastructure (Wanjari et al. 2016).

In the IoT, the sensors and gateway act as the neurons and backbone of the system. The cloud acts as the brain of the system. To store and access the massive amount of data, a cloud-based platform or data centers are used. Present-day remote monitoring systems apply deep learning and artificial intelligence to categorize and classify health data. The IoT has a variety of applications, such as agriculture monitoring, health monitoring, traffic monitoring, smart grid and energy-saving management, water supply management, fleet management, maintenance management and the like (Cheol Jeong et al. 2018).

In the healthcare industry, the IoT plays an important and crucial role. The automated features of the IoT in the medical industry are called the Internet of Medical Things. There are two types of applications in the health industry: (1) E-health and (2) M-health. Generally, the Internet of Medical Things is defined as a group of medical devices like blood pressure sensors, pressure sensors and the like that connect to a network of computers through gateways. E-health is implemented with the help of electronic devices and communication. M-health, or mobile health, is defined as the sharing of medicine prescriptions through mobile applications. The basic device used in IoT is radio-frequency identification (RFID). The RFID automatically identifies and tracks an object with a tag. In the RFID method, the chip is enabled with an RFID tag to support the patients. The chip is implanted in the person’s body to monitor the health parameters.

In Software Development Life Cycle technology, two different phases are used: (1) the planning phase and (2) the implementation phase. It is a type of waterfall methodology. In this method, each stage is parted into small groups. Testing and implementation are carried out through each stage. Five different phases are implemented: (1) Planning: In planning phase, the requirements are gathered to reach the goal. (2) Analysis: There are two steps in this stage: the preliminary stage and system analysis. In the preliminary stage, the problem, objectives and goals should be defined. In system analysis, the information is gathered, interpreted and used for a diagnosis. (3) Design: In the design stage, the perception of each stage is visible to the users. (4) Implementation: In this stage, the program is developed. (5) Maintenance: The performance of the system is monitored continuously.

1.2 CONTRIBUTIONS IN THIS CHAPTER

• This chapter discusses state-of-the-art healthcare systems. It also explores the recent technologies used in healthcare systems.
• Furthermore, cloud platforms, which have deep support towards data population and organization, are also discussed.
• The industrial healthcare platforms used in the present-day market are also discussed, along with geriatric healthcare services.

1.3 HEALTHCARE MONITORING IN IoT USING BIG DATA ANALYTICS

In general, big data refers to the collection of a large volume of structured and unstructured data such as unstructured medical notes, medical imaging data, genetic data, DNA sequences and the patient’s behavioral data received from the sensors. Big data is used when large volumes of data cannot be processed by using traditional methods. The large volume of data is deployed in the cloud, which can be accessed anytime using the Internet (Viceconti et al. 2015). The definition of big data consists of three Vs: (1) Volume: The data can be collected from various sources, leading to storage insufficiency. By using a cloud platform like deep lakes and Hadoop, the burden of storage can be reduced. (2) Velocity: Due to the large accumulation of data, the speed of data is reduced. By using RFID tags and sensors, the speed of data transfer is increased in real time. (3) Variety: Data can be in different forms such as structured and unstructured formats. The main objective of the system is to collect data from the sensor an...

Table of contents

• Cover Page
• Half Title Page
• Title Page
• Copyright Page
• Contents Page
• Preface Page
• Editors Page
• Contributors Page
• Section I Big Data in Healthcare
• Section II Medical Imaging
• Section III Computational Genomics
• Section IV Applications in Clinical Diagnosis
• Section V Issues in Security and Informatics in Healthcare
• Index