Deep Learning, Machine Learning and IoT in Biomedical and Health Informatics

Techniques and Applications

Sujata Dash, Subhendu Kumar Pani, Joel J. P. C. Rodrigues, Babita Majhi, Sujata Dash, Subhendu Kumar Pani, Joel J. P. C. Rodrigues, Babita Majhi

• Éditeur: CRC Press
• Année: 2022
• ISBN:
  9781000534054
• Édition: 1
• Sujet: Computer Science
• Sous-sujet: Computer Science General

Biomedical and Health Informatics is an important field that brings tremendous opportunities and helps address challenges due to an abundance of available biomedical data. This book examines and demonstrates state-of-the-art approaches for IoT and Machine Learning based biomedical and health related applications. This book aims to provide computational methods for accumulating, updating and changing knowledge in intelligent systems and particularly learning mechanisms that help us to induce knowledge from the data. It is helpful in cases where direct algorithmic solutions are unavailable, there is lack of formal models, or the knowledge about the application domain is inadequately defined. In the future IoT has the impending capability to change the way we work and live. These computing methods also play a significant role in design and optimization in diverse engineering disciplines. With the influence and the development of the IoT concept, the need for AI (artificial intelligence) techniques has become more significant than ever. The aim of these techniques is to accept imprecision, uncertainties and approximations to get a rapid solution. However, recent advancements in representation of intelligent IoTsystems generate a more intelligent and robust system providing a human interpretable, low-cost, and approximate solution. Intelligent IoT systems have demonstrated great performance to a variety of areas including big data analytics, time series, biomedical and health informatics. This book will be very beneficial for the new researchers and practitioners working in the biomedical and healthcare fields to quickly know the best performing methods. It will also be suitable for a wide range of readers who may not be scientists but who are also interested in the practice of such areas as medical image retrieval, brain image segmentation, among others.

• Discusses deep learning, IoT, machine learning, and biomedical data analysis with broad coverage of basic scientific applications

• Presents deep learning and the tremendous improvement in accuracy, robustness, and cross- language generalizability it has over conventional approaches

• Discusses various techniques of IoT systems for healthcare data analytics

• Provides state-of-the-art methods of deep learning, machine learning and IoT in biomedical and health informatics

• Focuses more on the application of algorithms in various real life biomedical and engineering problems

Part I Machine Learning Techniques in Biomedical and Health Informatics

1 Effect of Socio-economic and Environmental Factors on the Growth Rate of COVID-19 with an Overview of Speech Data for Its Early Diagnosis

Soumya Mishra, Tusar Kanti Dash, and Ganapati Panda

CV Raman Global University, Bhubaneswar, India

DOI: 10.1201/9780367548445-2

Contents

1. 1.1 Introduction
   • 1.1.1 Motivation and Research Objective
2. 1.2 Databases and Socioeconomic, Environmental Features
   • 1.2.1 Temperature (f1)
   • 1.2.2 Happiness Index (f2)
   • 1.2.3 Cleanliness Index (f3)
   • 1.2.4 Gross Domestic Product (f4)
   • 1.2.5 Pollution Index (f5)
   • 1.2.6 Number of Caregivers/Nurses per 1000 People (f6)
   • 1.2.7 Number of Physicians per 1000 People (f7)
   • 1.2.8 Diabetes Prevalence (f8)
   • 1.2.9 Population Aged over Sixty-five (f9)
   • 1.2.10 Smokers above Age Fifteen (f10)
3. 1.3 Growth Rate Calculation and Feature Selection
   • 1.3.1 Growth Rate Calculation
   • 1.3.2 Feature Selection
4. 1.4 COVID-19 Speech Analysis
5. 1.5 Conclusion
6. References

1.1 Introduction

A major concern for human health is the development of the 2019-nCoV virus in China. It has been reported as a pandemic by the World Health Organization (WHO) [1]. By 19th September 2020, there have been 30,295,744 humans affected by the 2019-nCoV [2]. The growth rate of the virus is rapid compared to two of its ancestors, SARS-CoV and Middle East respiratory syndrome coronavirus (MERS-CoV) [3]. Numerous researchers are working to identify the clinical predictors and features of mortality based on the analysis of data patients from Wuhan, China [4, 5]. The virus has become a challenge for critical care in the whole world, due to the small number of available resources [6]. In this section, a brief literature review of the COVID-19 growth rate analysis is presented.

The growth rate of this virus is varying from country to country. Research is ongoing to identify the factors that can influence the growth rate. In [7], the authors have given an analysis of the growth rate of COVID-19 on the Diamond Princess cruise ship and found out that the median (with 95% confidence) interval of the growth rate of COVID-19 was 2.28 during the early stage and that probable outbreak size is largely dependent on the change of growth rate. The growth rate of COVID-19, as compared with the SARS coronavirus, shows that the reproduction number of COVID-19 is considerably higher than the SARS coronavirus [8]. In another interesting paper [9], the authors have estimated the basic growth rate, and the infection, recovery, and mortality rates by using the Susceptible-Infected-Recovered-Dead model. They have found out that there is a gap between the actual and the reported cases. The numbers of infected cases are twenty times more than the reported ones. Similarly, the recovered cases are forty times higher as compared to the reported cases. For an estimation of the growth pattern of the COVID 2019, a novel fractional time delays dynamic system with fractional derivative is used in [10]. The simulation results proved that the proposed model provides a satisfactory estimation of the available actual data. In [11], Susceptible-Exposed-Infected-Removed model is used to generate the dynamics of the widespread diffusion of COVID-19 by incorporating the daily intercity migration data of China. The factors used in the analysis are the rate of infection and recovery, with the final percentage of the infested residents for more than 350 cities in China, and predicted that the growth rate will crown from the middle of February to the initial week of March 2020 for China.

The three mathematical models (i.e., Logistic model, Bertalanffy model, and Gompertz model) are used in [12] for analysis of the number of people expected to be affected by COVID-19 in Wuhan and the non-Hubei areas of China. They have predicted that the COVID-19 infection will be over by late April 2020 in Wuhan and by the end of March 2020 in Non-Hubei areas. In [13], the multivariate logistic regression and sensitivity analyses were carried out to identify the risk factors for developing severe Novel Coronavirus Pneumonia and found out that the early admission and surveillance by CT should be used for the improvement of clinical outcomes. By using the susceptible-exposed-infected-removed compartment model, the growth rate of COVID-19 was calculated in [14] and found out to be between 2.8 and 3.3 for China and between 3.2 and 3.9 for the international cases. The simulation results proved that the proposed model provides a satisfactory estimation of the available actual data.

1.1.1 Motivation and Research Objective

From the brief literature review, it has been observed that in the recent past, research has been mainly carried out on the influencing factors of the medical field, but very little data is available on the analysis of the socio-economic and environmental factors. To provide an accurate analysis of the effect of these factors on the growth rate of COVID-19, the research has been carried out in this chapter. The main research objectives are:

• Calculation of the growth rate of COVID-19 for different countries and classification of them into the low, medium, and high categories.
• Preparation of the Socioeconomic and Environmental factors based on the data set of different countries.
• Application of feature selection algorithms to identify the importance of each of these factors on the growth rate of COVID-19
• Analysis of different techniques and databases used for Non-invasive COVID-19 detection using speech signals

The remaining sections of the chapter have been organized as follows: Section 1.2 deals with the preparation of the database and extraction of the Socioeconomic and Environmental features. The Growth rate calculation and feature selection methods are dealt with in Section 1.3. In Section 1.4, speech-based COVID-19 detection is analyzed in detail. Section 1.5 concludes the chapter.

1.2 Databases and Socioeconomic, Environmental Features

In this Section, the details of the collection and preparation of databases along with the Socioeconomic, Environmental Features are discussed. Several online resources are used for this purpose [15–24]. The details are listed in Table 1.1 Thus, the factors possibly influencing lung health, and immunity have been collected.

Table 1.1 Details of Features and Data Collection Sources

Sl. No	Feature Name...

Table des matières

• Cover Page
• Half Title Page
• Series Page
• Title Page
• Copyright Page
• Contents
• Preface
• Acknowledgements
• Editors
• Contributors
• Part I Machine Learning Techniques in Biomedical and Health Informatics
• Part II Deep Learning Techniques in Biomedical and Health Informatics
• Part III Internet of Things (IoT) in Biomedical and Health Informatics
• Index