Advanced AI Techniques and Applications in Bioinformatics

Loveleen Gaur, Arun Solanki, Samuel Fosso Wamba, Noor Zaman Jhanjhi, Loveleen Gaur, Arun Solanki, Samuel Fosso Wamba, Noor Zaman Jhanjhi

• Publisher: CRC Press
• Year: 2021
• ISBN:
  9781000463019
• Edition: 1
• Topic: Technology & Engineering
• Subtopic: Biomedical Science
• Index: Technology & Engineering

The advanced AI techniques are essential for resolving various problematic aspects emerging in the field of bioinformatics. This book covers the recent approaches in artificial intelligence and machine learning methods and their applications in Genome and Gene editing, cancer drug discovery classification, and the protein folding algorithms among others. Deep learning, which is widely used in image processing, is also applicable in bioinformatics as one of the most popular artificial intelligence approaches. The wide range of applications discussed in this book are an indispensable resource for computer scientists, engineers, biologists, mathematicians, physicians, and medical informaticists.

Features:

• Focusses on the cross-disciplinary relation between computer science and biology and the role of machine learning methods in resolving complex problems in bioinformatics

• Provides a comprehensive and balanced blend of topics and applications using various advanced algorithms

• Presents cutting-edge research methodologies in the area of AI methods when applied to bioinformatics and innovative solutions

• Discusses the AI/ML techniques, their use, and their potential for use in common and future bioinformatics applications

• Includes recent achievements in AI and bioinformatics contributed by a global team of researchers

1 An Artificial Intelligence-based Expert System for the Initial Screening of COVID-19

Anurag Sharma, Hitesh Marwaha, Vikrant Sharma and Love Kumar

DOI: 10.1201/9781003126164-1

Contents

1.1 Introduction

1.2 Review of Literature

1.3 Material and Method

1.3.1 Hierarchical Fuzzy System

1.3.2 Methodology

1.4 Results

1.4.1 Fuzzy Inference System

1.4.2 Membership Functions

1.4.3 Rule Editor

1.4.4 Fuzzification and Defuzzification

1.4.5 Rule Viewer

1.4.6 Surface Viewer

1.4.7 Graphical User Interface

1.5 Conclusion

Bibliography

1.1 Introduction

Diverse types of viruses exist; unlike other biological entities, several viruses, like poliovirus, have RNA genomes, herpes virus has DNA genomes, and the influenza virus has a single-stranded genome, while others like smallpox have double-stranded genomes [1]. Similarly, coronaviruses consist of a positive-sense single-stranded RNA genome and a nucleocapsid of helical symmetry. During the occurrence of an extremely infectious virus with human-to-human transmission, hospitals and doctors have increased workloads and inadequate resources to cure and hospitalize suspected patients. COVID-19 is a type of harmful virus which in severe cases results in death. There is no vaccine to date to cure this deadly virus. The only way to protect ourselves from COVID-19 is to quarantine ourselves and maintain social distance. An increased number of patients with coronavirus were diagnosed in Wuhan, Hubei Province, China from December 2019. Apart from China, COVID-19 cases were also reported in other parts of the world and it became a pandemic. WHO was informed of patients with pneumonia of unidentified etiology found in the city of Wuhan in China on 31st December 2019. By 3rd January 2020, a total of 44 patients suffering from pneumonia were reported by WHO. In this report, the causative agent was unidentified [2].

On 11th and 12th January 2020, WHO acknowledged a thorough report from the National Health Commission in China that stated that the outburst was linked with the exposure of seafood in a market in Wuhan. The Chinese authorities recognized it as a coronavirus, which was out of control on 7th January 2020. Further, on 12th January 2020, the government of China reported the genetic sequence of the novel coronavirus for countries to use in making specific diagnostic kits. By 20th January 2020, 286 confirmed cases had been reported from China, Japan, Thailand, and the Republic of Korea. The WHO report of China by 10th April 2020 reached up to 83,305 confirmed cases of COVID-19 [2].

The WHO reported 1,610,909 cases (92,798 deaths) of COVID-19 globally by 10th April 2020. Till now, the highest numbers of cases were found in the United States of America. It has reached up to 425,889 confirmed cases of COVID-19 (14,665 deaths) by 11th April 2020. To date, the number of patients suffering from COVID-19 is increasing at a higher rate. Due to the unavailability of effective vaccines for COVID-19, precautions like personal hygiene and quarantine measures have become most important to break this series of communication and slow down the pace of the outbreak [3].

Therefore, the Indian government declared a complete lockdown for three weeks on 22nd March 2020 to combat the spread of COVID-19 in India. By 10th April 2020, India reached up to 6,412 COVID-19 cases (199 deaths). As COVID-19 is increasing rapidly, the Indian government decided to extend the lockdown period till 30th April 2020. By 24th July 2020, the total cases reported in India were more than 12,000,000 [4].

This work aims to develop an expert system based on fuzzy logic which can be used for initial self-screening of COVID-19. Moreover, the hierarchical fuzzy technique has been used to reduce the complexity of the designed system. The idea is to reduce the burden on hospitals and doctors by developing a self-screening tool for the initial diagnosis of COVID-19.

Figure 1.1 shows the global data of confirmed COVID-19 cases and deaths over time till 18th May 2020. The graph shows the perpetuating data of individuals day by day.

FIGURE 1.1 WHO 2020 report of confirmed COVID-19 cases and deaths [3].

1.2 Review of Literature

To the best of our knowledge there is not much literature and researchers have not contributed much on COVID-19 to date; some researchers have reviewed and discussed COVID-19, summarized as follows:

Chatterjee et al. [5] reviewed the budding proofs to assist and guide the public health reaction, primarily in India. This research has outlined the criteria to be used to produce significant knowledge for recommendations to prevent and control the deadly virus. The paper concluded with the imperative outcomes of continuing hard work to avert and contain COVID-19. Moreover, it points out the requirements for investment in health systems, improved health policy mechanisms, and the requirements for preparations as well as worldwide health security. Agarwal et al. [6] described the essentials of the design of such components as space, waste disposal, infection control, and the protection of healthcare personnel, associates concerned in planning and preparation, which can be adjusted to the framework for a new structure or makeshift construction on top of an accessible formation. The researchers finished by giving explicit requirements such as infection avoidance and control to slow the COVID-19 pandemic. Deng et al. [7] explained that in a solitary center case sequence of 138 confirmed patients hospitalized in Wuhan, China, the presumed hospital-connected diffusion of coronavirus was suspected in 41% of patients and 26% were in the intensive care unit (ICU), with the deaths of 4.3%. However, gender- and age-based outcomes were also revealed during this study. Further, the authors explained that hospital-linked communication was alleged as the presumed method of infection for health professionals. According to the authors, the median time from contact to first symptom was 5 days and to hospital admission was 7 days. Mandal et al. [8] explained that pinpointing quarantine would identify and quarantine 50% of symptomatic persons within 3 to 4 days of increasing symptoms. Moreover, the researchers concluded that the detection of travelers at the access of any harbor with indicative clinical features and from COVID-19-affected countries would achieve a delay in the entry of the virus into the community. Mohammed M.N. et al [9] have developed a system for the detection of COVID-19 based on thermal imaging techniques without much human interaction. The authors used internet of things technology to design this system. The purpose of this technology is to detect a human being with a high body temperature even in a crowded place. However, thermal imaging technology is not an appropriate technique for the detection of coronavirus as there are many reasons for high body temperature. Therefore a more efficient system is the need of the day.

Mohammed M.N. et al [10] have also designed a helmet-based system which works on the principle of thermal camera technology. This thermal camera is integrated into a helmet and performs a real-time screening process. However, the system is equipped with facial recognition technology to identify people with higher body temperatures. The constraint of this approach is the same as earlier discussed in previous paragraphs. Chamola V. et al. [11] presented a review of various technologies that are being used to manage the impact of coronavirus. The authors highlighted the impact of the internet of things (IoT), machine learning, unmanned aerial vehicles (UAV), drones, and 5G networks in combating the pandemic. The authors put significant efforts into providing an overview of these technologies and their role in the current virus condition. Waheed A. et al. [12] presented a new method, namely synthetic chest X-ray (CXR) with the help of Auxiliary Classifier Generative Adversarial Network (ACGAN). This method is named CovidGAN. This is proposed to improve CNN for coronavirus detection. The authors claim 95% accuracy with the proposed scheme compared to 85% with CNN. Basset et al. [13] proposed an improved marine predators algorithm (IMPA) with an X-ray segmentation-based hybrid COVID-19 detection model. The authors incorporated a ranking-based diversity lessening model to improve the performance of existing IMPA. The authors validated the results with the chest X-ray by assuming the thresho...