REVIEW ARTICLE Year : 2022  |  Volume : 11  |  Issue : 5  |  Page : 168-172

Role of artificial intelligence in the diagnosis of COVID-19: A mini review

PK Rajeesh Mohammed1, Saakshi Gulati2, Shivangi Gupta3
1 Department of Oral Pathology and Microbiology, KMCT Dental College, Kerala, India
2 Department of Oral Medicine and Radiology, Sathyabama Dental College and Hospital, Chennai, India
3 Department of Periodontics and Implantology, MMCDSR Deemed to be University, Mullana, Ambala, India

Date of Submission17-Jul-2022Date of Decision15-Aug-2022Date of Acceptance21-Sep-2022Date of Web Publication14-Oct-2022

Correspondence Address:
Shivangi Gupta
Department of Periodontics and Implantology, MMCDSR Deemed to be University, Mullana, Ambala, India
India

Source of Support: This study received no extramural funding, Conflict of Interest: None

DOI: 10.4103/2221-6189.357454

The ongoing COVID-19 pandemic due to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection has resulted in a significant public health care system crisis. This disease has resulted in devastating damage to human lives and significant disruption in economies. Use of “machine-learning” algorithms as tools of artificial intelligence may help identify a suspected or infected individual with an estimation of chances of survival. These algorithms make use of recorded observational data including medical histories, patient demographics as well as any related data on COVID-19.

Keywords: COVID-19; Artificial intelligence; Algorithm; Analysis

How to cite this article:
Mohammed PR, Gulati S, Gupta S. Role of artificial intelligence in the diagnosis of COVID-19: A mini review. J Acute Dis 2022;11:168-72
How to cite this URL:
Mohammed PR, Gulati S, Gupta S. Role of artificial intelligence in the diagnosis of COVID-19: A mini review. J Acute Dis [serial online] 2022 [cited 2022 Oct 16];11:168-72. Available from: https://www.jadweb.org/text.asp?2022/11/5/168/357454   1. Introduction 

Coronavirus Disease 2019 (COVID-19) has quickly turned into a pandemic involving the whole world with a very high exponential rate of growth and a lesser understood process of disease transmission. This virus stays in one’s body with little or nil symptoms, although it may lead to a rapid spread and progressive severity with a fatal variety of pneumonia that has been seen in 2% to 8% of infected individuals[1],[2],[3]. The exactness of mortality, disease prevalence as well as dynamics of disease transmission is ill-defined partly due to uniqueness of challenges caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection, for example, peak infectivity after or before onset of symptoms and poor understanding of its pathophysiology with multiple organs involvement[4].

The rapid spread of the novel coronavirus-19 (nCOVID-19) all over the world has raised alarms regarding this pandemic. As people continuously try to control the rapidly spreading of the virus, this pandemic has resulted in large numbers of deaths all around this world. Its rapid spreading has overwhelmed healthcare systems due to an acute shortage of important equipments along with a shortage of trained health providers. Also, there are various testing methodologies including reverse transcription-polymerase chain reaction (RT-PCR). The availability of rapid RT-PCR tests, high frequency of false negative results, delays in processing, variations in testing techniques as well as low sensitivity ranging 60%-70% have further resulted in a burden on the healthcare workers.

Computed tomography (CT) provides an insight into pathophysiology of this disease at various stages, its evolution, and detection. Although there are many challenges in diagnosing this disease, expert radiologists have reported a definitive pattern of infection characterized as “ground glass” opacities, round opacities, enlargement of intra-infiltrate vasculature, and greater consolidation which are signs of the progression of severe disease. CT scan can even detect COVID-19 at early stages in individuals with negative RT-PCR reports, in case the patient does not show any symptoms.

Thereby, a combination of CT imaging, entire genomic sequencing as well as electron microscopy have been co-adapted for screening and distinguishing SARS-CoV-2, the virus responsible for COVID-19. However, sometimes there may be a paucity of COVID-19 testing kits due to an increasing number of cases every passing day. Therefore, a framework that may enable a quick analytical substitute for containing the spread of COVID-19 among people all over the world must be devised. Vinod and Prabhaharan in their work devised a useful methodology or tool via artificial intelligence (AI) that may help identify COVID-19 among people by making use of a CT image scan along with chest x-ray imaging[5]. This strategy uses a data-set of chest X-ray images of SARS-CoV-2 infected and normal individuals and makes use of a ‘decision tree classifier’ for locating a person with COVID-19. The accurateness of an X-ray image can be analyzed for precision, re-call score, and ‘F1’ score. The result or outcome is dependent upon information that can be accessed in Kaggle and Open-I stores as per the uploaded chest X-rays and CT scanning images. This test methodology has demonstrated that this newly designed algorithm has good robustness, accurateness as well as precision[1][Figure 1].

Figure 1: Diagram showing artificial intelligence (AI)-based algorithm approach.

Click here to view

Advancements in AI provides an effective tool for facing diagnostic challenges. There is a large amount of information available throughout the initiation of information technology and continuous as well as increase in computational ability. AI has demonstrated excellent performance in tackling the aforementioned problems or challenges. The ability to extract various patterns as well as inter-relationships from available data has highlighted this research area, especially in various tasks that involve the description of available information as well as patient-related dynamics.

Thus, applications involving deep learning and machine learning techniques could recognize images as well as segmentation, forecast series of time, analyze sentiment, control systems, simulate dynamics, and self-operate robotic systems and have been proven an effective mechanism in maintaining a record of social contacts[6],[7]. All of these outcomes clearly explain attention on research activities all around the world and the use of AI is an effective tool for fighting the ongoing COVID-19 pandemic situation[8].

  2. AI methodology 

Temporal stages of various dynamics in current pandemic may act upon one’s environment, thereby, determining consequences over societal behavior as well as status of healthcare. Reliable strategies to address the dynamics of various issues need to be designed, and information concerning COVID-19 pandemic must be collected as well as analyzed. Monitoring infectious diseases, repurposing medicines or drugs used, quick and accurate analysis of data, performing analytics on social media as well as devising various methodologies for early detection are the areas wherein AI might be playing an important role[8]. Use of AI greatly impacts the prevention of further spread or new outbreaks by adopting methodologies along with generally used tools that can adapt to a variety of serious situations quickly.

AI methodologies can be applied in four main areas: action, detection, society, and health. The phase of detection precedes the phase of action. In the detection phase, data are collected along with organizations whereas, in the action phase, various models are used for producing a result over the surrounding environment, such as simplification of various diagnostic procedures as well as development of a predictable strategy for the diagnosis of COVID-19.

  3. Major applications of AI in COVID-19 pandemic 

AI can be divided into the processing of natural language, machine-based learning process, and computerized vision-based devices. It uses various models to implement and analyze data as well as make decisions[9]. Early diagnosis and recognition of infection is a major factor in applying AI. AI helps in identifying infrequent symptoms speedily, thereby and raises a red alarm to healthcare management system[10],[11]. It provides a quick decision at a low cost. It helps in developing management and diagnostic system for coronavirus by utilizing quick and good-quality algorithms and discovering underlying causes of infection via a variety of imaging tools such as magnetic resonance imaging scanning and CT.

By making controlled re-mediation, AI might help in constructing an intelligently designed framework with an automated configuration and also predicting the spread and propagation of COVID-19 pandemic. Thus, well-designed neural networking may also be developed following the observed characteristics of this virus[12],[13].

3.1. Tracing and identifying clusters

AI provides continued updates regarding patients and solutions. This technique can be used to trace contacts of infected individuals thereby locating the contagious pathogen. This will help in localizing disease clusters control further spread. Also, AI helps in estimating further spread of SARS-CoV-2 and its pre-existing forms.

3.2. Estimation of disease in a population

AI helps in estimating the total number of active cases in a given population and the associated rate of death in a given population. AI technology helps in identifying and forecasting presence of the virus, the further possibility of its spread, and its potential distribution by making use of social networks, available data in public domain, and those undergoing circulation in various media networking sites.

AI can also predict certain accuracy of estimated numbers of COVID positive cases and various clusters that might exist in any specific area. AI aids in identifying the most seriously affected areas, people as well as countries that can further help in taking effective steps against its spread.

3.3. Improving vaccines and treatment

AI can help in improving vaccines and treatment by utilizing drug-related research. This technique can be used to develop and expand distribution of vaccines and drugs. AI could speed up real-time recognition of the virus while traditional techniques take a long period and could accelerate research processes that are highly impractical for any human[12],[13]. It could accelerate drug and vaccine development of COVID-19, by formulating new tests at a much faster rate.

3.4. Reduction of workload

Healthcare staff has become overwhelmed due to a dramatic and unexpected rise in COVID-19 patients. AI significantly decreases the overall workload on healthcare workers during COVID-19 pandemic.

AI has been used for assisting healthcare workers to reduce work pressure[14]. AI can help in detection and treatment at early stages by using digitized tools and judgmental research work that provides up-to-date information[15]. AI may change the future of health medications and solve future related problems, thus, drastically decreasing workload of physicians.

3.5. Prediction of disease spread

AI could prevent COVID-19 disease by collecting real-time data and providing novel information. It could predict areas with focused clusters of infection, study the influx of the virus, and find out requirements for hospital beds as well as healthcare staff[15],[16]. AI aids in preventing the virus spread and disease outbreaks by making use of data and comparing data collected during different times. AI could describe disease characteristics, various infection sources, and transmission of infection[17]. It will be gradually proven to be critical for fighting this pandemic as well as any other epidemic condition in the future. This tool can be utilized to treat numerous illnesses as a preventive measure and to formulate different protocols for treatment. In near future, AI will play an important role in prevention and prediction[18],[19][Table 1].

Table 1: Application of artificial intelligence technologies in management of COVID-19.

Click here to view

  4. Conclusion 

Fast spread, high infectivity, and mortality rates of COVID-19 have exercised heavy strain on health care systems worldwide. Restricting one’s physical contact, early detection, and continuous monitoring of overall health status of patients with COVID-19 is important for management of COVID-19. Analyzing the available information related to COVID-19 using algorithms based on AI may contain the further progression and spread of the virus. AI technology can also help to develop anti-COVID-19 vaccines, perform contact tracing, predict outcomes of patients, spread awareness, and handle workload of associated medical personnel.

Conflict of interest statement

The authors declare no conflict of interest.

Funding

This study received no extramural funding.

Authors’ contributions

R.M.P.K., S.G. (Saakshi Gulati), and S.G. (Shivangi Gupta): concepts, manuscript preparetion, manuscript review; R.M.P.K. and S.G. (Shivangi Gupta): design, data acquisition, manuscript editing, guarantor; S.G. and S.G.: literature search.

Publisher’s note

The Publisher of the Journal remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

 

  References 
1.Hoehl S, Rabenau H, Berger A, Kortenbusch M, Cinatl J, Bojkova D, et al. Evidence of SARS-CoV-2 infection in returning travelers from Wuhan, China. New Engl J Med 2020; 382: 1278-1280.  
    2.Lai CC, Liu YH, Wang CY, Wang YH, Hsueh SC, Yen MY, et al. Asymptomatic carrier state, acute respiratory disease, and pneumonia due to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2): Facts and myths. J Microbiol Immunol Infect 2020; 53: 404-412.  
    3.Sohrabi C. World Health Organization declares global emergency: A review of the 2019 novel coronavirus (COVID-19). Int J Surg 2020; 76; 71-76.  
    4.He X, Lau EHY, Wu P, Deng XL, Wang J, Hao XX, et al. Temporal dynamics in viral shedding and transmissibility of COVID-19. Nat Med 2020; 26: 672-675.  
    5.Vinod DS, Prabhaharan SRS. Data science and the role of Artificial Intelligence in achieving the fast diagnosis of Covid-19. Chaos Solitons Fractals 2020; 140: 110182-110189.  
    6.Shorten, C., Khoshgoftaar, T.M., Furht, B. Deep Learning applications for COVID-19. J Big Data 2021; 8(1): 18.  
    7.Nayak J, Naik B, Dinesh P, Vakula K, Rao BK, Ding W, et al. Intelligent system for COVID-19 prognosis: A state-of-the-art survey. Appl Intell 2021; 51: 2908-2938.  
    8.Picialli F, di Cola VS, Giampaolo F, Cuomo S. The role of artificial intelligence in fighting the COVID-19 pandemic. Inf Syst Front 2021; 23: 1467-1497.  
    9.Kumar A, Gupta PK, Srivastava A. Review of modern technologies for tackling COVID-19 pandemic. Diabetes Metab Syndr 2020; 14: 569-573.  
    10.Maghdid HS, Ghafoor KZ, Sadiq AS, Curran K, Rabie K, A novel AI-enabled framework to diagnose coronavirus COVID 19 using smartphone embedded sensors: Design study. arXiv 2020; 1-7; DOI: http://arxiv.org/abs/2003.07434.  
    11.Wang CJ, New T, Sun F. Response to COVID-19 in Taiwan big data analytics. new technology, proactive testing. JAMA 2020; 323: 1341- 1342.  
    12.Ferguson NM, Laydon D, Nedjati-Gilani G. Impact of non-pharmaceutical interventions (NPIs) to reduce COVID-19 mortality and healthcare demand. London, UK: Medical Research Council, The Royal Society, Imperial College London; 2020, p. 3-20.  
    13.Shammam AA, Hadi AA, Ramul AR, Zahra MMA, Gheni HM. The artificial intelligence (AI) role for tackling against COVID-19 pandemic, Mater Today Proc 2021. DOI: https://doi.org/10.1016/j.matpr.2021.07.357.  
    14.He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. IEEE Comput Soc Conf Comput Vis Pattern Recogn 2016; 1-12.  
    15.Nemati E, Rahman MM, Nathan V, Vatanparvar K, Kuang J. A comprehensive approach for classification of the cough type. Annu Int Conf IEEE Eng Med Biol Soc 2020; 2020: 208-212.  
    16.Verde L, de Pietro G, Ghoneim A, Alrashoud M, Al-Mutib KN, Sannino G. Exploring the use of artificial intelligence techniques to detect the presence of coronavirus covid-19 through speech and voice analysis. IEEE Access 2021; 9: 65750-65758.  
    17.Xu X. A deep learning system to screen novel coronavirus disease 2019 pneumonia. Engineering 2020; 6(10): 1122-1129.  
    18.Hossain MS, Muhammad G. Deep learning based pathology detection for smart connected healthcare. IEEE Netw 2020; 34(6): 120-125.  
    19.Gravina R, Alinia P, Ghasemzadeh H, Fortino G. Multi-sensor fusion in body sensor networks: State-of-the-art and research challenges. Inf Fusion 2017; 35: 68-80.  
    

Publisher's note The Publisher of the Journal remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

  [Figure 1]
 
 
  [Table 1]