key: cord-0979694-tpqipceg
authors: Alhasan, Mustafa; Hasaneen, Mohamed
title: Digital Imaging, Technologies and Artificial Intelligence Applications during COVID-19 pandemic
date: 2021-05-15
journal: Comput Med Imaging Graph
DOI: 10.1016/j.compmedimag.2021.101933
sha: 021c8db65da4f9db88ecee22f974fb3d59221861
doc_id: 979694
cord_uid: tpqipceg

The advancement of technology remained an immersive interest for humankind throughout the past decades. Tech enterprises offered a stream of innovation to address the universal healthcare concerns. The novel coronavirus holds a substantial foothold of planet earth which is combatted by digital interventions across afflicted geographical boundaries and territories. This study aims to explore the trends of modern healthcare technologies and Artificial Intelligence (AI) during COVID-19 crisis, define the concepts and clinical role of AI in the mitigation of COVID-19, investigate and correlate the efficacy of AI-enabled technology in medical imaging during COVID-19 and determine advantages, drawbacks, and challenges of artificial intelligence during COVID-19 pandemic. The paper applied systematic review approach using a deliberated research protocol and Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow chart. Digital technologies can coordinate COVID-19 responses in a cascade fashion that extends from the clinical care facility to the exterior of the pending viral epicenter. With cases of healthcare robotics, aerial drones, and the internet of things as evidentiary examples. PCR tests and medical imaging are the frontier diagnostics of COVID-19. Computed tomography helped to correct the accuracy variation of PCR tests at a clinical sensitivity of 98%. Artificial intelligence can enable autonomous COVID-19 responses using techniques like machine learning. Technology could be an endless system of innovation and opportunities when sourced effectively. Scientists can utilize technology to resolve global concerns challenging the history of tangible possibility. Digital interventions have enhanced the responses to COVID-19, magnified the role of medical imaging amid the COVID-19 crisis and have exposed healthcare professionals to the opportunity of contactless care.

Humanity had endured the difficulty of pandemics since time immemorial. Historians chronicle the cohabitation between pathological infestations and humankind. Specifically, the viral disease documentation, disease containment, facilitated clinical research, and the digitization of healthcare systems across established healthcare facilities (Allam, 2020; Kim et al., 2020) . Regarding COVID-19, computer scientists strived to assist healthcare professionals in the war against COVID-19. Their contributions included developing contact tracing applications and home quarantine surveillance technologies via smartphones, wearables, and interactive maps. The interactive and publicly available global tracing map was mainly launched by the Center for Systems Science and Engineering at Johns Hopkins University on January 20 th , 2020. Additional applications included thermal scanners, robotics, depth cameras were introduced to detect the respiratory pattern of COIVD-19 patients, and the development of artificial intelligence (AI) (Adly et al., 2020) .

Labeled as the father of artificial intelligence (AI), Professor Marvin Minsky is amongst the scientists pool in artificial intelligence invented by John McCarthy in 1956 . Computer scientists explained the principle of artificial intelligence as the ability of machines and programmed computer systems to mirror the human capacity to learn, interpret, and develop resolutions to a particular issue (Adly et al., 2020; Madurai Elavarasan & Pugazhendhi, 2020) . Public health professionals received AI with great enthusiasm in 2018 due to the limitless capabilities provided by AI. Mixture of Numerous technology sought the chance to pioneer the movement of implementing AI in the mitigation of COVID-19; explicitly, BlueDot and Metabiota .

BlueDot is a web-based startup venture established by Dr. Kamran Khan in the during the SARS outbreak in 2003. The company rose to the publicity during the Swine flu outbreak of 2009. The reputation was credited to the accurate prediction framework designed by the company to track the progression of the virus. The framework created an exact forecast of the infectious trends of the virus using global air data. BlueDot solidified its leadership further during the Ebola outbreak of 2014 by disseminating its copyright risk assessment modules that predicted the Ebola virus's eventual spread in three Western African countries. The corporation continued its ambitious projects in the current catastrophe of COVID-19. Eventually, computing a detailed projection of possible COVID-19 targets nine days before the WHO's initial announcement. Compared to BlueDot, Metabiota was confined to human and animal research during its early ventures, especially in the African continent. Metabiota projected a steady growth and captivated the US government's attention that deployed the startup venture in the mitigation of the Ebola pandemic. Artificial intelligence techniques being machine learning, deep learning, and natural language processing algorithms sharpened the company's studies in human behavioral analysis through social media. Metabiota implemented the findings to determine future COVID-19 targets a week before the official announcements in Japan, Thailand, and Hong Kong. Artificial intelligence evidence a significant advantage in the management of COVID-19 (Allam, 2020) . It has delivered numerous applications in diagnostics, therapeutics, medical imaging, and reduction of the physical interactions between healthcare professionals and COVID-19 patients. Henceforth, diminishing the occupational exposure of healthcare workers to COVID-19 (Adly et al., 2020) .

Currently, there are many commercial AI applications available for health care systems. For example, PATHAI (improving diagnostic pathology) helps pathologists for more accurate diagnosis of samples. PAGER (proactive healthcare management) is another application that improves patient treatment through making health care recommendations. For drug discovery, J o u r n a l P r e -p r o o f ATOMWISE application can identify chemical structures and compounds. Lastly, the Quibim platform uses medical imaging and AI for early detection of COVID-19 (Builtin, 2020; MDDI, 2020) .

Clinical imaging has demonstrated a significant role in the early detection of Although the differentiation of other viral pneumonia from COVID-19 pneumonia is challenging and difficult on Computed tomography (CT) scan alone (Feng et al., 2020) , it has reported a high sensitivity of 97% specifically in patients with clinical suspicions of COVID-19 but with negative Reverse transcription polymerase chain reaction (RT-PCR) test results (Fechner et al., 2020) . Additionally, to assess the lung changes while avoiding the patient's movement suspected to have COVID-19, the American College of Radiology (ACR) suggested placing portable x-ray machines in the department. Therefore, portable chest x-ray (CXR) can be considered as a firstline imaging tool to evaluate the abnormalities present in symptomatic patient lungs (Cellina et al., 2020) .

Moreover, diagnosing COVID -19 patients through images is a bit challenging as there are some limitations of some imaging modalities. A study of four cases was done in Italy to compare Chest x-ray to chest CT scan in diagnosing COVID-19. The chest x-ray images for the four patients came negative without any sign of abnormality. However, the chest CT scan showed some COVID-19 imaging features such as ground-glass opacity (GGO), crazy paving pattern and consolidation. The study shows a high probability of having a false negative chest x-ray for a positive COVID-19 patient. Although chest x-ray is the better option because it is a faster and extensively accessible tool for evaluating lung parenchyma, most COVID-19 imaging research papers are currently focusing on chest CT because of its higher sensitivity (Cellina et al., 2020) .

Another study conducted on 234 patients addressed other CXR findings; they observed that 55 out of 234 patients had nodules and 156 had reticular-nodular opacities in their chest radiograph and reported that CXR has a sensitivity of 68.1%, with variability between 69% to 90% (Cozzi et al., 2020) . Additionally, the sensitivity of CXR in the current literature ranges from 25% to 69% (Oterino Serrano et al., 2020) . has been linked to other features such as vascular enlargement and air Broncho-gram (Meiler et al., 2020) . Another study addressed chest CT images showing additional signs of COVID-19 lesions, such as interlobular thickening of septa in a crazy-paving pattern, air trapping and reverse halo sign (Y. . Also, a study aimed to correlate the chest CT findings of COVID-19 patients with the course of the disease. The authors found that crazy paving, bronchus distortion, bronchiectasis, air trapping, and mediastinal and hilar lymph node enlargement were more profound in the Intensive Care Unit (ICU) patients compared to non-ICU and discharged patients (Erturk et al., 2020) . One of the studies mentioned that in severe cases chest CT showed flaky or widespread lesions .

Chest CT has been used to identify the stage of the lung lesions. About 30 patients were found to be in early stage, 49.1% of 110 cases in the progressive stage and 26 patients were in a severe category. For chest CT follow-up patients, 12 out of 16 patients (75%) demonstrated disease progression, 25% of 16 patients were of a mild degree and five moderate cases were seen on CT . Among those patients, CT findings showed disease extent as minimal in two patients, 4 patients had moderate progression, and two patients showed severe disease extent.

Furthermore, some of the reviewed articles discussed the significance of chest CT in identifying COVID-19 lung lesion distribution. In a study, an initial CT scan of the chest performed on 51 patients showed that the majority of the patients (74.5%) had an infection in all five lobes of the lung while the remaining had an infection in other lobes individually or bilaterally (Y. . CT can detect peripheral lung lesions in 43.6% of 110 patients and lesions in both central and peripheral zones in 56.4% of cases. Additionally, Chest CT was used to image lesion distribution in different lung lobes. About 5.5% of 110 patients displayed multiple lung lesions while 9.1% showed individual lobe lung lesions. The Majority of patients (72.7%) were seen to have lesions in multiple lobes of the right and left lung . The distribution of pulmonary lesions on CT of COVID-19 patients was mostly at all stages with the highest rate (66.6%) at stage 2. A few patients have central lesions (2.1%) (Ding et al., 2020) .

Another study showed that a 38-year-old symptomatic patient did RT-PCR test four times. The results came negative four times. However, the CT chest images were susceptible as they showed patchy ground-glass opacities (GGO) as a typical imaging feature of COVID-19 (Feng et al., 2020) . A review article stated that according to ACR, chest CT should not be used for screening as a first-line for the diagnosis of COVID-19 and should rather be used for symptomatic patients with clinical indications (Asefi & Safaie, 2020) . Moreover, the Chinese critical care ultrasound study group and Italian academy of thoracic ultrasound recently suggested that ultrasound of the lung (LUS) is an accurate and appropriate tool to identify lung involvement in patients with COVID-19, specifically for pregnant women; it is the safest option and can be done several times during pregnancy due to radiation-free technique characteristic and the ability to detect pneumonia related to COVID-19 (Buonsenso et al., 2020) .

Additionally, Ultrasound is one of the imaging modalities that has been used to scan positive COVID-19 patients. The sonographers have faced a unique challenge as their work requires close patient contact for a long period, which increases the risk of COVID-19 transmission. Therefore, the protocol for the scan was done by the radiologist to reduce the examination time. In addition, the use of the US scan was considered the last option for COVID-19 patients due to the increased infection risk for the sonographers (Eastgate et al., 2020) . A study stated that LUS findings among 10 pediatric patients were mainly vertical B-lines, white lung signs, irregularities of pleura, and subpleural consolidation. LUS has been demonstrated to be an important imaging tool in assessing moderate and severe COVID-19 pneumonia (Sultan & Sehgal, 2020) .

Furthermore, case studies reviewed from Canada and Turkey described lung US findings of a health care worker and pregnant patient. LUS, as a radiation free modality, could be safely used to image pregnant women. The pregnant patient displayed diffuse and thick B-lines on either side of the lungs and had basal posterior involvement, while the health care worker demonstrated multifocal B-lines, thickening of pleura, and subpleural consolidation (Sultan & Sehgal, 2020) . Moreover, a research paper on 4 pregnant patients with COVID-19 was conducted to observe lung involvement by ultrasound. All four patients showed signs of COVID-19 pneumonia on J o u r n a l P r e -p r o o f LUS including irregular pleural line, white lung, large sub-pleural consolidations, and multiple vertical artifacts (Buonsenso et al., 2020) .

Additionally, numerous studies reported that the appearance of air Broncho-gram sign in COVID-19 patients. Air Broncho-gram sign was mostly visible in the early and progressive stages (Sultan & Sehgal, 2020) . Three other articles reported that LUS is superior to chest CT in detecting smaller peri-pulmonary lung lesion along with pleural and peri-pulmonary effusion (Bosso et al., 2020; Convissar et al., 2020; Sultan & Sehgal, 2020) .

Although PCR test is routinely used, it has many limitations, including possible false negative results that require repeating the test, the result accessibility can take 2 days and the shortage of test kit. Additionally, CT scan alone can show normal images of COVID-19 patients. Therefore, integrating CT scan with AI using useful data such as clinical symptoms has improved the detection sensitivity similar to an expert radiologist level. Using this approach, COVID-19 cases were accurately detected compared to the wrong classification of the radiologist alone . Another study reported that using a trained deep learning algorithms scored 90.8% accuracy, 84% sensitivity, and 93% specificity in classifying COVID-19 pneumonia (Harmon et al., 2020) . Moreover, a study has compared the DeepCOVID-XR AI algorithm's performance to five radiology experts on 300 images. The accuracy of DeepCOVID-XR's accuracy was higher (82%) than the consensus (81%) of all five radiologists (Wehbe et al., 2020) .

The current status of combining AI, digital techniques and clinical imaging technologies concentrates on three main areas; positive or negative COVID-19 classification, lung lesions quantity identification, and differentiating COVID-19 from other types of Pneumonia. Although many studies have reported an acceptable accuracy, sensitivity, and specificity of such techniques, it is difficult to identify COVID-19 patients with a negative PCR test. Therefore, the clinical decision cannot only rely on AI in the presence of valuable lab tests. In order to improve the current role of such technologies, more useful data regarding certain diseases like diabetes, liver and kidney lesions, and age can be incorporated to improve the overall diagnostic performance (L. Summers, 2020) .

The resilience of modern digital technologies and artificial intelligence during COVID-19 created a great disagreement among healthcare professionals. Forming a divide of acceptance and rejection. Such debate motivated the foundation of this paper to answer a collection of research questions; (1) How did the pandemic help promoting the digital uprising of modern healthcare technologies and artificial intelligence? (2) What are the clinical applications of artificial intelligence during COVID-19? (3) What is the efficacy of AI-enabled technology in the disciplines of clinical imaging? (4) What are the pitfalls and challenges of artificial intelligence encountered during COVID-19?

The study targets to answer the previous questions by achieving the following aims: (1) Explore the trends of modern healthcare technologies and artificial intelligence during COVID-19 crisis (2) Define the concepts and clinical role of artificial intelligence in the mitigation of COVID-19 (3) Investigate and correlate the efficacy of AI-enabled technology in medical imaging during COVID-19 (4) Determine the advantages, limitations , and challenges of artificial intelligence during COVID-19 pandemic.

This research paper was developed based on the principles of systematic literature reviews. Systematic literature reviews are distinguished publications that maintain a profound societal influence in scientific explorations and dedicated healthcare research (O. S. . Such literary pieces designed the standard module of generalizing research literacy in the scientific community and the healthcare workforce; through guided collection and categorization of the current body of literature relevant to a particular research question. The compiled literature is eventually analyzed following rigorous eligibility criteria hypothesized around an expansive research protocol (Liberati et al., 2009) . The research aims to collate the latest full-text and peer-reviewed publications covering the scope of modern digital technologies and artificial intelligence during COVID-19.

Correspondingly, the research methodology utilized an extensive literature search constituting four databases: Google Scholar, PubMed, IEEE Xplore, and Fatima College of Health Sciences (FCHS) learning center database (EBSCOhost); to inspect all original publications associated with the topic of interest. The database selection rationale articulated the abundance of scientific literature from multiple databases presented by Google Scholar. Google Scholar functions as an umbrella server for a range of databases as exemplified by Elsevier. PubMed and IEEE Xplore were synonymous in terms of diversity and validity of interdisciplinary publications on COVID-19. The FCHS learning center database was employed to provide a supplementary portal for additional articles associated with technology, artificial intelligence, and COVID-19. The candidacy of the extracted literature cohort was determined through specific eligibility criteria;

(1) Articles written in the English language (2) Full-text and peer-reviewed articles published between March 2020 and September 2020 (3) Publications and studies conducted during the COVID-19 pandemic (4) Publications that include keywords connected to technology, artificial intelligence, and COVID-19. Further article stratification was concluded by a subsidiary exclusion criterion; (1) Pre-print publications and articles that await peer-review (2) Publications with denied access (3) Publications incompatible with the research scope (4) Articles that lack a revision timeline including submission, review, and publication dates (5) Articles that exhibit subjective opinion or bias about the proposed review topic.

This review adopted a four-phase research procedure. Phase one originated from the preliminary search on September 3 rd , 2020 of the Google Scholar database. The search identified 53 articles that expressed a tendency of variety in the acquired search results. Thus, prompting the phase two initiation on the following day by conducting a secondary search using the Google Scholar database to validate the quality of the displayed results. The secondary search yielded 40 publications focusing on artificial intelligence contributions in the healthcare industry during COVID-19. Phase three commenced on September 5 th , 2020, and marked the inspection of PubMed and IEEE Xplore. September 6 th , 2020 dated the investigation of the FCHS learning database, namely, EBSCOhost. Statistically, the selected data incorporated 12 articles from PubMed, 3 articles from IEEE Xplore, and 2 articles from the FCHS learning center database. Table ( 1) and figure (1) stratify the database search protocol and acquired results. Congruently, J o u r n a l P r e -p r o o f the accumulated literature cohort tallied a total of 110 articles assessed for eligibility using the pre-determined eligibility criteria. Nineteen publications were eliminated on account of incompatibility with the research scope during the primary title and abstract inspections. Duplication was evident in two publications obtained from EBSCOhost. Eleven publications were removed due to ineligibility in terms of the requisite criteria. Eventually, the systematic literature review included a total of 54 that fulfilled the inclusion criteria. (((("All Metadata": artificial intelligence ) AND "All Metadata":COVID-19) AND "All Metadata": COVID-19 ) AND "All Metadata": AI) 6/9/2020 FCHS learning database (EBSCOhost) 2 (((("All Metadata": artificial intelligence ) AND "All Metadata":COVID-19) AND "All Metadata": technologies ) AND "All Metadata": AI) Total 110 

Digital technologies supported the world in addressing the tragic impact of the novel coronavirus pandemic (Kumar et al., 2020) . COVID-19 painted a picture of resistance to the routineness of human life with weakened global economies and three billion people confined in compulsory public health lockdowns as the proprietary tones of COVID-19 (Lin & Hou, 2020) . Digital health societies, symposiums, and contributors lead the international scene in the consumption of technological interventions to circumvent the ripple of COVID-19. Examples of such interventions entailed the management of case detection, socioeconomic impressions streamed by the circulation of the virus, and the institution of public health mandates to tackle the outbreak within the affected territories (Kapoor et al., 2020; Whitelaw et al., 2020) . Researchers continue to interpret the interlinked combinations provided by digital technologies in response to COVID-19. This section of the systematic literature review discusses a variety of digital technologies associated with targeted COVID-19 responses from the perspective of the included literature; (1) Digital interventions within clinical facilities (2) Digital interventions outside clinical facilities (3) Digital interventions within the World Wide Web.

The global invasion of COVID-19 triggered tragic deficiencies in medical supply chains of personal protective equipment (PPE). Frontline responders require protective facial masks, goggles, and protective shields to tend to COVID-19 patients. PPE reduces the inherent occupational exposure by inhibiting the user from absorbing the viral droplets containing COVID-19. Three-dimensional printing remedied the emerging shortages by efficiently producing crucial medical supplies at accelerated speed using readily available material. The printing process involves applying a 3D laser to generate accurate measurements of the facial features of the potential user leading to the fabrication of a personalized facial mask. Opensource data contributes to the cost efficiency of three-dimensional printing as it facilitates the circulation of biodegradable material in the production of protective face shields (Tsikala Vafea et al., 2020) .

Healthcare robotics represents a specimen of the technological interventions promoted by the spread of COVID-19. Dedicated COVID-19 treatment facilities resorted to healthcare robotics to establish remote treatment protocols and viral infection control. Facilities stationed the robots in various central departments that housed COVID-19 patients analogous to isolation wards and intensive care units. Deployed units directed the autonomous completion of lab tests, vital signs acquisition, facility-based disinfection, and waste management. Logistical delivery of essential patient supplies to in-bound patient rooms was also attainable through the stationed robotic units within the hospital premises (Mastaneh & Mouseli, 2020) . The Providence Regional Medical Centre in Washington exercised an assistant robot capable of remote vital signs acquisition through a built-in stethoscope. Nevertheless, the robot functioned as a communication channel between physicians and COVID-19 patients allowing the former to communicate with patients through live stream video conferences (HealthManagement.org & Team, 2020).

Telemedicine is witnessing a period of profound use due to the novel coronavirus pandemic. The prospect of remote healthcare services is one of the gold standards of telemedicine that increases accessibility to medical services through the Information and Communication Technology platform or ICT. ICT platforms authorized physicians to launch teleconsultations using telecommunication networks or video conferences to address patient complaints and presented cases. In terms of COVID-19, telemedicine can provide real-time patient observation of quarantined individuals through selfreported medical information using smartphones or other digital devices. Telemedicine systems can stream continuous notification to healthcare providers reporting the patient's symptoms and disease progression. Such data can simplify the generalization of remote population management across the afflicted state (Fagherazzi et al., 2020; Kapoor et al., 2020; Mastaneh & Mouseli, 2020) . Technology-enabled critical care services or tele-critical care (TCC) signifies an expanded perspective of traditional intensive care units' conveniences. TCC can coordinate the transfer of critically ill patients in hospitals with limited occupancy or capabilities. The proper patient provision ensures the conservation of PPE reserves, and minimizes occupational exposure in paramedics and medical ambulances. Medical ICUs can witness significant advantages in the maximization of ICU efficacy in terms of early monitoring of possible respiratory complications associated with COVID-19 and conscious use of ICU resources like unit occupancy and available ventilators .

Drones played a vital role in the technological response to COVID-19. The drones provided aerial case detection, diagnostics, and supervision of the mandated public health regulations in terms of efficiency and population compliance. Healthcare professionals can operate drones to execute frequent supply deliveries of patient laboratory samples and relevant essentials quickly and efficiently (Mastaneh & Mouseli, 2020) . Drones can enforce population compliance through visual and audio prompts to the offenders for violating the public health mandates and omitting the use of a protective face mask. This practice was notable in China as drones patrolled rural localities with diagnosed cases and reported public health policies violations to the relevant authorities (HealthManagement.org & Team, 2020). High-resolution cameras, specialized thermal scanners, and high-performance computer systems are among the repertoire of tools offered by drones to limit the spread of COVID-19 in airports and public transportation hubs. Thermal scanning of places of populated congregation and remote disinfection of hotspots are examples of drone capabilities adopted in multiple countries worldwide (HealthManagement.org & Team, 2020; Kapoor et al., 2020) .

Geographic information systems or GIS systems are mobile computer-based frameworks that compile, archive, and categorize spatial data related to planet earth's surface. The embedded data cohorts could consist of street signs, geographical locations, population census, and characteristic vegetation within the system area coverage (Drew et al., 2020; Mbunge, 2020) . GIS systems can guide technological interventions for COVID-19 diagnostics through data reports that entail the location of diagnosed cases, hospital occupancy, and patient compliance with public health regulations. The system proceeds to perform a complex spatial analysis of COVID-19 related data that assists the authorities in maintaining constant surveillance of the virus spread within the infected country and the world (Mastaneh & Mouseli, 2020) .

The internet of things is a composite system that resembles the taxonomy of the human organ system. IoT frameworks are an entity of subsidiary systems functioning as a central component in the system network. Computer scientists stimulated IoT systems capabilities to become a vault of accessible public data available to the general population with examples ranging from daily case statistics to global fatality rates around the world (Adly et al., 2020) . Generalized internet of things frameworks can create communication channels between medical devices and the World Wide Web. Therefore, increasing the accessibility to vital information and medical services determines the disease progression and possible prognosis (Mastaneh & Mouseli, 2020) .

Technology-driven responses witnessed universal adaptation on a global scale to limit the spread of COVID-19. Researchers and stakeholders investigated the arguments reflecting the limitations of digital technologies in a variety of COVID-19 literature (Coombs, 2020) . Table ( 2) highlights a collection of provisions and limitations encountered with the adaptation of digital technology interventions.

Digital technologies rely on the availability of internet access which allows digital frameworks to reach full potential in terms of generalization and connectivity within the established server coverage. (Budd et al., 2020) Capital expenditure

Technological interventions require significant amounts of capital to invest in the development, maintenance, and adaptation of digital technologies in response to COVID-19. Such technologies encompass digital assistants like SIRI and ALEXA. (Keesara et al., 2020) Digital divide The disparity in accessibility to digital technologies can limit the coverage of technological interventions in some population groups that lack access to digital technologies.

(Ramsetty & Adams, 2020) 

Artificial intelligence is a discipline of many stratospheres and entwined intricacies. These intricacies mirror the coping mechanism of the human cerebral network with the fabrication of intelligent agents. Intelligent agents are the computerized counterpart of the human cerebral neurons as the artificial intelligence apparatus gradually develops an autonomous capacity to process the setting, the present data, and the parties involved to deduce the resolution of a particular scenario. Hence, reincarnating the architecture of human thought (Agbehadji et al., 2020) . Table ( 3) summarizes some of the common terminologies associated with artificial intelligence.

Terminology Definition

The computation of the human independent critical thinking, interpretation, and resolution through the categorization of random data. (Salman et al., 2020) Machine learning The recognition of data through identified patterns. (Bansal et al., 2020) 

A deep learning algorithm that targets AI systems designated in the segregation of multimedia information (i.e. video, audio, and images) (Salman et al., 2020) Supervised learning Data comprehension through the use of cataloged data. Unsupervised learning Data comprehension through the use of uncatalogued data. Natural language processing (NLP)

The encoding of accumulated data obtained through natural linguistics and dialogue. Natural inspired computing (NICC)

The development of novel algorithms and hardware through the multidisciplinary integration of mathematics, computer science, theoretical physics, etc. (Agbehadji et al., 2020) Convolutional layers A set of predetermined parameters that shape the AI learning process. (Salman et al., 2020) Pooling

The learning blocks of CNN.

Interlinked sets of CNN neurons that transfer learning data.

The training of an AI platform through a specified set of rewards and penalties established by the human programmer. Big data

The total volume of heterogeneous computed data in the present databases.

(Agbehadji et al., 2020) Table 3 : Glossary of common artificial intelligence terminology.

The signature communicability and lethality of COVID-19 represented a physiological fortress for the medical group and scientists alike. This challenge is founded on the rapidly mutative sequelae of COVID-19. Rapidly, the alteration of COVID-19 sequelae is governed by preexisting lung conditions. Pulmonary atelectasis, hemorrhage, edema, and lung tissue neoplasia are illustrative examples capable of altering COVID-19 diagnosis and confusing the trained medical eye (Adly et al., 2020) . Public health experts state the standard diagnostics of COVID-19 as a methodology of two folds consisting of nasopharyngeal PCR swabs and diagnostic radiology (Salman et al., 2020) . PCR test analysis reported a characteristic disparity in the J o u r n a l P r e -p r o o f differential diagnosis yielding a percentage of 60 to 71% clinical sensitivity (Adly et al., 2020) . Therefore, comprising the prospect of fast and definite acquisition of essential diagnostics relevant to patient survival (Naudé, 2020) . Public health experts resorted to bio surveillance to rectify the disparity of present molecular testing kits and provide the basis for remote detection and prevention of COVID-19 (Bansal et al., 2020) .

Contact tracing is a mainstay public health prevention regulation that detects the exposure of a particular sample of individuals to an infectious contagion through a reverse revision of each member's activities and movements involved in the chain of transmission (Lalmuanawma et al., 2020) . Public health epidemiology dates the application of contact tracing in instances of global pandemics like SARS, MERS, flu, Ebola, and COVID-19 (Agbehadji et al., 2020; Lin & Hou, 2020) . This process is evident in COVID-19 prevention protocols that allow a stringent observation of infected cases during 14 days through cellular technologies like global positioning systems, Bluetooth, social activity maps, and network credit card transactions. In episodes of proper systematic deployment, contact tracing can provide means of real-time detection, identification, and prevention of severe COVID-19 epicenters (Lalmuanawma et al., 2020) . Hence, reducing predicted causalities and augmenting the success of COVID-19 responses employed in individual localities and governments as solidified in South Korea (Lin & Hou, 2020) . The antithesis of the beneficial role of contact tracing manifests in possible data security breaches, invasion of autonomous privacy, and mishandling of the heterogeneous data volume (Lalmuanawma et al., 2020) .

Artificial intelligence devices assisted researchers and stakeholders to broaden the spectrum of contact tracing through the integration of artificial intelligence, big data, and Machine learning to enhance the scale of detection, identification, and parameter of contact tracing technologies (Bansal et al., 2020; Bragazzi et al., 2020) . Big data systems enable the accumulation of structured and unstructured data computed in the global databases on the internet. Clinical data obtained through big data can encompass radiographic images, alternative diagnostics including, salivary testing kits, pharmacology, and digitized health care documentation (Agbehadji et al., 2020; Bragazzi et al., 2020) . Machine learning and artificial intelligence encouraged researchers to discover key biological metrics associated with the treatment paradigm of COVID-19 counting on radiographic image history and hematology of the reported cases (Lalmuanawma et al., 2020) .

Artificial intelligence introduced the concept of digital phenotyping. Digital phenotyping exploits cellular-based sampling of surveys, texts, multimedia, and screen patterns to generate a detailed phenotype of the user's movements, behavioral trends, and digital signature (Bansal et al., 2020) . The efficacy of AI-enabled contact tracing and digital phenotyping is determined by the subject's commitment to provide the required data for contact tracing analysis. Several nonmedical technology combinations like Google and Apple proposed a Bluetooth operated contact tracing application that configures the operator's phone to a visual platform reporting proximity to a known COVID-19 case. Furthermore, Bluetooth powered contact tracing applications can breed the incidence of false positives and false negative diagnosis of COVID-19 due to maskbased socialization between the user and other mask-wearing individuals or personal inner circle (Lin & Hou, 2020) . The tradeoffs of AI-integrated contact tracing contrast with cost-efficiency, higher clinical sensitivity, minimized occupational exposure, and minimal detection time compared to conventional human-operated screening protocols (Agbehadji et al., 2020) .

Contact tracing applications were an adopted public health practice in the Arabian Peninsula with the Tabaud and TraceCovid as prominent instances. The Saudi Data and Artificial Intelligence Authority (SDAIA) designed the Tabaud or distancing standard Arabic dialect to establish a contact tracing network for all involved contacts. TraceCovid is a contact tracing application adopted in the United Arab Emirates as a part of the smart solutions platform in response to COVID-19 (Hassounah et al., 2020; Lalmuanawma et al., 2020) .

Vaccines are synthetic pharmaceuticals that stimulate the body to produce antibodies to mitigate foreign viral bacteria invasion. The public health contributions of vaccines are emphasized in targeted disease prevention and elimination of infectious diseases through reproducible efficacy and long-term stability . COVID-19 genomics revealed an intricate physiological puzzle that compelled scientists to adopt a changed strategy to mitigate the spread of COVID-19. This strategy consisted of investigating the current repository of antiviral medications for potential repurposing, synthetic modification of present antivirals to accommodate COVID-19 biology, and the application of alternative interventions including, traditional Chinese medicine to expand the potential of drug synthesis and repurposing. Artificial intelligence assisted researchers in the knowledge of pharmacological therapeutics and immunizations of COVID-19 (Bragazzi et al., 2020) . Natural language processing enabled clinicians to assume and identify COVID-19 related pharmaceuticals through the linguistic analysis of syntax found in dedicated pharmacology references (Adly et al., 2020) .

Moreover, Machine learning provided accurate identification and suggestions of possible medications eligible for COVID-19 repurposing such as Atazanavir and Baricitinib (Naudé, 2020) . Baricitinib is an immunosuppressant prescribed for patients with rheumatoid arthritis and myelofibrosis (Adly et al., 2020) . The discovery of Baricitinib was the byproduct of the joint venture between numerous dedicated institutions like BenevolentAI, the Imperial College of London, and a university establishment (Bragazzi et al., 2020) . Machine learning extrapolated a projected timeline for vaccine development in an estimated 18 months (Naudé, 2020) . When paired with reverse vaccinology, machine learning determined six potential vaccinations that target the proteins of the COVID-19 molecule, as reported by the Hong Kong headquarters of Insilico Medicine (Bansal et al., 2020; Bragazzi et al., 2020) . Reverse vaccinology is a discipline that applies computed technology to sequencing genome information (Del Tordello et al., 2017) .

AI applications can be useful for COVID-19 treatment planning (Tayarani N., 2020; van der Schaar et al., 2021) . It was reported that an approach based on machine learning could analyze and differentiate between different types of treatment plans that can aid physicians in selecting an effective treatment plan (Pu et al., 2020) . Additionally, the patient's recovery prediction was evaluated using three machine learning techniques. The results showed that patients with fever and fatigue living in China did not recover (Ella Hassanien et al., 2020) . Also, AI was found useful in COVID-19 disease progression and the need for oxygen treatment prediction using machine learning (Obinata et al., 2020) . Another application is related to the ability of the AI to prioritizing patients based on their health conditions. This technique utilizes Multi-Biological Laboratory Examination Framework to classify patients into mild, moderate, and severe cases to receive the most effective treatment (A. S. . Moreover, recovered COVID-19 patients can develop antibodies that can help the immune system of other severe cases. Therefore, AI was utilized using machine learning to identify the suitable donors and most critical patients (O. S. .

COVID-19 patients have consumed emergency departments, intensive care units, and floor units to the maximum capacity worldwide (Debnath et al., 2020) . According to reported metrics in global clinical facilities, the approximate COVID-19 death rate measures at 4.5% amid confirmed infections as of March 26 th , 2020. Notwithstanding the interval of 0.2 to 15% death rate calculated in infected demographics determined by age and history of pre-existing conditions (Salman et al., 2020) . The Our World Data initiative is among the scientific publications aimed to deliver empirical statistics regarding COVID-19.

For instance, case fatality rate or CFR ratio is the death censuses divided by the total COVID-19 cases within an affected area. The case fatality rate pattern may fluctuate by the constant change in pandemics data trends comparable to daily case counts (Roser et al., 2020) . Treatment paradigms and protocols differed in many hospitals and specialized medical institutions as some facilities resorted to manual data spreadsheets containing patient-centered information.

Healthcare workers discard the manual spreadsheets following patient discharge. The practice of manual spreadsheets may lead to data mismanagement due to the constant surplus of COVID-19 patients with variable case characteristics and physiological phenotypes.

Artificial intelligence systems can facilitate the clinical workload imposed on healthcare professionals through digitized data collection, storage, and examination using state of the art portable devices and technology with installed electronic health records platforms (Adly et al., 2020; Debnath et al., 2020) . Mobile tablets and personal digital assistance devices can provide significant convenience in facilitating data access for healthcare professionals regardless of their location, presided tasks, and rotation within the campus of the hospital or specialized institutions. Smartphones and smart bracelets can develop a third-party medium for physicians to consume AI-powered real-time case surveillance of isolated infections and private settings including the patient's residence and social interactions (Adly et al., 2020) . Machine learning can redefine the environment of clinical decision making in COVID-19 emergency departments by calculating risk trajectories, disease progression, prognostic outcome, and the effects of ventilation on the generalized case therapeutics and viral phenotype. Thus, mapping the proper prioritization and resource allocation of depleted supplies using empirical probabilities resembling respiratory failure incidence (Debnath et al., 2020) .

The American dramatist Zora Neale Hurston describes research as a formalized curiosity and investigation with a purpose (Hoffman-Jeep, 2005) . The vague nature of COVID-19 promoted widespread collaborations between researchers, technology stakeholders, computer scientists, virologists, and clinicians from all professions to organize a systematic application and stratification of COVID-19 research data. Data mined publications encompassed AI-focused and non-AI focused research. Hence, creating a preprint trend noted in several prominent databases.

Artificial intelligence system developers sourced non-AI-focused articles to browse for characteristic COVID-19 information that may enrich the training of original AI platforms designated in the management of COVID-19 (Naudé, 2020 (Kricka et al., 2020) .

Artificial intelligence is perceived with lauded acceptance and opposition from all scientific aisles (Allam, 2020) . Intelligent AI agents opened the scientific retrospective view in terms of COVID-19 and the prospects of artificial intelligence. The majority of artificial intelligence research associations praised digitized economies, supply chains, and exports as future AI landscapes following the reduction of COVID-19 (Naudé, 2020) . Additional research findings project the construction of smart cities and the integration of robotics in human labor as anticipated artificial intelligence presentations post-COVID-19 (Bragazzi et al., 2020; Naudé, 2020) . Table (4) collects some encountered challenges and tradeoffs of artificial intelligence in consideration of COVID-19.

Logistical maintenance

Compulsory public health lockdowns can impede communication between machine learning data specialists, engineers, and healthcare personnel to ensure proper usage and maintenance of the AI algorithm. (Debnath et al., 2020) Existing AI technique limitations

The performance of deep learning is restricted by the provided human data, inherent computation of all caserelated probabilities, and the volume of training data. J o u r n a l P r e -p r o o f

Public health COVID-19 responses and guidelines mandated the use of private user data (i.e. contact tracing) (Naudé, 2020) Training data shortage Scarce training data can compromise the performance and efficiency of COVID-19 centered AI models.

The disparity in COVID-19 variables can influence the results of predictive AI models. Such variables include the incubation period, levels of dyspnea, and oxygen saturation. (Adly et al., 2020; Debnath et al., 2020) Chatbots Chatbots are conversation simulators created to offer an alternative platform of human-AI communication.

These systems require long term training and maintenance to minimize the incidence of deficient clinical diagnosis and false response inputs.

(Madurai Elavarasan & Pugazhendhi, 2020) 

PCR swabs surfaced as abstract instrumentations in discrete COVID-19 diagnostics except for disparate detection accuracy depending on the sample quality, storage, and manual handling . Medical imaging rose to compensate for PCR tests discrepancies and impede the chain of transmission of COVID-19 (Shaikh et al., 2020) . Radiographic imaging modalities contribute to the observation of disease progression (Greenspan et al., 2020) by using radiographic image evidence to support frontline responders in establishing a diagnostic strategy compatible with the presented patient data . The radiographic imaging modalities of COVID-19 were sorted into primary and secondary modalities in terms of rationale, relevant studies, and imaging characteristics of diagnosed COVID-19 cases.

3.3.1.1 Primary imaging modalities

Chest radiographs (CXRs) are routine imaging procedures that map the anatomical chest patterns through means of attenuation of ionizing radiation generated from an X-ray machine tube. The representation of pulmonary pathology in routine CXRs assisted healthcare professionals in the expository investigation of COVID-19 sequelae, explicitly, early pathological infestations of COVID-19 (Shaikh et al., 2020) . X-ray machines are everywhere in medical imaging departments due to their cost-effectiveness and staple reliability (Kundu et al., 2020) . COVID-19 patients can present normal CXR findings in response to prompt acquisition times, absence of pathological signs in the pulmonary tissue, and elusive radiographic findings (Manna et al., 2020) . In the medical imaging society, variations in test circumstances, symptomatology, virus incubation time, and false negatives started a paradox debating the merits of CXRs as diagnostic imaging tools to scan and diagnose COVID-19 (Kundu et al., 2020) .

Positive CXR radiographs demonstrate COVID-19 as blurred patches on the lung's bilateral surface with altered locations and severity depending on the infection stage. The lung opacifications intensify in texture, solidity, and severity with the continuation of COVID-19 inflammation in the afflicted lungs. Mild COVID-19 cases may contrast routine COVID-19 opacifications. The COVID-19 CXR findings are revealed in 10-12 days after the disease initial exposure (Manna et al., 2020) . COVID-19 CXRs share significant commonalities with viral pneumonia. Therefore, challenging radiologists in the differential diagnosis of COVID-19 compared to viral pneumonia (Chowdhury et al., 2020) . The radiographic technique of COVID-19 consists of bedside AP CXRs, PA, and lateral chest projections to evaluate the onset of COVID-19 in the respiratory system (Manna et al., 2020) .

Computed tomography defines the correlation between the rotatory imaging detectors and ionizing radiation to emulate cross-sectional slices of the chosen scan coverage. Thoracic computed tomography exceeded PCR tests' diagnostic accuracy and the false-negative readings of CXRs with a clinical sensitivity of 98%, in contrast to 71% in PCR swabs (Kundu et al., 2020; Nagpal et al., 2020) . Computed tomography operators begin the image acquisition by instructing the patient to maintain a fixed position on the gantry patient bed and adhere to a set of breathing instructions during scan acquisition. Radiologists undertake the image reading and interoperation processes using the comprehensive scan coverage of thoracic CT slices. Thoracic CT slices cover the pulmonary anatomy from the pulmonary apices to the costophrenic angles to evaluate thoracic pathologies depending on the patient's body habitus. Computed tomography operators reconstruct the raw scan data and transfer the images to PACS servers following computer-based post-processing of the preliminary raw data (Nagpal et al., 2020) .

Medical imaging stakeholders have reported a plethora of COVID-19 findings in thoracic computed scans including ground-glass opacities (GGOs), crazy paving patterns, lobular consolidations, and the reverse halo sign as radiographic findings of COVID-19 (Kundu et al., 2020) . GGOs are at the most encountered computed tomography findings at a predominance of 60-70% as stated in epidemiological studies from Wuhan city (Nagpal et al., 2020) . GGOs are round morphological developments that occur in response to extra levels of radiographic attenuation in the thorax without any caused pathological mutations to the respiratory systems' normal bronchial and vascular habitus (Manna et al., 2020) . The lower right lobe is the primary site of GGOs in COVID-19 patients. GGO lesions may diffuse in the periphery of the lungs with increased thickness and infiltration according to the stage of infection (Nagpal et al., 2020) .

Crazy paving patterns are a subtype of GGOs with notable septal line thickening in the lobular pulmonary borders that may accompany GGOs or lung consolidations. Therefore, modeling a shape analogous to irregular paved cobblestones (Nagpal et al., 2020) . Lobular lung consolidations are solid obstructive opacities located in the airway borders and vascular margins of the lungs. The lobular lung consolidations may lead to lower pulmonary lobes involvement in the posterior and inferior directions (Manna et al., 2020) . COVID-19 consolidations may express air bronchograms and lung cavitation in the infected tissue depending on disease stage at the time of image acquisition. The reverse halo sign or the atoll sign is a GGO enclosed by profuse marginal consolidations. Reverse halo findings subsidize the alveolar damage in the interstitium and airway of COVID-19 patients (Nagpal et al., 2020) .

Hematological laboratory tests can demonstrate the vascular pathophysiology of COVID-19. Ddimer tests and vascular thrombogenicity are diagnostic assessments of vascular pathophysiology (Shaikh et al., 2020) . Medical imaging uses halo sign to show COVID-19-induced vascular injury. Halo sign refer to masses or nodules superimposed by GGOs. The halo sign commonly arises in angiographic disorders like vasculitis and vascular hemorrhage (Nagpal et al., 2020) . Radiographers use pulmonary CT angiography or duplex venous sonography to investigate the vascular appearance of COVID-19. The computed tomography findings of COVID-19 mature within 6-13 days of the official disease cycle (Manna et al., 2020) .

Magnetic resonance imaging applies magnetic and radio waves oscillation principles to produce sharp resolution images of the organs and soft tissue fascia within the region of interest. Pulmonary MRI scan findings revealed episodes of increased signal intensity on T1 and T2 weighted COVID-19 pneumonia images congruous to GGOs and peripheral lung involvement on thoracic CTs and CXR scans. Supplementary magnetic resonance imaging series revealed subtle cardiac damage in COVID-19 patients with signs of cardiomyopathy and myocarditis. Cine cardiac MRIs can deliver real-time motion images showing COVID-19 myocarditis registered by cardiac wall motion anomalies and delayed contrast enhancement within the scanned myocardium (Manna et al., 2020) . Progressive viral research confirmed the neurological pathology stimulated by COVID-19 with recorded brain damage, great vessel occlusion, and pediatric encephalitis (Shaikh et al., 2020) .

Mobile chest sonography is a staple imaging investigation occurring at the patient's bedside during an emergency or intensive care setting. Sonography is popular in medical imaging due to the convenience, omission of ionizing radiation, affordability, and consistency in scan replication and diagnostic findings. Point of care sonograms cited interrupted brightness mode lines, pulmonary consolidations, and pleural line thickness as signs of COVID-19 pneumonia. Followup sonograms of recovered COVID-19 patients displayed horizontal amplitude lines determining the remission of COVID-19 from the lung parenchyma (Manna et al., 2020) .

Positron emission tomography is a nuclear imaging procedure that traces inflammatory and oncological processes within the body through radiographic compounds denoted as tracers.

Fluorodeoxyglucose is the leading radiotracer in positron emission tomography (Shaikh et al., 2020) . COVID-19 inflammation can lead to residual FDG uptake conforming themes of GGOs and variable attenuation levels within the anatomy of interest (Manna et al., 2020) . Positron J o u r n a l P r e -p r o o f emission tomography scans with excess radiation dosage and prolonged scan time can identify the radiotracer motion and uptake within the whole body (Shaikh et al., 2020) .

The ambiguity of COVID-19 aggravated the concern of occupational exposure to vulnerable frontline responders attending to patients with high viral shedding levels, leading to hospital personnel's consequent infection. Artificial intelligence accelerated the expansion of healthcare technologies to challenge the transmissibility of COVID-19 . Medical imaging departments sustained significant impairments in patient management, department occupancy, and capitalization of existing resources indigenousness to the clinical facilities and the administrative health boards in the country of interest (Manna et al., 2020) .

Artificial intelligence agents automated medical imaging workflows in global imaging procedures during the COVID-19 pandemic with the example of patient monitoring. Modern imaging equipment utilized AI-enabled high-resolution cameras in mobile CT units using visual detection technologies to compute the patient habitus and parameters allied with image acquisition. Forward-looking infrared cameras and time of flight imaging are visual sense technologies that use artificial intelligence to outline the patient's anthropometry in terms of medical imaging. Artificial intelligence visual sensors execute autonomous computation of radiographic scan coverage and radiographic isocenters. Radiographic isocenters are the intersection between the center of the scan target and the scanner .

COVID-19 affiliated studies incorporated a pool of experimental conditions to establish the linkage between AI technologies and COVID-19 mitigation. Researchers designed the collection of scientific studies using various imaging modalities, artificial intelligence platforms, hospitalization settings, and patient demographics (Kundu et al., 2020) . Tables (5A, 5B Table 5B : Objectives, results, advantages and disadvantages of the included novel COVID-19-AI imaging studies.

The successive waves of COVID-19 reiterated the paradox of artificial intelligence in the discipline of medical imaging (O. S. . Artificial intelligence is a topic of avidity in the medical imaging society. Medical imaging experts welcomed AI features in routine medical imaging tasks that may encompass oncological staging, radiographic image reconstruction, disease triage, and radiographic segmentation (Reyes et al., 2020) .

Radiographic image segmentation composites a computed deconstruction of fundamental features of COVID-19 findings. AI-driven segmentation extracts individual structural components of the human lung corresponding to the pulmonary lobes, bronchial systems, and incidental lung infection or present neoplasia. Thus, radiologists are fostering suitable conditions to utilize computer-based quantification and evaluation of COVID-19 . The limitations of artificial intelligence are the large number of metadata and diverse platforms that require clinical facilities to apply reliable protocols in AI technique selection, integration, and interpretation of the computed medical imaging results in terms of COVID-19 (O. S. .

The fourth industrial revolution established technology at the helm of the fight against COVID-19. Technology directed the scene of smart interventions in several technology-driven countries across the seven continents counting South Korea, Japan, Germany, Iran, Italy, China, Turkey, Thailand, Canada, and many more (Fields et al., 2020) . The adopted smart solutions diversified in concept, target, and technique ranging from social media repurposing to applying artificial intelligence platforms to maximize the scope of contactless responses to COVID-19 (Budd et al., 2020) . This paper aimed to undertake the literary expedition in the pool of intercontinental literature discussing and deciphering the significance of technology in the battle of the novel coronavirus pandemic.

The method of this review adopted the principles of systematic literature reviews using a selective research protocol of four leading web-based databases to mine for literature about technology, medical imaging, and artificial intelligence during the present times of COVID-19. Thus, yielding 54 articles for targeted inclusion associated with the research direction and objectives in terms of keywords, research scope, and publication results. the included articles were tabulated into three letter-coded batches (A) digital technologies and COVID-19 (B) Artificial intelligence and COVID-19 (C) Artificial intelligence and medical imaging during COVID-19, and summarized in terms of aims, strengths, limitations and conclusion (Appendix A-D).

Batch (A) articles entailed 18 publications about digital technologies' notions in the mitigation of COVID-19. The findings of the discussed articles lead to the exploration of home quarantine, technology researchers, and stakeholders undertaking the precedence of COVID-19 (Kapoor et al., 2020; Whitelaw et al., 2020) . Further reflection on the articles in question allowed to deliberate on categorizing of technological interventions in the included literature. The eventual conclusions classified the technological interventions as digital interventions within clinical facilities, digital interventions outside clinical facilities, and digital interventions within the World Wide Web. The category of digital interventions within clinical facilities reflected on the ability of 3D printing to compensate for depleted PPE supplies, the multipurpose field of healthcare robotics to protect frontline responders, and the applications of telemedicine addressing the redefined conveniences of healthcare consultations, ICU units, and ER departments (Mastaneh & Mouseli, 2020; Scott et al., 2020; Tsikala Vafea et al., 2020) . Aerial drones were the prime focus of digital interventions outside clinical facilities due to the reiterated applications of autonomous supply deliveries to clinical facilities charged with the care of COVID-19 patients (HealthManagement.org & Team, 2020; Kapoor et al., 2020; Mastaneh & Mouseli, 2020) . Batch (A) was finalized with digital interventions within the World Wide Web discussing Geographic Information Systems and the internet of things to secure the databases of COVID-19 data (Adly et al., 2020; Mastaneh & Mouseli, 2020) . The provisions of digital technologies are capital expenditure, Internet accessibility, and the digital divide among the compromised demographics to COVID-19 (Budd et al., 2020; Keesara et al., 2020; Ramsetty & Adams, 2020) .

Artificial intelligence was the exclusive focus in batches (B) and (C). Batch B focused on the definitions of AI concepts, COVID-19 applications, ending at the investigation of challenges and J o u r n a l P r e -p r o o f prospects of artificial intelligence using 15 included articles from the systematic search protocol results. In terms of COVID-19 applications, artificial intelligence offers remote detection and diagnostics, expedited COVID-19 vaccine and therapeutics, redefined clinical workflow and revolutionized knowledge dissemination in a continuous apparatus to systematically address the novelty of COVID-19 (Bansal et al., 2020; Kim et al., 2020; Naudé, 2020) . Batch (C) convened 21 articles based on artificial intelligence and medical imaging interactions during the novel coronavirus pandemic. Batch (c) articles continued exploring artificial intelligence and medical imaging by assessing the radiographic workup of COVID-19 to conclude chest radiography and computed tomography as the primary imaging modalities (Shaikh et al., 2020) . Ahead of magnetic resonance imaging, point of care sonography, and positron emission tomography as secondary imaging modalities of COVID-19 (Manna et al., 2020) . The topic of novel AI-driven medical imaging experiments was the research direction of 12 research publications targeting the use of artificial intelligence technologies to facilitate the imaging workload of COVID-19 patients. For instance, convolutional neural networks can identify the patterns within medical radiographs to simplify the differentiation of COVID-19 from pneumonia-affiliated disorders (Salman et al., 2020) .

COVID-19 is the latest addition to a pedigree of infectious pandemics that spanned the last two decades. A lethal virus that paralyzed the cycle of life and captured the modern world within the walls of a residence and the World Wide Web. Scientists of the present millennia are equipped with the innovation and resourcefulness of the ever-expanding tech industry. Technology is a universe of algorithms, platforms, and contexts that redefine the lane of possibility for humanity.

Healthcare technologies aided frontline responders to strategize the response against the elusive virus leading to the advanced sequencing of the viral genome at a record speed and discovering innovative diagnostics to detect, quarantine, and treat confirmed cases worldwide. Reverse transcriptase-polymerase chain reaction swabs offered a means of direct diagnosis of COVID cases but a compromised accuracy depending on the conditions of the collected samples. The successful use of medical imaging, digital technologies, and artificial intelligence compensated for PCR tests disparate accuracy.

Medical imaging modalities screen and detect COVID-19 using chest radiography and computed tomography as gold standards in diagnosing suspected cases. Radiologists can repurpose magnetic resonance imaging, point of care sonography, and positron emission tomography as supportive modalities to visualize the disease's subtle and long term features. Artificial intelligence is a debate of elation and opposition among medical imaging professionals.

The intelligent agents proved to deliver a consistent diversity of applications within healthcare facilities and beyond. Artificial intelligence technologies exposed healthcare professionals to the opportunity of contactless care. Contactless care allows frontline responders to tend to fragile COVID-19 patients and minimizes the likelihood of inherent occupational exposure to the depleting disease.

The opposition of artificial intelligence stems from the requirements and early research that qualify novel AI platforms for generalization and public use. Artificial intelligence agents require vigorous programming, investment, training, and research process to validate the platform's efficacy, application rationale, without the exception of the limits of predictability in terms of the platform's direction and future use.

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. To highlight and summarize the applications of AI and big data in the global efforts against COVID-19.

Explains the potential of AI and big data in the mitigation of COVID-19 using a hierarchal approach that includes short, mid, and long term application.

Lacks real-time applications that might provide additional illustration and understanding of AI and big data implementation in the COVID-19 response.

Big data and artificial intelligence can provide short, mid, and long term applications that may influence COVID-19 and future development concepts. 

Approaches Based on artificial intelligence and the internet of intelligent things to prevent the spread of COVID-19: Scoping review

Review of big data analytics, artificial intelligence and nature-inspired computing models towards accurate detection of COVID-19 pandemic cases and contact tracing

Multi-Biological Laboratory Examination Framework for the Prioritization of Patients with COVID-19 Based on Integrated AHP and Group VIKOR Methods

Helping doctors hasten COVID-19 treatment: Towards a rescue framework for the transfusion of best convalescent plasma to the most critical patients based on biological requirements via ml and novel MCDM methods

Systematic review of artificial intelligence techniques in the detection and classification of COVID-19 medical images in terms of evaluation and benchmarking: Taxonomy analysis, challenges, future solutions and methodological aspects

The Rise of Machine Intelligence in the COVID-19 Pandemic and Its Impact on Health Policy

Extracting Possibly Representative COVID-19 Biomarkers from X-ray Images with Deep Learning Approach and Image Data Related to Pulmonary Diseases

The Role of Chest CT Scan in Diagnosis of COVID-19

AI Augmentation of Radiologist Performance in Distinguishing COVID-19 from Pneumonia of Other Etiology

Utility of Artificial Intelligence Amidst the COVID 19 Pandemic: A

Lung ultrasound as diagnostic tool for SARS-CoV-2 infection. Internal and Emergency Medicine

How Big Data and Artificial Intelligence Can Help Against COVID-19

Digital technologies in the public-health response to COVID-19

Clinical role of lung ultrasound for diagnosis and monitoring of COVID-19 pneumonia in pregnant women

False negative chest X-Rays in patients affected by COVID-19 pneumonia and corresponding chest CT findings

Can AI Help in Screening Viral and COVID-19 Pneumonia?

Application of Lung Ultrasound during the COVID-19 Pandemic: A Narrative Review

Will COVID-19 be the tipping point for the Intelligent Automation of work? A review of the debate and implications for research

COVID-19) infection: findings and correlation with clinical outcome

Machine learning to assist clinical decision-making during the COVID-19 pandemic

Reverse Vaccinology: Exploiting Genomes for Vaccine Design

Chest CT findings of COVID-19 pneumonia by duration of symptoms

Rapid implementation of mobile technology for real-time epidemiology of COVID-19

COVID-19 Pandemicconsiderations and challenges for the management of medical imaging departments in Queensland

EasyChair Preprint № 3223 Artificial Intelligence Approach to Predict the COVID-19 Patient's Recovery Artificial Intelligence Approach to Predict the COVID-19 Patient's Recovery

Covid-19: Correlation of early chest computed tomography findings with the course of disease

Digital health strategies to fight COVID-19 worldwide: Challenges, recommendations, and a call for papers

COVID-19 and the role of imaging: Early experiences in Central Switzerland

A case report of COVID-19 with false negative RT-PCR test: necessity of chest CT

Coronavirus Disease 2019 (COVID-19) diagnostic technologies: A country-based retrospective analysis of screening and containment procedures during the first wave of the pandemic

Black Death and the silver lining: Meaning, continuity, and revolutionary change in histories of medieval plague

History of the Plague: An Ancient Pandemic for the Age of Covid-19

Position paper on COVID-19 imaging and AI: From the clinical needs and technological challenges to initial AI solutions at the lab and national level towards a new era for AI in healthcare

Artificial intelligence for the detection of COVID-19 pneumonia on chest CT using multinational datasets

Digital Response During the COVID-19 Pandemic in Saudi Arabia

COVID-19: What Can Healthcare Learn? HealthManagement

Creating ethnography: Zora Neale Hurston and Lydia Cabrera

AI Techniques for COVID-19

Implementation of a deep learning-based computeraided detection system for the interpretation of chest radiographs in patients suspected for covid-19

Digital healthcare: The only solution for better healthcare during COVID-19 pandemic?

Covid-19 and health care's digital revolution

The influenza pandemic of 1918-1919 in the British caribbean

A Review for Artificial Intelligence Proving to Fight Against COVID-19 Pandemic and Prefatory Health Policy

COVID-19 pneumonia diagnosis using a simple 2d deep learning framework with a single chest CT image: Model development and validation

Artificial intelligence-powered search tools and resources in the fight against covid-19

A review of modern technologies for tackling COVID-19 pandemic. Diabetes and Metabolic Syndrome

How Might AI and Chest Imaging Help Unravel COVID-19's Mysteries?

Applications of machine learning and artificial intelligence for Covid-19 (SARS-CoV-2) pandemic: A review

Using Artificial Intelligence to Detect COVID-19 and Community-acquired Pneumonia Based on Pulmonary CT: Evaluation of the Diagnostic Accuracy

Coronavirus disease 2019 (COVID-19): Role of chest CT in diagnosis and management

From community-acquired pneumonia to COVID-19: a deep learning-based method for quantitative analysis of COVID-19 on thick-section CT scans

The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration

Combat COVID-19 with artificial intelligence and big data

Restructured society and environment: A review on potential technological strategies to control the COVID-19 pandemic

Multimodality Review of Radiologic Techniques, Clinical Utility, and Imaging Features

Technology and its Solutions in the Era of COVID-19 Crisis: A Review of Literature. Evidence Based Health Policy

Integrating emerging technologies into COVID-19 contact tracing: Opportunities, challenges and pitfalls. Diabetes and Metabolic Syndrome

Quibim Combines AI and Imaging to Detect COVID-19

Artificial intelligence-enabled rapid diagnosis of COVID-19 patients

Can CT performed in the early disease phase predict outcome of patients with COVID 19 pneumonia? Analysis of a cohort of 64 patients from Germany

Imaging of COVID-19 pneumonia: Patterns, pathogenesis, and advances

Artificial intelligence vs COVID-19: limitations, constraints and pitfalls

A deep learning approach to characterize 2019 coronavirus disease (COVID-19) pneumonia in chest CT images

Can artificial intelligence predict the need for oxygen therapy in early stage COVID-19 pneumonia?

Pediatric chest x-ray in covid-19 infection

Automated detection of COVID-19 cases using deep neural networks with X-ray images

Machine learning-based method for interpreting the guidelines of the diagnosis and treatment of COVID-19

Impact of the digital divide in the age of COVID-19

Diagnosis of COVID-19 Pneumonia Using Chest Radiography: Value of Artificial Intelligence

On the Interpretability of Artificial Intelligence in Radiology: Challenges and Opportunities

Coronavirus (COVID-19) Cases -Statistics and Research -Our World in Data

COVID-19 Detection using Artificial Intelligence

Advanced Digital Health Technologies for COVID-19 and Future Emergencies

Current Landscape of Imaging and the Potential Role for Artificial Intelligence in the Management of COVID-19

COVID-19: Is it the black death of the 21st century? Health Promotion Perspectives

Review of Artificial Intelligence Techniques in Imaging Data Acquisition, Segmentation and Diagnosis for COVID-19

End-to-end automatic differentiation of the coronavirus disease 2019 (COVID-19) from viral pneumonia based on chest CT

A Review of Early Experience in Lung Ultrasound in the Diagnosis and Management of COVID-19

Artificial Intelligence of COVID-19 Imaging: A Hammer in Search of a Nail

Applications of artificial intelligence in battling against covid-19: A literature review

Emerging Technologies for Use in the Study, Diagnosis, and Treatment of Patients with COVID-19

Interpretable artificial intelligence framework for COVID-19 screening on chest X-rays

How artificial intelligence and machine learning can help healthcare systems respond to COVID-19

Imaging manifestations and diagnostic value of chest CT of coronavirus disease 2019 (COVID-19) in the Xiaogan area

DeepCOVID-XR: An Artificial Intelligence Algorithm to Detect COVID-19 on

Applications of digital technology in COVID-19 pandemic planning and response. The Lancet Digital Health

WHO Coronavirus Disease (COVID-19) Dashboard. World Health Organization

A Deep Learning System to Screen Novel Coronavirus Disease

Rapid identification of COVID-19 severity in CT scans through classification of deep features

Trade routes and plague transmission in preindustrial

COVID-19 pathophysiology: A review

Automated detection and quantification of COVID-19 pneumonia: CT imaging analysis by a deep learning-based software

Deep transfer learning artificial intelligence accurately stages COVID-19 lung disease severity on J o u r n a l P r e -p r o o f portable chest radiographs

Emerging Technologies for Use in the Study, Diagnosis, and Treatment of Patients with COVID-19

Advanced Digital Health Technologies for COVID-19 and Future Emergencies

Digital Response During the COVID-19 Pandemic in Saudi Arabia

The authors would like to thank the Radiography and Medical Imaging senior students (Aisha Yunuis Mohamed Hasan Al Hammadi, Emtithal Yousif Mohamed Osman, Fatimah Mirza Mohamed Saleh and Sarah Ahmed Khameis Almotawa) for providing help during the research.J o u r n a l P r e -p r o o f

Lacks other studies that might provide some additional perspective in the scope of AI and COVID-19.The field of medical AI applications remains at early stages which was evident in the lack of literature pertaining to medical AI applications (i.e. resource allocation and experimental therapeutics).