COVID-19 Diagnostics, Tools, and Prevention

Abstract

The Coronavirus Disease 2019 (COVID-19), caused by the severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), outbreak from Wuhan City, Hubei province, China in 2019 has become an ongoing global health emergency. The emerging virus, SARS-CoV-2, causes coughing, fever, muscle ache, and shortness of breath or dyspnea in symptomatic patients. The pathogenic particles that are generated by coughing and sneezing remain suspended in the air or attach to a surface to facilitate transmission in an aerosol form. This review focuses on the recent trends in pandemic biology, diagnostics methods, prevention tools, and policies for COVID-19 management. To meet the growing demand for medical supplies during the COVID-19 era, a variety of personal protective equipment (PPE) and ventilators have been developed using do-it-yourself (DIY) manufacturing. COVID-19 diagnosis and the prediction of virus transmission are analyzed by machine learning algorithms, simulations, and digital monitoring. Until the discovery of a clinically approved vaccine for COVID-19, pandemics remain a public concern. Therefore, technological developments, biomedical research, and policy development are needed to decipher the coronavirus mechanism and epidemiological characteristics, prevent transmission, and develop therapeutic drugs.

Keywords: 3D printing; COVID-19; SARS-CoV-2; digital tracking; do-it-yourself; immunity; machine learning; pandemic policy; rapid testing; vaccines.

Figure 1

COVID-19 origins, transmission, immunity, and diagnostics. (a) Animals such as bats are a potential natural reservoir, and other intermediates facilitate the infection of coronavirus to humans. Human-to-human transmission is a common characteristic in three types of coronaviruses, severe acute respiratory syndrome coronavirus (SARS-CoV), SARS-CoV-2, and Middle East respiratory syndrome coronavirus (MERS-CoV). Even though the fatality rate is not lower than SARS-CoV and MERS-CoV, vaccines and treatments are needed; (b) Herd immunity plays an essential role in controlling the transmission of the disease. Without herd immunity, the virus will be available to get transferred from infected patients to any susceptible individual, resulting in a high infection rate and high mortality. On the other hand, herd immunity interrupts this chain of infection, resulting in better outcomes with less infected individuals and a significantly reduced the mortality rate; (c) The three main categories of COVID-19 diagnostics and the time-frame of each: laboratory test centers for current infection, medical screenings, antibody tests to determine immunity. Test centers use samples from nasopharyngeal swabs and test for COVID-19 infection via molecular or reverse transcription-polymerase chain reaction (RT-PCR) tests, with RT-PCR tests taking longer. Healthcare clinics and hospitals offer quick medical screenings via computed tomography (CT) scans or ultrasounds; ultrasounds are the safer option due to no radiation. The next step in rapid testing will be to test COVID-19 patients for antibodies against SARS-CoV-2 to see if the patient has the potential to fight reinfection and determine if the individual was previously sick.

Figure 2

SARS-CoV-2 air travel, mask’s effect on transmission, and temperature dependence. (a) The spread of virus particles depending on coughing and sneezing, as well as particle size. The mode of ejection affects velocity and distance traveled, while the size affects the ability to stay in the air; (b) The difference in the spread of viral load and aerosol droplets with a mask and without a mask; (c) The relationship between the mean replication rate of SARS-CoV-2 and the temperature of 24 different countries is demonstrated. There is a moderate negative correlation between the mean replication rate and temperature, with a p value < 0.001. Data were taken from Caspi et al. [55] “Climate Effect on COVID-19 Spread Rate: An Online Surveillance Tool” with permission.

Figure 3

Protection tools against COVID-19 in the form of face masks, shields, and sanitizers. (a) The N95 Respirator is Food and Drug Administration (FDA)-approved and is currently the best-known form of personal protective equipment (PPE) available to healthcare professionals; (b) FDA-approved surgical masks are to be used by healthcare workers in the absence of N95 respirators; (c) The regular method for tying surgical masks compared to the newly proposed method that can secure the side openings of surgical masks; (d) A surgical mask design modified with filter performance rating (FPR) 9-10 air-conditioning filters; (e) The Stanford full-face, snorkel-inspired Pneumask. The Pneumask has deployed 2600 units all across the United States (US), with concentrations in New York, Massachusetts, and California; (f) The team at Duke has fully modified a powered air-purifying respirator (PAPR) design, which is attached to the surgical helmets (left) and the 3D-printed component of the PAPR (right); (g) The 3D printable Stopgap Mask from the Veterans Health Administration that is approved by the FDA to protect clinical personnel from COVID-19; (h) The 3D printable Maker Mask that can be used by non-clinical personnel when there are no FDA-approved masks available; (i) The Prusa Printer face shield that can be used to protect clinical personnel when no FDA-approved face shields are available; (j) The Georgia Tech 3D-printed rigid frame and a disposable face shield design; (k) The MIT origami disposable face shield. The Georgia Tech and MIT labs have collaborated to produce almost one million face shields to be used by hospitals around the US; (l) An illustration of the transparent acrylic barrier concept for public settings where cross-contamination is frequent. Aviointeriors has designed a possible solution for plane seats, while Safetell and Umdasch have designed and implemented their solution for retail stores; (m) The sanitizer station is a representation of the Umdasch Hygiene Station that could be distributed to all types of public facilities in the future as a norm.

Figure 4

COVID-19 control by gloves and ventilators. (a) Gloves must be changed regularly and properly disposed of in biosafe disposal units. Not all gloves block the penetration of the virus. Thicker gloves offer a thicker barrier and better defense against COVID-19 droplets; (b) (1) The general design for new ventilators includes a breath sensor that senses the timing and size of breaths, a pressure pump that pumps air into and out of the lungs, and a tracheal tube that pumps air directly into the lungs; (2) Depiction of the Massachusetts Institute of Technology (MIT) New “Bridge” ventilator which uses a mechanical pump and a bag-valve-mask to administer air (left). The National Aeronautics and Space Administration (NASA) VITAL uses sensors and a mechanical pump to deliver safe breaths. Their design is licensable to accepted manufacturers (middle). A standard intensive care unit (ICU) ventilator is fully equipped with vitals monitoring, air delivery, and a computer that measures vitals and pressure for safe breathing. The safest option for patients in critical condition but are more expensive to produce are presented (right); (3) The Food and Drug Administration (FDA)’s Emergency Use Authorization (EUA) for mechanical ventilators helped drastically reduce the cost of new devices and increased the number of new device designs; (4) Newly developed ventilators are termed “bridge” ventilators and are intended for an intermediate use between hospitalized patients and critically ill patients. “Bridge” ventilators are used when standard ICU ventilators are not available, and standard ICU ventilators are still the safest option for patients in critical condition.

Figure 5

Machine learning analysis of COVID-19 dynamics. (a) Protein unfolding prediction. The first step is a multiple sequence alignment that compares a protein’s sequence with a similar one, suggesting a close location in the folded protein. The neural network learns to accurately predict the distance by training on precisely measured distances in proteins. In parallel, another neural network is trained to predict the angles of the joints in the folded protein chain. Finally, gradient descent is used to find a physically possible folding of a protein; (b) A deep-learning framework on a CT scan. First, the lung region is extracted using a segmentation method. Then, the corresponding image of the lung region is passed onto the deep-learning model that predicts the risk value based on the CT scan; (c) Machine Learning pipeline for patient risk forecasting; (d) Pandemic forecasting using machine learning under various quarantine rules. The model predicted the infected case count of the US with quarantine control and the exponential increase of cases if there were no quarantine measures. It also shows predicted case counts under various quarantine measures, as implemented in Wuhan, Italy, and South Korea. All of them led to a plateau of cases sooner than the US model; (e) COVID Scholar website for scientific article search. Searching for the term “rapid test” gives all journal articles related to the term. This can be further filtered depending on the year, the type of document, whether they were peer-reviewed, and whether they are specific to COVID-19.

Figure 6

Simulation tools for COVID-19 pandemic modeling and predictions. (a) Comparison of epidemic curves with (red) and without (blue) intervention, as suggested by the CDC; (b) The sequence of events initiated by utilizing the results of a simulator is shown; (c) The main factors considered for the COVID-19 simulators, such as measures of social distancing and probability of community spread; (d) Comparison of projected curves resulting from various simulations that cover different scenarios of social distancing, for the months of March to January in the US; (e) Projection of the number of active cases, those hospitalized, and deaths from COVID-19, simulated between May 3 to May 31 in the US; (f) A rendition of the billiards balls model by the Washington Post, to illustrate the benefits of social distancing. (LEFT) Simulation rendition which depicts effects caused if social distancing is not followed. (RIGHT) Simulation rendition, if social distancing is followed. The simulation renditions are followed by epidemic curve graphs; (g) A platform provided by the COVD-19 Visualizer website, which updates the number of cases worldwide in real-time; (h) A rendition of the comparison of mortality cases related to COVID-19, influenza, and pneumonia in the US, as visualized by the CDC; (i) A rendition of the various visualizations on COVID-19 spread worldwide, provided by the Johns Hopkins website; (j) Screenshot of an interactive simulator involving various metrics that predicts the severity of COVID-19 spread.

Figure 7

Digital monitoring tools of COVID-19 spread by online and cellphone applications. (a) General smartphone app-based contact tracing—the social encounters (distance less than 6 feet) of all the users are recorded by the app, using methods such as GPS information, QR code-tracking, and Bluetooth communication. An asymptomatic patient could infect others in their family, commute, or workplace. The patient takes the COVID-19 test (using the app) when individuals show signs of symptoms, such as fever. When the patient tests positive, instant notifications are sent to individuals who could be infected by the virus. Individuals are asked to take the COVID-19 test and self-isolate themselves for 14 days. The places where the patient went before testing positive are also thoroughly disinfected to prevent the spread of the disease; (b) Bluetooth-based contact tracing, as proposed by the MIT Lincoln Laboratory, does not require personal information or location to be shared. The app keeps a record of Bluetooth signals it has encountered, along with information about the distance between the devices and duration for which other individuals were nearby. When a user tests positive, the app sends the contact record to the database of anonymized contacts, which then sends exposure notifications to the users who might have been in close contact with the patient. This app works in a privacy-sensitive manner, without revealing user information to other individuals and health authorities; (c) Examples of apps that use surveys to track the symptoms of users to monitor the prevalence of the disease and identify potential hotspots: screenshots from a generic symptom tracking app to show sample survey questions, COVID Symptom Study (by Massachusetts General Hospital, the Harvard T.H. Chan School of Public Health, King’s College London and Stanford University School of Medicine, working with ZOE—a health science company) and the Healthy Together (by State of Utah, USA); (d) Examples of contact tracing apps: screenshots from a generic contact tracing app to show user interface and official government-developed apps—TraceTogether (Singapore) and COVIDSafe (Australia). These apps use Bluetooth-based contact tracing and provide region-specific healthcare guidelines and information.