Abstract: Artificial intelligence is a quickly spreading branch of computer science that deals with solving problems without the help of humans. It has played a huge role in solving numerous problems that were previously impossible to do without the help of machines. Artificial intelligence has not only helped in education and business but it has also played a huge role in medicine. Many patients have received the treatment that previously they were not getting. The most important use of artificial intelligence in today's world is during the COVID-19 crisis. Artificial intelligence has helped us detect, control, and suggest treatment methods for affected people. Although artificial intelligence has faced many challenges in its development, it is rapidly gaining importance in all fields. AI has helped us understand how COVID-19 affects the human body, thus enabling us to better identify how to fight it. Artificial intelligence is sometimes difficult to understand without an expert. In cases where there is no data reference for the AI to look at, it is not able to produce an appropriate answer for a query. Artificial intelligence is the most powerful weapon we have in our fight against COVID-19.

Use of Artificial Intelligence (AI) in dealing with COVID-19 cases

Image by Sasin Tipchai from Pixabay 

Artificial intelligence is a branch of computer science that deals with the development of machines that can perform tasks which were typically meant to be performed by humans. It is the simulation of the thought process of humans by machines, especially by computer systems.

Artificial intelligence is composed of three different cognitive skills:

a) learning -The learning process of AI works by acquiring data about the task to be performed and creating certain rules called “algorithms” which provide instructions on how the task is to be completed.

b) reasoning - The training process is responsible for finding the easiest and most effective algorithm to complete the task.

c) self-correction - The self-correction process is responsible for fine-tuning the algorithms and correcting the errors, if there are any, to ensure that they give the best results.

Stuart Russel and Peter Norvig state in their book, Artificial Intelligence-A Modern Approach, - “ AI is the study of agents that receive percepts from the environment and perform actions .” Norvig and Russel have explained four different approaches that have played a great role in defining the role of artificial intelligence :

  1. thinking humanly
  2. thinking rationally
  3. acting humanly
  4. acting rationally

The first two ideas comprise thought processes and reasoning, while the other two deal with behavior.

The first use of artificial intelligence in medicine was seen in the 1960s and 1970s when researchers at Stanford University produced the first expert system ( a computer system that imitates the decision-making ability of a human) called Dendral. While it was originally meant to be used for applications in organic chemistry, it proved to be the basis for a subsequent system called MYCIN which was later considered to be one of the earliest used of artificial intelligence in medicine. Researchers and developers started to realise that artificial intelligence systems in healthcare would prove to be immensely helpful in compensating for the absence of perfect data systems and to increase the expertise of physicians.

In the time during 1960s to 1970s, the applications of artificial intelligence in healthcare have grown to include:

  • growth of genome sequencing
  • enhancing the precision of robot-assisted surgery c) worldwide implementation of electronic healthcare systems, etc.

Artificial intelligence is used in six healthcare systems – hospitals, pharmaceuticals, diagnostics, medical equipment and supplies, medical insurance and telemedicine. Artificial intelligence applications can help doctors tackle problems such as uneven doctor-patient ratio, by providing high-quality healthcare to rural populations and by training doctors and nurses in handling complex medical procedures.

Artificial intelligence has had a huge impact in the healthcare industry. Its success can be largely attributed to the rapidly growing accessibility of healthcare data. Artificial intelligence is also responsible for advancements such as organizing better treatment plans, effective data analyzation,  and proper monitoring of treatments. It is also able to identify the early signs of diseases in X-rays, CT scans, and ultrasounds quickly and more accurately which is really beneficial for patients who previously had to wait for their diagnosis report. Now, the process can be done in a few hours, thus accelerating the treatment process.

AI can enable the next generation of radiology tools to be more accurate and detailed and it will also replace the need for tissue samples in surgery thus reducing the risk of infection. AI also helps in providing healthcare benefits to the underserved and the developing regions where the healthcare facilities are not adequate .

AI has also enabled patients to get answers to their queries instantly without having to wait in long appointment queues. Doctors can also keep in touch with their patients and remind them to take their medication and to take proper precautions.

In medical procedure, the most important factor to be considered is patient safety. The progress of artificial intelligence in healthcare, while extremely beneficial, also had some serious drawbacks. Doctors, physicians and nurses, though extremely grateful for all the help AI has given them, also care deeply about their patients and they prioritise the time spent with their patients face-to-face. And they are not completely satisfied with the new electronic method of communicating. Face-to-face talks between the doctor and patient play a huge role in boosting the patient’s spirits.

Using artificial intelligence also carries security risks. One of the biggest reasons for this risk is that most algorithms require massive amounts of data for processing. And storing all this data in a single

repository makes it a very attractive target for hackers and cyber criminals. Due to this reason, consumers aren't exactly jumping on board with the idea of sharing their details online.

Another disadvantage of AI in healthcare is the shift in distribution. While entering data in the EHS (electronic health systems), a slight change or mismatch in data due to environmental changes and circumstantial changes can result in errors in the prediction of AI. While the error may appear to be insignificant, it may lead to disparities in training and operational order.

Oversight is also one of the problems of AI. Since it is capable of performing many tasks and activities, the monitoring and maintenance of such machinery is a nearly impossible task.

Artificial intelligence is also of great help in detecting pandemic before they go global. In the case of infectious diseases; prevention, surveillance and rapid-response efforts are a great help in slowing or stopping outbreaks. It also creates huge challenges for health officials in gathering data about the outbreak, so this is where AI comes in. AI algorithms can detect pandemics before they reach global proportions by mining through news report and online content and detecting anomalies and helping experts in curbing the efforts of the outbreak before they get out of hand.

AI can help rid the world of some of the most infectious diseases by successful implementation of predictive modelling (a process that seeks to predict future outcomes that are likely to occur based on recurring patterns.

AI and big data analytics are a huge help in detecting spread of disease. High resolution simulations allow scientists to track anomalies and inconsistencies in the genome sequence of human in a worldwide basis. This process is extremely helpful in detecting warning signs before it is too late.

But too much dependence on AI is not entirely wise. Too much confidence in AI capabilities leads to ill-informed decisions which will lead to the loss of funding or transfer of funds to the wrong place.

These days the crisis that has enveloped the globe is the COVID-19 pandemic. And the first step towards the fighting of any crisis is to know all the facts there are too know about it. So, if we want to prepare against fighting this crisis, we need to know all the facts about it.

Corona viruses are a large family of viruses which are responsible for various illnesses in animals and humans. In humans, these viruses cause respiratory infections ranging from the common cold to more severe diseases such as Middle East Respiratory Syndrome (MERS) and Severe Acute Respiratory Disorder (SARS).

COVID-19 was caused by SARS-CoV-2. It was first identified in December 2019 in Wuhan, and since then has spread globally.

The common symptoms of COVID-19 include fever, cough, fatigue, shortness of breath, and loss of smell and taste. The majority of cases result in mild symptoms, but some cases may progress to viral pneumonia, multi-organ failure or cytokine storm (a form of systemic inflammatory response syndrome that is easily triggered by infections).

These symptoms usually manifest themselves around five days but the time may also range from two to fourteen days. Older people, and those with high blood pressure; heart and lung problems or diabetes are at higher mortality risk if they get affected by COVID-19.

The virus is primarily spread between people by close contact, by sneezing or by inhaling cough droplets of an infected person. These droplets may also fall on surfaces such as table-tops, door knobs and when a person touches these surfaces, he or she is under risk of being infected.

According to WHO, there is no specific vaccine or treatment for COVID-19. But, preventive measures can be taken; such as frequent hand washing, maintaining physical distance from each other, covering your mouth while coughing, and keeping unwashed hands away from one's face.

There are different ways to manage patients who have been affected, which include fluid therapy (the medical practice of replacing the bodily fluid lost through sweating, bleeding, etc), oxygen therapy,  and supporting the vital affected organs. If the symptoms indicate an early stage of infection, then they can be treated with mild therapy.

Doctors are still unsure about how the virus is killing people. It is unclear whether the virus affects the person’s immune system or whether the effect of the virus on the body's vital organs has overwhelmed the body's response system to diseases.

The process of infection has been deducted from case studies and monitoring patients.

When an infected person regards virus-laden droplets and they are inhaled by a healthy person, the virus enters the lining of the nose and throat. Once it reaches there, the virus binds to a receptor in the cells called ACE2. This receptor regulates the blood pressure and also makes the cells vulnerable to attack. Once the virus gets inside the cell, it makes numerous copies of itself and spreads throughout the body. As the virus multiplies, the body begins to show symptoms like cough, fever, etc. If the immune system is not able to fight the virus at this stage, then it travels directly to the lungs and then into the blood vessels where it battles with the healthy white blood cells. This further disrupts the immune system leading to Acute Respiratory Distress Syndrome (ARDS) and eventually death.

Artificial intelligence has helped in the fight against COVID-19 by:

  1. early warnings and alerts
  2. tracking and predictions
  3. data dashboards
  4. diagnosis and prognosis
  5. treatments and cures
  6. social control

COVID-19 was first detected by an AI tool called BlueDot. It was founded by a startup investment of around USD 9 million and it is said to be able to out-predict humans in detecting infectious diseases. BlueDot predicted the outbreak of the infection in December 2019 before the World Health Organization predicted it on 9 January 2020. BlueDot researchers also published a notice in the Journal of Travel Medicine on 14 January 2020, where it mentioned the top 20 destination cities where passengers from Wuhan were most likely to arrive. The notice warned that these cities could be at the most risk during this pandemic.

There were two other cases of COVID-19 being detected. HealthMap, an AI-based model at Boston's Children's Hospital in the USA, published a notice about the virus a day earlier than BlueDot, and only 30 minutes later, a scientist working at the Program for Monitoring Emerging Diseases issued an alert about the same. The warning issued by the hospital was not given as much significance as the warning issued by the scientist.

It is therefore concluded that human input is also needed for the optimum appreciation of AI.

There are a few problems that AI faces in detecting COVID-19. These include a lack of historical data by which to train the AI on identifying it and also the characteristics of COVID-19 are different from the previous pandemics, so there is no reference on how to fight it.

As a result of lack of data, too much extraneous information and data hubris and noisy social media, AI forecasts about the spread of COVID-19 are not entirely reliable.

It is no doubt that AI is definitely a boon for us in these troubled times. But AI has not always been affordable to the society. During the past fifty years, AI was only available to an exclusive circle of researchers and scientists. That is steadily changing as packages of AI are being made available to all sorts of institutions, organisations, and start-ups.

In this global crisis, we need to keep ourselves prepared and take all necessary measures to keep ourselves and our families safe. Artificial intelligence is progressing greatly and holds the solutions to almost all our problems, and to the problems it doesn't provide a solution to, it provides us means to create the solutions ourselves.

.    .    .