top of page
Search
Priyanshu Shekhar

Artificial Intelligence for COVID-19: Applications, Limitations, and Constraints.


1. Introduction

The COVID-19 pandemic has harmed the entire globe. Artificial Intelligence (AI) could prove to be the tool that the entire humanity needs in these testing times. AI has been one of the most talked-about technologies in the detection and prevention of the spread of this deadly virus. This article provides an overview of the potential contribution and applications of AI against COVID-19. AI in its present form has been estimated to track the spread of this virus that can identify high-risk patients. But there have been certain challenges and constraints related to the use of AI against COVID-19. The AI Models require training data on which they can train so that desired results can be discerned. Can AI ensure Data Privacy along with the rapid influx of information? These include even lack of data, too much data which imposes another level of challenge in this regard.


To start with, let's discuss the contribution and potential applications of AI against COVID-19.


2. Contributions and applications of AI for COVID-19

In the ongoing world-health crisis caused by the COVID-19 pandemic, the medical industry is looking for technologies that can help to monitor and control the spread of COVID-19. AI has been identified as one of those technologies that have the potential to easily monitor the spread of this virus, identify the patients which are at high risk, and can be useful in evaluating the risk in real-time. Some studies have also predicted that AI can even improve the planning, treatment, and reported outcomes of the COVID-19 patients. The applications and contributions of AI in the COVID-19 pandemic are as follows:[3]


2.1 Improved Diagnosis

Accurate and improved diagnosis of COVID-19 will not only help in saving lives but will also control the transmission of the Virus and can even generate data on which AI Models will be trained. AI can help achieve improved diagnosis by estimating the irregular symptoms and thus will help in making the authorities aware of the situation. AI is helping the diagnosis of cases that have been infected by the use of medical imaging technologies which include Computed tomography (CT), Magnetic resonance (MRI) Scans.[2]


2.2 Contact tracing and prediction

AI can prove to be a very useful tool when it comes to contact tracing and predicting how the COVID-19 Virus will behave and spread over time. AI can even help in estimating the level of infection by finding "hot spots' ' of the disease and do contact tracing of the individuals. AI will eventually assess the future scope of this virus and its likely recurrence.[2]


2.3 Development of Vaccines and Cures

AI was even praised for its potential in developing a new drug and accelerating the entire research process. This was evaluated way before the arrival of COVID-19 that AI-based techniques will help in drug research and Vaccines which can hold 100 percent efficacy. AI will help in the entire process of drug development by analyzing the available data on COVID-19 and it will eventually speed up the produce that will not only reduce the time and efforts but will also ensure better treatment and cure against COVID-19.[4]


2.4 Workload Management of Healthcare Workers

As we all are aware of the fact that the increase in the number of patients infected by this virus has resulted in the extensive workload of the healthcare workers. AI with its ability that can help in early diagnosis and provide an efficient treatment which will eventually ease their work. It can also help in providing efficient training to students and doctors about the new disease and by tackling the future potential challenges which will help in managing the workload of healthcare workers.[4]


2.5 Prevention of Diseases

AI has been and can be used further in helping the prevention of not only in a present pandemic but also in future other pandemics as well. With the use of various data and by the implementation of proper analysis of them, AI models can be trained which will be helpful for future viruses. AI is estimated to become an even better technology for disease prevention of futures epidemics by developing a better algorithm with the help of data gathered from this present pandemic. [2]


3. Constraints and Limitations

As estimated and analyzed by various studies, AI has the potential to be an important and helpful tool in the fight against COVID-19 and similar other pandemics. AI-based techniques despite having enormous potential in the fight against COVID-19 and similar other pandemics had to face certain constraints upon which it has been put forward that it is still at a preliminary stage and it will some amount of time after which its measures can be seen visible efficiently.[1]


One of the first constraints and limitations that had been impacting is lack of sufficient data and on the other way around, too much data. AI Models require certain data, preferably historical on which they can be trained but it has been lacking. There are not enough datasets and models on which we can work leading to the problems of non-adjustment of algorithms and concrete information to work on.[1] Thus, it is certain that more training data is needed on COVID-19 so that we can utilize it better for future epidemics and Viruses. So, more openness on sharing of information will have to ascertain the required amount of data.


On the other hand, it's not only about lack of data but too much data is also hindering the AI techniques to show their full capability. As the news about COVID-19 and pandemic reaches everyone through Social Media and other means, too much noisy and outlier data accumulates which then gets difficult to comprehend the useful data and information on which we have to work. [5]


4. Conclusion

AI has the potential to be an effective tool against the fight against COVID-19 but it has been constrained and limited by certain other factors. It has to be concluded that AI techniques are still at a preliminary stage before their actual results can be visible. In one of the studies, it was concluded that very few of the reviewed AI Systems have functioning maturity at this present stage. Data is central to whether AI can be a beneficial tool against COVID-19 and future other pandemics. There has been an apprehension that public health fears will overcome the data privacy concerns. But the government is continuously looking and surveillance of the citizens even after the pandemic is over will result in positive outcomes in the future to come. Thus the continuous questioning of Data Privacy is justified. Given the public threat caused by the pandemic, European GDPR[4] ( General Data Protection Regulation) has allowed personal data collection and analysis keeping the mind that it should have a clear focus on public health, and handling of such personal data should be taken care of. Thus it has become quite necessary for them to ensure particular care in handling such data. Although AI's impact has been limited, the public policy and other responses associated with it can accelerate the digitalization of the economy. Also, innovation in AI technology may be the result of the present crisis but rapid progress will be required to lay down the appropriate mechanism.


References

  1. Naude W (April 2020). Artificial Intelligence Vs COVID-19: Tracking and Prediction; 764-765

  2. Vaishya R; Javaid M; Khan I; (April 2020). Artificial Intelligence (AI) applications for COVID-19 pandemic: Main applications of AI in COVID-19; 337-339

  3. Husain A; Bouachir O; AI (July 2020). AI Techniques for COVID-19: Comparison to other studies; 128776-128777

  4. Naude W (April 2020). Artificial Intelligence against COVID-19: An Early Review: Actual and potential Contribution of AI against COVID-19; 2-10

  5. Bhat S; Sirchar S; Dhama K; Malik K (November 2020). How Artificial Intelligence may help the COVID-19 pandemic: Pitfalls of AI; 5-8


 

By Priyanshu Shekhar













106 views0 comments

Comments


bottom of page