Artificial Intelligence and COVID-19 Vaccines

COVID-19 rapidly transformed into a global challenge, costing thousands of lives, overwhelming healthcare systems, and threatening the economy all around the globe. In this worldwide health crisis, the medical industry requires new technologies like Artificial Intelligence (AI), to monitor, control and look ahead against the spread of COVID-19 pandemic, which has become a powerful tool for diagnostic test designs and vaccination development.

Currently, COVID-19 vaccine has been heralded as a  big achievement, several vaccines against COVID-19 have been developed by Moderna, Pfizer and Johnson & Johnson within a year at unprecedented speed, and inoculation programs, together with improving patient treatments and regular testing, point to a way out of this pandemic. But, the work is far from done at the moment. Better and more vaccines may be required to combat the pandemic and provide a better return to normalcy.  For these purposes, artificial intelligence has been quite promising and may be one of the best tools to weather  the pandemic.

In regards to vaccine development, there are already options for AI programs, which can be used to analyze data and create better and better results the more data the AI receives through machine learning programs. It can predict mortality risk by adequately analyzing the previous data of the patients. AI can easily track the spread of COVID-19 virus, identifies high-risk patients, and is useful in controlling this infection in real-time, which provides opportunities for developing and improving new medications and vaccines at a much faster rate than usual.

In recent years, AI, in general, is  becoming a popular tool in the biological pharma industry, not  just in the case of COVID specifically. It has been able  to support drug development, saving time and money in the process. AI can analyze tons of data in moments and assist in decision-making, determine the therapy a patient might need, manage clinical data, choose the right active ingredients, and  propose  alternatives among existing drugs. ⁽¹⁾ A centralized collection of worldwide COVID-19 patient data will be beneficial for future artificial intelligence and machine learning research to develop predictive, diagnostic, and therapeutic strategies against COVID-19 and similar pandemics in future. 

There have been new applications of AI in COVID-19 pandemic. DeepVacPred is one program that was developed by the Southern California Viterbi School of Engineering and was focused specifically on SARS-CoV-2.⁽²⁾ The AI can realize  various vaccine design cycles in minutes  and was able to eliminate 95% of the candidate compounds when normally these could take years. It can work to stay ahead do the  virus as it mutates. 

The current vaccines have a proven effectiveness against a specific type of COVID-19, but the virus can mutate and change, so that a new vaccination will be needed. This point is very important. By using an AI, it is possible to ensure that the vaccines that are developed can tackle the new RNA mutations of the virus. (3) This is not something human researchers can do quickly, but the DeepVacPred might be able to. This can help ensure that the world is ahead of the coronavirus and not responding to it, as happened at the start of the pandemic.

Specifically, in the case of DeepVacPred, the method has yielded 26 potential vaccines against the coronavirus, and scientists selected the best 11 to start working on a multi-epitope vaccine,  which can attack the spike proteins that the coronavirus uses to bind and penetrate a host cell. The dataset  involved around 700 thousand different proteins. ⁽⁴⁾

AI has also been utilized and proposed as a tool to facilitate vaccine distribution.  At the moment, the challenge is pretty complex, as populations around the globe need to receive vaccines according to priority, with more vulnerable populations coming first, and with special care being taken to avoid ethnic or other forms of bias. AI can be  used effectively and give the results much quicker than an individual or a group of researchers. This could mean that entire population groups would be excluded or disadvantaged, but this means mostly that the AI is a solution that needs to be employed with care and oversight. ⁽⁵⁾

At the moment, AI is being proposed as an effective solution for many problems  tied  to the COVID-19 vaccinations. Specifically, several new variants of SARS-CoV-2 include B.1.1.7 in the UK, B.1.351 in South Africa, and P.1 in Brazil have been discovered. These variants are spreading across the world now, and they each contain multiple mutations. AI can be used to develop more vaccinations and stay ahead of the virus and the potential mutations it can show. The possibility of predicting mutations and anticipating vaccines is tremendously valuable  to ensure better management of the situation. 

AI can play a role in specific areas in regards to vaccination, which can help establish the impact and triage, that is, see which areas need the vaccine first and which population groups should receive priority attention. Moreover, it can also be used for managing vaccine production, distribution, tracking the effects, manage the supply chain, establishing useful patterns and surveys for adverse effects post-vaccination. AI offers many distinct benefits and can manage these tasks more effectively and detect patterns that a researcher would not be able to see. ⁽⁶⁾

Additionally, AI can produce results very quickly, hundreds of times  faster than a human could, analyzing huge amounts of data right away and giving results that would help take action on a timely basis, rather than reacting to events after they unfold. However, there are  also risks in choosing this approach. It is possible that the AI that is being employed is using faulty mechanisms or is biased in some way. While there is the perception that an AI is unbiased by its nature, the failure to include or consider certain factors could make the results flawed. For example, there is a debate on whether researchers should use race and ethnicity as variables for their AI program in regards to vaccine distribution or whether this is not as essential and the possible problems this can cause. 

At the same time, AI’s other benefits are not to be ignored, even if it cannot be always trusted blindly.  For instance, outsourcing tasks to an AI could help free time and can be an excellent way of providing faster, better, and more accurate results with the  potential of preventing and anticipating various issues.

AI has been applied to many subfields of drug discovery and vaccine development. In the future, it seems likely that AI will serve more goals in regards to prevention and support. However, the technology needs to be employed carefully and with oversight to avoid any of the downsides linked to it. Furthermore, the safety and efficacy of these vaccines have not been fully tested in human clinical trials, which could be a major concern. Therefore, novel vaccines and vaccination strategies are needed to enhance the efficacy and safety of COVID-19 vaccine development. 

References

1、Drug Discovery Today, (2021) 26(1), 80–93. https://doi.org/10.1016/j.drudis.2020.10.010

2、ClinicalOmics. (2021). https://www.clinicalomics.com/topics/patient-care/therapeutics/vaccines/ai-program-rapidly-designs-optimal-covid-19-vaccines/

3、Sci Rep (2021)11, 3238. https://doi.org/10.1038/s41598-021-81749-9

4、Open Government. (2021). https://www.openaccessgovernment.org/covid-vaccine-candidates/103053/#:~:text=The%20AI%2Dassisted%20method%20predicted,and%20penetrate%20a%20host%20cell.

5、Jercich, K. (2021). https://www.healthcareitnews.com/news/ai-has-advantages-covid-19-vaccine-rollout-potential-dangers-too

6、Kahn, J. &  Vanian, J. (2021). https://fortune.com/2021/01/05/a-i-covid-19-vaccination-drive/

About the Author

Lin Zhang, M.D., senior director of a health care industry company in the United States. With the experience in clinical medicine, biotechnology, health industry and other fields, he is responsible for the research and development of plant medicine, functional food and health products. He was a clinician and worked for the National Cancer Institute, FDA and the National Cancer Center of Japan for many years.