It is not hyperbolic to say that the world is at war – and the enemy is a virus. Governments, medics, scientists, policymakers, businesses and individuals alike are racing to develop the most effective means of battling COVID-19. Biotechnology innovation and artificial intelligence (AI) are critical elements in this race.
Let’s take a closer look at how.
Accelerating drug discovery
Drug discovery is perhaps the most vital piece of the medical research picture when it comes to effectively battling COVID-19. Researchers are attempting to uncover not only an effective vaccine, but also drugs which can improve treatment and outcomes for those who are suffering.
However, drug discovery has traditionally been a slow process, for a number of different reasons. There are two broad options available to researchers – they can discover entirely new drugs, or they can repurpose existing ones.
The problem with the former strategy is that brand-new drugs typically then require an extensive period of testing and certification to be approved as safe. This could take years – and in the war against COVID-19, that’s too long a wait.
But the problem with the latter strategy is that involves a vast amount of information processing. Repurposing existing drugs means sifting through huge quantities of data on existing drug actions and side-effects to identify patterns, trends, connections and relationships which suggest new applications. Whilst there are thousands of approved drugs on the market, these can be combined into millions and even billions of further multi-drug combinations – which then need to be screened against key targets related to COVID-19.
This is where AI technology comes in. Machine learning algorithms can dramatically speed up and streamline these scanning processes, automatically undertaking the vast analysis needed to uncover new combinations and identify possibilities for effective treatment. One example, as reported in the Financial Times, involves identifying drugs with an anti-inflammatory action which could reduce certain COVID-19 systems, but cross-referencing this with the ability to inhibit enzymes that viruses use to enter human cells.
Diagnostics and patient assessment
AI and machine learning have plenty of vital capabilities in the fight against COVID-19 beyond the drug discovery and testing cycle. The same Financial Times article references, for example, how Cambridge researchers have developed a mobile phone app that can detect whether someone is suffering from the disease based on the sounds of their breathing.
By the same token, AI and machine learning technologies can be used to crunch data collected from COVID-19 patients and healthcare practitioners globally, unlocking new insights in terms of how different treatments and interventions work, and how patients with different pre-existing conditions are likely to respond. As ever, information is power – but being able to make sense of huge quantities of information requires serious computing power. AI programs have, for example, been used to examine thousands of COVID-19 patients’ X-rays, rapidly identifying signs of lung damage.
The war against COVID-19 is hugely complex, and the stakes are enormously high. Artificial intelligence could prove a crucial weapon in our arsenal.