The clock is ticking for the deadly pandemic. Healthcare professionals and researchers have warned, for years, that the next Spanish Flu is on the horizon. Potentially causing widespread death, famine, economic turmoil and social chaos. Despite these warnings, we’re still not fully prepared for the next big one. We don’t know where it’ll come from, whether it develops from animals like SARS, or if it’s a bioweapon, artificially created to be as fatal as possible.
150 million at risk
Indeed, one recent “plausible scenario” simulation by the Johns Hopkins Center for Health Security (CHS) estimates that a deadly pandemic, with the right mix of high transmissibility and fatality, could kill as many as 150 million people globally.
Recent outbreaks, such as SARS, MERS and the novel Coronavirus, have highlighted how quickly a pathogen can spread from country-to-country. So far, however, these diseases have been limited by either their mortality or transmission rate. If a disease evolves (or is developed) that is both lethal and can be transmitted quickly between humans, then the world is critically underprepared.
As Microsoft founder, Bill Gates forewarns, “There’s one area where the world isn’t making much progress, and that’s pandemic preparedness.”
A powerful weapon
But, unlike previous pandemics like Spanish Influenza, today’s medical researchers and healthcare professionals have a powerful weapon. Artificial intelligence (AI) is increasingly helping the healthcare industry across the full spectrum of disease prevention and cure. From drug creation and clinical trials to epidemiology and predicting disease spread.
Unfortunately, the world has also become more mobile and global since the Spanish Influenza. That means a pandemic can spread more rapidly than in 1918.
Tracking disease spread
AI offers a potential solution for this, by tracking the growth of human populations in specific disease-struck areas, linking this with travel information and flight paths, to forewarn of pandemic spread. Mining news reports and social media for mentions of a new disease can also provide an early warning system for governments and medical professionals. Canadian start-up BlueDot, for instance, predicted the outbreak of the novel Coronavirus at the end of December 2019. Days before the U.S. Center for Disease Control and Prevention (CDC) and the World Health Organization (WHO).
Time is of the essence to halt pandemic spread. So such a system can help researchers spring into action to find a cure and vaccine before the mortality rate rises. It also gives frontline health workers an early indication of a potential pandemic arising, so they can quarantine suspected infectious patients as quickly as possible.
Effective public education
Governments can use AI to understand what communications to spread to the public and how. Public health resources on the monkeypox virus, for example, would be relevant in the Democratic Republic of the Congo where public concerns are rising. The anxiety around the novel Coronavirus, however, is currently high in the U.S. and China. Tracking public sentiment around specific diseases can give a better insight into what’s worrying people and where Government-issued information will be valuable.
Additionally, in public health emergencies, AI can streamline processes like public health screening and travel control. Robin Li, CEO of Baidu, explains, “Big data and AI is not only instrumental in increasing city management efficiency and healthcare breakthroughs during public emergency events but can also empower all industries and become a driving force.”
The company’s AI technology was recently used in Beijing and Shenzhen to rapidly screen air and rail passengers for fever.
Machine learning (a type of AI) is particularly useful in pandemic situations because it can identify patterns in data that a human team will find impossible. Humans cannot physically sift through millions of data points on a local population, risk factors and treatment effects. But with machine learning, this can be done in a shorter timeframe and with a greater range of data.
In the early stages of a pandemic, it can be used for drug development, drug repurposing, vaccine creation and clinical trial selection. As the disease spreads, it can analyse treatment plans to assess what drugs are working – and detect possible resistance developing.
Of course, this use of AI has applications well beyond pandemic detection and control. The marriage of AI and healthcare, particularly in biotechnology, which is on its own exponential growth trajectory. As innovator Tej Kohli states, “AI and biotech are undergoing rapid development precisely because they have such far-reaching potential. As they move forward, we must keep looking for new combinations to unlock. I suspect that we will find we have underestimated their potential by considering them in isolation.”
The philanthropist is currently working on improving global organ donation through AI, matching organs with recipients based on location, availability and suitability – with a goal towards synthetic organ donation in the near future.
Likewise, AI is being used to screen certain types of cancer with greater accuracy than a human team, in robotic surgical techniques and diagnosing some neurological conditions.
However, for AI to be fully utilised in healthcare, the concerns around its ethics and potential for misuse must be addressed. AI in itself is not a magic bullet for the industry. It can only be as good as the data it’s provided with and its training. Biases, for instance, can be introduced during training that will skew its output. To prevent this, diversity in both data and data scientist must be considered – and the AI cannot be a ‘black box’. It must be able to explain why it has come to a specific decision or diagnosis. Likewise, human oversight will always be required when working with AI.
When any virulent disease appears, finding a cure swiftly is the number one priority. AI is making this easier for researchers in many ways. But human knowledge and experience will always be needed. AI will reach its full potential in the hands of people who understand both man and machine. So the onus is on healthcare professionals to understand the current capabilities of AI. Before the next pandemic hits.