How AI Can Improve Response Time to Potential Infectious Disease Outbreaks

Artificial intelligence is now being applied to public health and tracking infectious disease.

Bluedot is a company utilizing artificial intelligence to identify infectious disease threats quickly and work with stakeholders so they can move more efficiently when needed.

“We are using modern digital technologies and artificial intelligence to identify threats faster, to understand them, and assess their risks and to empower our clients to be able to respond effectively and efficiently,” Kamran Khan, MD, CEO and Founder at BlueDot said. “This includes government organizations, but also pharmaceutical and life sciences companies that manufacture vaccines, therapeutics, diagnostics, personal protective equipment—all the things that we need to effectively respond to an outbreak.”

Khan points to the potential lag time in identifying these threats that currently exists. “We do know that official sources of information that may come from public health agencies or governments around the world may not always be as timely as we would like,” Khan said. “That's a challenge, because when you're dealing with an outbreak, time is your most valuable asset. So, we are using the internet as a medium to detect signals of outbreaks around the world faster than governments may even detect or report on them.”

Bluedot achieves this by using AI and large language models to read through large amounts of unstructured text picking up clues about syndromes or respiratory viruses around the world. In this data, there might be a potential threat that public health officials and governments should be made aware of.

Many people familiar with AI would agree its strength lies in its ability to take large chunks of information and decipher it very quickly. Its limitations, thus far, has been AI’s inability to understand nuance. With this understanding, Khan explains it is the combination of human intuition and artificial intelligence that allows them to give more accurate information.

“We always have humans in the loop because we know that machines aren't perfect and they make errors,” Khan said. So we really are operating in a way where we're letting machines play to their strengths. They're able to process this data, 24 hours a day, 365 days a year. They don't need to sleep; they can read this and all the world's languages, but humans can apply our own intuition and our own understanding as to whether something might be misinformation.”

For example, one of the company’s latest business products is their influenza-like-illness (ILI) intelligence offering, which utilizes AI and human int to detect, assess, and respond to threats.

They also have a biweekly report, titled the ILI Pulse. This report is a compilation of timely information-based signals as well as official information.

Some of the other specific ILI offerings include:

• Early warning and expert assessment for high-consequence ILI activity
• Historical and annual cases and deaths
• ILI activity briefs to support risk communications and vaccination campaigns
• Global demographic, disease, and location drivers of burden
• Advanced warning of influenza seasons
• Rolling influenza forecasts for out-of-season and in-season activity

The company has been working with a variety of stakeholders including industry and government agencies. This past summer, Bluedot signed an agreement with the National Center for Global Health and Medicine (NCGM) of Japan. The collaboration is meant to aid Japan’s readiness and resilience to outbreaks of known and novel infectious diseases.

Under this new partnership, NCGM will utilize BlueDot’s global infectious disease intelligence platform to identify emerging threats earlier, bolster the efficiency of their global horizon scanning activities, and increase the depth and timeliness of their risk assessments.

Contagion spoke to Khan at the recent World AMR Congress and he offered insights about his company’s capabilities and how AI can be integrated into infectious disease monitoring.