Machine-learning analysis identified high blood glucose levels as a risk factor for COVID-19 patients, according to a study published in Frontiers in Public Health (1). Researchers believe this explains why older patients, often with a poor glucose metabolism due to other health conditions, are more vulnerable to this disease.
During the COVID-19 pandemic, researchers created the COVID-19 Open Research Dataset (CORD-19), with over 400,000 studies available to everyone. This has become the most extensive database about coronaviruses, making it the perfect place for AI platforms to start looking for new insights that can help our fight against this virus.
Not wanting to miss this opportunity, Prof Henry Markram, based at EPFL in Switzerland, believed his team had the perfect tool: The EPFL Blue Brain Project, which was developed to build the first digital reconstruction and simulation of the mouse brain.
With a few adaptations, the Blue Brain team built and trained this AI machine-learning tool to analyse hundreds of thousands of studies on COVID-19 and extract important information from these sources. Given the puzzling duality presented by this disease, where some patients have no symptoms while others endure life-threatening conditions, the team chose to ask why some people are more affected than others.
“Since early 2020, Blue Brain has been proactively contributing to the fight against COVID-19,” explains Prof. Henry Markram, Founder and Director of the Blue Brain Project. “With this call to action, we realised we could use our Machine Learning technologies and Data and Knowledge Engineering expertise to develop text and data mining tools required to try and help the medical community. Blue Brain set out to answer one of the most puzzling aspects of this pandemic – why some people get very sick, while others are completely unaffected”.
“Is it enough to just say that older people are more vulnerable? Why do some apparently healthy people die from COVID-19? Why do so many people die in the ICU?”, continued the researcher. “To answer these questions, we directed our AI to trace every step of the viral infection from the moment the virus enters the lungs until the time when the virus breaks out of the cells in the lungs and spreads throughout the body to infect the organs”.
This analysis identified high blood glucose as a significant risk for patients suffering from COVID-19. It turned out that elevated glucose can impair the first level of immune response in the lungs, creating the perfect conditions for the virus to spread. In addition, high glucose can facilitate other complications, such as hyper inflammation, acute respiratory distress syndrome, multi-organ failure and thrombotic events. For the researchers, a poor glucose metabolism – common in older patients – is the most likely reason to explain why this age group is more vulnerable to the infection. From these results, they suggest that the management of glucose levels should be a pivotal point to treat patients with COVID-19.
For Prof Markram, this is just one example of using computational tools like their Blue Brain AI system. There are millions of scientific studies across numerous disciplines, but it’s impossible for humans to read and understand all this information. But machines can. Potentially, machines can help solve the biggest challenges in our world.
(1) Logette E, Lorin C, Favreau C, Oshurko E, Coggan J, Casalegno F, Sy M, Monney C, Bertschy M, Delattre E, Fonta P, Kremlin J, Schmidt S, Keller D, Kerrien S, Scantamburlo E, Kaufman A and Markram H (2021) A Machine-Generated View of the Role of Blood Glucose Levels in the Severity of COVID-19. Front. Public Health, https://doi.org/10.3389/fpubh.2021.695139