Last month, the UEA hosted the “Long-Covid Datathon” – an exciting event aiming to use machine learning models to analyse large data sets relating to Covid and Long-Covid. Running from the 9^th to the 11^th of September, the event aimed to “bridge knowledge gaps and create innovative solutions for managing Long-covid”. Over the three days, people across many disciplines came together to learn about the condition of long covid, large data sets, and machine learning, before working in groups to develop preliminary models with which to analyse the data.  

Long-Covid is defined by the World Health Organisation (the WHO) as “the continuation or development of new symptoms 3 months after the initial SARS-CoV-2 infection, with these symptoms lasting for at least 2 months with no other explanation”. Being vague and a diagnosis of exclusion makes machine learning the ideal tool with which to learn more about this condition.  

According to the organiser, Dr Saber Sami, Ai has “been used [in healthcare] for a very long time”, with it’s pattern recognition being particularly key in this context. He stated AI “has either reached or surpassed human pattern recognition”. Mentioning his own specialty, he discussed how the medical school has been using AI to investigate how the brain changes before dementia develops, commenting that “only with AI and machine learning can we do this kind of work”.  

When asked about the outcome of the event, Dr Saber Sami said the results were “stimulating”, with some models suggesting some “very interesting” insights regarding age groups and the effects of different covid variants. When discussing the prominence of  Long-Covid, he emphasised it’s a ‘big public health issue’, which was “threatening the UK economy”, with those of working age some of the most effected, with estimates of 1.9-3.2 million people in the UK currently severely impacted. 

 The vague nature of the disease criteria also makes it more difficult to perform further research on it, and with most Covid data collection now finished, policies that Dr Sami described as “not really helpful” when “we know it’s a problem”, this makes it all the more challenging to follow up with those effected by the pandemic and the following waves.  

“Covid is still rife – still a huge problem” says Dr Sami, discussing a new and troubling variant with varied and severe symptoms. He states that it’s “not taken seriously enough”, hoping those who have been offered the next round of booster vaccines will take them.  

Vaccinations reduce the likelihood of contracting an illness, and the severity of symptoms if the disease is caught, but there is no evidence to suggest it has an effect on the chances of developing long-covid after infection. An interesting question raised in the event during a talk by Dr Paul Hunter on “The Changing Epidemiology of SARS-CoV-2 Infections During the Pandemic”, was how long these waves of booster vaccines were expected to continue, or if they will only adapt with concerning and resistant variants. The current research is aimed mostly towards an inclusive vaccine covering several different and seasonal illnesses (much like the MMR vaccine infants receive in the UK to protect against measles, mumps, and rubella): “watch this space” says Dr Sami.  

Dr Sami encouraged interested students to keep a look out for activities which they could get involved with from the medical school, such as a very sci-fi sounding investigation into playing Tetris using your mind! The experiment uses AI to process brainwaves to control the game, which could have huge implications for those with limited mobility. Keep watch for more exciting events!  

Photo Credit: Unsplash