Joana Barros of the SFI Insight Centre for Data Analytics and NUI Galway is using machine learning and social media to track the spread of infectious diseases.
Joana Barros completed her undergraduate degree in biology at the University of Porto before moving to Lisbon to pursue a master’s in bioinformatics and computational biology. This was when she first experienced the development of machine learning models. Specifically, she worked on the prediction of drug interactions with a potassium channel crucial for the heart’s normal function.
Barros moved to Ireland in 2016 to start her PhD at the SFI Insight Centre for Data Analytics at NUI Galway, looking at public health monitoring through digital health.
‘Synthesising information from digital sources to improve the care for chronic illnesses and rapidly detect infectious outbreaks is highly important’
– JOANA BARROS
What inspired you to become a researcher?
I have always been curious about life, how our human body works and how we behave. While studying biology, I very often felt limited by what I could do in the lab; there had to be a way to see the bigger picture! That was when I discovered bioinformatics and computer science. It was a gradual eye opening to all the possibilities when disciplines are crossed.
Can you tell us about the research you’re currently working on?
In the last few years I have been focusing on public health monitoring with the aid of search engines and social media. A lot of our communication is done in written words. We write how we feel when we are sick and how we are dealing with it.
With the advent of social media, we also share this information with the world. This is the focus of my PhD. My goal is to find out how we can use this information to improve the health of the population and what is its potential for shaping current health policies.
In your opinion, why is your research important?
Chronic illnesses are the leading causes of death and disability worldwide, with cardiovascular diseases, cancers, respiratory illnesses and diabetes already accounting for more than 80pc of these premature deaths.
At the same time, emerging infections are a constant concern. Think of the recent coronavirus pandemic and the widespread global impact of the disease Covid-19. As a whole, infectious diseases are the estimated cause of 25pc of annual deaths globally.
Therefore, synthesising information from digital sources to improve the care for chronic illnesses, as well as to rapidly detect infectious outbreaks to swiftly enable containment measures, is highly important.
Preserving and fostering a healthy population is pivotal not only to maintain it, but also to ensure strong economies and an overall quality lifestyle.
What commercial applications do you foresee for your research?
Monitoring the population’s health is extremely valuable. Currently, there are already a few companies (BlueDot, for example) developing models for disease outbreak detection and monitoring based on big data coming from social media, news articles, flight paths etc. I see this line of research having a greater impact when used in cooperation with public health entities.
What are some of the biggest challenges you face as a researcher in your field?
Without a doubt, privacy is one of the key concerns. Big data must not lead to unrestricted research. I understand and, most of all, share this concern. This is why I believe GDPR is a positive step for ethical and transparent research.
Are there any common misconceptions about this area of research?
That any model will be able to exactly tell how a disease will spread. The field is constantly improving and we can produce quality simulations.
However, it is paramount to understand that there are many uncontrollable variables. Models can be updated to reflect the evolving conditions but we should always consider the margin for error.
What are some of the areas of research you’d like to see tackled in the years ahead?
Our health is complex and needs to be tackled from multiple fronts. We now have a myriad of alternative sources to monitor public health, but we cannot disregard traditional methods.
That is why I’m keen to see increased cooperation among public health officials and computer scientists as this is fundamental to introduce improved monitoring tools, ensuring better care for the population’s health.
Are you a researcher with an interesting project to share? Let us know by emailing email@example.com with the subject line ‘Science Uncovered’.