Data science offers solutions to public health challenges

The use of data science is revolutionizing research in many scientific disciplines, including medicine. The analysis of large volumes of all kinds of medical data (from diagnostic images to blood tests or questionnaires) offers multiple opportunities in the personalized treatment of diseases, the generation of biomarkers, the improvement of diagnosis, the prediction of the evolution of certain pathologies, etc.

ADaS Lab (Applied Data Science Lab), a research group at the Universitat Oberta de Catalunya (UOC)'s Faculty of Computer Science, Multimedia and Telecommunications, develops artificial intelligence-based solutions to improve people's health and quality of life. This laboratory is affiliated with the eHealth Center and was created in March 2020. It is led by researchers Ferran Prados and Jordi Casas who, together with a team of five other people, are working to apply the innovations in data science to the real challenges of public health.

"The main drive behind our research is to see the interest of doctors in putting it to clinical use and the fact that it'll have a positive impact on patients", explained Jordi Casas, a member of the Faculty of Computer Science, Multimedia and Telecommunications and the director of the UOC's Master's Degree in Data Science. For Ferran Prados, "using data to find solutions to health problems and being able to apply these solutions to improve people's lives is very rewarding". Prados is an engineer with a PhD in Computer Science who specializes in medical imaging. In addition to doing research at the UOC, he also collaborates with the Centre for Medical Image Computing at University College London.

Artificial intelligence solutions for innovative healthcare centres

The ADaS Lab team is focused on optimizing the analysis of clinical data, mainly of medical images, in collaboration with prestigious healthcare centres. One of the projects it is carrying out in conjunction with IDIBAPS (a research centre affiliated with the Hospital Clínic in Barcelona) aims to identify biomarkers in order to determine the progression of multiple sclerosis patients based on the analysis of multimodal magnetic resonance imaging. "Our work's based on the needs expressed by doctors. It's important that we understand these issues so that we can provide artificial intelligence-based solutions that are genuinely useful for professionals," explained Jordi Casas. The results of this research, which is expected to be completed next spring, could be applied to improve the selection of patients participating in clinical trials.

In another ADaS Lab project, carried out in conjunction with the Hospital Universitari Vall d'Hebron, researchers are developing a solution to facilitate the diagnosis and study of chronic fatigue through a predictive analysis solution. At present, there are no biomarkers for diagnosing this disease and the only means of doing so are through lengthy questionnaires which have to be filled out by patients. ADaS Lab researchers are developing a solution that will allow a more reliable diagnosis that does not involve so much effort by patients and professionals.

They are currently designing a solution, with the collaboration of the Catalan Agency for Health Quality and Assessment (AQUAS), which will have a major impact on being able to identify patients suffering from COVID-19 through the sole use of medical images of chest X-rays, rather than having to carry out the current diagnostic test (PCR or antigen test). "We have 100,000 X-rays of patients from all over Catalonia, which we're using to train an artificial intelligence model in order to identify whether a patient has COVID-19 using only the medical image, in a similar way to how pneumonia is diagnosed," explained Prados.

AI challenges: from hype to biases

Artificial intelligence is one of the trending technologies and there is a certain 'hype' around it", admitted Prados and Casas. However, "it's undeniable that current advancements are happening at a much faster rate than in previous years. Before, AI was simply a course subject; now, there are whole degrees dedicated to the discipline and it covers a huge amount of different areas. The future is clear, and it will help decision-making enormously in a wide array of spheres", assured Casas.

One of the controversies surrounding artificial intelligence is the biases present in the data, which in turn cause the decisions that are made with the support of this technology to be biased as well. We have seen this during recruitment processes (such as when Amazon's algorithm was predisposed to choose men over women), in the granting of bank loans, and so on. For researchers, artificial intelligence is not to blame: "technology's impartial; it's the data which can present biases. Data have to be used very carefully, obtaining only those which are strictly necessary and making the AI models auditable and transparent. We're increasingly seeing that models show what lies behind AI decisions. This is an ongoing debate".

The eHealth Center, the academic centre for E-Health

The eHealth Center is an academic centre open to the world whose goal is to educate and empower professionals and ordinary citizens, through the use of technologies, to lead the paradigm shift in health. Its focus is on people and, through research, education and guidance, it aims to contribute towards social progress and well-being.

^{UOC R&I} 

^{The UOC's research and innovation (R&I) is helping overcome pressing challenges faced by global societies in the 21st century, by studying interactions between technology and human & social sciences with a specific focus on the network society, e-learning and e-health.} 

^{Over 500 researchers and 52 research groups work among the University's seven faculties and two research centres: the Internet Interdisciplinary Institute (IN3) and the eHealth Center (eHC).}

^{The University also cultivates online learning innovations at its eLearning Innovation Center (eLinC), as well as UOC community entrepreneurship and knowledge transfer via the Hubbik platform.}

^{The United Nations' 2030 Agenda for Sustainable Development and open knowledge serve as strategic pillars for the UOC's teaching, research and innovation. More information: research.uoc.edu #UOC25years}