University of Barcelona

The University of Barcelona (UB) was founded in 1450, is one of the oldest universities in Spain and is the largest university in Catalonia. UB is ranked the first Spanish university according to several rankings (QS World University Rankings 2018, ARWU/Shanghai Ranking 2018).

This project will be carried out at UB’s Artificial Intelligence in Medicine Lab (BCN-AIM), which is an essential part of the Department of Mathematics and Computer Science. BCM-AIM is composed of 10 passionate researchers in artificial intelligence, computer vision, medical imaging, machine/deep learning, and health-related applications. BCN-AIM’s current research focuses on the development of data science and machine learning approaches for the analysis of large-scale biomedical data, including imaging, biological, clinical, lifestyle and mobile data. Moreover, the research team has an established track record in coordination and participation in national, European and international projects on data science and AI, for example euCanSHare and EarlyCause.

Role in LongITools

UB is leading work package 7, building and testing a proof-of-concept healthcare app for risk monitoring. It is also participating in work packages 2 and 3, linking data for model validation and building exposome-based predictive modelling of cardiometabolic risk.

Dr Karim Lekadir

Principal Investigator (PI) and WP7 Leader

Karim is Ramon y Cajal Researcher and Director of BCN-AIM at the University of Barcelona. His research interests focuses on the application of artificial intelligence to build predictive models from multi-source big data (clinical, imaging, biological and lifestyle). The software he developed during his PhD is now used in more than 250 clinical centres worldwide. Karim is currently coordinating the euCanSHare and EarlyCause H2020 projects on big data sharing and analytics with application to personalised medicine research. He is also an Associate Editor of IEEE Transactions on Medical Imaging. In LongITools, he will lead the implementation of the exposome data analytics toolbox (WP2-3), and will coordinate the development of the healthcare app for cardio-metabolic risk assessment (WP7).


Catherine Gallin

Project Management

Catherine Gallin is a project manager at the BCN-AIM Lab at the University of Barcelona. She has a background in languages and the social sciences and holds a PhD in Food & Nutrition (social anthropology) from the University of Barcelona. Catherine is passionate about international learning and research and has experience in health & intercultural, academic, and linguistic project management and research. She is interested in interdisciplinary approaches to health and patient care, especially in terms of maternal, child, and reproductive health as well as in the factors that impact the development of autoimmune diseases and how they are treated.

Angélica Atehortúa

Research Team

Angélica is a postdoctoral researcher with a background in analysis and processing of medical image data using machine learning methods. Her PhD research focused on the design of new spatio-temporal saliency models to locally characterise the dynamic of the heart. Angélica’s current research interest focuses on the application of artificial intelligence to build predictive models of cardio-metabolic risk from exposome and phenotypic data.

Vien Ngoc Dang

Research Team

Vien is a PhD researcher at the University of Barcelona. A data scientist, she is especially passionate about the challenges that medical data analysis poses. She worked on a medical imaging project at Eurecom - Graduate School & Research Center in France, in collaboration with the Centre for Medical Image Computing, Department of Medical Physics and Bioengineering at the University College London. As part of this project, Vien worked on developing AI software systems to help physicians make informed decisions in inflammatory diseases of the central nervous system. At BCN-AIM, and as part of the LongITools project, Vien is developing and validating machine learning models for disease prediction from heterogeneous longitudinal data.