Loading...
 

 

Gestión de conocimiento y Data Mining

This research activity is devoted to biomedical data mining research, developing efficient and scalable methods for mining interesting patterns in large biomedical databases.

The research interest of GBT spans across the whole process of converting data into intelligence, in an efficient and effective manner. GBT specializes in developing (1) algorithms that learn to recognize complex patterns within rich and massive data and (2) technologies for selection, pre-processing, processing, interpretation and evaluation within data mining context. GBT has experience with several data mining tools: decision trees, neural networks, support vector machine, amongst others.

Some of the main data mining applications have been in the field of chronic disease and care data management (HIV, Diabetes),  neurorehabilitation and knowledge extraction from MRI brain images.

The application of data mining techniques in HIV patient data aims to identify the causes of HIV treatment failure and to better understand the changes in the patients’ outcomes. GBT researchers have applied temporal data mining techniques to the analysis of the data collected since 1981 by the Infectious Diseases Unit of the Hospital Clínic in Barcelona, Spain.

In particular, a precedence temporal rule extraction algorithm has been run on three different temporal periods of the AIDS epidemic, looking for two types of treatment failures: viral failure and toxic failure, corresponding to events of clinical interest to assess the treatment outcomes. The central goal is to find interesting temporal rules between complex patterns found in a set of time series. The analysis allowed to extract different typical patterns related to each period and to meaningfully interpret the previous and current behaviour of HIV therapy.

During the last years, GBT has participated in several projects where data mining techniques have featured prominently:

  • Evaluation of impact of system information in the care of HIV infected patients. FIS, Sanitary Research Fund (PI09/90899). 2010-2013.
  • NEUROCONTENT 2.0: Open environment for the generation and commercialization of therapeutic content to improve processes and increase tele-cognitive rehabilitation. Implementation in acquired brain injury, mental health and ICU patients. Ministry of Science and Innovation (IPT-300000-2010-30). 2010-2013
  • REHABILITA: Disruptive technologies for the rehabilitation of the future. Ministry of Science and Innovation (CENIT-E- CEN-20091043). 2009-2012
  • Neurolearning. Ministry of Industry, Tourism and Trade AVANZA PLAN: Digital Citizen Subprogram (TSI-020501-2008-0154). 2008-2011
  • Mobiguide: “Guiding patients anytime everywhere”- FP7-ICT 287811, 2011-2015
  • SineDie: “Intelligent systems for education and control of diabetes diagnosed during pregnancy”-Fondo de Investigaciones Sanitarias, 2011/2013
  • PERSONA: “Personalized Decision Support for Enhanced Control in Pervasive Healthcare Platforms”. CIBER-BBN- Biomedical Research Networking Center in Bioengineering, Biomaterials and Nanomedicine (CIBER-BBN), 2011/2012
  • A PRIORI: “Predictive analysis for the adjustment of insulin treatment and optimization of close-loop”, Fondo de Investigaciones Sanitarias, 2010/2012

EnglishEspañol

LinkedInTwitterGoogle+Facebook