The wealth of medical-related information that can be accessed today gives rise to a multidimensional source of knowledge for the health care domain. Research discoveries published in prestigious venues, data that lies in electronic-health records, discharge summaries, clinical notes, etc., all represent important medical information that can assist in the decision making process. The challenge that comes with accessing such vast and diverse sources of data stands in the ability to distil and extract reliable and relevant information. Computer-based tools that use natural language processing and machine learning techniques have been proven to help addressing such challenges.

In this talk, I describe and present experimental results for four problems that represent important pieces in the puzzle of building a computer-based personalized medicine. I address the problem of automatically identifying medical cases from textual data presented in discharge summaries and the problem of identifying, from published-research discoveries, reliable information to certain medical topics. I show how genetic information can be extracted and added to the information a health-care provider can use when assessing a medical case; and last but not the least, I describe experimental results for identifying and classifying semantic relations between medical entities in technical data, this, with the aim of prescribing individual-tailored therapies.