Mining electronic health records: challenges and impact

Big data applications in the Healthcare Sector can provide a high potential for improving the overall efficiency and quality of care delivery. In the health care sector though, big data analytics has still to address several technical requirements, being unstructured data analysis one of them. Unstructured data represents a powerful untapped resource-one that has the potential to provide deeper insights into data and ultimately help drive competitive advantage. In this talk some of the most common challenges of processing such data to extract useful knowledge will be analyzed. In particular, we will deal with the following challenges: i) clinical narratives preprocessing using NLP, ii) name entity recognition, iii) semantic enrichment, iv) integration of the results. We will focus on the real use cases in which we are working in the frame of a European H2020 project called IASIS. In fact, we will analyze the challenges of analyzing reports and notes of patients suffering from Alzheimer's disease disease and lung cancer to extract patterns (survival, treatment, antecedents, ... ) that can help physicians to get insights for better management of the disease.