Medical data are, most of the times, very complex both in form and content. One of the greatest challenges for the IT community in healthcare is to enable the full utilization of these data by information systems. This explicit variety combined with the fact that data usually derives from diverse systems are great obstacles to this task. The result is that data stored in medical information systems usually do not accurately represent reality. In order to eliminate the fallacy between stored and real data, specialized applications that facilitate and accelerate data import into information systems must be developed. This is the goal of Natural Language Processing, the scientific field that combines computer science and linguistics. As a result NLP systems use applications for the coding and standardization of information, known as controlled medical vocabularies. The result of these processes is data that can be used by various technologies, such as clinical data warehouses and decision support systems, the functionality of which is fully dependable on the completeness and accuracy of the data on which their analysis is imposed.