In this article we consider application of Natural Language Processing (NLP) techniques to the task of automatic detection of misrepresentation (« spin ») of research results in scientific publications from the biomedical domain. Our objective is to identify inadequate claims in medical articles, i.e. claims that state the beneficial effect of the experimental treatment to be greater than it is actually proven by the research results. After analyzing the problem from the point of view of NLP, we present methods that we consider applicable for automatic spin identification. We analyze the state of the art in similar or related tasks and we present our first results obtained with basic methods (local grammars) for the task of recognising entities specific for our goal.