Fuzzy rule base classification systems have been the focus of increased attention in recent years, due to their unique capability of providing human experts with outcomes by means of linguistic rules. In the same time period classifier fusion approaches have been shown to enhance the performance of pattern recognition systems. In the present study we applied a hybrid random subspace fusion scheme that constructs a set of different fuzzy classifiers utilizing different subsets of both the feature space and the sample domain, combining the results of these classifiers using appropriate decision functions. Experimental results using two protein mass spectra datasets of ovarian cancer demonstrate the usefulness of this approach in comparison to other classifier fusion approaches.