The understanding of dose/side-effects relationships in prostate cancer radiotherapy is crucial to define appropriate individual's constraints at the therapy planning. Most of the existing methods to predict side effects do not fully exploit the rich spatial information conveyed by the three-dimensional planned dose distributions. We propose a new classification method for three-dimensional individuals' doses, based on a new semi-nonnegative ICA algorithm to identify patients at risk of presenting rectal bleeding from a population treated for prostate cancer. The method firstly determines two bases of vectors from the population data (the first basis corresponds to rectal bleeding patients and the second one characterizes the non rectal bleeding patients). The classification is then achieved by projecting a new three-dimensional individual planned dose onto both subspaces spanned by the two bases. A given patient is thus classified by calculating its distance to the two subspaces. The results, obtained on a cohort of 87 patients (at two years follow-up) treated with radiotherapy, showed high performance in terms of sensitivity and specificity.