The dldalgorithms package contains a suite of methods based on density level detection and support vector machines (SVM) that we have used in our vascular lesion research. The method is sufficiently generic so that it can be applied to the detection of other type of lesions or to other tasks such as segmentation.

The version 1.0 of the package contains all the necessary routines to apply an LPU (learning from positive and unlabeled samples) algorithm by means of SVM.

If you are planning to use the dldalgorithms package, we would be grateful if you would be kind enough to cite:
[1] MA. Zuluaga et al.. Learning from Only Positive and Unlabeled Data to Detect Lesions in Coronary CT Images. MICCAI. vol. LNCS 6893, Part III, Toronto, Canada, pp. 9-16, September, 2011.