Supervised learning methods for pathological arterial pulse wave differentiation: A SVM and neural networks approach

•The arterial pulse pressure waveform (APW) provides an adequate description of the arterial system behaviour..•The development of techniques based on the automatic analysis of biomedical signals could be crucial for a reliable cardiovascular assessment.•An APW database comprising signals from 213 patients acquired with a novel optical system was used here.•Support Vector Machines (SVM) and Neural Networks were compared for differentiating between noisy waveforms, healthy and pathologic APWs.•SVM showed a higher accuracy possibly due to its ability to deal with the non-linearity and high-dimensionality degree of APW signal.