In this paper, we propose an automated method to assess wall motion in Left Ventricle (LV) function in cardiac cine-Magnetic Resonance Imaging (MRI) based on Dictionary Learning (DL) classification with a parametric approach. Time-signal intensity curves (TSICs) are identified in spatio-temporal profiles extracted from differents anatomical segments in a cardiac MRI study. Parameters are extracted from TSICs that present a decreasing then increasing shape reflecting dynamic information of the LV contraction. Several parameter combinations are used as input atoms to train a sparse classifier based on kernel DL and results are compared with Support Vector Machines. Best classification performance is obtained with an accuracy about 94%.