Reconstruction of state spaces is usually the first step in the analysis of chaotic systems. The application of reconstruction techniques is of great importance to understand the dynamics of the time series. For this reason, in this work, we evaluate the behavior of different embedding techniques, both uniform and non-uniform, in electrophysiological time series. In addition, the quality of the reconstruction was evaluated using the L-statistic and the Pecuzal cost function. It was evidenced that for the electrocardiography time series, Hankel Singular Value Decomposition method showed the better reconstructions, with an L-statistic close to 2.00. Contrary to what was seen in the electroencephalography time series, where all methods have good results, and the electromyography time series, where the False Nearest Neighbors and Unsupervised Maximum Relevance - Minimum Redundancy methods showed the best results. Therefore, it is important to consider the origin and dynamics of the signals for the selection of the best embedding method.