Introduction. Cryo-ablation is a common procedure used in hospitals to eliminate certain arrhythmias, such as Atrial Fibrillation (AF). Although it is sufficiently proven, this procedure sometimes involves interventions in very close areas to the phrenic nerve, with the consequent risk of damaging it. Currently, clinical practice incorporates manual safety guarantee protocols in its ablation procedures, which consist of placing the doctor's hand on the diaphragm to assess the possible affectations while simultaneously proceeding with the electro-stimulation of the phrenic nerve. Materials and methods. In this work, we propose the development of an automated quantifier that facilitates the clinical evaluation of possible conduction disorders. To achieve this goal, polygraph signals extracted during the ablation process of ten patients were used. Specifically, polygraph signals were captured during phrenic nerve stimulation when the sensor was located on the phrenic nerve (capture of activation) and when the sensor was not on the phrenic nerve (no capture). Experiments and results. Signal processing, including pre-processing, noise filtering, and delineation, was applied to every available situation and signal. To unmask the residue of muscle cellular potential during the phrenic nerve stimulation process, the results of the two situations described (capture and non-capture) were compared to quantify the muscle activity from nerve activation using a linear classifier. Conclusions. The results obtained in this work have confirmed that it is possible to automatically classify the level of muscle activity from the phrenic nerve with an accuracy of ACC=100\%, validating the hypothesis of establishing an automated and unbiased protocol for the measurement of the eventual deterioration of phrenic nerve conduction as a result of ablation in the vicinity.