Rotor ablation has been proposed as the effective treatment for human atrial fibrillation (AF). However, existing mapping systems usually take too much time to visualize activation sequences.Methods: We have developed a novel real-time phase mapping system called “ExTRa Mapping” to detect unstable rotors during AF. Based on bipolar signals of a 20-pole spiral-shaped ring catheter, action potential temporal dynamics was numerically estimated. Results: Unstable rotors during in silico AF were reconstructed by ExTRa Mapping (Fig). ExTRa Mapping could identify unstable rotors during clinical AF by creating phase maps on a real-time basis.Conclusion: ExTRa Mapping could open a new avenue for identifying unstable rotors during AF ablation.