We present in this paper a preliminary study for detecting early pattern of Cheyne-Stokes Breathing using a single electrocardiogram signal in patients with severe heart failure. Two ECG-derived respiration signals, namely Heart-Rate and R-Wave Amplitude, are computed and jointly used to estimate different respiratory events, respiratory rate and amplitude modulation. Three patients whose respiration goes from normal to severe CSB are used to test our method. Results show good performance for the detection of breathing cycles compared with the ventilation signal and the final classification based on respiratory events, AHI, amplitude modulation reveals exact correlation with the expert.