Disclosed is an apparatus for detecting a presence of Cheyne-Stokes respiration from a respiratory flow signal. The apparatus comprises a memory for storing respiratory data associated with the respiratory flow signal and a processor coupled with the memory. The processor is configured to (a) assess the respiratory data to detect apnea and/or hypopnea events, (b) calculate a raw feature selected from a cycle length, a shape feature and a duty cycle, (c) normalise the raw feature and (d) classify the raw feature.
