ITBM-RBM 2002 ; 23 : 333-339 0 2002 kditions scientifiques et mkdicales Elsevier SAS. Tous droits r&sew% Original article Heart rate variability analysis using correlation dimension and detrended fluctuation analysis R. U. Acharya*, C. M. Lim, P.Joseph Department of Electronics and Computer Engineering, Ngee Ann Polytechnic, 535 Clementi Road, Singapore 599489, Singapore Received 8 October 2001; revised 20 June 2002; accepted 1 September 2002 Abstract The ECG is a representative signal containing information about the condition of the heart. The shape and size of the P-QRS-T wave, the time intervals between its various peaks, etc., may contain useful information about the nature of disease afflicting the heart. However, the human observer can not directly monitor these subtle details. Besides, since biosignals are highly subjective, the symptoms may appear at random in the time scale. Therefore, the heart rate variability signal parameters, extracted and analysed using computers, are highly useful in diagnostics. This paper deals with the classification of certain diseases using correlation dimension (CD) and detrended fluctuation analysis (DFA). 0 2002 Editions scientifiques et medicales Elsevier SAS electrocardiogram I heart rate I correlation dimension R&urn6 - Analyse automatique de la variabilit6 de la frCquence cardiaque en utilisant la dimension de cortilation et la stationnarisation des fluctuations. L’ECG est un signal contenant des informations sur Mat du muscle cardiaque. La forme et /‘amplitude du complexe P-QRS-T, /es interval/es de temps entre ses divers pits, etc., peuvent contenir des infonnations &/es sur la nature de la pathologique cardiaque. Cependant, l’observateur humain ne peut pas analyser ces d&ails sub- tils. En outre, puisque les biosignaux sent forfement subjectifs, les sympt6mes peuvent apparaitre de facon aMatoire dans le temps. En consdquence, l’enregistrement et l’analyse automatique de la varia- bilit6 de la frdquence cardiaque sent fr& utiles au diagnostic. Cet article Waitela classification de cer- taines pathologi?s en utilisant la dimension de corrk!ation (CD) et la stationnarisation des fluctuations (DFA). 0 2002 Editions scientifiques et medicales Elsevier SAS 6lectrocardlogramme I rythme cardiaque I dimension de cortilation INTRODUCTION Electrocardiography dealswith the electrical activity of theheart. Monitored by placing sensors at the limb extremities of the subject,electrocardiogram (ECG) is a recordof the origin andpropagation of the elec- tric potentialthroughthecardiac muscle.It is consid- ered a representative signal of cardiac physiology, useful in diagnosing cardiac disorders. The stateof cardiac health is generally reflected in the shape of ECG waveform andheartrate. It may containimpor- tant pointers to the nature of diseases afflicting the heart. However, biosignals being non-stationary signals, this reflection may occur at random in the time scale(i.e., the disease symptomsmay not show up all the time, but would manifest at certainirregu- lar intervals during the day). Therefore, for effective diagnosis,the study of ECG pattern and heart rate variability (HRV) signal (instantaneous heart rate *Correspondence and reprints. E-mail address: aru@np.edu.sg (RU Acharya).