IOSR Journal of Engineering May. 2012, Vol. 2(5) pp: 1265-1269 ISSN: 2250-3021 www.iosrjen.org 1265 | P a g e Automated ECG Diagnosis Upasani D.E. 1 , Dr.R.D.Kharadkar 2 1 (Researcher Bharati Vidyapeeth, Pune) 2 (Principal, GHRIET, Pune) ABSTRACT Myocardial ischemia & other cardiac disorder diagnosis using long duration electrocardiographic recordings is a simple and non-invasive method that needs further development in order be used in the everyday medical practice. Several techniques that automate ischemia& other cardiac detection have been proposed during the last decade which are under evaluation. They are based on different methodological approaches, which include digital signal analysis, rule-based techniques, fuzzy logic methods and artificial neural networks, with each one of them exhibiting its own advantages and disadvantages. Most recent systems employ hybrid system (combination of ANN & Fuzzy logy) to perform diagnosis since they have demonstrated great consistency in producing accurate results. The performance of the developed detection systems is very promising but they need further evaluation Keywords:- ANN, arrhythmia, ECG, Ischemia, tachycardia, I. INTRODUCTION Clinically, biomedical signals are primarily acquired for monitoring (detecting or estimating) specific pathological/physiological states for purposes of diagnosis and evaluating therapy. In some cases of basic research, they are also used for decoding and eventual modelling of specific biological systems. Furthermore, current technology allows the acquisition of multiple channels of these signals. This brings up additional signal-processing challenges to quantify physiologically meaningful interactions among these channels. Goals of signal processing in all these cases usually are noise removal, accurate quantification of signal model and its components through analysis (system identification for modelling and control purposes), feature extraction for deciding function and prediction of future pathological or functional events as in substitute devices for heart. Typical biological applications may involve the use of signal-processing algorithms for more than one of these reasons. The monitored biological signal in most cases is considered an additive combination of signal and noise. Noise can be from instrumentation (sensors, amplifiers, filters, etc.), from electromagnetic interference (EMI), or in general, any signal that is asynchronous and uncorrelated with the noise characteristics, which will eventually lead to an appropriate choice of signal-processing method. The main objective of automated ECG analysers is to assist in making a diagnosis. Obviously, they will not replace medical experts, but their decision can be considered as an unbiased second opinion. The automated diagnosis systems developed so far are not reliable in detection of ischemic episodes, fibrillations, and recognition of arrhythmias and are having number of limitations and short comings. In recent years, computer assisted ECG interpretation has played an important role in automatic diagnosis of heart abnormalities [1, 3]. The ECG wave forms ; reflects the physical condition of human heart and is the most important biological signal to study and diagnose cardiac abnormalities, disturbances. So, it is important to record the patient‟s ECG for a long period of time for clinical diagnosis. The clinical significance diagnosis depends on different parameters of ECG; complex QRS, wave P, frequency, Heart Rate Variability R-R, etc. In these applications, it is more important to develop signal processing methods that permit real time feature extraction and de-noising of the ECG characteristic. The extracted parameters are used for the classification of the cardiac pathologies and make an automatic tool of diagnosis in the services of doctors before the arrival of a quantified patient. Table 1. ECG Characteristics: Assoc iated Wave Duration (Sec) Amplitu de (mV) Frequency Mechani cal Action P wave < 0.12 ≤ 0.3 10 Atrial Depolari zation QRS Comp lex 0.08 to 0.12 0.5 to 2 20 to 50 Depolari zation of Ventricle s T wave 0.2 0.2 5 Repolari zation of Ventricle s II. REVIEW OF LITERATURE: Sokolow et al., (1990), indicate that the state of cardiac health is generally reflected in the shape of the ECG waveform and heart rate. Cuiwei Li et al., (1995) showed that it is easy with multi scale information / decomposition in wavelets transformation to characterize the ECG waves. Khadra et al., (1997) proposed a classification of life threatening cardiac arrhythmias using wavelet transforms. MG Tsipouras et al., (2004) used time frequency analysis for classification of atrial tachya arrhythmias. Al-Fahoum and Howit (1999) joint radial basis neural networks to wavelet transformation to classify cardiac arrhythmias.