Detection and Classification of Heart Defects

Heart Defects are immensely threatful to human beings and can cause death. Improvements in diagnosis and treatment tools are welcome by the medical community and have proven to be one of the most useful diagnostic tools for heart patients, one of it to be mentioned would be the Electrocardiogram. Traditional technique of visual analysis of ECG is complicated for doctors, time consuming and requires expertise. Hence, computer based classification and detection of diseases can be immensely useful in diagnostics. This project has been inspired by the need to find an efficient method for ECG signal analysis and classification which is simple yet has good accuracy and less computation time. It deals with the study and analysis of ECG signal processing by means of MATLAB tool effectively for classification and detection of heart defects using Lead-II Configuration. Study of ECG signal includes reading and plotting of the ECG signal, acquisition of real time ECG data, ECG signal filtering and processing, feature extraction and detection of certain parameters, decoding, comparison, classification of the required features. In this thesis, we first find out the characteristics to classify a normal ECG and then pass any random signal to check whether the features or the values determined fall within the specified range with the ones characterized to be a normal ECG. If it does, then we classify it as normal ECG else we classify it as an abnormal ECG using Lead-II Configuration