Cardiac Arrhythmia Classification using Cartesian Genetic Programming Evolved Artificial Neural Network

Some of the major diseases that have a high impact on the society are Cardiovascular diseases (CVDs). An important category of CVDs are the Cardiac Arrhythmias. Conventional methods of diagnosis for the disease are prone to errors and need experience on part of the diagnosing physician. For automatic detection of Cardiac Arrhythmia we developed an algorithm that first applies digital signal processing and logical operations to the time domain ECG signal and hence detects the fiducial points of an ECG complex. From the fiducial points, the lengths and slopes of a number of segments and amplitudes of peaks are determined. These parameter values are applied to the fast learning evolutionary algorithm of Cartesian Genetic Programming Evolved Artificial Neural Network (CGPANN) to classify the beats. A number of setups were experimented with. In these experiments, the CGPANN was first trained with known data from the popular MIT-BIH database and then tested with another part of the known data and the results found highly satisfactory.