A genetic algorithm-neural network wrapper approach for bundle branch block detection

A Bundle Branch Block (BBB) is a delay or obstruction along electrical impulse pathways in the heart. The automated detection and classification of a BBB is important for prompt, accurate diagnosis and treatment of heart conditions, especially in accurate identification, of left BBB. This work proposes a new wrapper based hybrid approach for the detection of BBB that uses a Genetic Algorithm (GA) in combination with Artificial Neural Networks (ANN) to improve classification accuracy. Nineteen temporal features and three morphological features were extracted and normalized for each heartbeat from standard ECG recordings obtained from the MIT-BIH Arrhythmia database. The GA-ANN Hybrid resulted in improved sensitivity, specificity and accuracy (98%, 98% and 98% respectively) compared to the Principal Components Analysis (PCA)-ANN method (55%, 98% and 77% respectively) in the presence of noise. The GA-ANN Hybrid provides a better, more accurate identification for presence of BBB from ECG recordings leading to more timely diagnosis and treatment outcomes.