Detección de Soplos Cardíacos usando Medidas Derivadas del Análisis Acústico en Señales Fonocardiográficas

The cardiac mechanical information can be inferred from the sounds generated by the heart beats, which can be analyzed by signals known as phonocardiograms (PCG). A methodology based on acoustic analysis of digitized PCG signals is presented, oriented to detection of cardiac murmurs originated by valvular pathologies. Initially, a filtration system based on the wavelet transform is developed to reduce the disturbances that usually appear in the acquisition stage, adjusted and validated according to the clinical requirements. A between-beats segmentation algorithm is developed which uses information of the ECG signal previously acquired in a synchronous way to hook the beginning of the QRS complex with the beginning of the S1 sound of the PCG signal. Intra-beat segmentation is proposed for detecting S1, S2, systole and diastole based on the relationship analysis of energy and threshold. Features derived from the acoustic analysis are extracted on the segments. Feature effectiveness is evaluated by a linear Bayesian type classification model for separating the classes: normal and murmur. The used database of phonocardiographic records belongs to the National University of Colombia, having 164 records as follows: 81 records labelled as “normal” and 83 labelled as “murmur”. Finally, 360 representative beats were chosen by specialist, 180 normal and 180 with evidence of cardiac murmurs. Classification precision, sensitivity and specificity results are obtained. The best result of classification precision was 93,1% with sensitivity and specificity values equals to 93,3% and 92,8%, respectively, using Bayesian classifier and cross-validation procedure of 10 folds.