Least Mean Square (LMS) based neural design and metric evaluation for auscultation signal separation

Abstract

Physical examination of patients usually involves inspection, palpation, percussion and auscultation. Auscultation is an important key for the physical (respiratory and circulatory) examination and helpful in diagnosing various disorders. Auscultation is performed for the purposes of examining the circulatory and respiratory sounds and gastrointestinal system (bowel sounds). Besides inconsistencies in the propagation of the normal sounds, there are also several types of specific irregularities that can be heard in respiratory sounds. Commonly known abnormal sounds in lung sounds are wheezes, stridor, Squawks, Rhonchi and Crackles and heart sounds (Heart murmurs). However, detection of abnormal sounds during auscultation needs extensive training and experience. So, the separation of these heart sound signals (HSS) and the lung sound signals (LSS) is of great research interest. In this work, the authors proposed a novel design used for Auscultation Analysis. For this purpose, proposed design has been trained using different networks training algorithms which resulted in better mean square error reduction within compact time. Proposed design can be used for bio signal processing and physicians in the signal analysis.