Shape-Based Retrieval of Heart Sounds for Disease Similarity Detection

Abstract

Abstract. Retrieval of similar heart sounds from a sound database has applications in physician training, diagnostic screening, and decision support. In this paper, we exploit a visual rendering of heart sounds and model the morphological variations of audio envelopes through a constrained non-rigid translation transform. Similar heart sounds are then retrieved by recovering the corresponding alignment transform using a variant of shape-based dynamic time warping. Results of similar heart sound retrieval are demonstrated for various diseases on a large database of heart sounds.

Skeleton-based audio envelope shape analysis

“…Another possible solution is to employ the computer vision methods to some classes of audio signals [6]. Though the shape-based method for audio analysis has been proposed in [6] for a specific signal (heart sounds), there is a room for further search for a generic audio envelope shape descriptor.…”