Automatic wheezing recognition in recorded lung sounds

Signal analysis of medical acoustic sounds with applications to chest medicine

“…Another approach is to use a wavelet-based filter to remove heart sound noise from lung sounds [14]. Various other wavelet-based algorithms have been successful in applications specific to particular lung conditions [15][16][17]. Most research focuses on feature recognition, but [18] discusses classification approaches, using neural networks, from manually identified features.…”

“…Currently, there are two main types of sensors available, microphone or piezo electric transducers. Both have been used by signal processing researchers, including both commercially available devices [16,19,25,26] and laboratory constructed ones. None of these is yet suitable for use by a lay person.…”

Two-level coarse-to-fine classification algorithm for asthma wheezing recognition in children's respiratory sounds

“…TP, FP, TN and FN are the numbers of the signal segments (samples) classified as true positive, false positive, true negative and false negative, respectively. Reliability is the evaluation measure similar to the measure known in literature as performance [20][21][22] or g-mean, which is calculated as a square root of Eq. (2).…”

Fundamentals of Respiratory Sounds and Analysis

“…One should note that even detection of wheezes from vesicular sounds when the wheezes are faint can become challenging and for sure automated classification of the sounds within the group would be very difficult. For a detailed description of the employed wheeze detection techniques, the interested reader may look at [67][68][69][70][71].…”