A novel entropy-based wheeze detection algorithm offers reduced computational complexity suitable for low-power wearable monitoring systems.
May enable low-power wearable wheeze monitoring; leaves open clinical validation before adoption.
Wheezy breath is a general sign for airway obstructive diseases like asthma and chronic obstructive pulmonary disease. This paper presents a novel wheeze detection method that can be used to detect and identify wheezes automatically. The proposed method exploits ``entropy" to describe the pattern of the frequency spectrum of the wheezing signal and the wheeze identification can be conducted using only one or two entropy-based features. Therefore the proposed method has considerably reduced computational complexity and is able to operate with the low power consumption constraint under the wearable condition. The performance of the proposed method has been evaluated using lung sounds from patients and normal subjects under different Signal-to-Noise Ratio (SNR).
No takes yet. Share an insight, caveat, or question.
Zhang et al. (2009) studied this question.