Low-Power Wearable Respiratory Sound Sensing

Abstract

Building upon the findings from the field of automated recognition of respiratory sound patterns, we propose a wearable wireless sensor implementing on-board respiratory sound acquisition and classification, to enable continuous monitoring of symptoms, such as asthmatic wheezing. Low-power consumption of such a sensor is required in order to achieve long autonomy. Considering that the power consumption of its radio is kept minimal if transmitting only upon (rare) occurrences of wheezing, we focus on optimizing the power consumption of the digital signal processor (DSP). Based on a comprehensive review of asthmatic wheeze detection algorithms, we analyze the computational complexity of common features drawn from short-time Fourier transform (STFT) and decision tree classification. Four algorithms were implemented on a low-power TMS320C5505 DSP. Their classification accuracies were evaluated on a dataset of prerecorded respiratory sounds in two operating scenarios of different detection fidelities. The execution times of all algorithms were measured. The best classification accuracy of over 92%, while occupying only 2.6% of the DSP's processing time, is obtained for the algorithm featuring the time-frequency tracking of shapes of crests originating from wheezing, with spectral features modeled using energy.

A residential maintenance-free long-term activity monitoring system for healthcare applications

“…Many of these platforms do not pay the necessary attention to the energy consumption of the hardware used, resulting in wearable devices that require regular recharging, similarly to commercial products. Other works present low-power wearable systems with different types of sensors, such as bioimpedance [17], sound [18], electroencephalogram (EEG) [19] and inertial sensors [20]. Although relatively low power, these platforms target different body sensing applications that require sensing elements that are significantly energy-consuming compared to the sensing elements employed in our platform.…”

DesigningWearable Sensing Platforms for Healthcare in a Residential Environment

“…These platforms use off-the-shelf hardware and do not focus on their power consumption, resulting to wearable devices that require regular recharging. Other works present low power hardware that target various body sensing applications by incorporating different types of sensors, such as bio-impedance sensors [18], microphones [21] and inertial sensors [15].…”

Detection and Classification of Abnormal Respiratory Sounds on a Resource-constraint Mobile Device

“…The experimental results showed high accuracy of wheeze detection with low computational complexity. Authors in [6] focused on the analysis of computational complexity of common features extracted from Short Term Fourier Transform (STFT) using decision tree classification model. Four different algorithms and their detection accuracies were evaluated on a dataset from prerecorded respiratory sounds -wheezing in particular, with different efficiency metrics.…”

“…Recent studies [3], [4] and [5] have recorded major breakthrough in the development and improvement of computerized wheeze detection and analysis using sophisticated algorithms which involve detection of Fourier peaks and image analysis of the resulting spectrogram. Because these algorithms require enormous computational and sustainable power resources, the analyses were basically run on PCs and servers with few deliberate attempts on wearable low-power devices [6]. Besides, the approaches used did not include detection and evaluation of other abnormal sounds which may present similar morphologies alongside wheeze formation in a respiratory signal.…”