Smart wearable sensors are essential for continuous health-monitoring applications and detection accuracy of symptoms and energy efficiency of processing algorithms are key challenges for such devices. While several machine-learning-based algorithms for the detection of abnormal breath sounds are reported in literature, they are either too computationally expensive to implement into a wearable device or inaccurate in multi-class detection. In this paper, a kernel-like minimum distance classifier (K-MDC) for acoustic signal processing in wearable devices was proposed. The proposed algorithm was tested with data acquired from open-source databases, participants, and hospitals. It was observed that the proposed K-MDC classifier achieves accurate detection in up to 91.23% of cases, and it reaches various detection accuracies with a fewer number of features compared with other classifiers. The proposed algorithm’s low computational complexity and classification effectiveness translate to great potential for implementation in health-monitoring wearable devices.
1 May 2020 in Biomedical Signal Processing and Control
Remote monitoring of vital signs in infectious patients minimizes the risks of viral transmissions to healthcare professionals. Donning face masks could reduce the risk of viral transmissions and is currently practiced in medical facilities. An acoustic-sensing device was attached to face masks to assist medical facilities in remotely monitoring patients’ respiration rate and wheeze occurrence. Usability and functionality studies of the modified face mask were evaluated on 16 healthy participants. Participants were blindfolded throughout the data collection process. Respiratory rates of the participants were evaluated for one minute. The wheeze detection algorithm was assessed by playing 176 wheezes and 176 normal breaths through a foam mannequin. No discomfort was reported from the participants who used the modified mask. The mean error of respiratory rate was found to be 2.0 ± 1.3 breath per minute. The overall accuracy of the wheeze detection algorithm was 91.9%. The microphone sensor that was first designed to be chest-worn has been proven versatile to be adopted as a mask attachment. The current findings support and suggest the use of the proposed mask attachment in medical facilities. This application can be especially helpful in managing a sudden influx of patients in the face of a pandemic.
Remote monitoring of vital signs in infectious patients minimizes the risks of viral transmissions to healthcare professionals. Evidence indicates that donning face masks reduces the risk of viral transmissions and is now the norm in medical facilities. We propose attaching an acoustic-sensing device onto face masks to assist medical facilities in monitoring patients' respiration remotely. Usability and functionality studies of the modified face mask were evaluated on 16 healthy participants, who were blindfolded throughout the data collection. Around half of the participants noticed the difference between the modified and unmodified masks but they also reported there was no discomfort in using the modified mask. Respiratory rates of the participants were evaluated for one minute and the mean error of respiratory rate was found to be 2.0 +/- 1.3 breath per minute. As all participants were healthy, the wheeze detection algorithm was assessed by playing 176 wheezes and 176 normal breaths through a foam mannequin. The recordings were played at three different times to account for varying environmental noise. The overall accuracy of the wheeze detection algorithm was 91.9%. The current findings support and suggest the use of the mask attachment in medical facilities.