Artificial Intelligence Techniques Used for Wheeze Sounds Analysis: Review
Wheezes are acoustic, adventitious, continues and high pitch pulmonary sounds produce due to airway obstruction, these sounds mostly exist in pneumonia and asthma patients. Artificial intelligence techniques have been extensively used for wheeze sound ana
- PDF / 285,709 Bytes
- 4 Pages / 595 x 791 pts Page_size
- 19 Downloads / 204 Views
1 School of Mechatronic Engineering, Universiti Malaysia Perlis (UniMAP), Arau, Perlis, Malaysia Faculty of Electronics and Computer Engineering, Universiti Teknikal Malaysia Melaka (UTeM), Durian Tunggal, Melaka, Malaysia 3 School of Electronics Engineering (SENSE), Vellore Institute of Technology (VIT), Tamil Nadu, India 4 Department of Anesthesiology, Hospital Tengku Ampuan Rahimah (HTAR), Klang, Selangor, Malaysia 5 Faculty of Manufacturing Engineering, Universiti Malaysia Pahang, Pekan, Pahang, Malaysia
Abstract— Wheezes are acoustic, adventitious, continues and high pitch pulmonary sounds produce due to airway obstruction, these sounds mostly exist in pneumonia and asthma patients. Artificial intelligence techniques have been extensively used for wheeze sound analysis to diagnose patient. The available literature has not yet been reviewed. In this article most recent and relevant 12 studies, from different databases related to artificial inelegance techniques for wheeze detection has been selected for detailed review. It has been noticed that now trend is going to increase in this area, for personal assistance and continues monitoring of patient health. The literature reveals that 1) wheezes signals have enough information for the classification of patients according to disease severity level and type of disease, 2) significant work is required for identification of severity level of airway obstruction and pathology differentiation.
defines exact type of related disease. Wheezes monitoring identifies asthma attack and results of therapies, and wheezes characteristics recognize severity level of obstruction (4). Wheezes have multiple tones, typically 4-6 and usually 1-6 in the spectrum 100–1200 Hz. The home medical personal assistance to asthma patients is important so research is being carried out, trying to explore low computational, low power consumption and user friendly devices to maintain the health (5). Conventionally wheeze detection algorithms depend on wheeze amplitude (4). But wheeze amplitude dependent on flow rate, efforts are going to construct algorithms without sound attenuation. So currently authors are focusing to identify wheeze through artificial intelligence techniques related to classifiers and feature extraction (4).
Keywords— Wheeze, Wheeze Sounds, Respiratory Sounds, Airway Obstruction, Wheeze Analysis
I.
INTRODUCTION
Wheezes are continuous, adventitious and abnormal respiratory sounds. These are musical sounds super imposed on normal breath sounds mostly produce in asthma and pneumonia patients. The automatic wheezes recognition history started in 1980s. Physicians in the field of pulmonary medicine use a stethoscope to listen the respiratory sounds with the goal of diagnosing respiratory disorders and abnormalities. Oscillation of bronchi wall produces turbulent flow of breath which generates acoustic signal. Wheeze mostly produce in pneumonia and asthmatic patients. According to the WHO, 235 million individuals are suffering from asthma (2013), and 15% of children die due to p
Data Loading...
