Detection of water in lung: algorithm study

• Main Author: Chew, Fang Yu
• Other Authors: Ser Wee
• Format: Final Year Project
• Language: English
• Published: 2014
• Subjects: DRNTU::Engineering::Electrical and electronic engineering
• Online Access: http://hdl.handle.net/10356/61426
• Institution: Nanyang Technological University

Water retention is the build up of fluid in the body and its causes include congestive heart failure, severe chronic lung diseases and low protein levels in the blood. As the current lung water detection methods such as Computed Tomography (CT) scan, lung ultrasound and X-ray require expensive and bulky equipment, an alternative method, which is the acoustic-based technique, will be used to detect lung water retention. This method will be less expensive and the device will be more portable as compared to conventional detection methods and instruments. This project aims to create an algorithm that analyzes the audio signal recorded from patients with water retention in lungs and normal people whom do not have any water in their lungs. The lung sounds were provided at the start of the project and these samples were used for the various steps of the algorithm, which are feature extraction, classification and validation. From the spectral analysis of the lung sounds, there was a higher variation in magnitude of the signals for patients as compared to that of a normal person. In addition, it was observed that the classification accuracy rates were promising as they were both above 99%.