Noise cancellation algorithm for respiratory sound classification

The lung is an important organ and an important health indicator of a human. Respiratory diseases are one of the top causes of death and disabilities in the world [1]. It has therefore become vital that the doctor is able to properly diagnose the lung health of a patient as more people go to the cli...

Full description

Saved in:
Bibliographic Details
Main Author: Soh, Marianne Hui Ying
Other Authors: Ser Wee
Format: Final Year Project
Language:English
Published: Nanyang Technological University 2020
Subjects:
Online Access:https://hdl.handle.net/10356/139233
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:The lung is an important organ and an important health indicator of a human. Respiratory diseases are one of the top causes of death and disabilities in the world [1]. It has therefore become vital that the doctor is able to properly diagnose the lung health of a patient as more people go to the clinics and hospitals for their health check ups. Lung sounds such as wheezing and crackling indicates the presence of lung disease. With background noises and interferences, the doctor may not be able to give an accurate examination of the patient’s health. An adaptive filter has self-adjusting characteristics and adapts to changes in its input signals automatically. The objective of this project is to design an adaptive noise cancellation algorithm based on least mean square (LMS) filtering on MATLAB to improve the clarity of respiratory signals. The LMS algorithm, established by Widrow and Hoff, is commonly used to perform noise cancellation. After the design and implementation of the LMS algorithm, the project inputted a simulated signal into the adaptive noise cancellation algorithm and managed to output a clean signal. This project will present on the development and implementation of the LMS algorithm on MATLAB to remove undesired noises to get a clearer and cleaner signal. The results and findings will also be discussed. This algorithm will be able to aid doctors in providing a more accurate examination of a patient’s lung health.