Lung Sound Classification
Integrating Artificial Intelligence (AI) into lung sound classification has markedly improved respiratory disease diagnosis by analyzing intricate patterns within audio data. This paper presents a two-phase approach for enhanced diagnosis of respiratory diseases. In the initial binary classification phase, lung sounds are identified as healthy or abnormal, enabling early exits for normal cases. The second phase categorizes abnormal sounds into one of nine specific diseases. This project introduces the development of a user-friendly mobile application for seamless integration with the classification model. This application allows users to upload lung sound files, promptly receiving the top three potential diagnoses. The proposed two-phase strategy optimizes computational resources, while the mobile application extends the practical application of advanced diagnostic models, making the diagnosis more accurate and efficient. The study shows a system usability score above 70%.
Mobile application https://
Datasets:
A dataset of lung sounds recorded from the chest wall using an electronic stethoscope,
Sample data
GitHub code:
System Usability Study for the web application
User survey for results validation
Project Contributors: Ms. Sakuni Bandara , Mr. Supun Madusanka, Mr. Thinira Wanasinghe
Supervisors: Prof. Dulani Meedeniya, Mr.Meelan Bandara