Shin, Soogyeong (2024) Enhanced Disease Classification in Respiratory Sounds: A Transfer Learning Approach Utilizing ICBHI and Coswara Datasets. Master thesis, Voice Technology (VT).
|
PDF
MA-5661285-S-Shin.pdf Download (944kB) | Preview |
Abstract
The early and accurate detection of respiratory diseases, such as asthma, chronic obstructive pulmonary disease (COPD), and pneumonia, is critical for improving patient outcomes. This need has been emphasized after the COVID-19 pandemic. Traditional diagnostic methods, such as auscultation, depend on the experience of clinicians and are constrained by their subjective assessment. To address these limitations, this study seeks to increase the accuracy and efficiency of respiratory disease diagnosis using transfer learning through data augmentation. This research utilized publicly accessible respiratory sound datasets, ICBHI 2017 and Coswara, aiming to overcome the constraints of smaller datasets. A transfer learning strategy employing a Residual Networks (ResNet)-based model was implemented to enhance the accuracy of respiratory disease diagnosis. The models, initially trained on the ICBHI dataset, were fine-tuned with data from Coswara, enabling them to detect abnormal lung sound variations associated with specific respiratory conditions. The results indicate that the AI-enhanced model achieved accuracy peaks of 95.80\% and 93\% with the first and second fine-tuning strategies, respectively. These findings demonstrate the successful integration of two respiratory sound datasets through transfer learning. This study not only highlights the effectiveness of advanced AI techniques in medical diagnosis but also highlights the importance of dataset augmentation to ensure robust model performance. Integrating multiple datasets through transfer learning can be an effective strategy not only for respiratory diseases but also for developing diagnostic tools for a variety of conditions with small datasets.
| Item Type: | Thesis (Master) |
|---|---|
| Name supervisor: | Verkhodanova, V. and Coler, M.L. |
| Date Deposited: | 19 Jul 2024 09:21 |
| Last Modified: | 19 Jul 2024 09:21 |
| URI: | https://campus-fryslan.studenttheses.ub.rug.nl/id/eprint/533 |
Actions (login required)
 |
View Item |