Enhancing weakly supervised data augmentation networks for thyroid nodule assessment using traditional and doppler ultrasound images.
Authors
Affiliations (7)
Affiliations (7)
- Department of Computer Science, Faculty of Science, Ramkhamhaeng University, Bangkok, Thailand. Electronic address: [email protected].
- Department of Radiology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand.
- AI Center, Asian Institute of Technology, Pathumthani, Thailand.
- Logsig Co., Ltd, Pathumthani, Thailand.
- Harbor Branch Oceanographic Institute, Florida Atlantic University, FL, USA.
- Industrial Systems Engineering Department, Asian Institute of Technology, Pathumthani, Thailand.
- Information and Communication Technologies, Asian Institute of Technology, Pathumthani, Thailand.
Abstract
Thyroid ultrasound (US) is an essential tool for detecting and characterizing thyroid nodules. In this study, we propose an innovative approach to enhance thyroid nodule assessment by integrating Doppler US images with grayscale US images through weakly supervised data augmentation networks (WSDAN). Our method reduces background noise by replacing inefficient augmentation strategies, such as random cropping, with an advanced technique guided by bounding boxes derived from Doppler US images. This targeted augmentation significantly improves model performance in both classification and localization of thyroid nodules. The training dataset comprises 1288 paired grayscale and Doppler US images, with an additional 190 pairs used for three-fold cross-validation. To evaluate the model's efficacy, we tested it on a separate set of 190 grayscale US images. Compared to five state-of-the-art models and the original WSDAN, our Enhanced WSDAN model achieved superior performance. For classification, it reached an accuracy of 91%. For localization, it achieved Dice and Jaccard indices of 75% and 87%, respectively, demonstrating its potential as a valuable clinical tool.