Attention U-Net Based Ultrasound Nerve Segmentation with Efficient Net Classification
Attention U-Net Based Ultrasound Nerve Segmentation with Efficient Net Classification
Jyothi Medagam
2026 · DOI: 10.22214/ijraset.2026.81682
International Journal for Research in Applied Science and Engineering Technology · 0 Citations
TLDR
A deep learning-based approach that combines preprocessing, segmentation, and classification for automated ultrasound nerve analysis and results show improved Dice score, IoU, and classification accuracy compared to existing methods, indicating that the proposed method provides a reliable and efficient solution.
Abstract
Ultrasound imaging is commonly used to identify and study peripheral nerves, but image quality issues such as noise,
low contrast, and unclear structures make accurate analysis difficult. To address this, a deep learning-based approach isproposed that combines preprocessing, segmentation, and classification. CLAHE and SRAD are first applied to improve imagequality by enhancing contrast and reducing noise. Then, an Attention U-Net model is used to detect the nerve region by focusingon important areas while reducing background interference. The detected region is extracted and passed to an EfficientNet-B0model for classification. The system is trained and tested on the Kaggle ultrasound nerve dataset with around 5,600 images. Theresults show improved Dice score, IoU, and classification accuracy compared to existing methods, indicating that the proposedmethod provides a reliable and efficient solution for automated ultrasound nerve analysis.