| Title |
Needle Detection and Image Enhancement for Effective Ultrasound-guided Needle Biopsy |
| Authors |
이상혁(Sanghyuk Lee) ; 김성호(Seong Ho Kim) ; 송병철(Byung Cheol Song) |
| DOI |
https://doi.org/10.5573/ieie.2023.60.12.39 |
| Keywords |
Self-supervised learning; Biopsy needle detection; Segmentation |
| Abstract |
Existing deep learning-based needle detection models in ultrasound-guided needle biopsy rely on supervised learning. So, they still show limited performance in needle detection when labeled datasets are insufficient. To overcome this problem, we propose a novel needle detection model for effective ultrasound-guided needle biopsy by introducing a self-supervised learning methodology. Both qualitative and quantitative experimental results show that the proposed model detects needles more similar to actual needles than the existing models and provides better needle detection performance in terms of Dice and IoU. |