Masked autoencoder pretraining for peripheral nerve segmentation in ultrasound images.
Authors
Affiliations (7)
Affiliations (7)
- Chung-Ang University Industry-Academic Cooperation Foundation, Seoul, South Korea.
- Mellowing Factory Co. Ltd., Seoul, South Korea.
- Department of Dermatology, Korea University Guro Hospital, Seoul, Republic of Korea.
- Chung-Ang University College of Medicine, Seoul, Republic of Korea.
- Mellowing Factory Co. Ltd., Seoul, South Korea. [email protected].
- Chung-Ang University College of Medicine, Seoul, Republic of Korea. [email protected].
- Department of Physical and Rehabilitation Medicine, Chung-Ang University Gwangmyeong Hospital, Gwangmyeong-si, Gyeonggi-do, Republic of Korea. [email protected].
Abstract
Despite recent advances in medical image segmentation, deep learning applications in musculoskeletal ultrasound remain limited by small labeled datasets and a narrow clinical focus. Most existing work centers on the median nerve, leaving other clinically relevant peripheral nerves understudied. This study explores masked autoencoder (MAE) pretraining as a scalable approach to overcome data scarcity in peripheral nerve segmentation. A dataset of 10,603 ultrasound images across eight nerves was constructed, spanning over 1,500 patients with normal nerve conduction studies. A SegFormer-based encoder was pretrained on unlabeled ultrasound images using MAE and subsequently fine-tuned on nerve-specific segmentation tasks. Pretraining was applied per cross-validation fold to prevent data leakage. Compared to randomly initialized baselines, MAE-pretrained models achieved consistent performance gains across all nerves. Improvements were most pronounced for low-resource targets such as the sural (0.144 → 0.399 Dice) and lateral femoral cutaneous (0.280 → 0.550 Dice) nerves. Nerves with more abundant training data, like the median (0.890 → 0.897 Dice) and ulnar (0.862 → 0.869 Dice), showed only marginal improvements. These findings indicate that MAE pretraining particularly strengthens feature extraction for small, difficult-to-label structures in ultrasound segmentation. This underscores the potential of self-supervised learning to ease annotation demands and support the development of more comprehensive clinical AI tools.