Welcome to my homepage!
I am a PhD student in Okutomi Lab at Tokyo Institute of Technology. Currently I am working with Masatoshi Okutomi.
My research interest includes medical imaging and computer vision.
π₯ News
- 2026.05: Β ππ New paper on Pipeline Magnetic Flux Leakage Imaging is accepted by SIGKDD2026.
- 2026.05: Β ππ New paper on Segmentation of Bone Erosion is accepted by IEEE EMBC2026.
- 2025.10: Β ππ New paper on Segmentation of Wrist Carpal Bones is accepted by IEEE BIBM2025.
- 2025.10: Β ππ The extension of IEEE EMBC2024 paper is accepted by BSPC2026.
- 2025.09: Β ππ New paper on Multi-task X-Ray Rheumatoid Arthritis Database is accepted by NeurIPS2025.
- 2024.06: Β ππ We achieved Top-5 in the CVPR 2024: Segment Anything In Medical Images On Laptop.
- 2024.04: Β ππ New paper on MRI reconstruction is accepted by IEEE EMBC2024.
π Featured Datasets
RAM-H1200
A unified hand radiograph dataset for rheumatoid arthritis analysis, covering whole-hand bone instance segmentation, pixel-level bone erosion masks, SvdH joint regions, and joint-level BE/JSN scoring.
π Publications

PipeMFL-240K: A Dataset and Benchmark for Object Detection in Pipeline Magnetic Flux Leakage Imaging
Tianyi Qu*, Songxiao Yang*, Haolin Wang*, Huadong Song, Xiaoting Guo, Wenguang Hu, Guanlin Liu, Honghe Chen, Yafei Ou

BESegNet: A Bone Erosion Segmentation Network for Rheumatoid Arthritis in Conventional Radiography (to appear)
Songxiao Yang, Haolin Wang, Tamotsu Kamishima, Masayuki Ikebe, Yafei Ou, Masatoshi Okutomi

Songxiao Yang, Haolin Wang, Masayuki Ikebe, Tamotsu Kamishima, Yafei Ou, Masatoshi Okutomi

PixelINR: Scan-specific self-supervised MRI reconstruction based on implicit neural representations
Songxiao Yang, Yafei Ou, Masatoshi Okutomi

RAM-W600: A Multi-Task Wrist Dataset and Benchmark for Rheumatoid Arthritis
Songxiao Yang, Haolin Wang, Yao Fu, Ye Tian, Tamotsu Kamishima, Masayuki Ikebe, Yafei Ou, Masatoshi Okutomi

A Light-weight Universal Medical Segmentation Network for Laptops Based on Knowledge Distillation
Songxiao Yang, Yizhou Li, Ye Chen, Zhuofeng Wu, Masatoshi Okutomi

Instance-Wise MRI Reconstruction Based on Self-Supervised Implicit Neural Representation
Songxiao Yang, Yizhou Li, Masatoshi Okutomi

Hammer pdf: An intelligent pdf reader for scientific papers
Sheng-Fu Wang, Shu-Hang Liu, Tian-Yi Che, Yi-Fan Lu, Song-Xiao Yang, Heyan Huang, Xian-Ling Mao

U-SDRC: a novel deep learning-based method for lesion enhancement in liver CT images
Zhongshu Zheng, Ling Ma, Songxiao Yang, Said Boumaraf, Xiabi Liu, Xiaohong Ma
π Honors and Awards
- 2025.11 I received the Student Travel Grant from IEEE EMBC 2026.
- 2025.11 I received the Student Award from BIBM 2025.
- 2025.10 I received the Scholar Award from NeurIPS 2025.
- 2024.06 I received the βMeritorious Winner Awardβ(Top5) from CVPR 2024: Segment Anything In Medical Images On Laptop.
- 2023.09 I received the βICT Outstanding Student Award (Masters)β from the Department of Information and Communication Engineering.
- 2022.06 I received the βBest Research Proposal Presentation Awardβ from the Department of Information and Communication Engineering.
βοΈ Academic Service
- Reviewer of NeurIPS. (2026)
- Reviewer of IEEE International Conference on Bioinformatics and Biomedicine. (2024~)
π Educations
- 2023.09 - present, Tokyo Institute of Technology, System and Control Engineering, Doctor of Engineering.
- 2021.09 - 2023.09, Tokyo Institute of Technology, Information and Communication Engineering, Master of Engineering.
- 2017.09 - 2021.06, Beijing Institute of Technology, Computer Science and Technology, Bachelor of Engineering.
π» Internships
- 2025.01 - present, AI Research Engineer, Splink,inc., Minato, Tokyo, Japan.
- 2025.06 - 2025.12, Computer Vision Algorithm Engineer, Huawei Japan Research Center, Shinagawa, Tokyo, Japan.
- 2024.01 - 2024.06, Computer Vision Algorithm Engineer, Oppo Japan Research Center, Yokohama, Kanagawa, Japan.
- 2022.09 - 2023.04, Research Assistant, Tokyo Institute of Technology, Yokohama, Kanagawa, Japan.
- 2021.11 - 2022.06, Research Assistant, Beijing Institute of Technology, Beijing, China. Project URL: HammerPDF