High-Fidelity Functional Ultrasound Reconstruction via A Visual Auto-Regressive Framework

Functional ultrasound (fUS) imaging provides exceptional spatiotemporal resolution for neurovascular mapping, yet its practical application is significantly hampered by critical challenges. Foremost among these are data scarcity, arising from ethical considerations and signal degradation through the cranium, which collectively limit dataset diversity and compromise the fairness of downstream machine learning models.

@article{chen2025_2505.21530,
  title={ High-Fidelity Functional Ultrasound Reconstruction via A Visual Auto-Regressive Framework },
  author={ Xuhang Chen and Zhuo Li and Yanyan Shen and Mufti Mahmud and Hieu Pham and Chi-Man Pun and Shuqiang Wang },
  journal={arXiv preprint arXiv:2505.21530},
  year={ 2025 }
}