Denoising of ASL Data Using Deep Learning Priors Generated From Distribution Remapping
Ziyang Xu, Rong Guo, Ziwen Ke, Yudu Li, Yibo Zhao, Wen Jin, Ruihao Liu, Ziyu Meng, Yao Li, Zhi‐Pei Liang · 2026
Read Denoising of ASL Data Using Deep Learning Priors Generated From Distribution Remapping by Ziyang Xu, Rong Guo, Ziwen Ke, Yudu Li, Yibo Zhao, Wen Jin, Ruihao Liu, Ziyu Meng, Yao Li, Zhi‐Pei Liang (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13421004.txt.
Description
Ziyang Xu and colleagues build a deep learning cleaner for arterial spin labeling scans, a brain blood flow method whose images are notoriously noisy. Weak signal makes flow numbers jumpy and can sink the method's clinical promise. Standard learning-based cleaners overfit when training data run scarce, as they do here.
Their fix reshapes the data distribution first, stretching limited examples into a richer training ground. Cleaned high-quality outputs then feed a probability model that rebuilds features outside the training range under sparse assumptions. Simulations showed the remapping step was the key to making augmentation work at all in the small-data setting. Steadier flow maps could let this radiation-free scan serve more patients, including children and repeat studies.
Catalog record
| CREATOR | Ziyang Xu, Rong Guo, Ziwen Ke, Yudu Li, Yibo Zhao, Wen Jin, Ruihao Liu, Ziyu Meng, Yao Li, Zhi‐Pei Liang |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 38406 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |