St-RegSeg: an unsupervised registration-based framework for multimodal magnetic resonance imaging stroke lesion segmentation
Chengzhi Gui, Xingwei An, Tingting Li, Shuang Liu, Dong Ming · 2024
Read St-RegSeg: an unsupervised registration-based framework for multimodal magnetic resonance imaging stroke lesion segmentation by Chengzhi Gui, Xingwei An, Tingting Li, Shuang Liu, Dong Ming (2024) 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/pmc-11651977.txt.
Description
Finding a stroke lesion on MRI quickly and precisely guides urgent treatment, but scans come in several modes that must be aligned before a computer can outline the damage. Chengzhi Gui and four co-authors build St-RegSeg, a framework that registers multimodal scans without supervision and then segments the lesion with a proven network.
On a three-center challenge dataset, their registration beat the standard tool combination clearly: overlap scores rose about 25 percent and image errors fell. Segmentation accuracy improved modestly, while the whole pipeline ran about 41 times faster than the standard pairing. Against a newer rival it still won on both accuracy and speed. The authors offer the framework as a practical step toward faster stroke imaging workflows. The study appears in 2024.
Catalog record
| CREATOR | Chengzhi Gui, Xingwei An, Tingting Li, Shuang Liu, Dong Ming |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 54 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |