Multi-Scale Feature Fusion with Image-Driven Spatial Integration for Left Atrium Segmentation from Cardiac MRI Images
Bipasha Kundu, Zixin Yang, Richard Simon, Cristian Linte · 2025
Read Multi-Scale Feature Fusion with Image-Driven Spatial Integration for Left Atrium Segmentation from Cardiac MRI Images by Bipasha Kundu, Zixin Yang, Richard Simon, Cristian Linte (2025) 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/arxiv-2502-06615.txt.
Description
Bipasha Kundu, Zixin Yang, Richard Simon, and Cristian Linte present a 2025 model that outlines the heart's left atrium in MRI scans. The outline guides ablation care for a common rhythm disorder. Hand tracing is slow and varies by reader. Their model pairs a broad vision backbone with a U-shaped decoder built for medical cuts. The catch is that broad models blur fine detail, which hearts demand.
Learned weights pick the most useful layers from the backbone for each case. The raw image rejoins the flow late in decoding, which restores crisp edges lost in shrinking steps. Fine fusion across scales sharpens small structures. Tests weigh the design against set rivals. Sharp auto outlines could cut planning time for heart procedures.
Catalog record
| CREATOR | Bipasha Kundu, Zixin Yang, Richard Simon, Cristian Linte |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 17 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2502-06615