.png){kind=icon}
.png){kind=icon}
Return
Susceptibility artifact correction in magnetic resonance thermometry based on deep learning with simulation-derived thermal physics priors
.png){kind=icon}
PRE
AI
Y
Y
X
L
J
DOI：10.1016/j.ijheatmasstransfer.2026.129346.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• A deep learning framework with physical priors for MRT artifact correction. • Three-stage network decouples artifacts from temperature fields. • Trained in simulated temperature fields to overcome clinical data scarcity. • Superior performance over previous methods of MRT artifact correction. • The reconstructed error is on average less than 1 °C, and the speed is 0.18 s.
Journal
IF:
5.8
Papers:
2.5W
Citations:
10.2W
Organization
Cited Papers