.png){kind=icon}
.png){kind=icon}
Return
An efficient multi-scale attention and transformer framework for osteoarthritis diagnosis and surgical decision support using 3D MRI
.png){kind=icon}
PRE
AI
M
A
M
K
DOI：10.1016/j.compeleceng.2026.111331.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• An Enhanced AMWPMFA denoises MRI scans while preserving cartilage and bone structures for reliable analysis. • A task-aware transformer captures both global joint morphology and local degeneration patterns for accurate OA grading. • A hybrid convolutional attention network with multi-scale residual learning models complex 3D MRI features efficiently. • The framework is tested on the Landspitali and OAI datasets, with expert input confirming its reliability and clinical relevance.
Journal
C
IF:
4.9
Papers:
6.7K
Citations:
1.3W
Organization
Cited Papers