Hierarchical transformers like SwinUNETR show better robustness to perturbations in a two-stage fluence map prediction pipeline for IMRT, with smooth degradation under moderate changes but sharp failures under severe ones, and SSIM fails to capture clinically relevant errors.
Variable generalization performance of a deep learning model to detect pneumonia in chest radiographs: a cross-sectional study,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Robustness of Transformer-Based Fluence Map Prediction Under Clinically Realistic Perturbations
Hierarchical transformers like SwinUNETR show better robustness to perturbations in a two-stage fluence map prediction pipeline for IMRT, with smooth degradation under moderate changes but sharp failures under severe ones, and SSIM fails to capture clinically relevant errors.