Diaphragm motion was less reliable than fiducials for markerless tumor tracking
Tumor–diaphragm geometry varied by approximately 4.5 mm during breathing, substantially more than tumor–fiducial relationships.
Tumor–diaphragm geometry varied by approximately 4.5 mm during breathing, substantially more than tumor–fiducial relationships.
ETLD tracked 106,000 cine-MRI frames with submillimetre error, while integrated segmentation achieved a mean global Dice score of 87.7%.
SGRT triggered verification in 15% of head and neck fractions, but almost half of alerts were not confirmed on X-ray imaging.
Airbag compression reduced upper-abdominal respiratory motion comparably to a rigid plate but produced larger left–right setup errors.
Patient-specific reinforcement learning improved moving-target coverage over static GTV plans while reducing some normal-tissue exposure compared with ITV planning.
Rigid couch correction cannot fully resolve regional posture mismatch, immobilization failure or progressive anatomical change during head and neck radiotherapy.
A 4.5 mm isotropic margin covered most observed small-bowel occupancy despite craniocaudal displacements exceeding 20 mm in some patients.
Single-field optimization improved proton LATTICE motion robustness
VMAT-based lattice SFRT achieved cross-platform deliverability with gamma pass rates above 90% and preserved peak-to-valley dose ratios.
Cherenkov light imaging tracked gold implant motion in water with submillimetre accuracy during static and volumetric modulated arc therapy delivery.
After 66% of couch-angle changes, motion exceeded tolerance, and 0.5-millimetre and 0.5-degree errors reduced near-minimum target dose by 7.6%.
Mean breast target displacement was below 1.3 mm in every direction, with calculated respiratory margins of approximately 1 mm.
Gastrointestinal motility reduced single-fraction target coverage and broadened organ-at-risk dose ranges, while conventional fractionation substantially attenuated the interplay effect.
The patient-specific framework outperformed three comparison methods and reconstructed respiratory anatomy in 15.6 milliseconds per frame.
Hybrid phase-amplitude gating restored dose agreement during regular motion, while irregular breathing increased treatment time and produced inconsistent accuracy.