Synthetic MRI from planning CT shows limited head and neck tumor contour agreement
CT-generated synthetic MRI achieved mean Dice scores of 0.58 for primary and 0.73 for nodal tumor contours in ten test patients.
CT-generated synthetic MRI achieved mean Dice scores of 0.58 for primary and 0.73 for nodal tumor contours in ten test patients.
The vendor-neutral ARTEMIS platform automated radiotherapy planning workflows and reduced planner interaction time for independent dose checking by 42%.
Spectral CT adds quantitative tissue information across planning, particle range estimation and adaptive imaging, but prospective outcome evidence remains limited.
Upright gantry-less RT could reduce system complexity and improve access, but imaging, immobilization and limited clinical evidence remain major barriers.
In a diagnostically difficult breast cancer case, FES PET identified occult ER-positive nodal disease and materially changed radiotherapy volume and dose.
Energy-layer-wise compensation restored millimetre-level proton targeting despite magnetic-field beam shifts approaching 30 mm in an MR-integrated prototype.
The pre-plateau modulation factor preserved plan quality while further reduction created idle delivery time without meaningful efficiency gain.
A severity-weighted action-priority framework identified the same seven highest RT workflow risks while avoiding 23 additional traditional FMEA flags.
AI can automate particle planning, dose prediction, adaptation and quality assurance, but limited data and validation still constrain clinical adoption.
uTPS generated Halcyon VMAT plans with generally comparable dosimetry to Eclipse and greater than 97% gamma passing rates.
D95% and D98% PTV prescriptions produced substantial interpatient variability, while GTV- and ITV-based median or mean doses were more accurate.
Brainlab Elements commissioning was completed in one day, while Monte Carlo generally agreed with end-to-end measurements within 3%.
Automated proton plans reproduced clinical dysphagia and xerostomia estimates closely while reducing optimization time from days to about one hour.
Individual target optimization lowered normal-brain V12 and improved dose gradients while maintaining over 99% coverage, at the cost of higher monitor units.
Deep learning produced machine-deliverable lymphoma plans across heterogeneous anatomy, but target coverage and hot spots remained inferior to reference planning.