Protons produced greater subclinical lung changes after breast nodal irradiation
Proton therapy produced greater CT-detected lung changes than photons despite lower lung dose and no increase in clinical pulmonary toxicity.
Tag
Technical relevance
Fifteen years of radiotherapy data were converted into an AI-ready resource
An open-source workflow structured 15 years of radiotherapy data, although reirradiation prediction showed only modest discrimination.
Dose-guided positioning optimized proton target coverage within three minutes
Automated dose-guided positioning matched anatomical alignment overall and improved primary target coverage by 1.0% when the clinical goal was initially missed.
Transformer dose engine achieved near-Monte Carlo accuracy for carbon ion therapy
The Transformer calculated carbon-ion pencil-beam doses in 14 milliseconds with 98.0% gamma agreement at stringent 1%/1 mm criteria.
NanOx incompletely predicted laryngeal cancer cell survival across a helium Bragg peak
NanOx predictions differed from measured laryngeal cancer cell survival by 20–37% and overestimated the distal biological gradient.
Tomotherapy improved lung sparing in whole-pleural irradiation for disseminated thymic tumors
Tomotherapy improved target coverage and reduced bilateral lung exposure compared with VMAT and fixed-field IMRT in 20 whole-pleural irradiation cases.
Automated lattice radiotherapy evaluation cut analysis time nearly eightfold
An Eclipse-based tool reduced lattice plan evaluation from 15.4 to 2.0 minutes while maintaining close agreement with manual measurements.
Respiratory gating reduced lung SBRT target volumes and normal-tissue dose
A 30–70% gating window reduced planning target volume by 34% and mean lung dose by 0.75 Gy without compromising coverage.
Geometry-based scaling standardized central target definition for SCART planning
A phantom-derived model predicted SCART central target radius with R² values of 0.991–0.999, but clinical validation remains absent.
Helium pencil beam commissioning achieved submillimetre range accuracy and robust dose agreement
A scanned helium beamline achieved submillimetre range accuracy and generally ≤3% dose agreement, supporting further translation toward clinical helium therapy.
Protons reduced spinal cord and bowel dose in single-fraction spine SBRT plans
Proton plans preserved 20-Gy target coverage while reducing spinal cord, oesophageal, lung, and bowel dose compared with photon VMAT.
Combined dose–LET model predicted temporal lobe necrosis after skull-base proton therapy
Temporal lobe volumes receiving high dose or moderate dose with high LET jointly predicted grade 2 or higher necrosis.
Abdominal airbag matched plate compression but increased lateral setup error
Airbag compression reduced upper-abdominal respiratory motion comparably to a rigid plate but produced larger left–right setup errors.
ECHO automated lung planning across seven fractionation schedules in routine practice
ECHO generated consistent lung IMRT and VMAT plans while reducing active planner time by approximately 90 minutes per case.
AutoAdapt produced acceptable prostate MR-guided plans with minimal planner input
AI-generated constraints produced clinically acceptable adaptive prostate plans, although only seven of ten met every prespecified planning objective.