AI-generated synthetic CT shows promising dose accuracy in prostate patients with hip implants
In 20 prostate patients with hip replacements, implant-trained synthetic CT models closely reproduced CT dose, but two cases with rectal air had outliers.
In 20 prostate patients with hip replacements, implant-trained synthetic CT models closely reproduced CT dose, but two cases with rectal air had outliers.
Among 36 MR-Linac consortium experts, 86% reported access to the technology, but only 28% treated head and neck cancer on the platform.
Across 149 cervical radiotherapy fractions, adding ultrasound to CBCT improved uterocervix visibility but increased correct plan selection by only one percentage point.
In 376 spinal treatment sites, initial setup error was unrelated to pain severity but differed by spinal level and immobilization.
In 46 lung cancer patients, combined functional planning reduced high-functioning lung doses by approximately 1 Gy but modestly increased heart and spinal cord exposure.
In 22 spinal targets, a dedicated planning system reduced spinal cord D2% by 2.6–2.7 Gy versus general-purpose planning without compromising PTV coverage.
Across 250 prostate SBRT fractions, adaptive planning prevented major target undercoverage, while most scheduled plans remained close to the prescription dose.
AAPM MPPG 18.a sets daily output constancy at ±3%, annual field alignment at ±2 mm, and physicist-led commissioning requirements for superficial RT.
In 15 left-sided breast cancer cases, proton breath-hold reduced median heart dose from 2.1 to 0.8 Gy but increased lung exposure.
A self-supervised model detected clinical pelvic MRI anomalies with AUC 0.95, but missed all five hydrogel spacers and lacked downstream clinical validation.
In a constrained phantom target, robust optimization reduced worst-case dose loss but increased minimum LET loss from 11.7% to 16.7%.
Within 41 lung cancer patients, lower thoracic lymph nodes moved 5.2 versus 3.3 mm cranio-caudally, challenging uniform internal margins.
In 22 paired lattice plans, helical tomotherapy reduced valley-to-peak dose ratios and low-dose spill versus VMAT but required 2.5-fold longer beam-on time.
In 15 prostate nodal recurrence cases, IMPT maintained target coverage while substantially reducing kidney, bowel and integral dose compared with VMAT.
In 33 right-sided breast RT patients, larger target dimensions correlated with higher ipsilateral lung and cardiac doses, but proposed thresholds remain unvalidated.