Serial skin photographs predict significant breast radiation dermatitis by treatment week three
A dual-view AI model predicted grade ≥2 breast radiation dermatitis with AUC 0.91 overall and approximately 0.80 by week three.
A dual-view AI model predicted grade ≥2 breast radiation dermatitis with AUC 0.91 overall and approximately 0.80 by week three.
In 40 breast and chest wall plans, DynamicArc improved coverage and robustness in challenging anatomies but increased thyroid dose and hotspots.
In 142 patients with limited-stage SCLC, CBCT-reconstructed dose improved pneumonitis and esophagitis prediction compared with planned dose, but discrimination remained modest.
Automated cardiac contouring was considered acceptable with at most minor corrections for 93.8% of 630 structures, although coronary arteries remained challenging.
Practical pelvic contouring guidance illustrates how cervical involvement, extensive LVSI and bladder motion influence postoperative endometrial cancer target volumes.
Adding solid-water extensions restored PTV gamma agreement to 98–100% at 1%/1 mm, whereas unextended wide-field calibration performed poorly.
New ESTRO-EPTN recommendations define validated, risk-based pathways for introducing independent dose calculation and machine log-file QA in scanned particle therapy.
A nonclinical time-driven PBS configuration delivered 25 Gy in approximately 0.6 seconds at 42 Gy/s while preserving proton range and scanning accuracy.
Across 1,442 head and neck cancer patients, Bayesian ridge achieved the best imputation accuracy, while advanced methods showed similar downstream NTCP performance.
The vendor-neutral ARTEMIS platform automated radiotherapy planning workflows and reduced planner interaction time for independent dose checking by 42%.
Across more than 20,000 evaluated plans, integrated dose calculation, log-file and portal dosimetry methods identified complementary planning and delivery errors.
A PTV-dependent V50% threshold flagged 10% of clinical SRS plans, with replanning reducing excess intermediate-dose exposure by a median 14%.
In 68 cervical cancer patients, pelvic marrow dose correlated with hematologic toxicity, supporting stricter exploratory constraints for 44 Gy in 20 fractions.
Three representative liver cases illustrate breath-hold IMPT, compression-based proton SBRT and spatially fractionated proton therapy with detailed planning and QA workflows.
Uncertainty-guided post-processing increased MRI-based liver metastasis detection sensitivity from 84.6% to 98.1%, while generating additional false-positive candidates.