Carboplatin-paclitaxel showed comparable outcomes in cisplatin-ineligible head and neck cancer
Retrospective outcomes with carboplatin-paclitaxel were not significantly different from cisplatin, supporting its use when cisplatin is contraindicated.
Retrospective outcomes with carboplatin-paclitaxel were not significantly different from cisplatin, supporting its use when cisplatin is contraindicated.
Mini-LATTICE created more compact high-dose vertices and greater peak-to-valley separation than conventional LATTICE across six planning cases.
Automated LAD contours reproduced the association between coronary dose and survival in 460 stage III NSCLC patients, although performance varied by model.
Eight of ten patients completed simulation-free single-fraction adaptive lung SBRT, with treatment taking a median 72 minutes.
Six-month symptom control was similar with advanced pneumatic compression and usual care, although some clinician-rated lymphedema measures favored compression.
A hybrid deep learning–registration workflow reduced median daily prostate MRgRT contouring time from 7.8 to 3.1 minutes while maintaining high contour agreement.
ETLD tracked 106,000 cine-MRI frames with submillimetre error, while integrated segmentation achieved a mean global Dice score of 87.7%.
Two blinded oncologists often assigned different peak toxicity grades, including grade 3 mucositis rates of 40.5% versus 17.6%.
Low CT tumour shrinkage strongly predicted poorer locoregional control, while high interim FDG-PET uptake predicted both poorer control and survival.
The rs7720298 risk allele increased LncDNAH5 expression, promoting TP53BP1 degradation and worsening radiation-induced bladder injury in mice.
Mean pelvic marrow dose above approximately 24 Gy and V40 above 13% identified higher hematologic toxicity risk during rectal chemoradiotherapy.
Automated pediatric CSI contours achieved a mean Dice score of 0.959 and reduced target-delineation time from approximately eight hours to 30 minutes.
DIR reduced estimated liver maximum dose from 123.8 to 83.2 Gy, but accuracy deteriorated markedly for bowel and stomach.
A hybrid peri-tumoural CT model achieved an external AUC of 0.828 for predicting grade 2 or higher radiation pneumonitis.
Reference-free LLM planning improved hotspot control and boost-target conformity while maintaining organ-at-risk sparing comparable to clinical plans.