Planning computed tomography radiomics predicted recurrence in human papillomavirus-positive oropharyngeal cancer
A multi-regional clinical-radiomics model reached an external-test area under the curve of 0.908 for 5-year recurrence prediction.
Tag
technical-relevance
Magnetic resonance imaging radiomics model stratified survival risk after Gamma Knife for brain metastases
A multimodal imaging, radiomics, and clinical deep learning model showed moderate external validation for survival risk after Gamma Knife radiosurgery.
Magnetic resonance-only simulation supported prostate magnetic resonance-guided stereotactic body radiotherapy
Magnetic resonance-only prostate simulation achieved 96.80% local gamma agreement using synthetic computed tomography for dose calculation.
Dose-dependent linear energy transfer constraints supported proton toxicity interpretation
Dose-dependent linear energy transfer constraints aligned with published proton toxicity data better than fixed threshold approaches.
Photon-counting computed tomography estimated physical density with 5.5% phantom error
Photon-counting computed tomography generated effective atomic number and physical density images directly from virtual monoenergetic images.
Rectal air did not compromise prostate online adaptive radiotherapy robustness
Rectal air or air-filled balloons did not cause clinically relevant dose increases when posterior tissue shifts stayed within 5 millimetres.
Feature-wise scanning network improved liver radiotherapy segmentation efficiency
Feature-wise scanning Mamba improved liver radiotherapy segmentation accuracy while reducing model size and inference time.
Unshielded silicon diode detector supported clinical electron beam dosimetry
The SunSILICON detector directly measured electron beam depth doses, profiles, and output factors with close agreement to reference detectors.
Deep learning dose prediction met clinical constraints in esophageal radiotherapy planning
A conditional generative adversarial network predicted clinically acceptable esophageal radiotherapy dose distributions, but systematic target underdosage limited clinical use.
Cherenkov light tracked gold implant motion with submillimetre accuracy in water
Cherenkov light imaging tracked gold implant motion in water with submillimetre accuracy during static and volumetric modulated arc therapy delivery.
Upright computed tomography showed stable numbers for proton therapy planning
Upright computed tomography showed stable computed tomography numbers and greater than 99% proton dose gamma agreement versus conventional computed tomography.
Computed tomography geometric distortion was measurable across radiotherapy planning scanners
All eight radiotherapy planning computed tomography scanners showed measurable geometric distortion, with maximum values up to 1.88 millimetres.
Automated report-based chart checking detected residual errors after physics review
An automated report-based chart checker detected residual errors in 37.1% of manually reviewed external beam radiotherapy plans.
Optical surface imaging supported Monte Carlo dosimetry for total skin electron therapy
Photogrammetry-derived body surface models enabled Monte Carlo evaluation of total skin electron therapy dose homogeneity and hot-cold spot locations.
Volumetric modulated arc therapy complexity surveillance predicted patient-specific quality assurance results
Aperture irregularity and modulation complexity score were the strongest plan-complexity predictors of patient-specific quality assurance pass rate.