Development and evaluation of machine learning models for voxel dose predictions in online adaptive magnetic resonance guided radiation therapy
Source check: Source checked. Not independent scientific confirmation.
Finding: Artificial neural networks predicted patient-specific three-dimensional dose distributions from target and organ-at-risk structures for online adaptive MR-guided radiotherapy. Cross-validation on plans from 53 patients showed agreement with selected dose metrics and flagged 25 plans as potentially inferior. The retrospective planning analysis did not test effects on treatment decisions or patient outcomes.
Technology: Artificial neural networks predict patient-specific three-dimensional dose distributions from anatomical structures
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- Artificial neural networks predict patient-specific three-dimensional dose distributions from anatomical structures
- Test and result
- Cross-validation on more than 300 abdominal-cancer treatment plans reported R² values of 0.97, 0.95 and 0.95 for V95, V100 and D95; prediction intervals identified about 10% (25) of plans as lower quality.
- Limitation / still unknown
- Retrospective plan analysis; clinical integration and improved patient outcomes were not established.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://pmc.ncbi.nlm.nih.gov/articles/PMC7386189/
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.
Review date: 22 Nov 2026. Review scope: Primary source full text, including Methods and Results; first result-bearing version checked against primary publication metadata..
No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.
Editorial assessment: Impact 2/5 · Validation 2/5 · Novelty 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.