Conventional vs machine learning–based treatment planning in prostate brachytherapy: Results of a Phase I randomized controlled trial
Bibliographic record
Abstract
Purpose The purpose of this study was to evaluate the noninferiority of Day 30 dosimetry between a machine learning–based treatment planning system for prostate low-dose-rate (LDR) brachytherapy and the conventional, manual planning technique. As a secondary objective, the impact of planning technique on clinical workflow efficiency was also evaluated. Materials and Methods 41 consecutive patients who underwent I-125 LDR monotherapy for low- and intermediate-risk prostate cancer were accrued into this single-institution study between 2017 and 2018. Patients were 1:1 randomized to receive treatment planning using a machine learning–based prostate implant planning algorithm (PIPA system) or conventional, manual technique. Treatment plan modifications by the radiation oncologist were evaluated by computing the Dice coefficient of the prostate V 150% isodose volume between either the PIPA—or conventional—and final approved plans. Additional evaluations between groups evaluated the total planning time and dosimetric outcomes at preimplant and Day 30. Results 21 and 20 patients were treated using the PIPA and conventional techniques, respectively. No significant differences were observed in preimplant or Day 30 prostate D 90% , V 100% , rectum V 100 , or rectum D 1cc between PIPA and conventional techniques. Although the PIPA group had a larger proportion of patients with plans requiring no modifications (Dice = 1.00), there was no significant difference between the magnitude of modifications between each arm. There was a large significant advantage in mean planning time for the PIPA arm (2.38 ± 0.96 min) compared with the conventional (43.13 ± 58.70 min) technique ( p >> 0.05). Conclusions A machine learning–based planning workflow for prostate LDR brachytherapy has the potential to offer significant time savings and operational efficiencies, while producing noninferior postoperative dosimetry to that of expert, conventional treatment planners.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.003 | 0.002 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.004 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".