Research map / Architectures and efficiency / RNNs and normalization
D2T2 - Multimodal Automated Planning for Brachytherapy
Lance C. Moore, Aranyo Mitra, Ryan Truong, Karoline Kallis, Kelly Kisling, Sandra M. Meyers, Nuno Vasconcelos
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace post-hoc dose optimization with a physics layer predicting dwell times end-to-end (D2T2) | brachytherapy dose prediction · Brachytherapy test set (overall) | Dose MAE (%): 4.41 -> 3.54 | Improved |
| replace post-hoc dose optimization with a physics layer predicting dwell times end-to-end (D2T2) | brachytherapy dose prediction · Brachytherapy test set (overall) | Dwell time MAE (s): 11.79 -> 7.41 | Improved |
| replace post-hoc dose optimization with a physics layer predicting dwell times end-to-end (D2T2) | brachytherapy dose prediction · Brachytherapy test set (overall) | Mean time (s): 35.07 -> 0.10 | Improved |
| use D2T2 instead of OPT on low-data Other sites | D2T2 · Other-site plans | Dose MAE (%): 5.09 -> 3.75 | Improved |
| use D2T2 instead of OPT on vaginal plans | D2T2 · Vaginal plans | Dose MAE (%): 6.55 -> 6.51 | No clear effect |
| train with gamma loss instead of dose L2 loss (ablation) | D2T2 · Brachytherapy test set (loss ablation) | Dose MAE (%): 3.68 -> 3.64 | Improved |
| add dwell time L2 loss to the dose loss (ablation) | D2T2 · Brachytherapy test set (loss ablation) | Dose MAE (%): 3.68 -> 4.67 | Hurt |
| add dwell time L2 loss to the dose loss (ablation) | D2T2 · Brachytherapy test set (loss ablation) | Dwell time MAE (s): 7.41 -> 7.49 | No clear effect |
| train on dwell time L2 loss alone (ablation) | D2T2 · Brachytherapy test set (loss ablation) | Dose MAE (%): 3.68 -> 4.34 | Hurt |
| add gamma loss to dose loss without weighting (ablation) | D2T2 · Brachytherapy test set (loss ablation) | Dwell time MAE (s): 7.41 -> 7.85 | Hurt |
| weight gamma loss with lambda 0.05 in the combined loss (ablation) | D2T2 · Brachytherapy test set (lambda ablation) | Dose MAE (%): 3.69 -> 3.54 | Improved |
| weight gamma loss with lambda 0.05 in the combined loss (ablation) | D2T2 · Brachytherapy test set (lambda ablation) | Dwell time MAE (s): 7.85 -> 7.41 | Improved |
Closest papers by meaning
- Randomized Sparse Matrix Compression for Large-Scale Constrained Optimization in Cancer Radiotherapy (NeurIPS 2024 · similarity 0.81)
- Patient-Level Anatomy Meets Scanning-Level Physics: Personalized Federated Low-Dose CT Denoising Empowered by Large Language Model (CVPR 2025 · similarity 0.80)
- E2ENet: Dynamic Sparse Feature Fusion for Accurate and Efficient 3D Medical Image Segmentation (NeurIPS 2024 · similarity 0.80)
- Addressing Asynchronicity in Clinical Multimodal Fusion via Individualized Chest X-ray Generation (NeurIPS 2024 · similarity 0.80)
- Multi-Agent Reinforcement Learning Meets Leaf Sequencing in Radiotherapy (ICML 2024 · similarity 0.80)
- HEALNet: Multimodal Fusion for Heterogeneous Biomedical Data (NeurIPS 2024 · similarity 0.80)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
