Pipeline Progress & Blockers
Status as of 2026-03-23
§1 Route A Pipeline Status
Table 1. Route A pipeline steps
| # | Step | Status | Output | Date |
|---|---|---|---|---|
| 1 | DICOM → NIfTI conversion | Done | 114 patient directories in ReMIND_nifti/ | 2026-03-16 |
| 2 | First-surgery patient filtering | Done | first_surgery_patients.json (62 patients) | 2026-03-16 |
| 3a | Registration V1 (baseline) | Done | registration_results_v1.csv (204 rows, 6.9% conv.) | 2026-03-16 |
| 3b | Registration V2 (+mask, 3k iter) | Done | registration_results_v2.csv (114 rows, 56.1% conv.) | 2026-03-17 |
| 3c | A/B test: optimizer configs | Done | ab_test_20260317_075447.csv (70 rows, 7 configs) | 2026-03-17 |
| 3d | Registration V3 (2 mm dil.) | Done | registration_results_v3.csv (114 rows, 73.7% conv.) | 2026-03-17 |
| 3e | Rescue passes | Done | 84/114 valid after QC exclusion (73.7%) | 2026-03-18 |
| 4 | Pseudo-label generation | Done | 84 labeled volumes → 44,502 2D slices (43 patients) | 2026-03-17 |
| 5 | nnU-Net data formatting | Done | Dataset001_iUS (43 patients) | 2026-03-18 |
| 6 | nnU-Net training | Blocked | — | — |
| 7 | Evaluation on RESECT-SEG | Pending | — | — |
§2 Route B Pipeline Status
Table 2. Route B pipeline steps
| # | Step | Status | Output | Date |
|---|---|---|---|---|
| 1 | Virtual sweep generation (v3) | Done | 1,090 sweeps across 110 cases | 2026-03-18 |
| 2 | Best-effort deep tumor fix | Done | 4 cases verified (v3_deep_fix/) | 2026-03-18 |
| 3 | 2D→3D NIfTI conversion | Done | 1,878 MRI + 1,090 seg volumes (33.7 GB) | 2026-03-18 |
| 4 | MMHVAE data normalisation | Pending | — | — |
| 5 | MMHVAE inference | Blocked | — | — |
| 6 | Synthetic iUS → nnU-Net training | Pending | — | — |
| 7 | Evaluation on RESECT-SEG | Pending | — | — |
§3 Current Blockers
Critical GPU / nnU-Net Environment Issue — Blocks Route A Step 6
Remote machine (x99-debian) has CUDA 12.2 / Driver 535.247.01 with PyTorch 2.5.1+cu124. GPU driver communication anomaly: nvidia-smi works but PyTorch reports RuntimeError: No CUDA GPUs are available. Root cause is a driver state issue requiring machine reboot.
Impact: Cannot start Route A nnU-Net training until GPU access is restored.
Mitigation: Reboot remote machine. After reboot, verify with python -c 'import torch; print(torch.cuda.is_available())', then start 5-fold cross-validation training (estimated 60–120 hours).
Blocked MMHVAE Pre-trained Weights Unavailable — Blocks Route B Step 5
The MMHVAE repository (github.com/ReubenDo/MMHVAE) documents pre-trained weights at pretrained/mmhvae_f0/models/CP_main_1000.pth but provides no download link. No GitHub Release exists.
Impact: Cannot run MMHVAE inference to synthesise iUS images. Options: (1) contact the author (R. Dorent) to request weights, (2) train from scratch (~72h GPU time on RTX 2080 Ti), or (3) try older MHVAE checkpoint if available.
Status: Email draft prepared (email_draft_dorent.md). Verify whether the email has actually been sent.
§4 Expected DSC Performance Roadmap
Fig. 1. DSC targets and literature references
Fig. 1: Expected Dice Similarity Coefficient (DSC) ranges for each route, based on published literature. Route A baseline (pseudo-label with registration noise): 0.58–0.62 (Faanes et al., 2025). Route B (MMHVAE synthesis): 0.74 (Dorent et al., 2025 TPAMI). Route A+B combined target: 0.84 (expert inter-rater agreement ceiling from ReMIND dataset). Supervised baseline on RESECT-SEG = 0.73 (Dorent et al. 2025).
Table 3. Literature DSC references
| Method | Dataset | DSC | Source |
|---|---|---|---|
| nnU-Net + noisy pseudo-labels (registration) | ReMIND (55 patients) | 0.58–0.62 | Faanes et al. 2025 |
| MMHVAE synthesis → SegResNet | RESECT-SEG | 0.74 | Dorent et al. 2025 (TPAMI) |
| Supervised baseline (w/ manual iUS labels) | RESECT-SEG | 0.73 | Dorent et al. 2025 (TPAMI) |
| Expert inter-rater agreement | ReMIND | 0.84 | Juvekar & Dorent et al. 2024 |