3D-сегментация dental lesions при ограниченной разметке
3D-сегментация dental lesions при ограниченной разметке
Ответить самому
Сначала сформулируйте ответ как на собеседовании, затем откройте разбор и оцените себя.
Короткий ответ
Use tooth-aware cropping to control 3D memory and class imbalance, train a 3D segmentation/detection baseline, convert masks to instances with connected components or detection heads, and evaluate per-instance recall/precision with clinical slices.
Полный разбор
A strong design starts with constraints. Full CBCT volumes can be too large for direct 3D U-Net training, and lesions are small and sparse. If tooth masks are available, crop around each tooth plus context. This makes inputs smaller, normalizes anatomy and turns one huge rare-object problem into many tooth-local problems.
For a baseline, train a 3D U-Net-style semantic segmentation model per lesion class on high-quality voxel masks. Use class-balanced sampling, hard-negative mining, focal/Tversky/Dice-style losses, and augmentations that preserve clinical validity. Tooth-level weak labels can help with pretraining, auxiliary heads or sample mining, but should not replace voxel labels for final localization quality.
To produce instances and probabilities, threshold class probability maps, run connected components in 3D, filter tiny islands, and aggregate voxel scores per component. A detection-first alternative is a 3D detector followed by local segmentation. Evaluation should include instance-level sensitivity/precision at IoU or overlap thresholds, per-class and per-tooth slices, lesion-size slices, false-positive burden per scan and clinician review.
Теория
3D medical CV design is dominated by memory, sparsity, annotation quality and clinically meaningful evaluation.
Типичные ошибки
- Feed the full 1000^3 volume into a network without memory planning.
- Use only pixel IoU and miss instance-level false positives.
- Ignore tooth-local class imbalance.
Как отвечать на собеседовании
- Ask what labels exist and whether tooth masks are available.
- Offer a baseline first, then weak-label and instance-refinement improvements.