# Three-Phase Scribble-Adaptive Curriculum Learning for autoPETV Grand Challenge

arXiv: https://arxiv.org/abs/2608.22096 · HTML: https://mlsanity.com/paper/2608.22096/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 7 of its 14 references that are in the map.
- First posted on arXiv 22 Aug 2026.

## Builds on (most cited first, 7 of 7)

- [nnU-Net Revisited: A Call for Rigorous Validation in 3D Medical Image Segmentation](https://mlsanity.com/paper/2404.09556/) (2024)
- [Reviving Iterative Training with Mask Guidance for Interactive Segmentation](https://mlsanity.com/paper/2102.06583/) (2021)
- [DeepIGeoS: A Deep Interactive Geodesic Framework for Medical Image Segmentation](https://mlsanity.com/paper/1707.00652/) (2017)
- [Deep Interactive Segmentation of Medical Images: A Systematic Review and Taxonomy](https://mlsanity.com/paper/2311.13964/) (2023)
- [Panoptica -- instance-wise evaluation of 3D semantic and instance segmentation maps](https://mlsanity.com/paper/2312.02608/) (2023)
- [Sliding Window FastEdit: A Framework for Lesion Annotation in Whole-body PET Images](https://mlsanity.com/paper/2311.14482/) (2023)
- [The autoPET3 Challenge: Automated Lesion Segmentation in Whole-Body PET/CT unicodex2013 Multitracer Multicenter Generalization](https://mlsanity.com/paper/2605.05775/) (2026)

## Abstract (arXiv)

This report describes Libo Zhang's algorithmic solution to autoPETV Grand Challenge on interactive lesion segmentation in whole-body PET/CT. Interaction is encoded as two additional input channels that rasterize the accumulated foreground and background scribbles, and a residual-encoder U-Net of about 140 million parameters is trained with a three-phase curriculum over 4000 epochs: the network first learns fully automatic segmentation with silent interaction channels, then observes ground-truth-derived scribbles under randomly sampled visibility modes, and finally adapts to its own mistakes through online simulation of up to five error-driven correction steps. Training draws on 1811 autoPET and DeepPSMA studies, and the submission ensembles the best and final checkpoints of five folds by logit averaging. In interactive five-fold cross-validation with six interaction steps, the final checkpoints reach a mean AUC-Dice of 3.836 and a mean AUC-DMM of 3.869, improving monotonically in every fold, with roughly half of the total gain delivered by the first corrective scribble. Our code and trained model checkpoints are available on https://github.com/Libo1023/autoPETV-Curriculum.