# Adaptive Post-Processing Drives Instance-Level Detection in Stroke Lesion Segmentation

arXiv: https://arxiv.org/abs/2608.16377 · HTML: https://mlsanity.com/paper/2608.16377/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 6 of its 11 references that are in the map.
- First posted on arXiv 17 Aug 2026.

## Builds on (most cited first, 6 of 6)

- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Attention U-Net: Learning Where to Look for the Pancreas](https://mlsanity.com/paper/1804.03999/) (2018)
- [ISLES 2022: A multi-center magnetic resonance imaging stroke lesion segmentation dataset](https://mlsanity.com/paper/2206.06694/) (2022)
- [Panoptica -- instance-wise evaluation of 3D semantic and instance segmentation maps](https://mlsanity.com/paper/2312.02608/) (2023)
- [Voxels Intersecting along Orthogonal Levels Attention U-Net for Intracerebral Haemorrhage Segmentation in Head CT](https://mlsanity.com/paper/2208.06313/) (2022)
- [Examining Deployment and Refinement of the VIOLA-AI Intracranial Hemorrhage Model Using an Interactive NeoMedSys Platform](https://mlsanity.com/paper/2505.09380/) (2025)

## Abstract (arXiv)

Instance-level lesion detection has been an increasingly larger focal point in medical image segmentation besides the more standard voxel-level overlap. Still, most pipelines are trained and post-processed for voxel overlap alone. In particular, the mismatch is most pronounced for small lesions, where a near-miss prediction---substantial overlap that falls just short of the instance-matching threshold---scores the same as a complete miss. In our ISLES'26 submission, we found that closing this gap mattered far more in post-processing than in architecture design. Our Volume-Conditioned Adaptive Post-Processing (VCAP) scheme adjusts component-size thresholds to each case's predicted lesion burden, improving Lesion-F1 by 0.032 (unbiased cross-fold estimate)---approximately 6 times larger than any architectural change we tested. A resolution-aware attention architecture (Viola2Plus), designed for small-lesion segmentation, shows why the distinction matters: it left small-lesion Dice unchanged but raised small-lesion detection rate by 3.7%, a real effect voxel-overlap metrics alone would have missed. Under 5-fold cross-validation on the 1,453-case training set, our post-processed two-architecture ensemble achieves Dice 0.651 and Lesion-F1 0.614, versus 0.644 and 0.573 for the unprocessed single-model baseline.