Exact Dice Degradation Envelopes for Label-Free Segmentation Update Selection
eess.IV, cs.LG
Submitted: 2026-10-01
Updated: 2026-10-01
Comments: 18 pages, 7 figures, 6 tables. Includes mathematical proofs and a three-seed study on ClinicDB, Massachusetts Buildings, and Kvasir-SEG
License: http://creativecommons.org/licenses/by/4.0/
The gist: Test-time segmentation updates can improve an unlabeled input while damaging another.
Terminology
Abstract
Test-time segmentation updates can improve an unlabeled input while damaging another. Confidence and transformation consistency can rank candidates, but do not in themselves bound the loss caused by returning a changed mask. We derive the exact largest decrease in binary Dice between a baseline and a candidate, maximized over every possible unknown reference mask on the same pixels. A maximizing reference always lies inside baseline foreground: three observable mask counts reduce the problem to at most four reference-cardinality checks. Accepting an output only when its envelope is below a budget gives a deterministic, per-case Dice degradation limit relative to the baseline. A finite grid of logit interpolations permits partial candidate outputs without extra network forwards. Exhaustive and randomized validation checks 5,026,592 reference masks with no discrepancy. We evaluate six independently trained MK-UNet sources, three seeds per task, on ClinicDB, Massachusetts Buildings and the external Kvasir-SEG cohort, including appearance corruptions. Across 17,808 case-condition-seed evaluations, 0 certificate violations occur in the checked candidate comparisons. At a one-point Dice budget, exact versus loose interpolation changes the returned mask in the following fractions of episodes (ClinicDB: 51.0% versus 38.7%; Buildings: 62.1% versus 53.1%; Kvasir-SEG: 52.3% versus 34.9%). Mean Dice effects and useful-update rejection are reported separately from admissibility. An exploratory oracle-presence-bit ablation examines how weak supervision relaxes the empty-baseline restriction. The envelope quantifies relative mask-change risk; it does not certify anatomical correctness, preserve other metrics, or guarantee adaptation benefit.
Sources
- Tent: Fully Test-time Adaptation by Entropy Minimization
- Efficient Test-Time Model Adaptation without Forgetting
- Towards Stable Test-Time Adaptation in Dynamic Wild World
- Continual Test-Time Domain Adaptation
- HD-TTA: Hypothesis-Driven Test-Time Adaptation for Safer Brain Tumor Segmentation
- MK-UNet: Multi-kernel Lightweight CNN for Medical Image Segmentation
- Kvasir-SEG: A Segmented Polyp Dataset
- A note on the triangle inequality for the Jaccard distance
Related papers
- HDNet: High-resolution Dual-domain Learning for Spectral Compressive Imaging
- Mask-guided Spectral-wise Transformer for Efficient Hyperspectral Image Reconstruction
- FDIR: Fidelity-Anchored Decompressed Image Restoration for Human and Machine Perception
- A Disease-Centric Vision-Language Foundation Model for Precision Oncology in Kidney Cancer
- CADS: A Comprehensive Anatomical Dataset and Segmentation for Whole-Body Anatomy in Computed Tomography
- Task-Aware Bitrate Thresholds for Codec-Robust Ultrasound Segmentation