Superpixel-Guided Graph-Attention Boundary GAN for Adaptive Feature Refinement in Scribble-Supervised Medical Image Segmentation
Bibliographic record
Abstract
Fully supervised medical image segmentation still relies on labor-intensive, pixel-level annotations, which limits scale across cohorts and imaging settings. Scribble supervision reduces this burden, yet many CNN-based methods struggle under sparse labels due to weak global context and poor boundary handling. We address these issues with SGGAB-GAN, a scribble-supervised framework that uses adversarial learning, residual attention, and an enhanced feature pipeline built upon two modules: the Superpixel-Guided Graph-Attention Boundary (SGGAB) block and the Adaptive Feature Refinement Block (AFRB). First, the SGGAB block propagates limited scribble cues over a superpixel graph and reinjects boundary information, yielding crisp edges even with few annotations. Second, the AFRB fuses global context with local detail and works with residual attention gates to focus on anatomically relevant regions. On ACDC and MSCMRseg, SGGAB-GAN attains average Dice scores of 0.902 and 0.871, respectively, outperforming scribble-based methods such as ScribFormer and CycleMix while narrowing the gap to full supervision to under 2%. These results indicate that SGGAB-GAN provides high-quality segmentation at a fraction of the labeling cost, making it a scalable choice.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".