GazeRefine is a training‑free framework that uses eye‑gaze data as an inference‑time prompt for zero‑shot medical image segmentation. It converts sparse, duration‑weighted fixations into foreground and background priors that initialize semantic prototypes in a frozen DINOv3 feature space, then iteratively refines these prototypes through discrimination, affinity propagation, and anchoring to the gaze guidance. The method achieves strong results on colonoscopy polyp segmentation and competitive performance on prostate MRI, demonstrating that gaze‑guided prototype refinement can enable segmentation without dense expert annotations or model fine‑tuning.
arXiv:2607.12896v3 Announce Type: replace Abstract: Medical image segmentation foundation models are expected to generalize across diverse clinical scenarios, yet existing universal methods remain fr...
Pixel-level annotation remains a major bottleneck in medical image segmentation, making weak supervision an attractive yet under-constrained alternative. We propose OBBSeg, an intermediate supervision paradigm guided by Oriented Bounding Boxes (OBBs) that bridges the gap between full and weak supervision.
FoRIS is a training‑free in‑context segmentation framework that refines foreground masks through a coarse‑to‑fine process. It operates in three stages—Foreground Purification, Localization, and Consolidation—to suppress background noise, pinpoint target regions, and reconstruct complete foreground structures. The method achieves state‑of‑the‑art performance, improving mIoU by 4.5 and 4.8 points in 1‑shot and 5‑shot settings respectively.
arXiv:2606. 04705v1 Announce Type: cross Abstract: Semantic segmentation in medical imaging is a critical yet challenging task due to data scarcity and high variability across modalities.
PR-MaGIC is a training‑free, test‑time framework that refines prompts for in‑context image segmentation by using gradient flow from the mask decoder of the Segment Anything Model (SAM). It integrates seamlessly into existing in‑context segmentation pipelines and employs a simple top‑1 selection strategy to stabilize performance. Extensive evaluations show that PR‑MaGIC consistently improves segmentation quality across multiple benchmarks without requiring additional training or architectural changes.
arXiv:2607.10851v2 Announce Type: replace Abstract: Medical image classification models are ideally expected to identify diagnostically relevant regions while making predictions, yet standard classif...
SAM-DCE is a framework that adapts the Segment Anything Model (SAM) for fully automatic, prompt‑free medical image segmentation. It introduces Spatial Representation Compensation (SRC) to recover fine anatomical details from intermediate Vision Transformer features and Semantic Token Calibration (STC) to enhance the consistency of foreground mask tokens. By refining spatial and semantic representations before using SAM’s original token‑to‑mask mechanism, SAM‑DCE achieves superior performance on the Synapse multi‑organ CT segmentation dataset, with a mean Dice score of 83.86% and an HD95 of 16.04 mm, outperforming several CNN, Transformer, and SAM baselines.
arXiv:2608. 09818v1 Announce Type: cross Abstract: Reliable medical image understanding requires models to connect clinical language and visual reasoning with pixel-level grounding.
arXiv:2611.05874v1 Announce Type: new Abstract: Humans can easily recognize the same objects across different viewpoints of a 3D scene, whereas teaching models such multi-view-consistent perception t...
arXiv:2610.03474v1 Announce Type: new Abstract: Federated learning (FL) enables collaborative training of medical image segmentation models without sharing raw patient data, yet existing approaches a...
arXiv:2610.06938v1 Announce Type: new Abstract: Medical image segmentation remains fragmented along two axes: segmentation paradigms and data dimensionality. Existing methods are typically developed...
arXiv:2510. 15042v3 Announce Type: replace-cross Abstract: In the 3D medical image domain, vision-language pre-training is used to create vision-language encoders (VLEs) that can support radiologists by retrieving patients with similar abnormalities, predicting likelihoods of abnormality, or, with downstream adaptation, generating radiological reports.