Supervise Less, See More: Training-free Nuclear Instance Segmentation with Prototype-Guided Prompting

2025-11-25

AI Summary

In computational pathology, nucleus instance segmentation traditionally relies on labor-intensive dense annotations and costly fine-tuning, motivating the exploration of training-free, annotation-free approaches. This paper introduces SPROUT—the first fully training-free and annotation-free framework for nucleus instance segmentation. Its core innovation lies in constructing slice-specific prototypes grounded in histological priors, aligning cross-domain features via partial optimal transport, and generating positive/negative point prompts to guide the Segment Anything Model (SAM) for precise segmentation. SPROUT establishes a prototype-guided zero-shot prompting mechanism, pioneering a scalable, unsupervised paradigm for pathological segmentation. Evaluated across multiple benchmarks, SPROUT achieves performance on par with state-of-the-art unsupervised methods—without any fine-tuning—thereby substantially lowering deployment barriers and enhancing clinical applicability.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Calibration & Uncertainty Quantification
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Security and Privacy: Data transparency and provenance
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Accurate nuclear instance segmentation is a pivotal task in computational pathology, supporting data-driven clinical insights and facilitating downstream translational applications. While large vision foundation models have shown promise for zero-shot biomedical segmentation, most existing approaches still depend on dense supervision and computationally expensive fine-tuning. Consequently, training-free methods present a compelling research direction, yet remain largely unexplored. In this work, we introduce SPROUT, a fully training- and annotation-free prompting framework for nuclear instance segmentation. SPROUT leverages histology-informed priors to construct slide-specific reference prototypes that mitigate domain gaps. These prototypes progressively guide feature alignment through a partial optimal transport scheme. The resulting foreground and background features are transformed into positive and negative point prompts, enabling the Segment Anything Model (SAM) to produce precise nuclear delineations without any parameter updates. Extensive experiments across multiple histopathology benchmarks demonstrate that SPROUT achieves competitive performance without supervision or retraining, establishing a novel paradigm for scalable, training-free nuclear instance segmentation in pathology.

Problem

• Developing training-free nuclear instance segmentation without dense supervision
• Creating annotation-free framework using prototype-guided prompting for pathology
• Enabling precise nuclear delineations without parameter updates or retraining

Innovation

• Training-free prompting framework for nuclear segmentation
• Prototype-guided feature alignment via optimal transport
• Transforms features into prompts for SAM model