Whole-Slide Image Analysis under Realistic Few-Shot Annotation Protocols
2026-08-31 • Computer Vision and Pattern Recognition
Computer Vision and Pattern RecognitionArtificial Intelligence
AI summaryⓘ
The authors focus on improving how computers analyze detailed cancer images called whole-slide images, which contain many small sections or patches. Current methods look at each patch separately and struggle because real slides have uneven cancer types and complex layouts. They created SlideCRF, a new technique that considers the relationships between patches using biological and spatial information, even when some cancer types aren't present. They also designed new ways to mimic how pathologists provide feedback through limited clicks or markings. Their approach significantly improved accuracy compared to earlier methods across several datasets.
Whole-slide imagesVision-language modelsZero-shot predictionFew-shot transductionConditional random fieldsClass imbalancePatch-level analysisAnnotation protocolsSpatial relationshipsMacro F1 score
Authors
Tiffanie Godelaine, Maxime Zanella, Karim El Khoury, Benoit Macq, Christophe De Vleeschouwer
Abstract
Automating the analysis of whole-slide images has high clinical value, since characterizing cancers requires examining them in detail. Such analysis increasingly relies on vision-language models that provide patch-level zero-shot predictions. However, these predictions remain noisy and must be refined with a few annotations. A promising paradigm for this refinement is few-shot transduction. Rather than treating each patch independently, these methods leverage the relations between patches, together with a few annotations, to refine all predictions jointly. However, current transductive methods are evaluated under conditions that overlook key properties of whole-slide images: (i) datasets consist of independent patches extracted from multiple slides, ignoring the complex tissue organization; (ii) datasets are mostly balanced, whereas a single whole-slide image exhibits severe class imbalance, with several classes absent; and (iii) annotations are sampled at random, without reflecting how a pathologist annotates a limited number of regions. To align the transduction paradigm to realistic whole-slide settings, we introduce the following contributions. First, we propose SlideCRF, which adapts conditional random fields for whole-slide images by combining spatial and biological cues while accounting for classes that may be absent from a given slide. Second, we provide a set of realistic annotation protocols, based on spatially localized clicks and scribbles, modeling different pathologist interactions, such as the iterative correction of model errors. Across four datasets, we show that SlideCRF outperforms current transductive methods in macro F1, improving over the zero-shot predictions by +24.2% and +37.5% with one and 16 clicks per present class, respectively.