PAINT: Pathology-Aware Integrated Next-Scale Transformation for Virtual Immunohistochemistry

TL;DR

PAINT introduces a structure-first autoregressive framework for virtual immunohistochemistry, improving molecular staining synthesis by grounding it in observed tissue morphology to enhance structural fidelity and clinical relevance.

Contribution

The paper proposes PAINT, a novel structure-aware autoregressive model that reformulates virtual IHC synthesis as a structure-guided generation task, outperforming existing methods.

Findings

Outperforms state-of-the-art methods in structural fidelity.

Enhances clinical downstream task performance.

Grounds synthesis in observed tissue morphology.

Abstract

Virtual immunohistochemistry (IHC) aims to computationally synthesize molecular staining patterns from routine Hematoxylin and Eosin (H\&E) images, offering a cost-effective and tissue-efficient alternative to traditional physical staining. However, this task is particularly challenging: H\&E morphology provides ambiguous cues about protein expression, and similar tissue structures may correspond to distinct molecular states. Most existing methods focus on direct appearance synthesis to implicitly achieve cross-modal generation, often resulting in semantic inconsistencies due to insufficient structural priors. In this paper, we propose Pathology-Aware Integrated Next-Scale Transformation (PAINT), a visual autoregressive framework that reformulates the synthesis process as a structure-first conditional generation task. Unlike direct image translation, PAINT enforces a causal order by resolving molecular details conditioned on a global structural layout. Central to this approach is the introduction of a Spatial Structural Start Map (3S-Map), which grounds the autoregressive initialization in observed morphology, ensuring deterministic, spatially aligned synthesis. Experiments on the IHC4BC and MIST datasets demonstrate that PAINT outperforms state-of-the-art methods in structural fidelity and clinical downstream tasks, validating the potential of structure-guided autoregressive modeling.