HAPS: Rethinking Image Similarity for Virtual Staining

TL;DR

This paper introduces HAPS, a domain-specific similarity metric for histopathology images, improving virtual staining quality assessment and training data filtering by aligning with expert evaluations.

Contribution

The work formalizes histology image similarity as a distinct problem, evaluates existing metrics, and proposes HAPS, a new perceptual similarity measure tailored for histopathology data.

Findings

HAPS correlates well with expert similarity scores.

Filtering training data with HAPS improves virtual staining model performance.

Existing metrics like SSIM and PSNR are inadequate for histological data.

Abstract

Virtual staining of histopathology images (e.g., H&E-IHC) is an emerging tool in digital pathology, enabling faster and cheaper workflows by synthesizing target stains from routinely acquired slides. Yet, the quality of virtual staining models is still predominantly assessed with generic metrics such as SSIM, PSNR, and LPIPS. Originally developed for natural images, these metrics are inherently misaligned with the domain-specific characteristics of histological data, failing to capture tissue morphology preservation and biomarker expression patterns. Consequently, a robust, domain-specific standard for quantifying similarity across diverse histological modalities remains a critical gap in the field. In this work, we formalize histology image similarity as a standalone problem and systematically evaluate a broad set of full-reference metrics against a dataset of H&E-IHC patch pairs annotated with expert similarity scores. We further analyze metrics sensitivity to controlled geometric distortions (shifts, rotations and non-rigid deformations) that mimic realistic registration errors between serial sections. Guided by these observations, we propose the Histology-Aware Perceptual Similarity (HAPS) metric. HAPS computes distances in the feature space of a frozen encoder pretrained on histopathology data, adding a linear head to aggregate feature-level differences into a final score that aligns with expert assessments. Finally, we demonstrate the practical value of HAPS for quality control of training data. By quantifying the similarity of training pairs in the MIST dataset and filtering low-scoring samples, we create a cleaner training set. Virtual staining models trained on this refined data outperform those trained on the original, unfiltered dataset.