WSI-INR: Implicit Neural Representations for Lesion Segmentation in Whole-Slide Images

TL;DR

WSI-INR introduces a patch-free implicit neural representation framework for whole-slide image lesion segmentation, preserving spatial continuity and robustness across resolutions, outperforming traditional patch-based methods.

Contribution

This work presents the first application of implicit neural representations to whole-slide image segmentation, enabling continuous, resolution-agnostic tissue modeling.

Findings

Improved Dice score by +26.11% at lower resolution levels.

Outperforms U-Net and TransUNet in robustness to resolution changes.

Enables segmentation of heterogeneous pathological lesions.

Abstract

Whole-slide images (WSIs) are fundamental for computational pathology, where accurate lesion segmentation is critical for clinical decision making. Existing methods partition WSIs into discrete patches, disrupting spatial continuity and treating multi-resolution views as independent samples, which leads to spatially fragmented segmentation and reduced robustness to resolution variations. To address the issues, we propose WSI-INR, a novel patch-free framework based on Implicit Neural Representations (INRs). WSI-INR models the WSI as a continuous implicit function mapping spatial coordinates directly to tissue semantics features, outputting segmentation results while preserving intrinsic spatial information across the entire slide. In the WSI-INR, we incorporate multi-resolution hash grid encoding to regard different resolution levels as varying sampling densities of the same continuous tissue, achieving a consistent feature representation across resolutions. In addition, by jointly training a shared INR decoder, WSI-INR can capture general priors across different cases. Experimental results showed that WSI-INR maintains robust segmentation performance across resolutions; at Base/4, our resolution-specific optimization improves Dice score by +26.11%, while U-Net and TransUNet decrease by 54.28% and 36.18%, respectively. Crucially, this work enables INRs to segment highly heterogeneous pathological lesions beyond structurally consistent anatomical tissues, offering a fresh perspective for pathological analysis.