HiLa: Hierarchical Vision-Language Collaboration for Cancer Survival Prediction

TL;DR

The paper introduces HiLa, a hierarchical vision-language framework that enhances cancer survival prediction from whole-slide images by leveraging multi-level features and improved alignment with linguistic descriptions.

Contribution

It proposes a novel hierarchical vision-language collaboration framework with optimal prompt learning, cross-level propagation, and mutual contrastive learning for better survival prediction from WSIs.

Findings

Achieves state-of-the-art performance on three TCGA datasets.

Effectively models hierarchical interactions in WSIs.

Improves vision-language alignment for survival prediction.

Abstract

Survival prediction using whole-slide images (WSIs) is crucial in cancer re-search. Despite notable success, existing approaches are limited by their reliance on sparse slide-level labels, which hinders the learning of discriminative repre-sentations from gigapixel WSIs. Recently, vision language (VL) models, which incorporate additional language supervision, have emerged as a promising solu-tion. However, VL-based survival prediction remains largely unexplored due to two key challenges. First, current methods often rely on only one simple lan-guage prompt and basic cosine similarity, which fails to learn fine-grained associ-ations between multi-faceted linguistic information and visual features within WSI, resulting in inadequate vision-language alignment. Second, these methods primarily exploit patch-level information, overlooking the intrinsic hierarchy of WSIs and their interactions, causing ineffective modeling of hierarchical interac-tions. To tackle these problems, we propose a novel Hierarchical vision-Language collaboration (HiLa) framework for improved survival prediction. Specifically, HiLa employs pretrained feature extractors to generate hierarchical visual features from WSIs at both patch and region levels. At each level, a series of language prompts describing various survival-related attributes are constructed and aligned with visual features via Optimal Prompt Learning (OPL). This ap-proach enables the comprehensive learning of discriminative visual features cor-responding to different survival-related attributes from prompts, thereby improv-ing vision-language alignment. Furthermore, we introduce two modules, i.e., Cross-Level Propagation (CLP) and Mutual Contrastive Learning (MCL) to maximize hierarchical cooperation by promoting interactions and consistency be-tween patch and region levels. Experiments on three TCGA datasets demonstrate our SOTA performance.