Linking spatial biology and clinical histology via Haiku

arXiv cs.LG / 5/5/2026

📰 NewsSignals & Early TrendsIdeas & Deep AnalysisModels & Research

共有:

Key Points

The paper introduces “Haiku,” a tri-modal contrastive learning model designed to jointly model spatial proteomics from multiplexed immunofluorescence alongside matched H&E histology and clinical metadata in a shared embedding space.
Haiku is trained on a large, multi-patient dataset (26.7M spatial proteomics patches from 3,218 tissue sections across 1,606 patients, 11 organ types) and enables three-way cross-modal retrieval between molecular, histology, and clinical information.
In multiple downstream tasks, Haiku improves over unimodal baselines, including cross-modal retrieval (Recall@50 up to 0.611), survival prediction (C-index 0.737, +7.91% relative), and zero-shot biomarker inference (mean Pearson correlation 0.718 over 52 biomarkers).
The authors also propose a counterfactual prediction framework that changes only clinical metadata while holding tissue morphology fixed, surfacing niche-specific molecular shifts linked to breast cancer stage and lung cancer survival, presented explicitly as exploratory, hypothesis-generating signals.
Overall, the work argues that tri-modal alignment can bridge spatial biology and clinical context to support integrative analysis and data-driven biological discovery rather than purely mechanistic conclusions.

Abstract

Integrating molecular, morphological, and clinical data is essential for basic and translational biomedical research, yet systematic frameworks for jointly modeling these modalities remain limited. Here we present Haiku, a tri-modal contrastive learning model trained on multiplexed immunofluorescence (mIF). It comprises 26.7 million spatial proteomics patches from 3,218 tissue sections across 1,606 patients spanning 11 organ types, with matched hematoxylin and eosin (H&E) histology and clinical metadata aligned in a shared embedding space. Haiku enables three-way cross-modal retrieval, improves downstream classification and clinical prediction tasks over unimodal baselines, and supports zero-shot biomarker inference through fusion retrieval conditioned on clinical metadata-only text descriptions. Across tasks, Haiku outperforms competing approaches, achieving cross-modal retrieval (Recall@50 up to 0.611 versus near-zero baseline), survival prediction (C-index 0.737, +7.91% relative improvement), and zero-shot biomarker inference (mean Pearson correlation 0.718 across 52 biomarkers). Furthermore, we introduce a counterfactual prediction framework in which modifying only clinical metadata while fixing tissue morphology surfaces niche-specific molecular shifts associated with breast cancer stage progression and lung cancer survival outcomes. In a lung adenocarcinoma case study, the counterfactual analysis recovers niche-specific shifts characterized by increased CD8 and granzyme B, reduced PD-L1, and decreased Ki67, broadly consistent with patterns reported for favorable outcomes. We present these counterfactual results as exploratory, hypothesis-generating signals rather than mechanistic claims. These capabilities demonstrate that tri-modal alignment via Haiku enables integrative analysis of spatial biology, bridging molecular measurements with clinical context for biological exploration.