Semi-Supervised Domain Adaptation with Latent Diffusion for Pathology Image Classification.

2026-01-23

arXiv.org

AI Summary

This study addresses the challenge of domain shift in computational pathology, where deep learning models often fail to generalize across institutions due to distributional discrepancies. Existing approaches struggle to effectively leverage unlabeled target-domain data or risk compromising histological structure integrity. To overcome these limitations, the authors propose a novel semi-supervised domain adaptation framework that integrates a conditional latent diffusion model with features from foundation models, along with slide-level metadata such as batch identity and staining information. This integration enables the generation of synthetic pathological images that preserve structural fidelity while aligning with target-domain characteristics, thereby enhancing classifier training. Evaluated on a lung adenocarcinoma prognosis task, the method improves the target-domain weighted F1 score from 0.611 to 0.706 and macro F1 score from 0.641 to 0.716, without degrading source-domain performance.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Computer Vision: Diffusion Models for Vision
• Application Domains: Humanities & Computational Social Science

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

Deep learning models in computational pathology often fail to generalize across cohorts and institutions due to domain shift. Existing approaches either fail to leverage unlabeled data from the target domain or rely on image-to-image translation, which can distort tissue structures and compromise model accuracy. In this work, we propose a semi-supervised domain adaptation (SSDA) framework that utilizes a latent diffusion model trained on unlabeled data from both the source and target domains to generate morphology-preserving and target-aware synthetic images. By conditioning the diffusion model on foundation model features, cohort identity, and tissue preparation method, we preserve tissue structure in the source domain while introducing target-domain appearance characteristics. The target-aware synthetic images, combined with real, labeled images from the source cohort, are subsequently used to train a downstream classifier, which is then tested on the target cohort. The effectiveness of the proposed SSDA framework is demonstrated on the task of lung adenocarcinoma prognostication. The proposed augmentation yielded substantially better performance on the held-out test set from the target cohort, without degrading source-cohort performance. The approach improved the weighted F1 score on the target-cohort held-out test set from 0.611 to 0.706 and the macro F1 score from 0.641 to 0.716. Our results demonstrate that target-aware diffusion-based synthetic data augmentation provides a promising and effective approach for improving domain generalization in computational pathology.

Problem

• domain shift
• computational pathology
• semi-supervised domain adaptation
• generalization
• unlabeled data

Innovation

• semi-supervised domain adaptation
• latent diffusion model
• computational pathology
• synthetic data augmentation
• domain generalization