Closing the Domain Gap in Biomedical Imaging by In-Context Control Samples

2026-04-22

AI Summary

This work addresses the degradation of cross-batch generalization in biomedical imaging caused by batch effects. To mitigate this issue, the authors propose Control-Stabilized Adaptive Risk Minimization via Batch Normalization (CS-ARM-BN), a novel approach that integrates negative control samples with meta-learning. By leveraging unperturbed control images inherently present in each experimental batch as stable contextual anchors, CS-ARM-BN enables context-aware in-domain adaptation through batch normalization, effectively removing technical systematic biases. Evaluated on the JUMP-CP dataset for mechanism-of-action (MoA) classification, the method improves accuracy on unseen batches from 0.862 ± 0.060 to 0.935 ± 0.018—nearly matching in-domain performance (0.939 ± 0.005)—and substantially outperforms existing techniques, thereby overcoming a critical bottleneck in batch-effect correction under strong domain shifts.

Technology Category

• Machine Learning: Adversarial Learning & Robustness
• Computer Vision: Adversarial Attacks & Robustness
• Multiagent Systems: Adversarial Agents

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

The central problem in biomedical imaging are batch effects: systematic technical variations unrelated to the biological signal of interest. These batch effects critically undermine experimental reproducibility and are the primary cause of failure of deep learning systems on new experimental batches, preventing their practical use in the real world. Despite years of research, no method has succeeded in closing this performance gap for deep learning models. We propose Control-Stabilized Adaptive Risk Minimization via Batch Normalization (CS-ARM-BN), a meta-learning adaptation method that exploits negative control samples. Such unperturbed reference images are present in every experimental batch by design and serve as stable context for adaptation. We validate our novel method on Mechanism-of-Action (MoA) classification, a crucial task for drug discovery, on the large-scale JUMP-CP dataset. The accuracy of standard ResNets drops from 0.939 $\pm$ 0.005, on the training domain, to 0.862 $\pm$ 0.060 on data from new experimental batches. Foundation models, even after Typical Variation Normalization, fail to close this gap. We are the first to show that meta-learning approaches close the domain gap by achieving 0.935 $\pm$ 0.018. If the new experimental batches exhibit strong domain shifts, such as being generated in a different lab, meta-learning approaches can be stabilized with control samples, which are always available in biomedical experiments. Our work shows that batch effects in bioimaging data can be effectively neutralized through principled in-context adaptation, which also makes them practically usable and efficient.

Problem

• batch effects
• domain gap
• biomedical imaging
• deep learning
• experimental reproducibility

Innovation

• batch effects
• meta-learning
• control samples
• domain adaptation
• biomedical imaging