Out-of-distribution data supervision towards biomedical semantic segmentation

2025-07-16

AI Summary

Biomedical image segmentation often suffers from foreground-background misclassification due to limited and imperfectly annotated training data. To address this, we propose Med-OoD, the first framework to explicitly incorporate out-of-distribution (OoD) data into fully supervised medical image segmentation. Crucially, Med-OoD leverages only unlabeled OoD samples—containing no foreground pixels—for supervision, requiring no external datasets, additional annotations, feature-level regularization, or architectural modifications. This establishes a novel “OoD-aware fully supervised learning paradigm” that inherently mitigates misclassification. Evaluated on the Lizard dataset, Med-OoD significantly reduces pixel-level misclassification rates and achieves a mean Intersection-over-Union (mIoU) of 76.1%, demonstrating both effectiveness and strong generalization capability under standard fully supervised settings.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Semi-Supervised Learning
• Data Mining & Knowledge Management: Anomaly/Outlier Detection

Application Category

• Security and Privacy: Data transparency and provenance
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Biomedical segmentation networks easily suffer from the unexpected misclassification between foreground and background objects when learning on limited and imperfect medical datasets. Inspired by the strong power of Out-of-Distribution (OoD) data on other visual tasks, we propose a data-centric framework, Med-OoD to address this issue by introducing OoD data supervision into fully-supervised biomedical segmentation with none of the following needs: (i) external data sources, (ii) feature regularization objectives, (iii) additional annotations. Our method can be seamlessly integrated into segmentation networks without any modification on the architectures. Extensive experiments show that Med-OoD largely prevents various segmentation networks from the pixel misclassification on medical images and achieves considerable performance improvements on Lizard dataset. We also present an emerging learning paradigm of training a medical segmentation network completely using OoD data devoid of foreground class labels, surprisingly turning out 76.1% mIoU as test result. We hope this learning paradigm will attract people to rethink the roles of OoD data. Code is made available at https://github.com/StudioYG/Med-OoD.

Problem

• Addresses misclassification in biomedical segmentation networks
• Proposes OoD data supervision without external sources
• Enhances segmentation performance on limited medical datasets

Innovation

• Uses OoD data supervision without external sources
• Integrates into networks without architectural changes
• Trains segmentation using OoD data only