Towards Visually Explaining Statistical Tests with Applications in Biomedical Imaging

2026-01-20

AI Summary

This work addresses the limited interpretability of existing deep two-sample tests and the inapplicability of mainstream post-hoc explanation techniques—which rely on class labels—to unsupervised statistical testing scenarios. The authors propose the first deep framework that integrates explainable AI with label-free statistical inference, delivering simultaneous sample-level and feature-level explanations. Their method identifies both the critical image regions driving distributional differences between groups and the high-influence individual samples contributing to the test outcome. By unifying deep two-sample testing, unsupervised attribution mechanisms, and medical image analysis, the approach successfully pinpoints disease-relevant anatomical structures and key instances in biomedical data, substantially enhancing the transparency, practical utility, and clinical interpretability of statistical hypothesis testing.

Technology Category

• Computer Vision: Interpretability, Explainability, and Transparency
• Machine Learning: Calibration & Uncertainty Quantification
• Humans and AI: Explainable AI (XAI) for Human Understanding

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Security and Privacy: Data transparency and provenance
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Deep neural two-sample tests have recently shown strong power for detecting distributional differences between groups, yet their black-box nature limits interpretability and practical adoption in biomedical analysis. Moreover, most existing post-hoc explainability methods rely on class labels, making them unsuitable for label-free statistical testing settings. We propose an explainable deep statistical testing framework that augments deep two-sample tests with sample-level and feature-level explanations, revealing which individual samples and which input features drive statistically significant group differences. Our method highlights which image regions and which individual samples contribute most to the detected group difference, providing spatial and instance-wise insight into the test's decision. Applied to biomedical imaging data, the proposed framework identifies influential samples and highlights anatomically meaningful regions associated with disease-related variation. This work bridges statistical inference and explainable AI, enabling interpretable, label-free population analysis in medical imaging.

Problem

• explainable AI
• deep two-sample test
• statistical testing
• biomedical imaging
• label-free analysis

Innovation

• explainable AI
• deep two-sample test
• label-free statistical testing
• biomedical imaging
• instance-wise explanation