AdURA-Net: Adaptive Uncertainty and Region-Aware Network

2026-02-27

AI Summary

This work addresses diagnostic uncertainty in multi-label chest X-ray datasets arising from ambiguous radiology reports and automated label extraction. To this end, the authors propose a geometry-driven, adaptive uncertainty-aware framework that integrates anatomical structure modeling with uncertainty learning. The approach employs adaptive dilated convolutions and a multi-scale deformable alignment module, and introduces a dual-head loss function combining masked binary cross-entropy with Dirichlet evidence learning. Implemented on a DenseNet backbone, the method enables joint optimization of region-aware feature representation and uncertainty quantification. Experiments on benchmark datasets such as CheXpert and MIMIC-CXR demonstrate that the proposed framework significantly improves both classification reliability and model calibration, thereby enhancing trustworthiness in high-stakes clinical settings.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Uncertainty Representations
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs

Abstract

One of the common issues in clinical decision-making is the presence of uncertainty, which often arises due to ambiguity in radiology reports, which often reflect genuine diagnostic uncertainty or limitations of automated label extraction in various complex cases. Especially the case of multilabel datasets such as CheXpert, MIMIC-CXR, etc., which contain labels such as positive, negative, and uncertain. In clinical decision-making, the uncertain label plays a tricky role as the model should not be forced to provide a confident prediction in the absence of sufficient evidence. The ability of the model to say it does not understand whenever it is not confident is crucial, especially in the cases of clinical decision-making involving high risks. Here, we propose AdURA-Net, a geometry-driven adaptive uncertainty-aware framework for reliable thoracic disease classification. The key highlights of the proposed model are: a) Adaptive dilated convolution and multiscale deformable alignment coupled with the backbone Densenet architecture capturing the anatomical complexities of the medical images, and b) Dual Head Loss, which combines masked binary cross entropy with logit and a Dirichlet evidential learning objective.

Problem

• uncertainty
• clinical decision-making
• multilabel classification
• thoracic disease
• radiology reports

Innovation

• Adaptive Uncertainty
• Region-Aware Network
• Dual Head Loss
• Dirichlet Evidential Learning
• Deformable Alignment