# Beyond Boundary Noise: Aggregated Aleatoric Uncertainty Fails to Capture Presence Ambiguity in 3D Lung Nodule Segmentation

arXiv: https://arxiv.org/abs/2608.14766 · HTML: https://mlsanity.com/paper/2608.14766/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 16 of its 28 references that are in the map.
- First posted on arXiv 14 Aug 2026.

## Builds on (most cited first, 16 of 16)

- [Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning](https://mlsanity.com/paper/1506.02142/) (2015)
- [Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles](https://mlsanity.com/paper/1612.01474/) (2016)
- [UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction](https://mlsanity.com/paper/1802.03426/) (2018)
- [Aleatoric and Epistemic Uncertainty in Machine Learning: An Introduction to Concepts and Methods](https://mlsanity.com/paper/1910.09457/) (2019)
- [A Probabilistic U-Net for Segmentation of Ambiguous Images](https://mlsanity.com/paper/1806.05034/) (2018)
- [Learning From Noisy Labels By Regularized Estimation Of Annotator Confusion](https://mlsanity.com/paper/1902.03680/) (2019)
- [Quantifying Aleatoric and Epistemic Uncertainty in Machine Learning: Are Conditional Entropy and Mutual Information Appropriate Measures?](https://mlsanity.com/paper/2209.03302/) (2022)
- [Benchmarking Uncertainty Disentanglement: Specialized Uncertainties for Specialized Tasks](https://mlsanity.com/paper/2402.19460/) (2024)
- [Disentangling Human Error from the Ground Truth in Segmentation of Medical Images](https://mlsanity.com/paper/2007.15963/) (2020)
- [Supervised Uncertainty Quantification for Segmentation with Multiple Annotations](https://mlsanity.com/paper/1907.01949/) (2019)
- [A review of uncertainty quantification in medical image analysis: probabilistic and non-probabilistic methods](https://mlsanity.com/paper/2310.06873/) (2023)
- [Spatially Varying Label Smoothing: Capturing Uncertainty from Expert Annotations](https://mlsanity.com/paper/2104.05788/) (2021)
- [LNDb: A Lung Nodule Database on Computed Tomography](https://mlsanity.com/paper/1911.08434/) (2019)
- [Uncertainty Quantification for Machine Learning in Healthcare: A Survey](https://mlsanity.com/paper/2505.02874/) (2025)
- [Confidence Contours: Uncertainty-Aware Annotation for Medical Semantic Segmentation](https://mlsanity.com/paper/2308.07528/) (2023)
- [Benchmarking Uncertainty and its Disentanglement in multi-label Chest X-Ray Classification](https://mlsanity.com/paper/2508.04457/) (2025)

## Built on by (most cited first, 1 of 1)

- [Uncertainty-Aware End-to-End AI Weather Forecasting: Disentangling Observation and Model Contributions](https://mlsanity.com/paper/2608.30795/) (2026)

## Abstract (arXiv)

Uncertainty estimation is critical for the safe clinical deployment of deep learning in medical image segmentation, with aleatoric uncertainty theoretically designed to capture irreducible data ambiguity. However, whether entropy-based measures reflect clinically meaningful ambiguity, i.e. case-level disagreement about whether a pathology is present at all, remains poorly understood. Contrary to most prior work, which focused on pixel-wise boundary disagreement, we systematically evaluate how well aleatoric uncertainty captures presence ambiguity. Our evaluation spans 3D lung nodule segmentation across four architectures with Monte Carlo dropout and deep ensembles, on LIDC-IDRI and an external validation cohort (LNDb). We find that entropy-based uncertainty maps align with boundary noise and minor drawing variation but carry insufficient discriminative signal for presence ambiguity. In contrast, a lightweight supervised ambiguity head trained on frozen segmentation features substantially outperforms all entropy-aggregation-based baselines across architectures, metrics, and both cohorts, and matches or exceeds methods that explicitly model ambiguity under disagreement supervision (Probabilistic U-Net, Annotator-Confusion 3D-UNet). A qualitative feature-space analysis shows that presence ambiguity is already encoded in the frozen encoder features of pixel-wise-trained networks, only to be discarded by the segmentation output and its entropy aggregation. Our findings expose a fundamental mismatch between the theoretical promise of aleatoric uncertainty and its practical behavior, and suggest that practitioners should not rely on entropy-based uncertainty as a proxy for clinical ambiguity in safety-critical applications.