# Uncertainty as a Proxy for Semantic Correctness in Diffusion-Based Medical Image Synthesis

arXiv: https://arxiv.org/abs/2610.03224 · HTML: https://mlsanity.com/paper/2610.03224/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Medical imaging and segmentation, one of 14 directions in Diffusion models.
- Builds on 21 of its 41 references that are in the map.
- First posted on arXiv 2 Oct 2026.

## Placed in

- Diffusion models / [Medical imaging and segmentation](https://mlsanity.com/diffusion/medical-imaging-and-segmentation/)

## Builds on (most cited first, 21 of 21)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [Denoising Diffusion Implicit Models](https://mlsanity.com/paper/2010.02502/) (2020)
- [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)
- [UNet++: A Nested U-Net Architecture for Medical Image Segmentation](https://mlsanity.com/paper/1807.10165/) (2018)
- [Attention U-Net: Learning Where to Look for the Pancreas](https://mlsanity.com/paper/1804.03999/) (2018)
- [A Review of Uncertainty Quantification in Deep Learning: Techniques, Applications and Challenges](https://mlsanity.com/paper/2011.06225/) (2020)
- [TotalSegmentator: robust segmentation of 104 anatomical structures in CT images](https://mlsanity.com/paper/2208.05868/) (2022)
- [Out-of-Distribution Detection with Deep Nearest Neighbors](https://mlsanity.com/paper/2204.06507/) (2022)
- [Diffusion Models for Medical Image Analysis: A Comprehensive Survey](https://mlsanity.com/paper/2211.07804/) (2022)
- [Being Bayesian, Even Just a Bit, Fixes Overconfidence in ReLU Networks](https://mlsanity.com/paper/2002.10118/) (2020)
- [Bayesian Hypernetworks](https://mlsanity.com/paper/1710.04759/) (2017)
- [Estimating Epistemic and Aleatoric Uncertainty with a Single Model](https://mlsanity.com/paper/2402.03478/) (2024)
- [BayesDiff: Estimating Pixel-wise Uncertainty in Diffusion via Bayesian Inference](https://mlsanity.com/paper/2310.11142/) (2023)
- [Aligning Synthetic Medical Images with Clinical Knowledge using Human Feedback](https://mlsanity.com/paper/2306.12438/) (2023)
- [Conformal Risk Control for Semantic Uncertainty Quantification in Computed Tomography](https://mlsanity.com/paper/2503.00136/) (2025)
- [Conformalized Generative Bayesian Imaging: An Uncertainty Quantification Framework for Computational Imaging](https://mlsanity.com/paper/2504.07696/) (2025)
- [Trustworthy image-to-image translation: evaluating uncertainty calibration in unpaired training scenarios](https://mlsanity.com/paper/2501.17570/) (2025)
- [Generating Synthetic Contrast-Enhanced Chest CT Images from Non-Contrast Scans Using Slice-Consistent Brownian Bridge Diffusion Network](https://mlsanity.com/paper/2508.16897/) (2025)
- [AortaDiff: A Unified Multitask Diffusion Framework For Contrast-Free AAA Imaging](https://mlsanity.com/paper/2510.01498/) (2025)
- [Evaluating and Calibrating Diffusion Model-derived Uncertainty for Quantitative MRI Mapping](https://mlsanity.com/paper/2608.11942/) (2026)

## Abstract (arXiv)

Diffusion models can synthesise contrast-enhanced CT (CECT) from non-contrast CT (NCCT), avoiding contrast administration and its environmental and patient-access costs. However, visually realistic images are not necessarily anatomically correct, and the pixel-intensity and feature-space similarity metrics used to assess generation quality do not directly measure anatomical correctness. In this work, we investigate whether uncertainty can serve as a proxy for semantic correctness in diffusion-based medical image synthesis. We study NCCT-to-CECT synthesis using AortaDiff, a multitask diffusion framework that jointly generates CECT images and lumen segmentations. The segmentation output provides an explicit representation of the generated vascular anatomy, enabling segmentation-derived errors to be used as a quantitative measure of generation correctness. Six methods spanning weight (Ensemble, HyperDiff, BayesDiff), architecture-perturbation (MCDropout), generative-stochasticity (RDS) and input-perturbation (TTA) uncertainty are compared at the pixel, region and image levels, and for detection of clinically relevant out-of-distribution (OOD) cases. Uncertainty proves informative at all three spatial scales, remains informative on an external multi-centre dataset under distribution shift, and supports OOD detection. MCDropout stands out among the six: it ranks among the leading methods at every scale, generalizes well on the external dataset, and can be enabled at inference on any model already trained with dropout, so reliable uncertainty comes at no extra training cost. Uncertainty reliably flags severe failures but discriminates poorly among already high-quality images. These findings support uncertainty as a practical and computationally economical signal for quality filtering, reliability assessment and OOD detection in NCCT-to CECT synthesis.