# PEEL: Physics-Enabled Evidential Learning for Identifiable Uncertainty in CT Imaging

arXiv: https://arxiv.org/abs/2609.29599 · HTML: https://mlsanity.com/paper/2609.29599/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 4 of its 4 references that are in the map.
- First posted on arXiv 27 Aug 2026.

## Builds on (most cited first, 4 of 4)

- [Predictive Uncertainty Estimation via Prior Networks](https://mlsanity.com/paper/1802.10501/) (2018)
- [Deep Evidential Regression](https://mlsanity.com/paper/1910.02600/) (2019)
- [The Unreasonable Effectiveness of Deep Evidential Regression](https://mlsanity.com/paper/2205.10060/) (2022)
- [ELVAE: Evidential Learning-Based Variational Autoencoder for Uncertainty-Aware Generation](https://mlsanity.com/paper/2608.10398/) (2026)

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

- [PEEL-DDPM: Physics-Enabled Evidential Learning for the Denoising Diffusion Probabilistic Model](https://mlsanity.com/paper/2609.31742/) (2026)

## Abstract (arXiv)

Normal-inverse-gamma (NIG) regression is not uniquely identifiable from its marginal Student-t likelihood: the likelihood determines three combinations of four NIG parameters and is constant along a one-dimensional fiber. We identify that fiber using independent physical measurement. As an initial embodiment, a reconstruction network receives one noisy filtered-backprojection (FBP) image and is first trained only by Student-t negative log-likelihood to estimate the three identifiable coordinates (gamma, alpha, c). The network is then frozen; repeated physical-noise realizations propagated through its reconstruction output form a Monte Carlo (MC) teacher label for output-domain aleatoric variance. An aleatoric head attached to frozen features learns this label, after which (beta, nu) are recovered algebraically. On 30 held-out simulated objects at five photon levels, one-image predictions achieved pooled Spearman correlations of 0.832-0.951 against independent 400-repeat references, median within-image correlations were 0.834-0.947, and 98.81-99.55% of evaluated pixels satisfied the algebraic admissibility condition. The method needs no KL term, reference prior, evidence regularizer, or cross-loss weight.