Conditional Variational Autoencoder for Learned Image Reconstruction

TL;DR

This paper introduces a flexible conditional variational autoencoder framework for learned image reconstruction that provides uncertainty quantification and handles various noise models, demonstrated on PET imaging.

Contribution

It develops a novel framework that approximates the posterior distribution in image reconstruction, enabling uncertainty quantification and transferability across datasets.

Findings

Generates high-quality samples compared to state-of-the-art methods.

Handles implicit noise models and priors effectively.

Provides efficient posterior sampling via feed-forward propagation.

Abstract

Learned image reconstruction techniques using deep neural networks have recently gained popularity, and have delivered promising empirical results. However, most approaches focus on one single recovery for each observation, and thus neglect the uncertainty information. In this work, we develop a novel computational framework that approximates the posterior distribution of the unknown image at each query observation. The proposed framework is very flexible: It handles implicit noise models and priors, it incorporates the data formation process (i.e., the forward operator), and the learned reconstructive properties are transferable between different datasets. Once the network is trained using the conditional variational autoencoder loss, it provides a computationally efficient sampler for the approximate posterior distribution via feed-forward propagation, and the summarizing statistics of the generated samples are used for both point-estimation and uncertainty quantification. We illustrate the proposed framework with extensive numerical experiments on positron emission tomography (with both moderate and low count levels) showing that the framework generates high-quality samples when compared with state-of-the-art methods.