Prior image-based medical image reconstruction using a style-based generative adversarial network

TL;DR

This paper introduces a novel MRI image reconstruction method that leverages a style-based GAN to incorporate prior images and disentangle image attributes, improving reconstruction quality in ill-posed inverse problems.

Contribution

It proposes a new approach using StyleGAN's latent space to regularize MRI reconstruction with prior images, focusing on disentangled style constraints.

Findings

Outperforms classical reconstruction methods based on traditional metrics.

Effective in scenarios with structurally similar prior images but different contrasts.

Demonstrates the potential of style-based generative models for medical image reconstruction.

Abstract

Computed medical imaging systems require a computational reconstruction procedure for image formation. In order to recover a useful estimate of the object to-be-imaged when the recorded measurements are incomplete, prior knowledge about the nature of object must be utilized. In order to improve the conditioning of an ill-posed imaging inverse problem, deep learning approaches are being actively investigated for better representing object priors and constraints. This work proposes to use a style-based generative adversarial network (StyleGAN) to constrain an image reconstruction problem in the case where additional information in the form of a prior image of the sought-after object is available. An optimization problem is formulated in the intermediate latent-space of a StyleGAN, that is disentangled with respect to meaningful image attributes or "styles", such as the contrast used in magnetic resonance imaging (MRI). Discrepancy between the sought-after and prior images is measured in the disentangled latent-space, and is used to regularize the inverse problem in the form of constraints on specific styles of the disentangled latent-space. A stylized numerical study inspired by MR imaging is designed, where the sought-after and the prior image are structurally similar, but belong to different contrast mechanisms. The presented numerical studies demonstrate the superiority of the proposed approach as compared to classical approaches in the form of traditional metrics.