Learned Primal Dual Splitting for Self-Supervised Noise-Adaptive MRI Reconstruction

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Janjusevic, Nikola, Khalilian-Gourtani, Amirhoussein, Wang, Yao, Feng, Li
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866908361435381760
author Janjusevic, Nikola
Khalilian-Gourtani, Amirhoussein
Wang, Yao
Feng, Li
author_facet Janjusevic, Nikola
Khalilian-Gourtani, Amirhoussein
Wang, Yao
Feng, Li
contents Magnetic resonance imaging (MRI) reconstruction has largely been dominated by deep neural networks (DNN); however, many state-of-the-art architectures use black-box structures, which hinder interpretability and improvement. Here, we propose an interpretable DNN architecture for self-supervised MRI reconstruction and denoising by directly parameterizing and learning the classical primal-dual splitting, dubbed LPDSNet. This splitting algorithm allows us to decouple the observation model from the signal prior. Experimentally, we show other interpretable architectures without this decoupling property exhibit failure in the self-supervised learning regime. We report state-of-the-art self-supervised joint MRI reconstruction and denoising performance and novel noise-level generalization capabilities, where in contrast black-box networks fail to generalize.
format Preprint
id arxiv_https___arxiv_org_abs_2504_15390
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Learned Primal Dual Splitting for Self-Supervised Noise-Adaptive MRI Reconstruction
Janjusevic, Nikola
Khalilian-Gourtani, Amirhoussein
Wang, Yao
Feng, Li
Image and Video Processing
Magnetic resonance imaging (MRI) reconstruction has largely been dominated by deep neural networks (DNN); however, many state-of-the-art architectures use black-box structures, which hinder interpretability and improvement. Here, we propose an interpretable DNN architecture for self-supervised MRI reconstruction and denoising by directly parameterizing and learning the classical primal-dual splitting, dubbed LPDSNet. This splitting algorithm allows us to decouple the observation model from the signal prior. Experimentally, we show other interpretable architectures without this decoupling property exhibit failure in the self-supervised learning regime. We report state-of-the-art self-supervised joint MRI reconstruction and denoising performance and novel noise-level generalization capabilities, where in contrast black-box networks fail to generalize.
title Learned Primal Dual Splitting for Self-Supervised Noise-Adaptive MRI Reconstruction
topic Image and Video Processing
url https://arxiv.org/abs/2504.15390