Saved in:
Bibliographic Details
Main Authors: Kabas, Bilal, Arslan, Fuat, Nezhad, Valiyeh A., Ozturk, Saban, Saritas, Emine U., Çukur, Tolga
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2412.09331
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909745866080256
author Kabas, Bilal
Arslan, Fuat
Nezhad, Valiyeh A.
Ozturk, Saban
Saritas, Emine U.
Çukur, Tolga
author_facet Kabas, Bilal
Arslan, Fuat
Nezhad, Valiyeh A.
Ozturk, Saban
Saritas, Emine U.
Çukur, Tolga
contents Medical image reconstruction from undersampled acquisitions is an ill-posed inverse problem requiring accurate recovery of anatomical structures from incomplete measurements. Physics-driven (PD) network models have gained prominence for this task by integrating data-consistency mechanisms with learned priors, enabling improved performance over purely data-driven approaches. However, reconstruction quality still hinges on the network's ability to disentangle artifacts from true anatomical signals-both of which exhibit complex, multi-scale contextual structure. Convolutional neural networks (CNNs) capture local correlations but often struggle with non-local dependencies. While transformers aim to alleviate this limitation, practical implementations involve design compromises to reduce computational cost by balancing local and non-local sensitivity, occasionally resulting in performance comparable to CNNs. To address these challenges, we propose MambaRoll, a novel physics-driven autoregressive state space model (SSM) for high-fidelity and efficient image reconstruction. MambaRoll employs an unrolled architecture where each cascade autoregressively predicts finer-scale feature maps conditioned on coarser-scale representations, enabling consistent multi-scale context propagation. Each stage is built on a hierarchy of scale-specific PD-SSM modules that capture spatial dependencies while enforcing data consistency through residual correction. To further improve scale-aware learning, we introduce a Deep Multi-Scale Decoding (DMSD) loss, which provides supervision at intermediate spatial scales in alignment with the autoregressive design. Demonstrations on accelerated MRI and sparse-view CT reconstructions show that MambaRoll consistently outperforms state-of-the-art CNN-, transformer-, and SSM-based methods.
format Preprint
id arxiv_https___arxiv_org_abs_2412_09331
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Physics-Driven Autoregressive State Space Models for Medical Image Reconstruction
Kabas, Bilal
Arslan, Fuat
Nezhad, Valiyeh A.
Ozturk, Saban
Saritas, Emine U.
Çukur, Tolga
Image and Video Processing
Computer Vision and Pattern Recognition
Medical image reconstruction from undersampled acquisitions is an ill-posed inverse problem requiring accurate recovery of anatomical structures from incomplete measurements. Physics-driven (PD) network models have gained prominence for this task by integrating data-consistency mechanisms with learned priors, enabling improved performance over purely data-driven approaches. However, reconstruction quality still hinges on the network's ability to disentangle artifacts from true anatomical signals-both of which exhibit complex, multi-scale contextual structure. Convolutional neural networks (CNNs) capture local correlations but often struggle with non-local dependencies. While transformers aim to alleviate this limitation, practical implementations involve design compromises to reduce computational cost by balancing local and non-local sensitivity, occasionally resulting in performance comparable to CNNs. To address these challenges, we propose MambaRoll, a novel physics-driven autoregressive state space model (SSM) for high-fidelity and efficient image reconstruction. MambaRoll employs an unrolled architecture where each cascade autoregressively predicts finer-scale feature maps conditioned on coarser-scale representations, enabling consistent multi-scale context propagation. Each stage is built on a hierarchy of scale-specific PD-SSM modules that capture spatial dependencies while enforcing data consistency through residual correction. To further improve scale-aware learning, we introduce a Deep Multi-Scale Decoding (DMSD) loss, which provides supervision at intermediate spatial scales in alignment with the autoregressive design. Demonstrations on accelerated MRI and sparse-view CT reconstructions show that MambaRoll consistently outperforms state-of-the-art CNN-, transformer-, and SSM-based methods.
title Physics-Driven Autoregressive State Space Models for Medical Image Reconstruction
topic Image and Video Processing
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2412.09331