{
  "abstract": "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 provides improved performance relative to state-of-the-art CNN-, transformer-, and SSM-based methods.",
  "abstract_shown": true,
  "affiliations": [
    "Bilkent University"
  ],
  "article_type": null,
  "arxiv": null,
  "author_total": 7,
  "authors": [
    {
      "affs": [
        0
      ],
      "name": "Bilal Kabaş",
      "orcid": "0009-0008-7177-0594"
    },
    {
      "affs": [
        0
      ],
      "name": "Fuat Arslan",
      "orcid": "0009-0005-7844-890X"
    },
    {
      "affs": [
        0
      ],
      "name": "Valiyeh Ansarian Nezhad",
      "orcid": "0009-0004-9223-410X"
    },
    {
      "affs": [
        0
      ],
      "name": "Kader K. Oguz"
    },
    {
      "affs": [
        0
      ],
      "name": "Şaban Öztürk",
      "orcid": "0000-0003-2371-8173"
    },
    {
      "affs": [
        0
      ],
      "name": "Emine Ülkü Sarıtaş",
      "orcid": "0000-0001-8551-1077"
    },
    {
      "affs": [
        0
      ],
      "name": "Tolga Çukur",
      "orcid": "0000-0002-2296-851X"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/cnxpzz03vexhx.bib",
  "cited_by": 40,
  "doi": "10.1109/tmi.2026.3716153",
  "fulltext_url": null,
  "id": "cnxpzz03vexhx",
  "issns": [
    "0278-0062",
    "1558-254X"
  ],
  "landing_url": "https://doi.org/10.1109/tmi.2026.3716153",
  "language": "en",
  "licence": "CC-BY",
  "licence_label": "CC BY",
  "licence_source": "journal",
  "licence_url": "https://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": null,
  "native_subjects": [
    [
      "openalex",
      "2741|Advanced MRI Techniques and Applications"
    ]
  ],
  "openalex": "W4405355158",
  "pdf_url": null,
  "pmcid": null,
  "pmid": "42490290",
  "published": "2026-07-23",
  "publisher": "Institute of Electrical and Electronics Engineers",
  "ris": "https://research.draconitic.com/paper/cnxpzz03vexhx.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T13:26:54Z",
      "native_id": "W4405355158",
      "source": "openalex",
      "url": "https://doi.org/10.1109/tmi.2026.3716153"
    }
  ],
  "subjects": [
    {
      "field": "medicine",
      "origin": "openalex:2741|Advanced MRI Techniques and Applications",
      "subfield": "imaging"
    }
  ],
  "title": "Physics-Driven Autoregressive State Space Models for Medical Image Reconstruction",
  "title_text": "Physics-Driven Autoregressive State Space Models for Medical Image Reconstruction",
  "type": "peer-reviewed",
  "updated_at": "2026-11-09T13:21:15Z",
  "url": "https://research.draconitic.com/paper/cnxpzz03vexhx",
  "venue": "IEEE Transactions on Medical Imaging",
  "versions": []
}