% IMPORTANT: The following is UTF-8 encoded.  This means that in the presence
% of non-ASCII characters, it will not work with BibTeX 0.99 or older.
% Instead, you should use an up-to-date BibTeX implementation like “bibtex8” or
% “biber”.

@ARTICLE{JuliusBlcker:307471,
      author       = {T. Julius Blöcker and N. Delopoulos and M. A. Palacios and
                      S. Klüter and J. Hörner-Rieber and C. Rippke and L.
                      Placidi and L. Boldrini and V. Frascino and N. Andratschke
                      and M. Baumgartl and R. Dal Bello and S. N. Marschner and C.
                      Belka$^*$ and S. Corradini and D. Dudas and M. Riboldi and
                      C. Kurz and G. Landry},
      title        = {{GTV} segmentation in {MRI} guided radiotherapy with
                      promptable foundation models.},
      journal      = {Physics in medicine and biology},
      volume       = {71},
      number       = {1},
      issn         = {0031-9155},
      address      = {Bristol},
      publisher    = {IOP Publ.},
      reportid     = {DKFZ-2025-03070},
      pages        = {015006},
      year         = {2025},
      note         = {PUBLISHED29 December 2025},
      abstract     = {Objective. Magnetic resonance imaging (MRI) guided
                      radiotherapy requires the delineation of gross tumor volumes
                      (GTV) in daily MRI from MRI-linacs. Specialized models have
                      been developed for this task for certain tumors. This study
                      investigated an alternative, using promptable foundation
                      models.Approach. Promptable foundation models were prompted
                      with six different sparse geometric prompt types (points,
                      boxes, 2D masks) to produce GTV segmentation masks,
                      including Segment-anything 2 (SAM2), SAM2 fine-tuned for
                      medical imaging (MedSAM2), and nnInteractive, an
                      nnUnet-based promptable model for medical imaging. A diverse
                      multi-institutional dataset of clinical GTV masks from the
                      abdomen, lung, liver, pancreas, and pelvis sites on MRI
                      scans from MRI-linacs was used to evaluate model outputs
                      using various metrics, including the Dice similarity
                      coefficient (DSC).Main results. The models produced
                      segmentation masks comparable or superior to those from
                      domain-specific models with median DSCs of up to 0.85
                      (nnInteractive-mask3 prompt). Prompts with more spatial
                      information yielded better results with lower variance, with
                      the effect reduced for nnInteractive and MedSAM2. These
                      produced overall better results (median DSC over all prompt
                      types 0.75 for nnInteractive, 0.70 for MedSAM2, 0.54 for
                      SAM2).Significance. This investigation showed that
                      promptable foundation models can in principle be used for
                      GTV segmentation in MRI across multiple tumor types,
                      although more research is necessary to reduce the variance
                      and improve model performance.},
      keywords     = {Magnetic Resonance Imaging / Humans / Radiotherapy,
                      Image-Guided: methods / Tumor Burden / Image Processing,
                      Computer-Assisted: methods / Neoplasms: radiotherapy /
                      Neoplasms: diagnostic imaging / Neoplasms: pathology / GTV
                      segmentation (Other) / MRI-linac (Other) / MRgRT (Other) /
                      deep learning (Other) / foundation models (Other) /
                      promptable models (Other)},
      cin          = {MU01},
      ddc          = {530},
      cid          = {I:(DE-He78)MU01-20160331},
      pnm          = {899 - ohne Topic (POF4-899)},
      pid          = {G:(DE-HGF)POF4-899},
      typ          = {PUB:(DE-HGF)16},
      pubmed       = {pmid:41401509},
      doi          = {10.1088/1361-6560/ae2db9},
      url          = {https://inrepo02.dkfz.de/record/307471},
}