Medical Foundation Model Features as Perceptual Loss for Brain MRI Contrast Dose Simulation

• URL: http://arxiv.org/abs/2608.28773v1
• Date: Fri, 28 Aug 2026 18:30:34 GMT
• Title: Medical Foundation Model Features as Perceptual Loss for Brain MRI Contrast Dose Simulation
• Abstract summary: Most perceptual losses are still computed with natural-image backbones such as VGG16 or ResNet50.<n>This mismatch may weaken supervision for anatomy, contrast enhancement, and acquisition variability.<n>We test whether medical foundation model features provide a more suitable perceptual loss for brain MRI contrast dose simulation.
• Score: 1.5123472914752591
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Perceptual losses are widely used in medical image synthesis because they encourage agreement in high-level structure beyond voxel-wise intensity similarity. In practice, most perceptual losses are still computed with natural-image backbones such as VGG16 or ResNet50, even when the target domain is magnetic resonance imaging (MRI). This mismatch may weaken supervision for anatomy, contrast enhancement, and acquisition variability. We test whether medical foundation model features provide a more suitable perceptual loss for brain MRI contrast dose simulation. The study has two stages. First, we compare RadImageNet, SegVol, and BrainIAC with ImageNet-pretrained VGG16 and ResNet50 as frozen feature extractors on four public medical imaging benchmarks: thyroid ultrasound, breast ultrasound, anterior cruciate ligament knee MRI, and meniscus knee MRI. RadImageNet achieves the lowest mean rank across the Stage I representation suite and is selected as $φ^\star$. Second, we replace only the VGG16 feature extractor in an existing iterative brain MRI dose simulation framework with $φ^\star$. The generator, reconstruction loss, adversarial loss, auxiliary losses, optimization schedule, and loss weights are kept unchanged. Standard metrics change modestly, with PSNR increasing from 41.63 to 41.74, SSIM from 0.9739 to 0.9754, RMSE decreasing from 0.1384 to 0.1369, and residual-uptake CNR from 0.0085 to 0.0082. The visual results show the main effect: RadImageNet reduces residual enhancement in marked structures, follows a more faithful dose-reduction trajectory, and remains close to the acquired 10% low-dose target. These results support domain-aligned radiology features as a practical perceptual feature space for MRI dose simulation, while leaving clinical equivalence and larger-cohort validation as future work.

Related papers

• A Unified Model for Compressed Sensing MRI Across Undersampling Patterns [69.19631302047569]
  We propose a unified MRI reconstruction model robust to various measurement undersampling patterns and image resolutions.<n>Our model improves SSIM by 11% and PSNR by 4 dB over a state-of-the-art CNN (End-to-End VarNet) with 600$times$ faster inference than diffusion methods.
  arXiv  Detail & Related papers  (2024-10-05T20:03:57Z)
• Volumetric Reconstruction Resolves Off-Resonance Artifacts in Static and Dynamic PROPELLER MRI [76.60362295758596]
  Off-resonance artifacts in magnetic resonance imaging (MRI) are visual distortions that occur when the actual resonant frequencies of spins within the imaging volume differ from the expected frequencies used to encode spatial information. We propose to resolve these artifacts by lifting the 2D MRI reconstruction problem to 3D, introducing an additional "spectral" dimension to model this off-resonance.
  arXiv  Detail & Related papers  (2023-11-22T05:44:51Z)
• Faithful Synthesis of Low-dose Contrast-enhanced Brain MRI Scans using Noise-preserving Conditional GANs [102.47542231659521]
  Gadolinium-based contrast agents (GBCA) are indispensable in Magnetic Resonance Imaging (MRI) for diagnosing various diseases. GBCAs are expensive and may accumulate in patients with potential side effects. It is unclear to which extent the GBCA dose can be reduced while preserving the diagnostic value.
  arXiv  Detail & Related papers  (2023-06-26T13:19:37Z)
• Reconstruction of Cardiac Cine MRI Using Motion-Guided Deformable Alignment and Multi-Resolution Fusion [4.828904401099519]
  A novel end-to-end deep learning network is developed to improve cardiac cine MRI reconstruction. The proposed method exhibits high-quality results with richer details and fewer artifacts for cardiac cine MRI reconstruction on different accelerations.
  arXiv  Detail & Related papers  (2023-03-09T01:03:35Z)
• Synthetic Low-Field MRI Super-Resolution Via Nested U-Net Architecture [0.0]
  The aim of this paper is to improve the SNR and overall image quality of low-field MRI scans to improve diagnostic capability. We propose a Nested U-Net neural network architecture super-resolution algorithm that outperforms previously suggested deep learning methods with an average PSNR of 78.83 and SSIM of 0.9551.
  arXiv  Detail & Related papers  (2022-11-28T04:09:21Z)
• Model-Guided Multi-Contrast Deep Unfolding Network for MRI Super-resolution Reconstruction [68.80715727288514]
  We show how to unfold an iterative MGDUN algorithm into a novel model-guided deep unfolding network by taking the MRI observation matrix. In this paper, we propose a novel Model-Guided interpretable Deep Unfolding Network (MGDUN) for medical image SR reconstruction.
  arXiv  Detail & Related papers  (2022-09-15T03:58:30Z)
• Automated SSIM Regression for Detection and Quantification of Motion Artefacts in Brain MR Images [54.739076152240024]
  Motion artefacts in magnetic resonance brain images are a crucial issue. The assessment of MR image quality is fundamental before proceeding with the clinical diagnosis. An automated image quality assessment based on the structural similarity index (SSIM) regression has been proposed here.
  arXiv  Detail & Related papers  (2022-06-14T10:16:54Z)
• Negligible effect of brain MRI data preprocessing for tumor segmentation [36.89606202543839]
  We conduct experiments on three publicly available datasets and evaluate the effect of different preprocessing steps in deep neural networks. Our results demonstrate that most popular standardization steps add no value to the network performance. We suggest that image intensity normalization approaches do not contribute to model accuracy because of the reduction of signal variance with image standardization.
  arXiv  Detail & Related papers  (2022-04-11T17:29:36Z)
• A Long Short-term Memory Based Recurrent Neural Network for Interventional MRI Reconstruction [50.1787181309337]
  We propose a convolutional long short-term memory (Conv-LSTM) based recurrent neural network (RNN), or ConvLR, to reconstruct interventional images with golden-angle radial sampling. The proposed algorithm has the potential to achieve real-time i-MRI for DBS and can be used for general purpose MR-guided intervention.
  arXiv  Detail & Related papers  (2022-03-28T14:03:45Z)
• Impact of loss functions on the performance of a deep neural network designed to restore low-dose digital mammography [8.041868861658]
  ResNet architecture, with hierarchical skip connections, is proposed to restore low-dose digital mammography. We extracted 256,000 image patches from a dataset of 400 images of retrospective clinical mammography exams. To validate the network in a real scenario, a physical anthropomorphic breast phantom was used to acquire real low-dose and standard full-dose images.
  arXiv  Detail & Related papers  (2021-11-12T14:15:08Z)
• Generative Adversarial Networks (GAN) Powered Fast Magnetic Resonance Imaging -- Mini Review, Comparison and Perspectives [5.3148259096171175]
  One drawback of MRI is its comparatively slow scanning and reconstruction compared to other image modalities. Deep Neural Networks (DNNs) have been used in sparse MRI reconstruction models to recreate relatively high-quality images. Generative Adversarial Networks (GAN) based methods are proposed to solve fast MRI with enhanced image perceptual quality.
  arXiv  Detail & Related papers  (2021-05-04T23:59:00Z)
• Towards Ultrafast MRI via Extreme k-Space Undersampling and Superresolution [65.25508348574974]
  We go below the MRI acceleration factors reported by all published papers that reference the original fastMRI challenge. We consider powerful deep learning based image enhancement methods to compensate for the underresolved images. The quality of the reconstructed images surpasses that of the other methods, yielding an MSE of 0.00114, a PSNR of 29.6 dB, and an SSIM of 0.956 at x16 acceleration factor.
  arXiv  Detail & Related papers  (2021-03-04T10:45:01Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.