Pith. sign in

REVIEW 17 cited by

Solving Inverse Problems in Medical Imaging with Score-Based Generative Models

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2111.08005 v2 pith:LO6Q7L4I submitted 2021-11-15 eess.IV cs.CVcs.LGstat.ML

Yang Song , Liyue Shen , Lei Xing , Stefano Ermon This is my paper
classification eess.IVcs.CVcs.LGstat.ML
keywords measurementmeasurementsimagesmedicalmodelgenerativeimaginginverse
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Reconstructing medical images from partial measurements is an important inverse problem in Computed Tomography (CT) and Magnetic Resonance Imaging (MRI). Existing solutions based on machine learning typically train a model to directly map measurements to medical images, leveraging a training dataset of paired images and measurements. These measurements are typically synthesized from images using a fixed physical model of the measurement process, which hinders the generalization capability of models to unknown measurement processes. To address this issue, we propose a fully unsupervised technique for inverse problem solving, leveraging the recently introduced score-based generative models. Specifically, we first train a score-based generative model on medical images to capture their prior distribution. Given measurements and a physical model of the measurement process at test time, we introduce a sampling method to reconstruct an image consistent with both the prior and the observed measurements. Our method does not assume a fixed measurement process during training, and can thus be flexibly adapted to different measurement processes at test time. Empirically, we observe comparable or better performance to supervised learning techniques in several medical imaging tasks in CT and MRI, while demonstrating significantly better generalization to unknown measurement processes.

Discussion (0). Sign in to comment.

Forward citations

Cited by 17 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. OpenAlex reports about 138 citations worldwide. Full citation record

  1. Strong Gravitational Lensing Posterior Sampling in Pixel-Space Using Diffusion Models and Recurrent Inference Machines

    astro-ph.IM 2026-07 conditional novelty 7.0 of 10

    DiRIM uses a diffusion model with recurrent score refinement to sample pixel-space joint posteriors of the lensed source and foreground mass map, reproducing mock strong-lens observations to the noise level.

  2. Physics Matters in PnP: Recovery Guarantees with the MMSE and NN Denoisers

    math.OC 2026-07 conditional novelty 6.0 of 10

    PnP-FBS with MMSE denoisers recovers the true signal with explicit pointwise and Wasserstein error bounds, provided the denoiser's noise covariance is matched to the preconditioned observation noise.

  3. Agentic Autoresearch for CT Reconstruction

    physics.med-ph 2026-07 conditional novelty 6.0 of 10

    An LLM agent autonomously benchmarked 26 CT reconstruction methods and showed that adding mild noise nearly inverts the clean-data ranking, with retraining partially restoring it.

  4. From Sparse X-rays to 3D CT: Training-Free Reconstruction with Diffusion Priors

    eess.IV 2026-06 unverdicted novelty 6.0 of 10

    TF-PRDiT uses a frozen 3D diffusion transformer prior with task-specific forward operators and predictor-corrector sampling to solve X-ray-to-CT reconstruction and other volumetric inverse problems without any retraining.

  5. Jacobian-Aware Posterior Sampling for Inverse Problems

    cs.CV 2025-11 conditional novelty 6.0 of 10

    AdaPS adaptively scales likelihood guidance in DDIM posterior sampling via agreement between two surrogates, improving LPIPS/PSNR trade-offs on image restoration tasks without task-specific tuning.

  6. Reverse Imaging for Wide-spectrum Generalization of Cardiac MRI Segmentation

    cs.CV 2025-08 conditional novelty 6.0 of 10

    Estimating tissue spin properties from a cardiac MRI and synthesizing new sequence contrasts via physics improves zero-shot segmentation across protocols.

  7. Zero-Shot Cyclic Peptide Design via Composable Geometric Constraints

    cs.LG 2025-07 conditional novelty 6.0 of 10

    CP-Composer trains a geometric diffusion model on linear peptides and imposes cyclization constraints at generation time, achieving 38-84% constraint satisfaction across four cyclization strategies.

  8. Self-supervised feature learning for cardiac Cine MR image reconstruction

    eess.IV 2025-05 conditional novelty 6.0 of 10

    Using only undersampled cardiac Cine data, SSFL-Recon with contrastive or VICReg feature pretraining reconstructs at up to 16x acceleration with quality comparable to supervised learning.

  9. Joint Reconstruction of Activity and Attenuation in PET by Diffusion Posterior Sampling in Wavelet Coefficient Space

    physics.med-ph 2025-05 unverdicted novelty 6.0 of 10

    A wavelet diffusion model combined with diffusion posterior sampling enables joint 3D activity-attenuation reconstruction in PET from emission data alone, outperforming MLAA on simulated TOF data.

  10. Lyapunov Guidance: A Unified Framework for Stabilizing Generative Flows

    cs.LG 2026-07 reject novelty 5.0 of 10

    Flow guidance is framed as Lyapunov control with a pseudo-projection for stability, but the projected flow is not shown to sample the target conditional distribution.

  11. Optimal Linear Baseline Models for Scientific Machine Learning

    cs.LG 2025-08 conditional novelty 5.0 of 10

    Closed-form rank-constrained linear estimators, derived from Bayes risk, unify forward modeling, inverse recovery, autoencoding, and denoising, and often match or beat trained neural networks.

  12. Solving Inverse Problems via Diffusion-Based Priors: An Approximation-Free Ensemble Sampling Approach

    cs.LG 2025-06 conditional novelty 5.0 of 10

    A weighted-particle sampler evolves the posterior through the diffusion model's reverse dynamics, with theoretical error bounds and improved image reconstructions.

  13. Tomographic Foundation Model -- FORCE: Flow-Oriented Reconstruction Conditioning Engine

    eess.IV 2025-06 conditional novelty 5.0 of 10

    FORCE, a PFGM++-based reconstruction framework with data-fidelity conditioning, outperforms several unsupervised CT reconstruction baselines on low-dose and sparse-view benchmarks, but its metal-artifact results are o...

  14. EgoAnimate: Generating Human Animations from Egocentric top-down Views

    cs.CV 2025-07 conditional novelty 4.0 of 10

    EgoAnimate synthesizes a frontal T-pose image from an egocentric top-down photo using a fine-tuned Stable Diffusion model, then animates it with off-the-shelf image-to-motion methods to produce an animatable avatar.

  15. Stochastic and Non-local Closure Modeling for Nonlinear Dynamical Systems via Latent Score-based Generative Models

    cs.LG 2025-06 conditional novelty 4.0 of 10

    Joint training of autoencoders with diffusion models in latent space gives stochastic turbulence closure accuracy close to physical-space diffusion models at roughly 5-7x lower cost.

  16. Adaptive Mask-guided K-space Diffusion for Accelerated MRI Reconstruction

    eess.IV 2025-06 reject novelty 4.0 of 10

    AMDM reconstructs undersampled MRI by masking k-space frequency components with adaptive masks inside a diffusion model, and reports large PSNR gains over baseline methods.

  17. ResPF: Residual Poisson Flow for Efficient and Physically Consistent Sparse-View CT Reconstruction

    eess.IV 2025-06 conditional novelty 4.0 of 10

    ResPF combines conditional PFGM++ sampling with hijacked initialization, ASD-POCS data consistency, and residual fusion to reconstruct CT images from sparse-view projections.

Pith tools