Bi-level Guided Diffusion Models for Zero-Shot Medical Imaging Inverse Problems

• URL: http://arxiv.org/abs/2404.03706v1
• Date: Thu, 4 Apr 2024 10:36:56 GMT
• Title: Bi-level Guided Diffusion Models for Zero-Shot Medical Imaging Inverse Problems
• Authors: Hossein Askari, Fred Roosta, Hongfu Sun,
• Abstract summary: Inverse problems aim to infer high-quality images from incomplete, noisy measurements. The Diffusion Models have recently emerged as a promising approach to such practical challenges. A central challenge in this approach is how to guide an unconditional prediction to conform to the measurement information. We propose underlinetextbfBi-level underlineGuided underlineDiffusion underlineModels (BGDM)
• Score: 4.82425721275731
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: In the realm of medical imaging, inverse problems aim to infer high-quality images from incomplete, noisy measurements, with the objective of minimizing expenses and risks to patients in clinical settings. The Diffusion Models have recently emerged as a promising approach to such practical challenges, proving particularly useful for the zero-shot inference of images from partially acquired measurements in Magnetic Resonance Imaging (MRI) and Computed Tomography (CT). A central challenge in this approach, however, is how to guide an unconditional prediction to conform to the measurement information. Existing methods rely on deficient projection or inefficient posterior score approximation guidance, which often leads to suboptimal performance. In this paper, we propose \underline{\textbf{B}}i-level \underline{G}uided \underline{D}iffusion \underline{M}odels ({BGDM}), a zero-shot imaging framework that efficiently steers the initial unconditional prediction through a \emph{bi-level} guidance strategy. Specifically, BGDM first approximates an \emph{inner-level} conditional posterior mean as an initial measurement-consistent reference point and then solves an \emph{outer-level} proximal optimization objective to reinforce the measurement consistency. Our experimental findings, using publicly available MRI and CT medical datasets, reveal that BGDM is more effective and efficient compared to the baselines, faithfully generating high-fidelity medical images and substantially reducing hallucinatory artifacts in cases of severe degradation.

Related papers

• Dimensionality Reduction and Nearest Neighbors for Improving Out-of-Distribution Detection in Medical Image Segmentation [1.2873975765521795]
  This work applied the Mahalanobis distance (MD) post hoc to the bottleneck features of four Swin UNETR and nnU-net models that segmented the liver. Images the models failed on were detected with high performance and minimal computational load.
  arXiv  Detail & Related papers  (2024-08-05T18:24:48Z)
• Partitioned Hankel-based Diffusion Models for Few-shot Low-dose CT Reconstruction [10.158713017984345]
  We propose a few-shot low-dose CT reconstruction method using Partitioned Hankel-based Diffusion (PHD) models. In the iterative reconstruction stage, an iterative differential equation solver is employed along with data consistency constraints to update the acquired projection data. The results approximate those of normaldose counterparts, validating PHD model as an effective and practical model for reducing artifacts and noise while preserving image quality.
  arXiv  Detail & Related papers  (2024-05-27T13:44:53Z)
• Discrepancy-based Diffusion Models for Lesion Detection in Brain MRI [1.8420387715849447]
  Diffusion probabilistic models (DPMs) have exhibited significant effectiveness in computer vision tasks. Their notable performance heavily relies on labelled datasets, which limits their application in medical images. This paper introduces a novel framework by incorporating distinctive discrepancy features.
  arXiv  Detail & Related papers  (2024-05-08T11:26:49Z)
• UNICORN: Ultrasound Nakagami Imaging via Score Matching and Adaptation [59.91293113930909]
  Nakagami imaging holds promise for visualizing and quantifying tissue scattering in ultrasound waves. Existing methods struggle with optimal window size selection and suffer from estimator instability. We propose a novel method called UNICORN that offers an accurate, closed-form estimator for Nakagami parameter estimation.
  arXiv  Detail & Related papers  (2024-03-10T18:05:41Z)
• B-Spine: Learning B-Spline Curve Representation for Robust and Interpretable Spinal Curvature Estimation [50.208310028625284]
  We propose B-Spine, a novel deep learning pipeline to learn B-spline curve representation of the spine. We estimate the Cobb angles for spinal curvature estimation from low-quality X-ray images.
  arXiv  Detail & Related papers  (2023-10-14T15:34:57Z)
• One Sample Diffusion Model in Projection Domain for Low-Dose CT Imaging [10.797632196651731]
  Low-dose computed tomography (CT) plays a significant role in reducing the radiation risk in clinical applications. With the rapid development and wide application of deep learning, it has brought new directions for the development of low-dose CT imaging algorithms. We propose a fully unsupervised one sample diffusion model (OSDM)in projection domain for low-dose CT reconstruction. The results prove that OSDM is practical and effective model for reducing the artifacts and preserving the image quality.
  arXiv  Detail & Related papers  (2022-12-07T13:39:23Z)
• MIRST-DM: Multi-Instance RST with Drop-Max Layer for Robust Classification of Breast Cancer [62.997667081978825]
  We propose the Multi-instance RST with a drop-max layer, namely MIRST-DM, to learn smoother decision boundaries on small datasets. The proposed approach was validated using a small breast ultrasound dataset with 1,190 images.
  arXiv  Detail & Related papers  (2022-05-02T20:25:26Z)
• Harmonizing Pathological and Normal Pixels for Pseudo-healthy Synthesis [68.5287824124996]
  We present a new type of discriminator, the segmentor, to accurately locate the lesions and improve the visual quality of pseudo-healthy images. We apply the generated images into medical image enhancement and utilize the enhanced results to cope with the low contrast problem. Comprehensive experiments on the T2 modality of BraTS demonstrate that the proposed method substantially outperforms the state-of-the-art methods.
  arXiv  Detail & Related papers  (2022-03-29T08:41:17Z)
• Solving Inverse Problems in Medical Imaging with Score-Based Generative Models [87.48867245544106]
  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. We propose a fully unsupervised technique for inverse problem solving, leveraging the recently introduced score-based generative models.
  arXiv  Detail & Related papers  (2021-11-15T05:41:12Z)
• Deep AUC Maximization for Medical Image Classification: Challenges and Opportunities [60.079782224958414]
  We will present and discuss opportunities and challenges brought by a new deep learning method by AUC (aka underlinebf Deep underlinebf AUC classification)
  arXiv  Detail & Related papers  (2021-11-01T15:31:32Z)
• A Lightweight Structure Aimed to Utilize Spatial Correlation for Sparse-View CT Reconstruction [6.8438089867929905]
  Severe imaging noise and streaking artifacts turn out to be a major issue in the low dose protocol. We propose a dual-domain deep learning-based method that breaks through the limitations of currently prevailing algorithms. Our method achieves the state-of-the-art performance by reaching the PSNR of 40.305 and the SSIM of 0.948, while ensuring high model mobility.
  arXiv  Detail & Related papers  (2021-01-19T13:26:17Z)

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.