arXiv 2408.036572024-08-07cs.CV

PHOCUS：基于物理的超声分辨率增强反卷积方法

PHOCUS: Physics-Based Deconvolution for Ultrasound Resolution Enhancement

Technical University of Munich（慕尼黑工业大学）；Munich Center for Machine Learning (MCML)（慕尼黑机器学习中心）；Helmholtz Zentrum Munich（慕尼黑亥姆霍兹中心）；Stanford University（斯坦福大学）；LUMA Vision GmbH（卢马视觉有限公司）；Rechts der Isar University Hospital（右岸大学医院）；German Centre for Cardiovascular Research（德国心血管研究中心）

Felix Duelmer, Walter Simson, Mohammad Farid Azampour, Magdalena Wysocki, Angelos Karlas, Nassir Navab

备注：: Accepted at the Workshop of Advances in Simplifying Medical Ultrasound at MICCAI 2024

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出PHOCUS方法，利用隐式神经表示和可微分物理渲染管线，直接从B模式超声图像中恢复连续回声强度图，实现分辨率增强，并在合成数据、体模及在体图像上验证了其优于传统反卷积方法的性能。

摘要：

超声在医学诊断中被广泛使用，能够提供便捷且强大的成像能力，但由于衍射和成像系统有限孔径导致的分辨率限制，其诊断应用受到制约。超声成像系统的冲激函数被称为点扩散函数（PSF），在图像形成过程中，它与反射体的空间分布进行卷积。通过去除卷积过程引起的图像畸变来恢复高分辨率的反射体分布，可以改善图像的清晰度和细节。传统上，反卷积技术试图校正成像系统相关的PSF，直接处理射频（RF）数据。然而，RF数据通常不易获取。因此，我们引入一种基于物理的反卷积过程，使用建模的PSF，直接处理更常见的B模式图像。通过利用隐式神经表示（INRs），我们学习从空间位置到其各自回声强度值的连续映射，有效补偿离散化的图像空间。我们的贡献包括一种新颖的方法，通过可微分的基于物理的渲染管线，直接从B模式图像中恢复连续的回声强度图，用于超声分辨率增强。我们在合成数据上对所提方法进行了定性和定量评估，证明其在PSNR和SSIM等指标上优于传统方法。此外，我们还展示了在超声体模和颈动脉在体采集图像上的定性增强效果。

英文摘要：

Ultrasound is widely used in medical diagnostics allowing for accessible and powerful imaging but suffers from resolution limitations due to diffraction and the finite aperture of the imaging system, which restricts diagnostic use. The impulse function of an ultrasound imaging system is called the point spread function (PSF), which is convolved with the spatial distribution of reflectors in the image formation process. Recovering high-resolution reflector distributions by removing image distortions induced by the convolution process improves image clarity and detail. Conventionally, deconvolution techniques attempt to rectify the imaging system's dependent PSF, working directly on the radio-frequency (RF) data. However, RF data is often not readily accessible. Therefore, we introduce a physics-based deconvolution process using a modeled PSF, working directly on the more commonly available B-mode images. By leveraging Implicit Neural Representations (INRs), we learn a continuous mapping from spatial locations to their respective echogenicity values, effectively compensating for the discretized image space. Our contribution consists of a novel methodology for retrieving a continuous echogenicity map directly from a B-mode image through a differentiable physics-based rendering pipeline for ultrasound resolution enhancement. We qualitatively and quantitatively evaluate our approach on synthetic data, demonstrating improvements over traditional methods in metrics such as PSNR and SSIM. Furthermore, we show qualitative enhancements on an ultrasound phantom and an in-vivo acquisition of a carotid artery.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。