---
title: "自监督物理信息隐式CEST重建：通过洛伦兹编码"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2607.06132v1"
permalink: "https://feed.astralor.com/items/itm_f61ee42a7ab52ca40839e356"
published: "2026-07-07T10:49:07.000Z"
score: 78
channel: "finance"
tags: ["CEST MRI","Lorentz Encoding","医学成像","物理信息神经网络","自监督学习","隐式神经表示"]
---

# 自监督物理信息隐式CEST重建：通过洛伦兹编码

## AI 点评

> 这篇论文让我眼前一亮：它将自监督学习与严格的物理模型融合，而不是简单堆叠神经网络。对医学影像AI领域的研究者来说，LE是一个优雅的范式，尤其适合资源受限的临床场景。CEST MRI是代谢成像的利器，LE让它的采集速度不再成问题——这对放射科医生和患者都是好消息。

## 速览

化学交换饱和转移（CEST）MRI虽然提供有价值的代谢信息，但临床受限于长采集时间。稀疏采样虽能减少扫描时间，但从有限数据重建高分辨率Z谱是一个病态逆问题。本文提出洛伦兹编码（LE），一种物理信息框架，通过隐式连续坐标学习将CEST重建转化为自监督任务。LE利用可学习洛伦兹基函数投影稀疏坐标到物理约束空间，有效去噪并保证物理一致性。在人体脑部数据上，39点采样下PSNR达57.58 dB，SSIM达0.9994，显著超越现有方法，并能准确映射代谢物（APT、NOE、MT）。

## 关键点

- 提出洛伦兹编码（LE），将CEST重建建模为自监督隐式神经表示学习，引入物理信息约束。
- LE利用可学习洛伦兹基函数将稀疏采样坐标投影到物理约束空间，抑制伪影和无效信号。
- 在人体脑部数据上39点采样下PSNR达57.58 dB、SSIM达0.9994，显著优于现有方法。
- 学到的编码形成连续、几何有序的潜在空间轨迹，支持定量代谢物映射（APT、NOE、MT）。

## 全文

Multi-Pool Chemical Exchange Saturation Transfer (CEST) MRI provides valuable metabolic information but is clinically limited by long acquisition times. Although sparse sampling reduces scanning time, reconstructing high-resolution Z-spectra from limited data remains an ill-posed inverse problem. Conventional interpolation and generic Implicit Neural Rep-resentations (INRs) often lack physical constraints, leading to spectral artifacts and physically invalid signals. To address this, we propose Lorentz Encoding (LE), a physics-informed framework that formulates CEST reconstruction as a self-supervised reconstruction task via implicit continuous coordinate learning. Unlike generic positional encodings, LE regularizes the continuous spectral mapping by projecting sparse coordinates into a physically constrained space governed by a combination of parametric Lorentzian profiles with learnable basis functions. This mechanism effectively reduces noise and enforces consistency with physical models. Experiments on in vivo human brain data demonstrate that LE significantly outperforms state-of-the-art methods. Specifically, under a 39-point sampling strategy, LE achieves a PSNR of 57.58 dB and an SSIM of 0.9994. Furthermore, the learned physics-informed encodings form a continuous, geometrically ordered trajectory in the latent space, ensuring accurate quantitative metabo-lite mapping (APT, NOE, MT).

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_f61ee42a7ab52ca40839e356