文章速递​

近日，上海交通大学医学院医学技术学院张孝勇研究员（原工作单位：复旦大学类脑智能科学与技术研究院）领衔的研究团队在《表型组学（英文）》(Phenomics)上发表了一篇题为“Hierarchical Convolution-Based Multilayer Perception for Denoising 3D MRI to Enhance Diagnostic Confidence in Cerebral Small Vessel Disease”的研究论文。

​

该研究提出了一种新的三维磁共振成像（3D-MRI）降噪方法，多个数据集的结果显示，该方法显著提升了MRI图像质量并更好恢复脑小血管病的细微结构，从而增强了疾病诊断的可靠性和临床应用价值。​

扫描二维码 | 下载PDF原文

论文DOI链接：

https://link.springer.com/article/10.1007/s43657-025-00219-8

论文引用格式：

Yang, H., Zhang, S., Han, X. et al. Hierarchical Convolution-Based Multilayer Perception for Denoising 3D MRI to Enhance Diagnostic Confidence in Cerebral Small Vessel Disease. Phenomics (2025). https://doi.org/10.1007/s43657-025-00219-8

研究背景

​

脑小血管病是人群常见的慢性进展性脑血管疾病，其影像学表现通常体积小、对比度低且分布多样，因此对 MRI 图像质量高度敏感。临床常规 MRI 在采集过程中易受到噪声影响，这会掩盖关键病灶特征，降低医生识别轻微结构变化的能力，进而影响诊断准确性和后续影像分析任务。

​

现有的 MRI 去噪方法虽然取得一定进展，但仍面临收敛困难、易导致结构过度平滑等问题，难以在保证去噪效果的同时完整保留细小的病灶信息。因此，开发一种兼具高效去噪和可靠结构恢复能力的3D-MRI 去噪方法，对提升脑小血管病的临床诊断具有重要意义。

​

研究方法

​

本研究提出了一种结合局部空间特征与全局语义特征的新型3D-MRI去噪框架（图1），核心思路是通过多层级特征提取机制，在抑制噪声的同时增强对细微结构的捕捉能力。方法通过体素级特征建模、分层卷积编码与残差结构设计，实现对噪声分布、结构特征及高维表征的协同建模，从而有效缓解过度平滑问题。

此外，研究在模拟不同噪声水平的3D-MRI上进行了系统训练，并在多来源、多模态的独立测试数据上进行了验证，以评估方法的稳健性和泛化性能。影像质量由客观指标（如 PSNR、SSIM）和临床放射科医师的主观评分共同评价。

图1 新型3D-MRI 去噪框架

研究结果

​

​实验结果显示，该方法在3D-MRI 去噪中取得了显著优势，在客观数值指标上相比多种代表性传统方法与深度学习方法均有大幅提升，特别是在高噪声条件下仍能保持稳定表现（图2）。

​

更重要的是，该方法在恢复脑小血管病关键影像特征方面表现突出，能够更准确地重建被噪声破坏的细小结构，如血管周围间隙、微小梗塞及局灶性缺血（图3）。多名经验丰富的放射科专家对去噪后图像评分，结果显示其病灶显著性和整体图像质量均明显优于对比方法，并具有统计学意义。

图2 模型在MRI图像中的降噪效果

图3 病灶恢复效果

(a)血管周围间隙；(b)腔隙性梗死；(c)局灶性缺血

研究意义

​

本研究的结果表明，通过融合多层级特征建模的新型3D-MRI 去噪策略，不仅有效抑制噪声，而且更好地保留脑小血管病的关键结构信息，为临床提供更高质量的影像。总体而言，该研究提供了一种性能稳定、结构信息保真度高的3D- MRI 去噪方案，尤其对脑小血管病的精准诊断有所帮助，具有良好的临床应用潜力与推广价值。

​

Abstract

​

Cerebral small vessel disease (CSVD) manifestations in magnetic resonance (MR) images play a pivotal role as essential indicators for accurate diagnosis. However, the presence of noise in MR images significantly degrades image quality, thus compromising the precision of lesion detection and disease diagnosis. Although deep learning with residual architectures has shown promise in MR denoising tasks, current methods face several challenges, such as issues with model convergence, limited generalization capabilities, and oversmoothing, all of which collectively hinder denoising performance. Our objective is to enhance denoising performance by introducing a new model named the Hierarchical Convolution-Based Multi-Layer Perceptron (HC-MLP), specifically designed to improve the diagnostic confidence of CSVD. Our HC-MLP framework comprises three primary components: 1) The inclusion of MLP modules mitigates bias caused by pure CNN models. 2) The straightforward structures of MLPs and CNNs simplify training and improve generalization. 3) The use of voxel-wise input and the integration of the residual MLP structure partially address the oversmoothing issue. Extensive experiments have been conducted on two public datasets (UK Biobank and ATLAS) and an external dataset to test the effectiveness of HC-MLP. A total of 120 brain MRI scans from the UK Biobank and 120 brain MRI scans from ATLAS with CSVD were randomly chosen for model training and testing. For external testing, all 29 subjects with various MRI features of CSVD were included. Peak signal-to-noise ratio (PSNR), structural similarity index measure (SSIM), and normalized mean squared error (NMSE) were used for model validation. The Mann-Whitney-Wilcoxon two-sided test was used for score comparisons. Furthermore, two senior radiologists scored the results of the denoising performance. The experimental results show that HC-MLP significantly outperforms several state-of-the-art denoising algorithms, achieving a substantial improvement in PSNR (6.91% increase on UK Biobank and 5.31% on ATLAS) and SSIM (3.67% increase on UK Biobank and 2.27% on ATLAS). The CSVD recovery results further illustrate the superior performance of HC-MLP. Moreover, the performance of HC-MLP has been confirmed by radiologists. The proposed HC-MLP not only achieves significant enhancements in denoising 3D MR images but also successfully restores key features of CSVD that may be compromised by simulated noise in MR images, thereby improving the diagnostic confidence of CSVD.

​

作者简介

​

通讯作者

张孝勇，上海交通大学医学院研究员，博士生导师。长期从事磁共振成像技术及智能算法研究，在Medical Image Analysis、IEEE TMI、Advanced Science等权威期刊发表学术论文60余篇，主持国家自然科学基金重点项目1项、面上项目3项，以及上海市“探索者计划”、市级重大专项子课题等多项省部级课题。担任中国老年病学、中华医学会放射学分会等多个国家一级学会的常务委员或委员职务，同时担任Phenomics、Magnetic Resonance Letters等多个SCI期刊的编委。​

第一作者

杨海波，复旦大学类脑智能科学与技术研究院博士研究生。主要研究方向为医学图像处理，化学交换饱和转移磁共振成像技术。