无需真实高分辨率图像，用自监督方法提升fMRI空间分辨率。
TV-based Deep 3D Self Super-Resolution for fMRI
- 结合深度网络与总变差正则化，实现无真实标签的自监督超分辨率。
- 在不依赖外部真值数据情况下，性能接近有监督方法。
- 适合脑功能成像研究者，尤其适用于难以获取高分辨真值的场景。
功能性磁共振成像（fMRI）虽能揭示认知过程，但其固有的空间分辨率限制制约了对大脑精细功能结构的分析。受扫描仪和序列参数限制，时间分辨率、空间分辨率、信噪比与扫描时间之间存在权衡。深度学习超分辨率（SR）方法可通过低分辨率（LR）图像生成高分辨率（HR）图像，从而缩短扫描时间。然而，现有多数方法依赖有监督训练，需真实高分辨率（GT）数据作为标签，而此类数据通常难以获取，且限制了超分辨率的上限。本文提出一种新颖的自监督深度学习超分辨率模型，融合深度网络与解析方法，并引入总变差（TV）正则化。该方法无需外部真值图像，性能可媲美有监督技术，同时有效保留功能图谱特征。
原文摘要 · Abstract (English)
While functional Magnetic Resonance Imaging (fMRI) offers valuable insights into cognitive processes, its inherent spatial limitations pose challenges for detailed analysis of the fine-grained functional architecture of the brain. More specifically, MRI scanner and sequence specifications impose a trade-off between temporal resolution, spatial resolution, signal-to-noise ratio, and scan time. Deep Learning (DL) Super-Resolution (SR) methods have emerged as a promising solution to enhance fMRI resolution, generating high-resolution (HR) images from low-resolution (LR) images typically acquired with lower scanning times. However, most existing SR approaches depend on supervised DL techniques, which require training ground truth (GT) HR data, which is often difficult to acquire and simultaneously sets a bound for how far SR can go. In this paper, we introduce a novel self-supervised DL SR model that combines a DL network with an analytical approach and Total Variation (TV) regularization. Our method eliminates the need for external GT images, achieving competitive performance compared to supervised DL techniques and preserving the functional maps.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。