School of Computing, Informatics, and Decision Systems Engineering, Arizona State University, Tempe, AZ, USA.
Division of Arts and Sciences, NYU Shanghai, Shanghai, China.
Brain Struct Funct. 2022 May;227(4):1507-1522. doi: 10.1007/s00429-022-02480-3. Epub 2022 Mar 24.
Retinotopic map, the mapping between visual inputs on the retina and neuronal responses on the cortical surface, is one of the central topics in vision science. Typically, human retinotopic maps are constructed by analyzing functional magnetic resonance responses to designed visual stimuli on the cortical surface. Although it is widely used in visual neuroscience, retinotopic maps are limited by the signal-to-noise ratio and spatial resolution of fMRI. One promising approach to improve the quality of retinotopic maps is to register individual subject's retinotopic maps to a retinotopic template. However, none of the existing retinotopic registration methods has explicitly quantified the diffeomorphic condition, that is, retinotopic maps shall be aligned by stretching/compressing without tearing up the cortical surface. Here, we developed Diffeomorphic Registration for Retinotopic Maps (DRRM) to simultaneously align retinotopic maps in multiple visual regions under the diffeomorphic condition. Specifically, we used the Beltrami coefficient to model the diffeomorphic condition and performed surface registration based on retinotopic coordinates. The overall framework preserves the topological condition defined in the template. We further developed a unique evaluation protocol and compared the performance of the new method with several existing registration methods on both synthetic and real datasets. The results showed that DRRM is superior to the existing methods in achieving diffeomorphic registration in synthetic and empirical data from 3T and 7T MRI systems. DRRM may improve the interpretation of low-quality retinotopic maps and facilitate applications of retinotopic maps in clinical settings.
视知觉图谱是视觉输入与大脑皮层神经元反应之间的映射关系，是视觉科学的核心主题之一。通常，人类视知觉图谱是通过分析大脑皮层表面设计的视觉刺激的功能磁共振响应来构建的。尽管它在视觉神经科学中得到了广泛应用，但视知觉图谱受到 fMRI 信号噪声比和空间分辨率的限制。提高视知觉图谱质量的一种有前途的方法是将个体的视知觉图谱与视知觉模板进行配准。然而，现有的视知觉配准方法都没有明确量化微分同胚条件，即视知觉图谱应该通过拉伸/压缩而不是撕裂皮层表面来进行对齐。在这里，我们开发了用于视知觉图谱的微分同胚配准（DRRM）方法，以在微分同胚条件下同时对齐多个视觉区域的视知觉图谱。具体来说，我们使用 Beltrami 系数来建模微分同胚条件，并基于视知觉坐标进行表面配准。整体框架保留了模板中定义的拓扑条件。我们进一步开发了一种独特的评估方案，并将新方法的性能与其他几种现有的配准方法在合成数据集和来自 3T 和 7T MRI 系统的真实数据集上进行了比较。结果表明，DRRM 在合成数据和来自 3T 和 7T MRI 系统的真实数据中，在实现微分同胚配准方面优于现有的方法。DRRM 可能会改善低质量视知觉图谱的解释，并促进视知觉图谱在临床环境中的应用。
