Michel Vincent, Eger Evelyn, Keribin Christine, Thirion Bertrand
PARIETAL Team, INRIA Saclay- Î le-de-France, 91191 Saclay, France.
Int J Biomed Imaging. 2011;2011:350838. doi: 10.1155/2011/350838. Epub 2011 Jun 23.
Inverse inference has recently become a popular approach for analyzing neuroimaging data, by quantifying the amount of information contained in brain images on perceptual, cognitive, and behavioral parameters. As it outlines brain regions that convey information for an accurate prediction of the parameter of interest, it allows to understand how the corresponding information is encoded in the brain. However, it relies on a prediction function that is plagued by the curse of dimensionality, as there are far more features (voxels) than samples (images), and dimension reduction is thus a mandatory step. We introduce in this paper a new model, called Multiclass Sparse Bayesian Regression (MCBR), that, unlike classical alternatives, automatically adapts the amount of regularization to the available data. MCBR consists in grouping features into several classes and then regularizing each class differently in order to apply an adaptive and efficient regularization. We detail these framework and validate our algorithm on simulated and real neuroimaging data sets, showing that it performs better than reference methods while yielding interpretable clusters of features.
反向推理最近已成为分析神经成像数据的一种流行方法，通过量化脑图像中关于感知、认知和行为参数所包含的信息量。由于它勾勒出了为准确预测感兴趣参数而传达信息的脑区，因此能够了解相应信息在大脑中是如何编码的。然而，它依赖于一个受维度诅咒困扰的预测函数，因为特征（体素）比样本（图像）多得多，因此降维是必不可少的一步。我们在本文中引入了一种新模型，称为多类稀疏贝叶斯回归（MCBR），与传统方法不同，它能自动根据可用数据调整正则化量。MCBR包括将特征分组到几个类别中，然后对每个类别进行不同的正则化，以便应用自适应且高效的正则化。我们详细介绍了这个框架，并在模拟和真实的神经成像数据集上验证了我们的算法，结果表明它比参考方法表现更好，同时还能产生可解释的特征簇。
