Sabuncu Mert R, Van Leemput Koen
Athinoula A. Martinos Center for Biomedical Imaging, Department of Radiology, MGH, Harvard Medical School, USA.
Med Image Comput Comput Assist Interv. 2011;14(Pt 3):99-106. doi: 10.1007/978-3-642-23626-6_13.
This paper presents the Relevance Voxel Machine (RVoxM), a Bayesian multivariate pattern analysis (MVPA) algorithm that is specifically designed for making predictions based on image data. In contrast to generic MVPA algorithms that have often been used for this purpose, the method is designed to utilize a small number of spatially clustered sets of voxels that are particularly suited for clinical interpretation. RVoxM automatically tunes all its free parameters during the training phase, and offers the additional advantage of producing probabilistic prediction outcomes. Experiments on age prediction from structural brain MRI indicate that RVoxM yields biologically meaningful models that provide excellent predictive accuracy.
本文介绍了相关性体素机器（RVoxM），这是一种贝叶斯多变量模式分析（MVPA）算法，专门设计用于基于图像数据进行预测。与通常用于此目的的通用MVPA算法不同，该方法旨在利用少量空间聚类的体素集，这些体素集特别适合临床解释。RVoxM在训练阶段自动调整其所有自由参数，并具有产生概率预测结果的额外优势。从结构性脑磁共振成像进行年龄预测的实验表明，RVoxM产生了具有生物学意义的模型，提供了出色的预测准确性。
