Confidence Sets for Cohen's d effect size images.
Confidence Sets for Cohen's d effect size images.
复制标题
DOI：
10.1016/j.neuroimage.2020.117477
复制
发表时间：
2021-02-01
期刊：
影响因子：
5.7
通讯作者：
Nichols TE
中科院分区：
文献类型：
--
作者：
Bowring A;Telschow FJE;Schwartzman A;Nichols TE
置信集（CSs）将置信区间的概念扩展到fMRI图。对于科恩阈值，上CS断言下CS在哪里。我们证明了在HCP主题级Cohen's d数据上的CSs方法。我们比较CS与标准的统计体素推理的结果。与传统的聚类测试不同，CS精确地量化了空间不确定性。目前的统计推断方法的任务功能磁共振成像遭受两个基本的限制。首先，重点仅仅是检测非零信号或信号变化，这是一个在大规模研究（例如英国生物样本库）中加剧的问题，其中“零假设谬误”导致即使微不足道的影响也被确定为显著的。其次，对于任何样本量，广泛使用的聚类推断方法只表明可以拒绝零假设的区域，而不提供任何关于激活的空间不确定性的概念。在这项工作中，我们解决这些问题，通过开发空间置信集（CS）的阈值科恩的效果大小图像中发现的集群。我们产生了一个上，下CS，使科恩的效果大小超过和低于非零阈值，分别对大脑区域的信心声明。CS传达了关于效应量的大小和可靠性的信息，这些信息通常在统计量和效应估计图中单独给出。我们扩大了我们以前的工作中开发的理论CSs的%粗体变化效果图使用最近的结果从自举文献。通过评估与2D和3D Monte Carlo模拟类似的fMRI数据的经验覆盖率，我们发现我们的方法是准确的样本量低至。我们计算科恩的CS为人类连接组计划的工作记忆任务功能磁共振成像数据，说明了一个可靠的科恩的反应，为给定的阈值的大脑区域。通过比较CS与从传统的统计体素推理获得的结果，我们突出了激活本地化的改进，可以获得与置信集。
Confidence Sets (CSs) extend the idea of confidence intervals to fMRI maps. For a Cohen’s threshold upper CS asserts where lower CS where . We demonstrate the CSs method on HCP subject-level Cohen’s d data. We compare the CSs with results from standard statistical voxelwise inference. Unlike traditional cluster tests, CSs precisely quantify spatial uncertainty. Current statistical inference methods for task-fMRI suffer from two fundamental limitations. First, the focus is solely on detection of non-zero signal or signal change, a problem that is exacerbated for large scale studies (e.g. UK Biobank, ) where the ‘null hypothesis fallacy’ causes even trivial effects to be determined as significant. Second, for any sample size, widely used cluster inference methods only indicate regions where a null hypothesis can be rejected, without providing any notion of spatial uncertainty about the activation. In this work, we address these issues by developing spatial Confidence Sets (CSs) on clusters found in thresholded Cohen’s effect size images. We produce an upper and lower CS to make confidence statements about brain regions where Cohen’s effect sizes have exceeded and fallen short of a non-zero threshold, respectively. The CSs convey information about the magnitude and reliability of effect sizes that is usually given separately in a -statistic and effect estimate map. We expand the theory developed in our previous work on CSs for %BOLD change effect maps using recent results from the bootstrapping literature. By assessing the empirical coverage with 2D and 3D Monte Carlo simulations resembling fMRI data, we find our method is accurate in sample sizes as low as . We compute Cohen’s CSs for the Human Connectome Project working memory task-fMRI data, illustrating the brain regions with a reliable Cohen’s response for a given threshold. By comparing the CSs with results obtained from a traditional statistical voxelwise inference, we highlight the improvement in activation localization that can be gained with the Confidence Sets.
登录
查看更多内容
影响因子：
22.4
作者：
ROZEBOOM, WW
通讯作者：
ROZEBOOM, WW
影响因子：
3.7
作者：
Cremers HR;Wager TD;Yarkoni T
通讯作者：
Yarkoni T
DOI：
10.1073/pnas.1121049109
发表时间：
2012-04-03
影响因子：
11.1
作者：
Gonzalez-Castillo, Javier;Saad, Ziad S.;Bandettini, Peter A.
通讯作者：
Bandettini, Peter A.
DOI：
10.1073/pnas.1222447110
发表时间：
2013-06-18
影响因子：
11.1
作者：
Cacioppo, John T.;Cacioppo, Stephanie;VanderWeele, Tyler J.
通讯作者：
VanderWeele, Tyler J.
影响因子：
--
作者：
LAUBSCHER, NF
通讯作者：
LAUBSCHER, NF