Unification of sparse Bayesian learning algorithms for electromagnetic brain imaging with the majorization minimization framework.
Unification of sparse Bayesian learning algorithms for electromagnetic brain imaging with the majorization minimization framework.
复制标题
DOI：
10.1016/j.neuroimage.2021.118309
复制
发表时间：
2021-10-01
期刊：
影响因子：
5.7
通讯作者：
Haufe S
中科院分区：
文献类型：
--
作者：
Hashemi A;Cai C;Kutyniok G;Müller KR;Nagarajan SS;Haufe S
已经证明了使用稀疏贝叶斯学习（SBL）的基于脑电或脑磁图（EEG/MEG）的脑源成像（BSI）的方法在具有少量不同活动源的情况下（诸如事件相关设计）实现优异的性能。本文从三个方面对SBL的理论和实践进行了拓展。首先，我们重新制定现有的SBL算法下的优化最小化（MM）的框架。这种统一的观点，不仅提供了一个有用的理论框架，比较不同的算法在其收敛行为，但也提供了一个原则配方，通过设计适当的边界的贝叶斯边际似然函数，构造新的算法具有特定的属性。其次，基于MM原理，我们提出了一种新的方法，称为LowSNR-BSI，在低信噪比（SNR）设置实现良好的源重建性能。第三，噪声水平的精确知识是精确源重建的关键要求。在这里，我们提出了一种新的原则性技术，以准确地学习噪声方差的数据，无论是联合内的源重建过程中，或使用两个建议的交叉验证策略之一。经验上，我们可以证明从MM理论预测的单调收敛行为在数值实验中得到证实。通过仿真，我们进一步证明了LowSNR-BSI在低SNR条件下优于传统SBL的优势，以及学习噪声水平优于来自基线数据的估计的优势。为了证明我们的新方法的实用性，我们显示神经生理上合理的源重建平均听觉诱发电位数据。
Methods for electro- or magnetoencephalography (EEG/MEG) based brain source imaging (BSI) using sparse Bayesian learning (SBL) have been demonstrated to achieve excellent performance in situations with low numbers of distinct active sources, such as event-related designs. This paper extends the theory and practice of SBL in three important ways. First, we reformulate three existing SBL algorithms under the majorization-minimization (MM) framework. This unification perspective not only provides a useful theoretical framework for comparing different algorithms in terms of their convergence behavior, but also provides a principled recipe for constructing novel algorithms with specific properties by designing appropriate bounds of the Bayesian marginal likelihood function. Second, building on the MM principle, we propose a novel method called LowSNR-BSI that achieves favorable source reconstruction performance in low signal-to-noise-ratio (SNR) settings. Third, precise knowledge of the noise level is a crucial requirement for accurate source reconstruction. Here we present a novel principled technique to accurately learn the noise variance from the data either jointly within the source reconstruction procedure or using one of two proposed cross-validation strategies. Empirically, we could show that the monotonous convergence behavior predicted from MM theory is confirmed in numerical experiments. Using simulations, we further demonstrate the advantage of LowSNR-BSI over conventional SBL in low-SNR regimes, and the advantage of learned noise levels over estimates derived from baseline data. To demonstrate the usefulness of our novel approach, we show neurophysiologically plausible source reconstructions on averaged auditory evoked potential data.
登录
查看更多内容
影响因子：
--
作者：
Dalal SS;Zumer JM;Guggisberg AG;Trumpis M;Wong DD;Sekihara K;Nagarajan SS
通讯作者：
Nagarajan SS
影响因子：
2.1
作者：
Bekhti, Yousra;Lucka, Felix;Gramfort, Alexandre
通讯作者：
Gramfort, Alexandre
影响因子：
5.7
作者：
Cai C;Hashemi A;Diwakar M;Haufe S;Sekihara K;Nagarajan SS
通讯作者：
Nagarajan SS
影响因子：
2.7
作者：
Cichocki, Andrzej;Amari, Shun-ichi
通讯作者：
Amari, Shun-ichi
DOI：
10.1111/j.2517-6161.1977.tb01600.x
发表时间：
1977-01-01
期刊：
JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-METHODOLOGICAL
影响因子：
--
作者：
DEMPSTER, AP;LAIRD, NM;RUBIN, DB
通讯作者：
RUBIN, DB