Regression-based machine-learning approaches to predict task activation using resting-state fMRI
Regression-based machine-learning approaches to predict task activation using resting-state fMRI
复制标题
DOI：
10.1002/hbm.24841
复制
发表时间：
2019-10-22
影响因子：
4.8
通讯作者：
Wang, Yang
中科院分区：
文献类型：
--
作者：
Cohen, Alexander D.;Chen, Ziyi;Wang, Yang
静息态 fMRI 已显示出通过使用通用线性模型 (GLM) 将静息态网络特征映射到激活 z 分数来预测个体任务激活的能力。问题仍然是相对简单的 GLM 是否是完成此预测的最佳方法。在这项研究中，比较了几种基于回归的机器学习方法，包括 GLM、前馈神经网络和随机森林引导聚合（bagging）。对 350 名人类连接组项目受试者的静息状态和任务数据进行了分析。首先，评估训练对象数量对预测精度的影响。此外，还比较了不同模型的预测精度和 Dice 系数。随着训练数量达到 200 名受试者，预测准确度有所提高；然而，大约 30-40 名训练对象的预测曲线发生了弯曲。所有模型在相关矩阵上都表现良好，该矩阵显示了所有受试者的实际任务激活和预测任务激活之间的相关性，所有任务都表现出强烈的对角线趋势。总体而言，神经网络和随机森林装袋技术优于 GLM。然而，这些方法需要额外的计算能力和处理时间。这些结果表明，虽然 GLM 表现良好，但任务激活的静息态 fMRI 预测可以受益于更复杂的机器学习方法。
Resting-state fMRI has shown the ability to predict task activation on an individual basis by using a general linear model (GLM) to map resting-state network features to activation z-scores. The question remains whether the relatively simplistic GLM is the best approach to accomplish this prediction. In this study, several regression-based machine-learning approaches were compared, including GLMs, feed-forward neural networks, and random forest bootstrap aggregation (bagging). Resting-state and task data from 350 Human Connectome Project subjects were analyzed. First, the effect of the number of training subjects on the prediction accuracy was evaluated. In addition, the prediction accuracy and Dice coefficient were compared across models. Prediction accuracy increased with the training number up to 200 subjects; however, an elbow in the prediction curve occurred around 30-40 training subjects. All models performed well with correlation matrices, which displayed correlation between actual and predicted task activation for all subjects, exhibiting a strong diagonal trend for all tasks. Overall, the neural network and random forest bagging techniques outperformed the GLM. These approaches, however, require additional computing power and processing time. These results show that, while the GLM performs well, resting-state fMRI prediction of task activation could benefit from more complex machine learning approaches.